**Meet the editor**

Prof. Jürgen Gegner studied Physics at the University of Erlangen-Nuremberg and completed his first degree ("Dipl.-Phys.") in 1989 with a diploma thesis on deuteron-deuteron scattering. He graduated his doctoral studies in Materials Science ("Dr. rer. nat.") at the University of Stuttgart, where he defended his dissertation on interfacial oxygen segregation in 1994 at the Max Planck

Institute for Metals Research. After postdoctoral research on ceramics and aluminum alloys at the Max Planck Institute for Microstructure Physics in Halle and the Institute of Applied Mechanics at the University of Erlangen-Nuremberg, respectively, he joined the Loctite Global Engineering Center at Garching near Munich in 1999 as Senior Engineer. Since 2000 he has been working for SKF in Schweinfurt, from 2005 as manager of the Material Physics laboratory. He defended his postdoctoral degree in Materials Science (venia legendi, "Dr. rer. nat. habil.") with a habilitation thesis on solid state diffusion and a lecture on adhesive bonding at the University of Siegen in 2005, where he was promoted to Adjunct Professor in 2012. In 2008, he was announced as Visiting Professor of Materials Research at Ariel University in Israel. Since 2007, he is head of the German Research Committee for Residual Stresses. His main fields of work are material use with a focus on rolling contact tribology, surface engineering, heat treatment, materials development, characterization and testing, microstructure analysis, mechanistic modeling and numerical simulation. He published 4 books and more than 150 papers, is inventor of 19 patents and committee member of leading international conferences, at which he gave 9 invited plenary and keynote lectures.

Contents

**Preface VII**

Walter Holweger

Chapter 2 **Lubrication and Lubricants 55**

**Friction Nodes 77** Maciej Paszkowski

**Section 1 Lubrication and Properties of Lubricants 1**

Chapter 1 **Fundamentals of Lubricants and Lubrication 3**

Nehal S. Ahmed and Amal M. Nassar

**Section 2 Boundary Lubrication Applications 107**

**Rocket Turbopumps 109**

**Section 3 Testing and Modeling 183**

Wulczynski

Masataka Nosaka and Takahisa Kato

Chapter 5 **Titanium and Titanium Alloys as Biomaterials 155** Virginia Sáenz de Viteri and Elena Fuentes

Chapter 6 **New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears 185**

> Remigiusz Michalczewski, Marek Kalbarczyk, Michal Michalak, Witold Piekoszewski, Marian Szczerek, Waldemar Tuszynski and Jan

Chapter 3 **Some Aspects of Grease Flow in Lubrication Systems and**

Chapter 4 **Cryogenic Tribology in High-Speed Bearings and Shaft Seals of**

### Contents



Remigiusz Michalczewski, Marek Kalbarczyk, Michal Michalak, Witold Piekoszewski, Marian Szczerek, Waldemar Tuszynski and Jan Wulczynski

Chapter 7 **Introduction of the Ratio of the Hardness to the Reduced Elastic Modulus for Abrasion 217** Giuseppe Pintaude

Preface

detrimental.

practitioners and help them solve their problems.

and covers structural as well as rheologic properties.

As the subject of tribology comprises lubrication, friction and wear of contact components highly relevant to practical applications, it challenges scientists from chemistry, physics and materials engineering around the world on today's sophisticated experimental and theoreti‐ cal foundation to complex interdisciplinary research. Recent results and developments are preferably presented and evaluated in the context of established knowledge. Consisting of eleven chapters divided into the four parts of Lubrication and Properties of Lubricants, Boundary Lubrication Applications, Testing and Modeling, and Sustainability of Tribosys‐ tems, this textbook therefore merges basic concepts with new findings and approaches. Tri‐ bology – Fundamentals and Advancements, supported by competent authors, aims to convey current research trends in the light of the state of the art to students, scientists and

The undesirable loss of energy and material by friction and wear cause damage amounting to several ten billion dollars. The high economical relevance of the demand for durable products has been driving the development and continuous refinement of the scientific and

The excellent invited Chapter 1 on *Fundamentals of Lubricants and Lubrication* opens the book by providing an expert introduction to the chemistry and basic properties of oils, additives and greases on the advanced level of a standard reference guide. The two following contri‐ butions complete the general background of the Part 1. Chapter 2 on *Lubrication and Lubricat‐ ing Oil* outlines the regimes of lubrication and deals with liquid, solid and gaseous lubricants. Oils take up the most space of this paper. Chapter 3 on *Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes* extends the discussion of grease lubrication

Part 2 turns the reader's attention to applications under adverse operating conditions. Chap‐ ter 4 on *Cryogenic High-Speed Bearings and Shaft Seals of Rocket Turbopumps* reviews promising approaches for reliable machine elements in future space transportation systems. A crucial material issue in aerospace engineering is associated with this task, for instance, due to the differences in the thermal alteration of the mechanical behavior. The optional use of ceram‐ ics in hybrid bearings at extremely low service temperatures and boundary lubrication indi‐ cates the importance of innovative materials for tribologically loaded components. Chapter 5 on *Titanium and Titanium Alloys as Biomaterials* makes this conclusion even more meaning‐ ful and extents it to mixed friction oriented surface engineering. Recent attempts to increase the wear resistance of orthopedic implants are discussed. Titanium and its alloys provide excellent biocompatibility and corrosion protection but the poor tribological properties are

engineering discipline of tribology since the middle of the last century.


### Preface

Chapter 7 **Introduction of the Ratio of the Hardness to the Reduced**

M. Tauviqirrahman, R. Ismail, J. Jamari and D.J. Schipper

Xana Fernández-Pérez, Amaya Igartua, Roman Nevshupa, Patricio Zabala, Borja Zabala, Rolf Luther, Flavia Gili and Claudio Genovesio

Manfred R. Mauntz, Jürgen Gegner, Ulrich Kuipers and Stefan

Chapter 9 **Friction and Wear of a Grease Lubricated Contact — An**

Chapter 10 **Innovative "Green" Tribological Solutions for Clean**

Chapter 11 **A Sensor System for Online Oil Condition Monitoring of**

**Elastic Modulus for Abrasion 217**

Chapter 8 **Artificial Slip Surface: Potential Application in**

Giuseppe Pintaude

**VI** Contents

**Lubricated MEMS 231**

**Energetic Approach 251**

**Section 4 Sustainability of Tribosystems 273**

**Operating Components 305**

**Small Engines 275**

Erik Kuhn

Klingau

As the subject of tribology comprises lubrication, friction and wear of contact components highly relevant to practical applications, it challenges scientists from chemistry, physics and materials engineering around the world on today's sophisticated experimental and theoreti‐ cal foundation to complex interdisciplinary research. Recent results and developments are preferably presented and evaluated in the context of established knowledge. Consisting of eleven chapters divided into the four parts of Lubrication and Properties of Lubricants, Boundary Lubrication Applications, Testing and Modeling, and Sustainability of Tribosys‐ tems, this textbook therefore merges basic concepts with new findings and approaches. Tri‐ bology – Fundamentals and Advancements, supported by competent authors, aims to convey current research trends in the light of the state of the art to students, scientists and practitioners and help them solve their problems.

The undesirable loss of energy and material by friction and wear cause damage amounting to several ten billion dollars. The high economical relevance of the demand for durable products has been driving the development and continuous refinement of the scientific and engineering discipline of tribology since the middle of the last century.

The excellent invited Chapter 1 on *Fundamentals of Lubricants and Lubrication* opens the book by providing an expert introduction to the chemistry and basic properties of oils, additives and greases on the advanced level of a standard reference guide. The two following contri‐ butions complete the general background of the Part 1. Chapter 2 on *Lubrication and Lubricat‐ ing Oil* outlines the regimes of lubrication and deals with liquid, solid and gaseous lubricants. Oils take up the most space of this paper. Chapter 3 on *Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes* extends the discussion of grease lubrication and covers structural as well as rheologic properties.

Part 2 turns the reader's attention to applications under adverse operating conditions. Chap‐ ter 4 on *Cryogenic High-Speed Bearings and Shaft Seals of Rocket Turbopumps* reviews promising approaches for reliable machine elements in future space transportation systems. A crucial material issue in aerospace engineering is associated with this task, for instance, due to the differences in the thermal alteration of the mechanical behavior. The optional use of ceram‐ ics in hybrid bearings at extremely low service temperatures and boundary lubrication indi‐ cates the importance of innovative materials for tribologically loaded components. Chapter 5 on *Titanium and Titanium Alloys as Biomaterials* makes this conclusion even more meaning‐ ful and extents it to mixed friction oriented surface engineering. Recent attempts to increase the wear resistance of orthopedic implants are discussed. Titanium and its alloys provide excellent biocompatibility and corrosion protection but the poor tribological properties are detrimental.

Part 3 of the book is dedicated to laboratory testing, modeling approaches and mathematical simulation. The *New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears,* proposed in Chapter 6, is applied to evaluate the performance of coatings. The examination of new lubricants requires gear testing as well. The prediction of the abrasion response of a tribosystem involves material characteristics of the counterbodies. It is an of‐ ten raised question what relationship holds between wear and hardness? *Introducing the Ra‐ tio of the Hardness to the Reduced Elastic Modulus for Abrasion* in Chapter 7 serves the purpose of modeling the wear coefficient. The suitability is demonstrated on the examples of the well-known tribosystems of glass‒bearing steel and alumina‒hard metal. The characteriza‐ tion of the cutting efficiency is particularly relevant to machining processes. Chapter 8, enti‐ tled *Artificial Slip Surface: Potential Application in Lubricated MEMS,* is another highlight of the book. It presents a novel lubrication model for moving parts in micro electro-mechanical systems, based on the numerical solution of a modified Reynolds equation, to improve the tribological performance of lubricated contacts. Chapter 9 on *Friction and Wear of a Grease Lubricated Contact ‒ An Energetical Approach* instructively specializes the discussion of Chap‐ ter 3. It elaborates an advanced energy balance model and includes rheometer tests. Grease degradation is correlated with the energetics of a tribosystem.

Part 4 finally addresses the emerging sustainability topic from the two essential aspects of environmentally friendly tribology and early failure detection. Chapter 10 *on Innovative "Green" Tribological Solutions for Clean Small Engines* considers the practical problems with biodegradable lubricants, such as abrasive wear, in this demanding operating environment. It investigates new biofuel-compatible eco-oils and materials or coatings that promote their use. Chapter 11 introduces *A Sensor System for Online Oil Condition Monitoring of Operating Components* with up-to-date web-based accessibility. The electrical oil sensor assists with in‐ creasing the level of machine availability, e.g., of off-shore wind turbines and can also be used to measure the percentage of biodiesel in a fuel blend.

The editor especially would like to thank the authors for their valuable work. The support of the publisher is also highly appreciated.

> **Jürgen Gegner** University of Siegen, Germany

**Section 1**

**Lubrication and Properties of Lubricants**

**Lubrication and Properties of Lubricants**

Part 3 of the book is dedicated to laboratory testing, modeling approaches and mathematical simulation. The *New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears,* proposed in Chapter 6, is applied to evaluate the performance of coatings. The examination of new lubricants requires gear testing as well. The prediction of the abrasion response of a tribosystem involves material characteristics of the counterbodies. It is an of‐ ten raised question what relationship holds between wear and hardness? *Introducing the Ra‐ tio of the Hardness to the Reduced Elastic Modulus for Abrasion* in Chapter 7 serves the purpose of modeling the wear coefficient. The suitability is demonstrated on the examples of the well-known tribosystems of glass‒bearing steel and alumina‒hard metal. The characteriza‐ tion of the cutting efficiency is particularly relevant to machining processes. Chapter 8, enti‐ tled *Artificial Slip Surface: Potential Application in Lubricated MEMS,* is another highlight of the book. It presents a novel lubrication model for moving parts in micro electro-mechanical systems, based on the numerical solution of a modified Reynolds equation, to improve the tribological performance of lubricated contacts. Chapter 9 on *Friction and Wear of a Grease Lubricated Contact ‒ An Energetical Approach* instructively specializes the discussion of Chap‐ ter 3. It elaborates an advanced energy balance model and includes rheometer tests. Grease

Part 4 finally addresses the emerging sustainability topic from the two essential aspects of environmentally friendly tribology and early failure detection. Chapter 10 *on Innovative "Green" Tribological Solutions for Clean Small Engines* considers the practical problems with biodegradable lubricants, such as abrasive wear, in this demanding operating environment. It investigates new biofuel-compatible eco-oils and materials or coatings that promote their use. Chapter 11 introduces *A Sensor System for Online Oil Condition Monitoring of Operating Components* with up-to-date web-based accessibility. The electrical oil sensor assists with in‐ creasing the level of machine availability, e.g., of off-shore wind turbines and can also be

The editor especially would like to thank the authors for their valuable work. The support of

**Jürgen Gegner** University of Siegen,

Germany

degradation is correlated with the energetics of a tribosystem.

used to measure the percentage of biodiesel in a fuel blend.

the publisher is also highly appreciated.

VIII Preface

**Chapter 1**

**Fundamentals of Lubricants and Lubrication**

Literature about lubricants is available in all public domains. Readers should search at those platforms in the case of special interests. Citations given here do not represent the full scale

Part of this chapter will be the basic chemical structure of lubricants including some property descriptions. Since literature in tribology is innumerous, the reader should check his special

**•** to transfer power in the application (hydraulic, automatic transmission, breaks). [6, 8]

Functionality of lubricants is defined by their chemical structure and their physical properties. Basics of lubrication are covered by organic chemistry to a major and inorganic chemistry to

Lubricants are regulated internationally and locally, e. g. by ASTM (American Standard of Testing Materials) or DIN (Deutsche Industrienorm). Regulation covers the physical, chemical and toxicological description of lubricants including safety guide lines and others. [2, 3]

> © 2013 Holweger; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,

© 2013 Holweger; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

distribution, and reproduction in any medium, provided the original work is properly cited.

Lubricants play a key role in machinery element safety. Their main tasks are

Additional information is available at the end of the chapter

but reflect an overview from a today's perspective. [1-7]

**•** to keep moving parts apart from each other,

**•** to take heat out of the contact by their through pass,

**•** to transport functional additives toward the surface and

Walter Holweger

**1. Introduction**

area of interest.

**•** to keep surfaces clean,

a minor extent. [2, 3]

http://dx.doi.org/10.5772/55731

#### **Chapter 1**

### **Fundamentals of Lubricants and Lubrication**

#### Walter Holweger

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/55731

**1. Introduction**

Literature about lubricants is available in all public domains. Readers should search at those platforms in the case of special interests. Citations given here do not represent the full scale but reflect an overview from a today's perspective. [1-7]

Part of this chapter will be the basic chemical structure of lubricants including some property descriptions. Since literature in tribology is innumerous, the reader should check his special area of interest.

Lubricants play a key role in machinery element safety. Their main tasks are


Functionality of lubricants is defined by their chemical structure and their physical properties. Basics of lubrication are covered by organic chemistry to a major and inorganic chemistry to a minor extent. [2, 3]

Lubricants are regulated internationally and locally, e. g. by ASTM (American Standard of Testing Materials) or DIN (Deutsche Industrienorm). Regulation covers the physical, chemical and toxicological description of lubricants including safety guide lines and others. [2, 3]

#### **2. Some basics**

The spatial structure of carbon chemistry defines all activities of the lubricants derived from them. The spatial structure of organic carbon chemicals is given by the binding state of carbon. [10]

Three main types are discussed. Two are essential for lubrication: single and double bonds.

#### **2.1. Single bonds: Tetrahedral binding**

In the tetrahedral binding state, reflecting the status of single bonds, carbon is placed in the centre of a pyramid with bindings into space from the centre to the corner (Figure 1).

**Figure 4.** Cyclic Structures by carbon-carbon binding

**Figure 5.** Energy rise in rotated structures

molecule rises.

**2.2. Double bonds**

Single bonds in hydrocarbons are free to rotate (Figure 5). Rotation leads to the situation that hydrogen atoms within the chain get close to each other. As a consequence the energy of the

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 5

Ground State - low in energy Rotated State - Higher energy

them closely together without giving time to relax. (Figure 6)

**Figure 6.** Excitation by pushing molecules to one another by shear stress

internal molecular energy, sometimes high enough to cut them.

the other as double, whereas the remaining bonding stays single. (Figure 7)

Similar to internal rotation, molecular energy rises if molecules get under stress by moving

Also the fact of putting or pressing molecules toward a surface may lead to a steep increase in

Carbon may also bind to others by double bonds, such that two of the four bindings attach to

**Figure 1.** Tetrahedral binding of carbon

Carbon is placed in the centre of the tetrahedral with four attached valences. Within chemical convention in order to abbreviate the structure denotation the atom symbols are neglected.

Carbon may bind to another one by corner to corner. (Figure 2)

**Figure 2.** Corner to corner binding state

Corner to corner binding leads to zigzag chains, where the angle of carbon to the hydrogen atoms is 108°. In general the hydrogen is neglected, leading to a skeleton drawing of the structure.

Beyond the fixed angle of 108° and the zigzag shape of such hydrocarbon structures, a high variety of structures arise due to the fact that those bindings may branch or bind to cyclic structures. (Figure 3 and Figure 4)

**Figure 3.** Branched structures by carbon to carbon binding

**Figure 4.** Cyclic Structures by carbon-carbon binding

**2. Some basics**

4 Tribology - Fundamentals and Advancements

of carbon. [10]

**2.1. Single bonds: Tetrahedral binding**

**Figure 1.** Tetrahedral binding of carbon

**Figure 2.** Corner to corner binding state

structures. (Figure 3 and Figure 4)

**Figure 3.** Branched structures by carbon to carbon binding

structure.

The spatial structure of carbon chemistry defines all activities of the lubricants derived from them. The spatial structure of organic carbon chemicals is given by the binding state

Three main types are discussed. Two are essential for lubrication: single and double bonds.

In the tetrahedral binding state, reflecting the status of single bonds, carbon is placed in the

Carbon is placed in the centre of the tetrahedral with four attached valences. Within chemical convention in order to abbreviate the structure denotation the atom symbols are neglected.

Corner to corner binding leads to zigzag chains, where the angle of carbon to the hydrogen atoms is 108°. In general the hydrogen is neglected, leading to a skeleton drawing of the

Beyond the fixed angle of 108° and the zigzag shape of such hydrocarbon structures, a high variety of structures arise due to the fact that those bindings may branch or bind to cyclic

Carbon may bind to another one by corner to corner. (Figure 2)

centre of a pyramid with bindings into space from the centre to the corner (Figure 1).

Single bonds in hydrocarbons are free to rotate (Figure 5). Rotation leads to the situation that hydrogen atoms within the chain get close to each other. As a consequence the energy of the molecule rises.

Ground State - low in energy Rotated State - Higher energy

**Figure 5.** Energy rise in rotated structures

Similar to internal rotation, molecular energy rises if molecules get under stress by moving them closely together without giving time to relax. (Figure 6)

**Figure 6.** Excitation by pushing molecules to one another by shear stress

Also the fact of putting or pressing molecules toward a surface may lead to a steep increase in internal molecular energy, sometimes high enough to cut them.

#### **2.2. Double bonds**

Carbon may also bind to others by double bonds, such that two of the four bindings attach to the other as double, whereas the remaining bonding stays single. (Figure 7)

*Hydrocarbons,* e.g. Structures that contain solely Hydrogen and Carbon (H, C)

from other precursors, such as phosphoric acid esters.

*Group I+*: Oils that are in a VI range of 103-108.

with viscosity index (VI) of 80 till 120.

*Group II+*: Oils in the VI range of 113-119.

*Group III+*: Oils providing a VI at least of 140.

Group IV and V is declared as synthetic oil.

**4. Saturated natural hydrocarbons**

Within a general scheme, base oils are identified as Groups.

binding state compared to Esters.

between 80 and 120.

during hydrogenation.

*Ester Oils*, e.g. Structures containing Hydrogen, Carbon and Oxygen. Some Esters are derived

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 7

*Polyglycoles*: Structure containing Hydrogen, Carbon and Oxygen but being different in

*Group I*: Those lubricants are built from saturated hydrocarbons, e.g. hydrocarbons without alkenes (hydrocarbons with double bonds)) (> 90%), obtained by solvent extraction processes and catalytic hydrogenation. Sulfur may part in amount of > 0.03%. Viscosity index (VI) is in

*Group II*: Hydrogenated (saturated) hydrocarbons (> 90%) and sulfur below 0.03% per weight

The base oil within this group is manufactured by hydrocracking, solvent extraction or

*Group III*: Oils with a saturation > 90%, sulfur < 0.03% and a viscosity index > 120. Those oils are produced by catalytic procedures with a concurrent rearrangement of the carbon backbone

*Group IV*: Poly-α-Olefins with sulfur content approximately 0, viscosity index 140–170, being produced by catalytic polymerization of low molecular weight end terminated olefins.

North America states Group III, IV and V as synthesized hydrocarbons (SHC) while in Europe

Saturated hydrocarbons are those who do not contain double bonds in their structure. They derive from the tetrahedral binding of carbon (bindings that point into corners of tetrahedron). The simplest structure is given by methane, ethane, propane, butane with carbons attached at the corners of the tetrahedral. These representatives are present in the natural gases, while methane is found in enormous quantities as methane-ice cluster. The gases themselves are not

catalytic dewaxing processes. Those oils are pale or water like colored.

*Group V*: All other oils, e.g. esters, polyglycoles, phosphate esters.

in use as lubricants but are components of fuels. (Figure 10)

**Figure 7.** Double bonding

Double bond shows a 120° neighborhood angle to the carbon. This angle is kept constant and will lead toward different structures in the double bond chemistry. (Figure 8)

Both structures differ in their energy. Double bonds are part of biodegradable additives (native oils) but also additives and thickeners in the case of greases. Z-Structures are dominant in native oils.

#### **2.3. Triple bonds**

Triple bonds are seldom found in tribology. They represent a high energy state in molecules with very high reactivity. As such, they are part of catalytic degradation processes in lubricants. Within a triple bond carbons attach to each other by a linear principle (Figure 9):

**Figure 9.** Triple bond present in a hydrocarbon

#### **3. Base oils in lubrication: General comments about specie and groups**

Hydrocarbon Base Oils for Lubrication derive from organic chemistry. Different categories are given by their chemical composition and structure. [2]

*Hydrocarbons,* e.g. Structures that contain solely Hydrogen and Carbon (H, C)

*Ester Oils*, e.g. Structures containing Hydrogen, Carbon and Oxygen. Some Esters are derived from other precursors, such as phosphoric acid esters.

*Polyglycoles*: Structure containing Hydrogen, Carbon and Oxygen but being different in binding state compared to Esters.

Within a general scheme, base oils are identified as Groups.

*Group I*: Those lubricants are built from saturated hydrocarbons, e.g. hydrocarbons without alkenes (hydrocarbons with double bonds)) (> 90%), obtained by solvent extraction processes and catalytic hydrogenation. Sulfur may part in amount of > 0.03%. Viscosity index (VI) is in between 80 and 120.

*Group I+*: Oils that are in a VI range of 103-108.

*Group II*: Hydrogenated (saturated) hydrocarbons (> 90%) and sulfur below 0.03% per weight with viscosity index (VI) of 80 till 120.

*Group II+*: Oils in the VI range of 113-119.

**Figure 7.** Double bonding

6 Tribology - Fundamentals and Advancements

**Figure 8.** E and Z structures in double bond

**Figure 9.** Triple bond present in a hydrocarbon

given by their chemical composition and structure. [2]

native oils.

**2.3. Triple bonds**

Double bond shows a 120° neighborhood angle to the carbon. This angle is kept constant and

Both structures differ in their energy. Double bonds are part of biodegradable additives (native oils) but also additives and thickeners in the case of greases. Z-Structures are dominant in

Triple bonds are seldom found in tribology. They represent a high energy state in molecules with very high reactivity. As such, they are part of catalytic degradation processes in lubricants.

**3. Base oils in lubrication: General comments about specie and groups**

Hydrocarbon Base Oils for Lubrication derive from organic chemistry. Different categories are

Within a triple bond carbons attach to each other by a linear principle (Figure 9):

will lead toward different structures in the double bond chemistry. (Figure 8)

The base oil within this group is manufactured by hydrocracking, solvent extraction or catalytic dewaxing processes. Those oils are pale or water like colored.

*Group III*: Oils with a saturation > 90%, sulfur < 0.03% and a viscosity index > 120. Those oils are produced by catalytic procedures with a concurrent rearrangement of the carbon backbone during hydrogenation.

*Group III+*: Oils providing a VI at least of 140.

*Group IV*: Poly-α-Olefins with sulfur content approximately 0, viscosity index 140–170, being produced by catalytic polymerization of low molecular weight end terminated olefins.

*Group V*: All other oils, e.g. esters, polyglycoles, phosphate esters.

North America states Group III, IV and V as synthesized hydrocarbons (SHC) while in Europe Group IV and V is declared as synthetic oil.

#### **4. Saturated natural hydrocarbons**

Saturated hydrocarbons are those who do not contain double bonds in their structure. They derive from the tetrahedral binding of carbon (bindings that point into corners of tetrahedron). The simplest structure is given by methane, ethane, propane, butane with carbons attached at the corners of the tetrahedral. These representatives are present in the natural gases, while methane is found in enormous quantities as methane-ice cluster. The gases themselves are not in use as lubricants but are components of fuels. (Figure 10)

hydrocarbons. Due to their high solidification point they are a threat if present in Diesel fuels

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 9

Apart from their function as hydrocarbon waxes they are not suitable as lubricants for machine

Suitable lubricants are derived from C16–C70 hydrocarbons with branched chains. Branching leads to low pour points (the point where the lubricant starts to get solid). Machine oils with low pour point, suitable for low temperature applications are branched in their carbon chain.

In general, the viscosity of a lubricant - as a measure for the ability to move across - increases with the molecular weight, e.g. the number of carbon atoms attached. Viscosity is measured by different techniques. Basically the lubricant is pushed or moved in between plates or by moving it in the gravity field. International convention states 16 classes of viscosity as an ISO Standard (ISO VG classes): ISO VG 5, 7, 10, 15, 22, 32, 46, 68, 100, 150, 220, 320, 460, 680, 1000 and 1500. Low numbers indicate low viscosity, higher numbers high viscosity. Since viscosity is strictly related to temperature, the ISO VG classification refers to 40°C as a standard temperature. The nature of measuring the viscosity leads to the physical value of an area per

*Low molecular weight*, branched hydrocarbons are often used in *pneumatic spraying*, due to their

*Low viscous hydrocarbons* from ISO VG 10, 15, 22, 32, 46, 68 and 100 are in use as *hydraulic oils.*

*Hydrocarbons with higher viscosities* are part of *machine oils*, carrying out the ordinary lubrication functions. Machine oil viscosities are in the range of ISO VG 68, 100, 150, 220, 320, 460. The

/s. Hence, ISO VG 68 for example denotes a viscosity of the lubricant, measured at

/s within a range of roughly 10% below and 10% beyond the given 68mm2

/s.

**Figure 14.** Representatives of saturated hydrocarbons as typical lubricants

viscosity range, starting at 2 (water-like), 5, 10 and 15.

number of carbons is in the range of 30–80 in the chain.

Common hydraulic oil viscosity is around ISO VG 32, 46 and 68.

by blocking filters.

oil circuits.

(Figure 14)

time: mm2

40°C within 68 mm2

**Figure 10.** Methane, Ethane, Propane, Butane

Starting from pentane the hydrocarbons get liquid and are the principal components of fuels, solvents, and raw materials for the chemical industry. To facilitate reading and drawing only the carbon backbone is drawn without explicitly showing hydrogen. (Figure 11)

**Figure 11.** Pentane, Hexane, Heptane

Binding of carbon to carbon may be realized in chains, but also in branched chains and different cycles (Figure 12).

**Figure 12.** Methylbutane (Isopentane), 2,2-Dimethylpropane (Neopentane), Cyclohexane

Hydrocarbons from C10 on till C14 are in use as solvents for cleaning (C11-C13 isoparaffines) (Figure 13):

**Figure 13.** C11-C13 Iso paraffines

From C16 on, hydrocarbons represent typical structures present in lubricants. As the linear hydrocarbons, beginning at C18 are solids, they are common in waxes and thickeners for liquid hydrocarbons. Due to their high solidification point they are a threat if present in Diesel fuels by blocking filters.

Apart from their function as hydrocarbon waxes they are not suitable as lubricants for machine oil circuits.

Suitable lubricants are derived from C16–C70 hydrocarbons with branched chains. Branching leads to low pour points (the point where the lubricant starts to get solid). Machine oils with low pour point, suitable for low temperature applications are branched in their carbon chain. (Figure 14)

**Figure 14.** Representatives of saturated hydrocarbons as typical lubricants

Methane

**Figure 11.** Pentane, Hexane, Heptane

cycles (Figure 12).

(Figure 13):

**Figure 13.** C11-C13 Iso paraffines

**Figure 10.** Methane, Ethane, Propane, Butane

H

H H

H

8 Tribology - Fundamentals and Advancements

H H H

H H H

Starting from pentane the hydrocarbons get liquid and are the principal components of fuels, solvents, and raw materials for the chemical industry. To facilitate reading and drawing only

Binding of carbon to carbon may be realized in chains, but also in branched chains and different

Hydrocarbons from C10 on till C14 are in use as solvents for cleaning (C11-C13 isoparaffines)

From C16 on, hydrocarbons represent typical structures present in lubricants. As the linear hydrocarbons, beginning at C18 are solids, they are common in waxes and thickeners for liquid

the carbon backbone is drawn without explicitly showing hydrogen. (Figure 11)

H

H

H H

Propane Butane

H

<sup>H</sup> <sup>H</sup>

H

<sup>H</sup> <sup>H</sup>

H H

> H H

H

Ethane

**Figure 12.** Methylbutane (Isopentane), 2,2-Dimethylpropane (Neopentane), Cyclohexane

H H H

> In general, the viscosity of a lubricant - as a measure for the ability to move across - increases with the molecular weight, e.g. the number of carbon atoms attached. Viscosity is measured by different techniques. Basically the lubricant is pushed or moved in between plates or by moving it in the gravity field. International convention states 16 classes of viscosity as an ISO Standard (ISO VG classes): ISO VG 5, 7, 10, 15, 22, 32, 46, 68, 100, 150, 220, 320, 460, 680, 1000 and 1500. Low numbers indicate low viscosity, higher numbers high viscosity. Since viscosity is strictly related to temperature, the ISO VG classification refers to 40°C as a standard temperature. The nature of measuring the viscosity leads to the physical value of an area per time: mm2 /s. Hence, ISO VG 68 for example denotes a viscosity of the lubricant, measured at 40°C within 68 mm2 /s within a range of roughly 10% below and 10% beyond the given 68mm2 /s.

> *Low molecular weight*, branched hydrocarbons are often used in *pneumatic spraying*, due to their viscosity range, starting at 2 (water-like), 5, 10 and 15.

> *Low viscous hydrocarbons* from ISO VG 10, 15, 22, 32, 46, 68 and 100 are in use as *hydraulic oils.* Common hydraulic oil viscosity is around ISO VG 32, 46 and 68.

> *Hydrocarbons with higher viscosities* are part of *machine oils*, carrying out the ordinary lubrication functions. Machine oil viscosities are in the range of ISO VG 68, 100, 150, 220, 320, 460. The number of carbons is in the range of 30–80 in the chain.

Some applications in heavy duty processes demand viscosities even higher in the range of ISO VG 680, 1000 and 1500.

the intermediate carbon radical is stabilized by the aromat and thus starts to stay persistent. As a fact, aromats may strongly boost oxidation of hydrocarbons if present in the mixture due

Aromats and naphtenics (containing unsaturated hydrocarbons and aromats) should be

PAO is dominating all synthetic hydrocarbons by amount of production and worldwide turnover. Syntheses start from Dec-1-ene, a linear C10 hydrocarbon with a double bond at the beginning of the molecule. Polymerization and hydrogenation leads to PAO, as a highly

Modern PAO may also start from a variety of hydrocarbons (C8-C12) by the same processes. PAO are the most prominent worldwide used hydrocarbons and found in all important applications, e.g. gear oils, circuit oil, hydraulic oil, base stock for automotive applications and

Stabilization Further Attack

O

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 11

to the mentioned persistency of the reactive intermediates. (Figure 17)

stabilized against oxidation.

O-O Metal (Iron)

**.**

**5. Synthetic hydrocarbons**

**5.1. Poly-α-Olefins (PAO)**

others. [2, 3, 6, 7]

**Figure 17.** Oxygen attack in the oxidation mechanism of Alkylaromats

branched and fully saturated hydrocarbon (Figure 18). [2, 4]

**.**

#### **4.1. Cyclic hydrocarbons (Naphtenes)**

Naphtenic hydrocarbons are derived from hydrocarbon cycles with more or less long chains attached to the cycle. Due to their high branching they are very common in low temperature applications (below -30°C) for hydraulics; low temperature greases. (Figure 15)

**Figure 15.** Principal Structure of Napthenic Hydrocarbons

#### **4.2. Aromatic hydrocarbons (Alkyl aromats)**

Aromatic Hydrocarbons (Alkyl Aromats) derive from the six-membered benzene ring system, attached by hydrocarbon chains. Aromatic hydrocarbons are in use for low temperature applications.

Alkyl Naphthalenes are a modern group of aromatic hydrocarbons. They may act as solvent improvers for synthetic oils, facilitators in generating greases, low temperature applications and much more (Figure 16):

**Figure 16.** Alkyl Aromats and Naphtalenes

Aromats and aromat-containing hydrocarbons are very vulnerable toward oxygen. Oxidation of aromats starts at the attached hydrocarbon chain, proximate to the aromat nucleus by a radical attack. This position is always very sensitive in similar structure, due to the fact, that the intermediate carbon radical is stabilized by the aromat and thus starts to stay persistent. As a fact, aromats may strongly boost oxidation of hydrocarbons if present in the mixture due to the mentioned persistency of the reactive intermediates. (Figure 17)

Aromats and naphtenics (containing unsaturated hydrocarbons and aromats) should be stabilized against oxidation.

**Figure 17.** Oxygen attack in the oxidation mechanism of Alkylaromats

#### **5. Synthetic hydrocarbons**

#### **5.1. Poly-α-Olefins (PAO)**

Some applications in heavy duty processes demand viscosities even higher in the range of ISO

Naphtenic hydrocarbons are derived from hydrocarbon cycles with more or less long chains attached to the cycle. Due to their high branching they are very common in low temperature

Aromatic Hydrocarbons (Alkyl Aromats) derive from the six-membered benzene ring system, attached by hydrocarbon chains. Aromatic hydrocarbons are in use for low temperature

Alkyl Naphthalenes are a modern group of aromatic hydrocarbons. They may act as solvent improvers for synthetic oils, facilitators in generating greases, low temperature applications

Aromats and aromat-containing hydrocarbons are very vulnerable toward oxygen. Oxidation of aromats starts at the attached hydrocarbon chain, proximate to the aromat nucleus by a radical attack. This position is always very sensitive in similar structure, due to the fact, that

applications (below -30°C) for hydraulics; low temperature greases. (Figure 15)

VG 680, 1000 and 1500.

10 Tribology - Fundamentals and Advancements

applications.

and much more (Figure 16):

**Figure 16.** Alkyl Aromats and Naphtalenes

**4.1. Cyclic hydrocarbons (Naphtenes)**

**Figure 15.** Principal Structure of Napthenic Hydrocarbons

**4.2. Aromatic hydrocarbons (Alkyl aromats)**

PAO is dominating all synthetic hydrocarbons by amount of production and worldwide turnover. Syntheses start from Dec-1-ene, a linear C10 hydrocarbon with a double bond at the beginning of the molecule. Polymerization and hydrogenation leads to PAO, as a highly branched and fully saturated hydrocarbon (Figure 18). [2, 4]

Modern PAO may also start from a variety of hydrocarbons (C8-C12) by the same processes. PAO are the most prominent worldwide used hydrocarbons and found in all important applications, e.g. gear oils, circuit oil, hydraulic oil, base stock for automotive applications and others. [2, 3, 6, 7]

Sulfurization with activated sulfur precursors lead toward sulfurized isobutenes (SIB) widely

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 13

Esters are in general reaction products between alcohols and acids. Their formation is also possible by means of other techniques, e.g. specific oxidation reactions, rearrangements in

Carboxylic Acid esters are created by the reaction of alcohols and carboxylic acids [A] and their derivatives, by trans–esterification (B), or catalytic reactions, e.g. epoxides with carbon dioxide

Mono-Esters derive from a monocarboxylic acid (Carboxylic Acid that contains only one acidic

Esters derived from this structure are seldom used as pure lubricants, more as solvents or dispersants. For example alcohol ethoxylates, formed by addition of alcohols to epoxides may be esterified by a monocarboxylic acid leading toward a dispersant or self-emulsifying solvent.

Di-Esters are synthesized by use of dicarboxylic acids, mainly adipaic or sebacaic acid and two molecules of an alcohole. 2-Ethylhexylalcohole (Iso Octanole), leading to Di-isooctyladipate

They constitute an important group of oils, with either the function of base oil by themselves

For Di-Esters the reaction of alcohols (A) with two hydroxyl groups and a monofunctional

For technical purposes the reaction product of Neo Pentylgylcole (3.3 Dimethyl-propane-1.4 diol) with oleic acid is important in lubrication technologies for use as a friction reducer and

Despite their high variety in structure esters are used in different categories: [1, 2, 4]

centre) and monofunctional alcohols (Alcohole with only one OH group). [10]

(DOA) or Di-isooctylsebacate (DOS, DEHS = Di-ethylhexylsebacate). (Figure 22)

but also as adjuvant to mineral oil or PAO formulations.

carboxylic acid (B, B') is also applicable. (Figure 23)

in minimal lubrication systems. (Figure 24)

used as extreme pressure (see also section about EP/AW additives).

organic molecules or different reactions. [10]

**6.2. Esters in lubrication technology**

**6. Ester oils**

**6.1. General**

(C) (Figure 20). [10]

**6.3. Mono-esters**

(Figure 21)

**6.4. Di-esters**

**Figure 18.** Principle of PAO formation

The extraordinary importance of PAO is due to its applicability at very low temperatures (pour points below –30°C) and, in the case of suitable antioxidant prevention also at higher temper‐ atures (> 120°C). While PAO is, by its structure, very common in low temperature applications, it is very poor in the contact with metal surfaces beyond 120°C if not properly additivated by antioxidants.

Principal antioxidants for PAO are Phenyl-α-Naphtylamine (PAN) and octyldiphenylamines (see antioxidants (AO)).

#### **5.2. Polyisobutenes (PIB)**

PIB are a sub class of polymerized olefins. They are widely used to boost low viscous oils to higher ISO VG grades or as functional additives to improve the viscosity index (VI): the attitude of the lubricant to lower its viscosity strongly by temperature is reduced by addition of PIB. Synthesis is carried out starting from isobutene by catalytic oxidation processes (Figure 19):

**Figure 19.** PIB formation by catalytic polymerization of Isobutene

Sulfurization with activated sulfur precursors lead toward sulfurized isobutenes (SIB) widely used as extreme pressure (see also section about EP/AW additives).

#### **6. Ester oils**

#### **6.1. General**

Esters are in general reaction products between alcohols and acids. Their formation is also possible by means of other techniques, e.g. specific oxidation reactions, rearrangements in organic molecules or different reactions. [10]

Carboxylic Acid esters are created by the reaction of alcohols and carboxylic acids [A] and their derivatives, by trans–esterification (B), or catalytic reactions, e.g. epoxides with carbon dioxide (C) (Figure 20). [10]

#### **6.2. Esters in lubrication technology**

Despite their high variety in structure esters are used in different categories: [1, 2, 4]

#### **6.3. Mono-esters**

The extraordinary importance of PAO is due to its applicability at very low temperatures (pour points below –30°C) and, in the case of suitable antioxidant prevention also at higher temper‐ atures (> 120°C). While PAO is, by its structure, very common in low temperature applications, it is very poor in the contact with metal surfaces beyond 120°C if not properly additivated by

PAO

Catalyst

Principal antioxidants for PAO are Phenyl-α-Naphtylamine (PAN) and octyldiphenylamines

PIB are a sub class of polymerized olefins. They are widely used to boost low viscous oils to higher ISO VG grades or as functional additives to improve the viscosity index (VI): the attitude of the lubricant to lower its viscosity strongly by temperature is reduced by addition of PIB. Synthesis is carried out starting from isobutene by catalytic oxidation processes (Figure 19):

Isobutene Polyisobutene

**Figure 19.** PIB formation by catalytic polymerization of Isobutene

Dec-1-ene

Polymerisation

Remove precursors

**Figure 18.** Principle of PAO formation

12 Tribology - Fundamentals and Advancements

n

antioxidants.

(see antioxidants (AO)).

**5.2. Polyisobutenes (PIB)**

Mono-Esters derive from a monocarboxylic acid (Carboxylic Acid that contains only one acidic centre) and monofunctional alcohols (Alcohole with only one OH group). [10]

Esters derived from this structure are seldom used as pure lubricants, more as solvents or dispersants. For example alcohol ethoxylates, formed by addition of alcohols to epoxides may be esterified by a monocarboxylic acid leading toward a dispersant or self-emulsifying solvent. (Figure 21)

#### **6.4. Di-esters**

Di-Esters are synthesized by use of dicarboxylic acids, mainly adipaic or sebacaic acid and two molecules of an alcohole. 2-Ethylhexylalcohole (Iso Octanole), leading to Di-isooctyladipate (DOA) or Di-isooctylsebacate (DOS, DEHS = Di-ethylhexylsebacate). (Figure 22)

They constitute an important group of oils, with either the function of base oil by themselves but also as adjuvant to mineral oil or PAO formulations.

For Di-Esters the reaction of alcohols (A) with two hydroxyl groups and a monofunctional carboxylic acid (B, B') is also applicable. (Figure 23)

For technical purposes the reaction product of Neo Pentylgylcole (3.3 Dimethyl-propane-1.4 diol) with oleic acid is important in lubrication technologies for use as a friction reducer and in minimal lubrication systems. (Figure 24)

**6.5. Tri-esters**

3CH

*6.5.1. Glycerol esters*

They are mainly represented by to major groups:

Tri-Esters are mainly created by the reaction of trivalent alcohols with monocarboxylic acids.

O

O

O

O

<sup>O</sup> CH3

O

+

O

O

O

CH3

O

CH3

CH3

CH3 DOS

OH CH3

O

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731

DOA

15

DEHS

O

**Figure 21.** Mono Ester Formation with the specialty of esterified alcohole ethoxilates

O

O

O

O

OH

3CH

3CH

3CH

**Figure 22.** DOA and DOS

3CH

3CH

O

Esters derived from glycerol as a trivalent alcohole leads to tri-Esters. (Figure 25)

O

**Figure 23.** Formation of Di-Esters by Di-Alcoholes (A) reacted with Monocarboxylic Acids (B, B')

O

OH OH +

<sup>B</sup> <sup>A</sup> B'

**Figure 20.** Examples for creation of esters. (A) Reaction of Carboxylic Acids with Alcohols, (B) Transesterfication, (C) Reaction of Expoxide to cyclic Esters.

**Figure 21.** Mono Ester Formation with the specialty of esterified alcohole ethoxilates

**Figure 22.** DOA and DOS

**Figure 23.** Formation of Di-Esters by Di-Alcoholes (A) reacted with Monocarboxylic Acids (B, B')

#### **6.5. Tri-esters**

Tri-Esters are mainly created by the reaction of trivalent alcohols with monocarboxylic acids. They are mainly represented by to major groups:

#### *6.5.1. Glycerol esters*

**Figure 20.** Examples for creation of esters. (A) Reaction of Carboxylic Acids with Alcohols, (B) Transesterfication, (C)

Reaction of Expoxide to cyclic Esters.

14 Tribology - Fundamentals and Advancements

Esters derived from glycerol as a trivalent alcohole leads to tri-Esters. (Figure 25)

As a fact of the presence of short chain carboxylic acids those esters are nutrients, biological

As a special glycerol ester, important for lubrication, ricinoleic acid esters have to be men‐

Within this group ricinoleic acid represents the group of 12-hydroxy substituted C18 carbox‐

Hence, alkaline cleavage of ricinoleic acid glycerol esters lead to 12-Hydroxi-oleic acid on the

Catalytic hydrogenation of 12-Hydroxioleic acid results in the formation of 12-Hydroxistearic acid, which is important for modern grease concepts. Sebacaic Acid on the other hand is a raw material for DOS (see above) but also for the production of complex greases. (Figure 26) [1, 2, 4]

Alkalines

O O

12-Hydroxioleic Acid

Octan-2-ole OH

H

OH

O

12-Hydroxi Stearic Acid

Sebacaic Acid

OH

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 17

O

OH O

OH

OH

one hand and to sebacaic acid on the other hand by degradation of the double bond.

OH

**Figure 26.** Cleavage of Glycerol – Ricinioleic Acid and hydrogenation to 12-Hydroxistearic Acid, Sebacaic Acid and Oc‐

Glycerole Esters with long chain carboxylic acids only, e.g. Glycerole Tristearate, are no longer nutrients and sparingly biodegradable. They are used as emulsifiers, consistency givers.

Glycerole Trioleate is a powerful friction reducer in tribological applications.

degradable and widely used as natural, biodegradable oils.

O

Glycerol-Ricinoleic Acid Ester

H2

O

OH

O O

H

tan-2-ole [10].

OH

O

O O

tioned.

ylic acids.

**Figure 24.** Neopentylgylcoledioleate (NPG-Dioleate)

**Figure 25.** Esterification of Glycerole to tri-Esters

Glycerol Tri-Esters represent the huge group of natural oils. Sunflower, rapseed oil are prominent representatives. A mixture of short chain carboxylic acids with unsaturated long chain acids is used.

As a fact of the presence of short chain carboxylic acids those esters are nutrients, biological degradable and widely used as natural, biodegradable oils.

As a special glycerol ester, important for lubrication, ricinoleic acid esters have to be men‐ tioned.

Within this group ricinoleic acid represents the group of 12-hydroxy substituted C18 carbox‐ ylic acids.

Hence, alkaline cleavage of ricinoleic acid glycerol esters lead to 12-Hydroxi-oleic acid on the one hand and to sebacaic acid on the other hand by degradation of the double bond.

Catalytic hydrogenation of 12-Hydroxioleic acid results in the formation of 12-Hydroxistearic acid, which is important for modern grease concepts. Sebacaic Acid on the other hand is a raw material for DOS (see above) but also for the production of complex greases. (Figure 26) [1, 2, 4]

**Figure 26.** Cleavage of Glycerol – Ricinioleic Acid and hydrogenation to 12-Hydroxistearic Acid, Sebacaic Acid and Oc‐ tan-2-ole [10].

Glycerole Esters with long chain carboxylic acids only, e.g. Glycerole Tristearate, are no longer nutrients and sparingly biodegradable. They are used as emulsifiers, consistency givers.

Glycerole Trioleate is a powerful friction reducer in tribological applications.

O

O

O

**Figure 25.** Esterification of Glycerole to tri-Esters

chain acids is used.

O

OH

**Figure 24.** Neopentylgylcoledioleate (NPG-Dioleate)

16 Tribology - Fundamentals and Advancements

OH

OH O OH

+

O

O

O

O

OCH2

O

CH2O

O

Glycerole Carboxylic Acid

Glycerol Tri-Esters represent the huge group of natural oils. Sunflower, rapseed oil are prominent representatives. A mixture of short chain carboxylic acids with unsaturated long

Glycerole Triester

OH

OH

O

*6.5.2. Triesters, derived from alcohols else than glycerole*

#### *6.5.2.1. Trimethylolpropane esters (TMP-esters)*

TMP-Esters are created out of Trimethylolpropane (TMP) by reaction with short chain carboxylic acids, e.g. the range from C6 to C10. (Figure 27)

COOH

COOH

OH

COOH

+

**Figure 28.** TM Esters by reaction of trimellitic acid with branched alcohole

**Figure 29.** PE Esters

OH

OH

O

OO

O

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 19

O

O

<sup>O</sup> <sup>O</sup>

O

O

O O

O

O

**Figure 27.** TMP Esters

TMP Trioleate is created by reaction of TMP with oleic acid or by trans-esterification.and commonly used as lubricant in minimal lubrication.

#### *6.5.2.2. Trimellitic esters (TM-esters)*

Apart from the described structures where trivalent alcohols get reacted with monocarboxylic acids, trimellitic Esters (TM-Esters) are products from Trimellitic Acid Anhydride with Mono alcohols. (Figure 28)

Due to the aromatic core those esters are high in thermal stability and widely used in high temperature applications.

#### **6.6. Tetra esters (Pentaerythrolesters, PE-esters)**

Pentaerythrole acts as a four-valent alcohole which may be esterified by four carboxylic acids. (Figure 29)

Carboxylic acids are in the range from C6 to C10.

Dipentaerythrol Esters (Di PE Esters) are formed starting from Dipentaerythrole as a six- valent alcohole reacted by six monocarboxylic acids in a Carbon Chain length from 6 to 10. (Figure 30)

#### **6.7. Polyesters**

In the past 20 years new groups of esters have been created by reaction of polycarboxylic polymers with alcohols. Those are reaction products of maleic acid anhydride (MSA) by Ene-Reaction with PAO precursors, leading to the PAO-backbone MSA addition product that might be esterified by butanole, leading to carboxylic complex esters (Figure 31).

**Figure 28.** TM Esters by reaction of trimellitic acid with branched alcohole

**Figure 29.** PE Esters

*6.5.2. Triesters, derived from alcohols else than glycerole*

carboxylic acids, e.g. the range from C6 to C10. (Figure 27)

O

O

O

O

<sup>3</sup>CH

commonly used as lubricant in minimal lubrication.

**6.6. Tetra esters (Pentaerythrolesters, PE-esters)**

Carboxylic acids are in the range from C6 to C10.

*6.5.2.2. Trimellitic esters (TM-esters)*

O

O

TMP-Esters are created out of Trimethylolpropane (TMP) by reaction with short chain

TMP Trioleate is created by reaction of TMP with oleic acid or by trans-esterification.and

Apart from the described structures where trivalent alcohols get reacted with monocarboxylic acids, trimellitic Esters (TM-Esters) are products from Trimellitic Acid Anhydride with Mono

Due to the aromatic core those esters are high in thermal stability and widely used in high

Pentaerythrole acts as a four-valent alcohole which may be esterified by four carboxylic acids.

Dipentaerythrol Esters (Di PE Esters) are formed starting from Dipentaerythrole as a six- valent alcohole reacted by six monocarboxylic acids in a Carbon Chain length from 6 to 10. (Figure 30)

In the past 20 years new groups of esters have been created by reaction of polycarboxylic polymers with alcohols. Those are reaction products of maleic acid anhydride (MSA) by Ene-Reaction with PAO precursors, leading to the PAO-backbone MSA addition product that

might be esterified by butanole, leading to carboxylic complex esters (Figure 31).

*6.5.2.1. Trimethylolpropane esters (TMP-esters)*

18 Tribology - Fundamentals and Advancements

**Figure 27.** TMP Esters

alcohols. (Figure 28)

(Figure 29)

**6.7. Polyesters**

temperature applications.

**7.1. Polar acitivity**

or 20 % esters per weight to create this effect.

**7.2. Low temperature (Pour point) properties**

High temperature applications in the use of esters are achieved by

**•** Sterical hindrance of the ß-Position in the Alcohol

**•** Use of Aromatic Nuclei in the Ester structure

**Figure 32.** Decomposition of esters via cyclic rearrangement

and hence the pour point rises.

side (Figure 32).

**7.3. High temperature properties**

Esters are polar by their nature due to the central element where a carboxylic acid tail binds toward an alcohol. Polarity gives some advantage but also disadvantage in the case esters are used. In general, esters enhance the solubility of functional additives and keep them away from fall-out. Esters also enhance the cleaning of metal surfaces in operation, preventing a formulation by creation of sludge. Esters are, as a fact of their polarity, aggressive toward sealings with a general tendency to shrink them. Plastics and elastomers under bending are susceptible toward stress corrosion cracking if attacked by esters. Hence, stress-corrosion cracking has to be considered explicitly in the case if esters are used. Since hydrocarbons, like PAO have a tendency to swell elastomers, the addition of esters may counteract such that the effect is neutralized. As a fact synthetic oils based on PAO are additivated by addition of 10

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 21

Di-Esters, e.g. DOA, DOS are very useful in temperature ranges that undergo -40°C. This effect might be explained by the lack of hydroxyl groups that might associate at low temperature via hydrogen bridging, but also as a consequence of the crystallization hindrance due to the spatial structure of esters which does not allow a dense crystal packing. In contrast, esters may be designed such that their low temperature properties are lost, just by changing their structure. Also, if the number of polar groups increase the tendency to molecular association increases,

As a specialty esters may rearrange within their structure via a preferred six-membered cyclic intermediate that creates an alkene on one side and a carboxylic acid on the other

> <sup>O</sup> <sup>H</sup> O

a ß

O OH

**Figure 30.** Di PE Esters

**Figure 31.** Complex Ester Formation by Ene-Reaction Sequences

Complex Esters from those structures are widely used to improve the additive solubility and performance. Their structure with shielding the carboxylic groups causes less aggressiveness toward sealings.

#### **7. Structure activity relationship in esters**

Esters are prominent representatives of lubricants where the chemical structure promptly leads to a specific tribological activity. However, if a tribological acitivity is demanded, the specific construction of esters may offer the solution.

#### **7.1. Polar acitivity**

Esters are polar by their nature due to the central element where a carboxylic acid tail binds toward an alcohol. Polarity gives some advantage but also disadvantage in the case esters are used. In general, esters enhance the solubility of functional additives and keep them away from fall-out. Esters also enhance the cleaning of metal surfaces in operation, preventing a formulation by creation of sludge. Esters are, as a fact of their polarity, aggressive toward sealings with a general tendency to shrink them. Plastics and elastomers under bending are susceptible toward stress corrosion cracking if attacked by esters. Hence, stress-corrosion cracking has to be considered explicitly in the case if esters are used. Since hydrocarbons, like PAO have a tendency to swell elastomers, the addition of esters may counteract such that the effect is neutralized. As a fact synthetic oils based on PAO are additivated by addition of 10 or 20 % esters per weight to create this effect.

#### **7.2. Low temperature (Pour point) properties**

Di-Esters, e.g. DOA, DOS are very useful in temperature ranges that undergo -40°C. This effect might be explained by the lack of hydroxyl groups that might associate at low temperature via hydrogen bridging, but also as a consequence of the crystallization hindrance due to the spatial structure of esters which does not allow a dense crystal packing. In contrast, esters may be designed such that their low temperature properties are lost, just by changing their structure. Also, if the number of polar groups increase the tendency to molecular association increases, and hence the pour point rises.

#### **7.3. High temperature properties**

H

n O O

O O

O

O O

**Figure 31.** Complex Ester Formation by Ene-Reaction Sequences

**7. Structure activity relationship in esters**

specific construction of esters may offer the solution.

O

toward sealings.

**Figure 30.** Di PE Esters

20 Tribology - Fundamentals and Advancements

O

O

O

n

O

O

O

O <sup>O</sup> <sup>O</sup>

O

O O

RO

RO

O

OR

O

OR

O O

O

O

O

O

ROH

Complex Esters from those structures are widely used to improve the additive solubility and performance. Their structure with shielding the carboxylic groups causes less aggressiveness

Esters are prominent representatives of lubricants where the chemical structure promptly leads to a specific tribological activity. However, if a tribological acitivity is demanded, the High temperature applications in the use of esters are achieved by


As a specialty esters may rearrange within their structure via a preferred six-membered cyclic intermediate that creates an alkene on one side and a carboxylic acid on the other side (Figure 32).

**Figure 32.** Decomposition of esters via cyclic rearrangement

Degradation of esters via such mechanisms takes place at ambient temperatures, e.g. by copper activation even at 70°C. The formation of carboxylic acids and alkenes may lead to corrosion and unfavorable deposits on metals. In the case of blocking the ß-position, as in the NPG and TMP esters, the cyclic rearrangement is blocked and the ester does not undergo the thermal degradation. Such oils are commonly used as turbine oils.

#### **7.4. Side reactions**

#### *7.4.1. Hydrolysis*

Ester Oils generally hydrolyze by interaction with water. The hydrolytic process is somehow the reverse reaction how esters form. The attack of water is enhanced if alkalinity is present but also acids may catalyze the hydrolysis. Common understanding states the attack of so called nucleophiles, like water at the carbonyl C-atom, followed by rearrangement sequences, leading to carboxylic acid and alcohols. (Figure 33)

**Figure 33.** water-based cleavage of Esters toward carboxylic acids and alcohols.

Catalytic hydrolysis of ester oils also take place at metal surfaces, e.g. under tribological conditions. Formation of carboxylic acids may lead to corrosion as a consequence.

phorous offers two main oxidation states (+III and +V) from which acids are derived. Depend‐ ing on the oxidation state and the alcohols, phosphoric esters are different in use. Also

Lactone

O

O

OH

O

<sup>O</sup> <sup>O</sup> OH

O

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 23

O P O

Phosphorus in the oxidation state (+V) creates a plenty of variant Acids, such as Orthophos‐ phoric Acid, Diphosphoric Acid, Triphosphoric Acids switching into each other. (Figure 36)

Phosphoric Acids are created by reaction of either phosphoric acid anhydride with alcohols or phosphoric acid derivatives, e.g. POCl3 (Phosphorous-Oxi-Chloride) with alcohols. Ali‐

O

phosphorous overtakes the role of anti-wear activity in such substances. [9]

A common representative is Trilaurylphosphite (Figure 35).

**Figure 34.** Lactone formation by side-chain oxidation of esters

**Figure 35.** Trilaurylphosphite as a representative of Phosphinic Acid Esters

phatic alcohols are in use, but also Phenols [10].

*7.5.1. Phosphoric acid esters*

Fe/O2

#### *7.4.2. Biodegradation*

Esters may decline under the interaction of bacteria and combust. Biodegradation is observed in the case of vegetable oils, e.g. glycerol esters, seldom on technical esters. In principal biodegradation cleaves esters, like water does to carboxylic acids. Biodegradation as a complex process does not stop there but lead to further products. Esters may oxidize as described in mineral oils and PAO at the organic tail. As a specialty they may undergo hydroxylation at a side position followed by trans-esterification to lactones. The lactone sequence is described already in the mineral oil section. Lactones are observed if esters, but also PAO are decomposed on iron at higher temperature. Infrared Spectra show absorption at 1800 -1760 cm-1 caused by lactone formation (see also chapter of antioxidants). (Figure 34)

#### **7.5. Other esters**

Esters may be created, as already mentioned by reaction of acids, in a different way as carboxylic ones. Prominent representatives are esters derived from phosphoric acid. Phos‐

**Figure 34.** Lactone formation by side-chain oxidation of esters

phorous offers two main oxidation states (+III and +V) from which acids are derived. Depend‐ ing on the oxidation state and the alcohols, phosphoric esters are different in use. Also phosphorous overtakes the role of anti-wear activity in such substances. [9]

A common representative is Trilaurylphosphite (Figure 35).

**Figure 35.** Trilaurylphosphite as a representative of Phosphinic Acid Esters

#### *7.5.1. Phosphoric acid esters*

Degradation of esters via such mechanisms takes place at ambient temperatures, e.g. by copper activation even at 70°C. The formation of carboxylic acids and alkenes may lead to corrosion and unfavorable deposits on metals. In the case of blocking the ß-position, as in the NPG and TMP esters, the cyclic rearrangement is blocked and the ester does not undergo the thermal

Ester Oils generally hydrolyze by interaction with water. The hydrolytic process is somehow the reverse reaction how esters form. The attack of water is enhanced if alkalinity is present but also acids may catalyze the hydrolysis. Common understanding states the attack of so called nucleophiles, like water at the carbonyl C-atom, followed by rearrangement sequences,

Catalytic hydrolysis of ester oils also take place at metal surfaces, e.g. under tribological

Esters may decline under the interaction of bacteria and combust. Biodegradation is observed in the case of vegetable oils, e.g. glycerol esters, seldom on technical esters. In principal biodegradation cleaves esters, like water does to carboxylic acids. Biodegradation as a complex process does not stop there but lead to further products. Esters may oxidize as described in mineral oils and PAO at the organic tail. As a specialty they may undergo hydroxylation at a side position followed by trans-esterification to lactones. The lactone sequence is described already in the mineral oil section. Lactones are observed if esters, but also PAO are decomposed on iron at higher temperature. Infrared Spectra show absorption at 1800 -1760 cm-1 caused by

Esters may be created, as already mentioned by reaction of acids, in a different way as carboxylic ones. Prominent representatives are esters derived from phosphoric acid. Phos‐

conditions. Formation of carboxylic acids may lead to corrosion as a consequence.

OH

OH

O

degradation. Such oils are commonly used as turbine oils.

leading to carboxylic acid and alcohols. (Figure 33)

O

OH2

**Figure 33.** water-based cleavage of Esters toward carboxylic acids and alcohols.

lactone formation (see also chapter of antioxidants). (Figure 34)

O

**7.4. Side reactions**

22 Tribology - Fundamentals and Advancements

*7.4.2. Biodegradation*

**7.5. Other esters**

*7.4.1. Hydrolysis*

Phosphorus in the oxidation state (+V) creates a plenty of variant Acids, such as Orthophos‐ phoric Acid, Diphosphoric Acid, Triphosphoric Acids switching into each other. (Figure 36)

Phosphoric Acids are created by reaction of either phosphoric acid anhydride with alcohols or phosphoric acid derivatives, e.g. POCl3 (Phosphorous-Oxi-Chloride) with alcohols. Ali‐ phatic alcohols are in use, but also Phenols [10].

**Figure 36.** Representation of Phosphoric Acid Ester

#### *7.5.1. Formation of phosphoric acid ester*

Reaction of aliphatic alcohols, e.g. hexanole, with phosphorous pentoxide leads to hexylphos‐ phate. In general some acidity remains due to insufficient esterifications. As a consequence those esters are often neutralized with amines to give amine phosphates. [10] (Figure 37)

Amine phosphates are widely used as anti-wear and anti-corrosion additives in all kinds of applications. Phosphoric Acid Esters derived from Phenoles are shown below. (Figure 38).

In a different reaction Scheme Phosphoroxichloride reacts with alcohols. Those reactions are convenient to come to aryl phosphoric acid esters. Arylphosphates are somehow used to come to non-flammable high temperature lubricants at temperatures beyond 200°C.

Apart from the use as base oil phosphoric esters like Tricresylphosphates, based on the reaction from Phosphorous Oxichloride with Cresol (Methylphenols) are common additives for lubricants in bearing industry.

Whilst TCP with the methyl group in the para-position is seen as hazardous, TCP isomers in the ortho is registered to be highly toxic. Also mixtures of TCP isomers, due to the content of the highly toxic ortho isomer are registered as highly toxic. TCP, despite its superior behavior as AW additive for bearing lubrication is restricted for use (Figure 39).

Use of Thiophosphorylchloride as precursor, the reaction with Phenole leads to EP/AW additives like Triphenylphosphorothionate (TPPT) and its derivatives. (Figure 40)

TPPT is widely used as non metal EP Additive as a substitute for Zn and Molybdenum Dithiophosphates. Due to its thermal stability, TPPT undergoes reactions at higher tempera‐ tures (>100°C). As to the fact that TPPT is ashless and starts to react at higher temperatures, it is a preferred additive in high temperature lubrication in combination with sterically hindered esters and PAO. In contrast to TCP, TPPT is not registered to be toxic, even more, the use of TPPT is allowed at level of 0.5% per weight for incidental food contact.

#### **8. Polyglycoles (PG)**

#### **8.1. General**

Synthesis of Polyglycoles starts from Epoxides, obtained by catalytic oxidation of Alkenes from Petrol- or hydrocarbon chemistry. Polymerization catalysed either by acids or alkaline result in the formation of polyglycoles. In the case of alkaline catalyst, e.g. alkoxides on half of the PG contains a hydroxyl group while the end is capped by an ether function. (Figure 41) [1, 2,

NH

Alcoholes

7, 10]

O

O

O

P P O O

**Figure 37.** Phosphate Esters and Amine phosphates

O

O

P <sup>O</sup> <sup>O</sup>

O

Phosphor (V) Oxide

P

O

Amines

O

P P O O

O

O

P <sup>O</sup> <sup>O</sup>

O

P

O

O

OP O

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 25

O OP OH

O OP O

NH2+

O

OH <sup>O</sup> <sup>n</sup>

OH <sup>O</sup> <sup>n</sup>

Phosphoric Acid Ester

Amine Phosphate

**Figure 37.** Phosphate Esters and Amine phosphates

*7.5.1. Formation of phosphoric acid ester*

**Figure 36.** Representation of Phosphoric Acid Ester

OH

OH OP OH

24 Tribology - Fundamentals and Advancements

lubricants in bearing industry.

**8. Polyglycoles (PG)**

**8.1. General**

Reaction of aliphatic alcohols, e.g. hexanole, with phosphorous pentoxide leads to hexylphos‐ phate. In general some acidity remains due to insufficient esterifications. As a consequence those esters are often neutralized with amines to give amine phosphates. [10] (Figure 37)

O O

OH

OH

P-O-P OH OH

O

HO

OH

P-O-P-O-P

O

OH

OH

O

Amine phosphates are widely used as anti-wear and anti-corrosion additives in all kinds of applications. Phosphoric Acid Esters derived from Phenoles are shown below. (Figure 38). In a different reaction Scheme Phosphoroxichloride reacts with alcohols. Those reactions are convenient to come to aryl phosphoric acid esters. Arylphosphates are somehow used to come

Apart from the use as base oil phosphoric esters like Tricresylphosphates, based on the reaction from Phosphorous Oxichloride with Cresol (Methylphenols) are common additives for

Whilst TCP with the methyl group in the para-position is seen as hazardous, TCP isomers in the ortho is registered to be highly toxic. Also mixtures of TCP isomers, due to the content of the highly toxic ortho isomer are registered as highly toxic. TCP, despite its superior behavior

Use of Thiophosphorylchloride as precursor, the reaction with Phenole leads to EP/AW

TPPT is widely used as non metal EP Additive as a substitute for Zn and Molybdenum Dithiophosphates. Due to its thermal stability, TPPT undergoes reactions at higher tempera‐ tures (>100°C). As to the fact that TPPT is ashless and starts to react at higher temperatures, it is a preferred additive in high temperature lubrication in combination with sterically hindered esters and PAO. In contrast to TCP, TPPT is not registered to be toxic, even more, the use of

Synthesis of Polyglycoles starts from Epoxides, obtained by catalytic oxidation of Alkenes from Petrol- or hydrocarbon chemistry. Polymerization catalysed either by acids or alkaline result

additives like Triphenylphosphorothionate (TPPT) and its derivatives. (Figure 40)

to non-flammable high temperature lubricants at temperatures beyond 200°C.

as AW additive for bearing lubrication is restricted for use (Figure 39).

TPPT is allowed at level of 0.5% per weight for incidental food contact.

in the formation of polyglycoles. In the case of alkaline catalyst, e.g. alkoxides on half of the PG contains a hydroxyl group while the end is capped by an ether function. (Figure 41) [1, 2, 7, 10]

**Figure 38.** Arylphosphates derived from Phosphor Oxide Chloride Reactions

ortho Tricresylphosphate highly toxic

**Figure 39.** TCP and some isomers

meta - Tricresylphosphate hazardous

para- Tricresylphosphate hazardous

The choice of either different alkenes (Group R) or alkoxides (R') leads toward a huge variety

**Figure 41.** General Formation of Polyglycoles by alkaline catalytic polymerization of Alkene Epoxides

O

R

<sup>O</sup> R'O

R

**Polyglycoles PEG PPG-PEG PPG PBG**

Physical Data Density approx. 1 0.95-0.98 0.95-0.98 0.95-0.98

Tribological Data Viscosity, 40°C Range 32 - 46 32- 100 32- 100 32- 100

ABS compatible

**•** *Polyethylene- Polypropylene Oxide Mixtures* started from mixtures or Ethylene and Propylene

Table 1 offers an overview across the most common PG their data and applicability.

Paintings not compatible

Technically only a couple of variances are produced in a larger scale, such as:

Hydrocarbons non miscible partially partially partially Ester Oils partially- full partially-full partially-full partially-full Other PG PPG partially- full partially-full partially-full partially-full

VT Coefficient Range 180-..> 200 180-..> 200 180-..> 200 180-..> 200

PBG partially- full partially-full partially-full partially-full

Water miscible(%) 100 partially

Mixed Polymers Ethylene/Prop ylene Oxide

O

R

Proyleneoxide Polymers

Butyleneoxid e Polymers

OH

27

n

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731

of PG, all of them with different chemical and physical properties.

R'OH

Flashpoint approx. Pourpoint approx.

VP Coefficient Range Others Seals NBR compatible

**•** *Polyethylenglycoles (PEG)* where Ethylene Oxide is the starter

**•** *Polypropyleneglycoles (PPG)* where Propylene Oxide is the starter

Description Ethyleneoxide Polymer

O

R

**Table 1.** PG and their data and applicability

Oxide

Chemical Data

**Figure 40.** Synthesis of TPPT

**Figure 41.** General Formation of Polyglycoles by alkaline catalytic polymerization of Alkene Epoxides

The choice of either different alkenes (Group R) or alkoxides (R') leads toward a huge variety of PG, all of them with different chemical and physical properties.


**Table 1.** PG and their data and applicability

P=O

26 Tribology - Fundamentals and Advancements

Phosphorousoxichloride

O P = O

O

O

ortho Tricresylphosphate highly toxic

**Figure 39.** TCP and some isomers

P=S

Phosphorousoxichloride

Cl

Cl Cl +

**Figure 40.** Synthesis of TPPT

OH

Cresol

**Figure 38.** Arylphosphates derived from Phosphor Oxide Chloride Reactions

OH

Phenol

O P = O

Tricresylphosphate TCP

O P = O

O

O

para- Tricresylphosphate hazardous

Triphenylphosphorothionate

TPPT

O

O

O P = O

O

O

meta - Tricresylphosphate hazardous

O P = S

O

O

Cl +

Cl

Cl

Technically only a couple of variances are produced in a larger scale, such as:


Table 1 offers an overview across the most common PG their data and applicability.

Single addition of long chain alcohols lead to the formation of fatty alcohol ethoxilates, for use as non-ionic detergants and dispersants in lubricant formulations, as silicone free defoaming and emulsifiers for lubricant formulae.

In general PG offer very high viscosity indices, mainly above 160 (compared to mineraloils at ranges from 20 (alkylnaphtalenes), napthenics (70), paraffine base solvates (110), Poly-α-

This high VI allows reducing the calculated viscosity in a given tribological application down

calculated for a mineral oil with a VI of 100, this viscosity maybe reduced by use of a PG down

of -30.. – 40 °C even. Reaching the Pour point, PG tend to form highly viscous liquid, however, crystallization-inhibited. As a fact of this huge increase, PG is not for use even at temperatures above the pour point. Realistically PG is not suitable in the vicinity of their pour point. Therefore they are not very good low temperature base oils compared, for instance, with esters or PAO fluids. Due to their chemical structure PG are somehow strong solvents, e.g. paintings. In the case PG is used, the system has to be checked whether the paintings of the tank, the machine housing or others are affected. Dissolved painting from the tank may cause severe problems in the oil circuit by blocking filters. Additive response, known from standard applications, may change seriously by use of PG due to their different solvent capability. Extreme Pressure Additives have to be checked in their performance if used in PG. Normally

As a fact of the presence of epoxides in PG and due to their cancerogenic potential, the use of

PEG, apart from its wide use in cosmetic industry is completely water miscible. Due to its water miscibility PEG is only or sparingly soluble in hydrocarbons. Compatibility of the PEG with a given fluid has to be checked before use. PEG, as facts of its water miscibility will uptake water without separation. In case of the use of PEG in applications within water environment the water ingress should be checked carefully. Effects of water ingress are increasing threat of

Water miscibility is of use in non-flammable hydraulics in coal mining industries, but also in applications of pharmacy and food processing. In general PEG is allowed within the FDA regulation to be safe for incidental food contact. Due to the positive effect of sliding especially in worm gears PEG is somehow recommended for use in such applications. [2, 6, 7, 9]

Polyethylene Glycols are made from ethylene oxide by polymerization (Figure 43).

/s. Pour points are low in the case of PG, very often in the range

O

OH OH

O

n

/s, at 40°C) is

29

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731

to one or two levels. For example, if in a given application ISO VG 320 (320 mm2

anti-wear and anti-friction additives may be decreased in their content.

O O

Olefines (140).

to 220 mm2

/s or even 150 mm2

PG formulations drops down.

**8.2. Polyethylene Glycols (PEG)**

**Figure 43.** PEG formation and structure

*8.2.1. Use of PEG*

corrosion and thinning due to the mixture.

In general PG are not thermally stable by themselves and tend to decompose by emission of volatile degradation products, e.g. low boiling compounds, such as aldehydes, ketones, acids and others. Due to this behavior PG are used in high temperature applications where the formation of polymers and lacquers due to heat induced degradation of lubricants is not convenient, for example high temperature chain lubrication.

Presence of alkalines, such as overbased sulphonates, widely used in motor oils, as corrosion inhibitor lead to multiple cross-reactions with the decomposition products of PG (aldol reactions): Results of the aldole reaction are tars, sludge and slurries in the system. In conse‐ quence corrosion resistance of PG should always be carried out by acidic corrosion inhibitors, such as succinic-esters, Zinc-Naphtenates or Phosphoric partial esters. (Figure 42)

**Figure 42.** Aldole sludge formation in PG by use of alkaline

It has to be considered that PG are poorly soluble even amongst themselves and should be carefully checked. In general their solubility in mineral oils is poor, better in esters (depending on the structure). However, PG needs to be stabilized by antioxidants in order to prevent the early thermal degradation. By doing so, the application of PG are enhanced significantly, such, that even applications temperatures > 160°C are approached.

Convenient stabilizers are Phenyl-α-Naphtylamine, Phenothiazines or Alkyldiphenylamines. The amount should be adapted to the application.

In general PG offer very high viscosity indices, mainly above 160 (compared to mineraloils at ranges from 20 (alkylnaphtalenes), napthenics (70), paraffine base solvates (110), Poly-α-Olefines (140).

This high VI allows reducing the calculated viscosity in a given tribological application down to one or two levels. For example, if in a given application ISO VG 320 (320 mm2 /s, at 40°C) is calculated for a mineral oil with a VI of 100, this viscosity maybe reduced by use of a PG down to 220 mm2 /s or even 150 mm2 /s. Pour points are low in the case of PG, very often in the range of -30.. – 40 °C even. Reaching the Pour point, PG tend to form highly viscous liquid, however, crystallization-inhibited. As a fact of this huge increase, PG is not for use even at temperatures above the pour point. Realistically PG is not suitable in the vicinity of their pour point. Therefore they are not very good low temperature base oils compared, for instance, with esters or PAO fluids. Due to their chemical structure PG are somehow strong solvents, e.g. paintings. In the case PG is used, the system has to be checked whether the paintings of the tank, the machine housing or others are affected. Dissolved painting from the tank may cause severe problems in the oil circuit by blocking filters. Additive response, known from standard applications, may change seriously by use of PG due to their different solvent capability. Extreme Pressure Additives have to be checked in their performance if used in PG. Normally anti-wear and anti-friction additives may be decreased in their content.

As a fact of the presence of epoxides in PG and due to their cancerogenic potential, the use of PG formulations drops down.

#### **8.2. Polyethylene Glycols (PEG)**

Single addition of long chain alcohols lead to the formation of fatty alcohol ethoxilates, for use as non-ionic detergants and dispersants in lubricant formulations, as silicone free defoaming

In general PG are not thermally stable by themselves and tend to decompose by emission of volatile degradation products, e.g. low boiling compounds, such as aldehydes, ketones, acids and others. Due to this behavior PG are used in high temperature applications where the formation of polymers and lacquers due to heat induced degradation of lubricants is not

Presence of alkalines, such as overbased sulphonates, widely used in motor oils, as corrosion inhibitor lead to multiple cross-reactions with the decomposition products of PG (aldol reactions): Results of the aldole reaction are tars, sludge and slurries in the system. In conse‐ quence corrosion resistance of PG should always be carried out by acidic corrosion inhibitors,

such as succinic-esters, Zinc-Naphtenates or Phosphoric partial esters. (Figure 42)

OH

OH

It has to be considered that PG are poorly soluble even amongst themselves and should be carefully checked. In general their solubility in mineral oils is poor, better in esters (depending on the structure). However, PG needs to be stabilized by antioxidants in order to prevent the early thermal degradation. By doing so, the application of PG are enhanced significantly, such,

Convenient stabilizers are Phenyl-α-Naphtylamine, Phenothiazines or Alkyldiphenylamines.

O

Aldehydes Ketones

OH OH <sup>n</sup>

O

H n

Aldol Products

n

H

and emulsifiers for lubricant formulae.

28 Tribology - Fundamentals and Advancements

O

R

<sup>O</sup> R'O

R

Heat

O

**Figure 42.** Aldole sludge formation in PG by use of alkaline

The amount should be adapted to the application.

that even applications temperatures > 160°C are approached.

Polyglycole

R

Catalysis <sup>O</sup>

convenient, for example high temperature chain lubrication.

Polyethylene Glycols are made from ethylene oxide by polymerization (Figure 43).

**Figure 43.** PEG formation and structure

PEG, apart from its wide use in cosmetic industry is completely water miscible. Due to its water miscibility PEG is only or sparingly soluble in hydrocarbons. Compatibility of the PEG with a given fluid has to be checked before use. PEG, as facts of its water miscibility will uptake water without separation. In case of the use of PEG in applications within water environment the water ingress should be checked carefully. Effects of water ingress are increasing threat of corrosion and thinning due to the mixture.

#### *8.2.1. Use of PEG*

Water miscibility is of use in non-flammable hydraulics in coal mining industries, but also in applications of pharmacy and food processing. In general PEG is allowed within the FDA regulation to be safe for incidental food contact. Due to the positive effect of sliding especially in worm gears PEG is somehow recommended for use in such applications. [2, 6, 7, 9]

#### **8.3. Polypropylene Glycoles (PPG)**

Polypropylene Glycols are made from Proplyene Oxide by polymerization by use of butoxides leading to a half ether structure (Figure 44):

**8.5. Alcohole Ethoxilates**

**Figure 46.** Alcohole Ethoxilates

O

lations.

**9. Siloxanes**

Si Cl Cl

**Figure 47.** Scheme of Silicone Oil formation

in the side chain (Figure 48).

**Figure 48.** Polymethyl-Aryl Siloxanes

Alcohole Ethoxilates are formed by a cross reaction of epoxides with alcohols. (Figure 46)

The use of long chain alcohols leads to alcohole ethoxilates being used as non-ionic surfactants, emulsifiers and dispersants in multiple applications, e.g. hydraulic oils with dispersant

Due to their non ionic nature alcohole ethoxilates are widely compatible in lubricant formu‐

Siloxanes (Silicone Oils) are common lubricants in multiple applications, such as food and pharmacy, but also in applications where special low friction properties are demanded [2, 3, 8]

> Si O Si

Side groups are methyl, but also phenyl groups leading to polydimethylsiloxanes or polyar‐ ylsiloxanes. Mixtures of methyl and arylsiloxanes are in use with different spreading between

O O

Si

<sup>O</sup> <sup>n</sup>

O O

Si

<sup>O</sup> <sup>n</sup>

In general silicones are the result of alkylchlorosilane hydrolyses [10] (Figure 47).

OH2 <sup>O</sup>

O Si O Si O

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731

OH

31

OH

capability, cutting fluids, and dispersants for applications where sludge is expected.

**Figure 44.** PPG Structure

Due to the additional methyl group in the structure water miscibility drops down (contrast to PEG) while the oil miscibility promotes. Also by choosing longer alkyl chain butoxides, PPG structures may be obtained with enhanced oil solubility.

While PEG is highly dissolved in water, PPG forms droplets immersed in the water. Due to this fact water separation out of PPG is difficult to achieve.

The partial solubility and immersion of PPG in water causes a very high fish toxicity. PPG should never be used in the case of its break out-in free lands or water

#### *8.3.1. Use of PPG*

PPG is commonly used as high temperature circuit oil, e.g. calandars, compressors, high temperature chain lubrication. All over PPG has to be stabilized by acid corrosion inhibitors, e.g. phosphoric partial esters and antioxidants like Phenyl-α-Naphtylamine [2] [3] [5] [8].

#### **8.4. Polybutylene Glycoles (PBG)**

PBG are seldom in use and made consequentially from Butylene Oxide polymerization by use of alkoxides, leading to half esters (Figure 45)

```
Figure 45. PBG Structure
```
#### *8.4.1. Use of PBG*

PBG is useful for enhancing the solubility of additives, boosting the viscosity index of mineral oil variants.

#### **8.5. Alcohole Ethoxilates**

**8.3. Polypropylene Glycoles (PPG)**

30 Tribology - Fundamentals and Advancements

O

3CH

**Figure 44.** PPG Structure

*8.3.1. Use of PPG*

**8.4. Polybutylene Glycoles (PBG)**

O

**Figure 45.** PBG Structure

*8.4.1. Use of PBG*

oil variants.

C2H5 R'OH

of alkoxides, leading to half esters (Figure 45)

leading to a half ether structure (Figure 44):

R'OH

structures may be obtained with enhanced oil solubility.

this fact water separation out of PPG is difficult to achieve.

should never be used in the case of its break out-in free lands or water

Polypropylene Glycols are made from Proplyene Oxide by polymerization by use of butoxides

Due to the additional methyl group in the structure water miscibility drops down (contrast to PEG) while the oil miscibility promotes. Also by choosing longer alkyl chain butoxides, PPG

While PEG is highly dissolved in water, PPG forms droplets immersed in the water. Due to

The partial solubility and immersion of PPG in water causes a very high fish toxicity. PPG

PPG is commonly used as high temperature circuit oil, e.g. calandars, compressors, high temperature chain lubrication. All over PPG has to be stabilized by acid corrosion inhibitors, e.g. phosphoric partial esters and antioxidants like Phenyl-α-Naphtylamine [2] [3] [5] [8].

PBG are seldom in use and made consequentially from Butylene Oxide polymerization by use

PBG is useful for enhancing the solubility of additives, boosting the viscosity index of mineral

O

C2H5

O <sup>O</sup> R'O OH

C2H5

n

C2H5

C2H5

O

CH3

O <sup>O</sup> R'O OH

CH3

CH3

n

CH3

Alcohole Ethoxilates are formed by a cross reaction of epoxides with alcohols. (Figure 46)

**Figure 46.** Alcohole Ethoxilates

The use of long chain alcohols leads to alcohole ethoxilates being used as non-ionic surfactants, emulsifiers and dispersants in multiple applications, e.g. hydraulic oils with dispersant capability, cutting fluids, and dispersants for applications where sludge is expected.

Due to their non ionic nature alcohole ethoxilates are widely compatible in lubricant formu‐ lations.

#### **9. Siloxanes**

Siloxanes (Silicone Oils) are common lubricants in multiple applications, such as food and pharmacy, but also in applications where special low friction properties are demanded [2, 3, 8]

In general silicones are the result of alkylchlorosilane hydrolyses [10] (Figure 47).

**Figure 47.** Scheme of Silicone Oil formation

Side groups are methyl, but also phenyl groups leading to polydimethylsiloxanes or polyar‐ ylsiloxanes. Mixtures of methyl and arylsiloxanes are in use with different spreading between in the side chain (Figure 48).

**Figure 48.** Polymethyl-Aryl Siloxanes

Siloxanes with different structures are not generally miscible amongst each other. Miscibility has to be checked carefully. As mentioned, siloxanes are widely used in lubrication technology due their exceptional properties concerning low friction capability, high temperature stability and low toxicity in various applications. Prominent applications are starter components in cars, valves in food industry, slow speed bearings and high temperature applications where arylsiloxanes are in use. Siloxanes creep widely across surfaces and may cause problems in coatings, lacquering and paintings.

High temperature combustion of PFPE may cause the emission of hydrogen fluoride and fluoro phosgene which makes PFPE formulations somehow corrosive, especially on steel alloy compositions. Due to this fact, the high temperature corrosiveness should be carefully taken

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 33

Additives in lubricants enhance base oil functionalities. Additive technology is in broad scale based on organic chemistry syntheses. From their origin they are found by chance, less than by a real scientific approach. Nevertheless, literature about their reactions is innumerous from

Modern additive technology commenced in the early 20th century and has progressed contin‐ uously due to advanced organic chemistry syntheses. Upcoming modern spectrometry has

Beyond the basic and industrial reaction mechanism studies, mixtures of additives have been studied extensively by industry and science over the years. Such studies reveal the mechanisms of compatibility and incompatibility of additives acting together at a given application. [9]

For example a functional mismatch is caused by diverse demanding addressed to additives, e.g.: additives acting against corrosion may interfere with additives that have to prevent metal

Modern additive technology is inevitable to reach the "for-life" goal of modern technologies. As "for-life" might be understood in a different manner by users, additive packages are developed during the decades adapted to a given customer demanding. For example, the demanding to get automatic transmission gear oil performances is achieved by additive packages that may not fit for wind turbine or paper mill applications. Hence, additives and

In general there are no rules up to now to predict additive performances at a given technical application. As a consequence formulations have to be tested in forecast extensively to assure

Additives may cover a distinct structure-property relationship. Since there are no scientific rules declaring on how a chemical structure of an additive causes a function all variations in

Additive technologies have been revised many times during their history, either due to a change in demanding or due to their toxicity. Toxicity is a severe problem in additive tech‐

Since the validation of those different chemical additive structures causes tremendous costs, it is a fact, that additive free technologies or additive technologies with marginal content level

its functionality. Such testing is addressed by international and national regulations.

nology, since no one knows their real long term biological and ecological effects. [9]

their mixtures have to be selected carefully for each purpose. [2, 3, 8, 9]

been used to clarify their structures and their reaction at different metal sites. [2, 3, 9]

into account in the case of PFPE use.

**11. Additives**

the very beginning. [2, 3, 9]

surfaces against fretting or welding.

additives have to be validated by tests.

are favored as future solutions.

### **10. Polyfluorinated Polyether (PFPE) base oil**

PFPE Base Oil is created by polymerization of Perfluoroepoxids. Structure of PFPE is similar to polyglycoles but with overall substitution of hydrogen by fluorine [2] [3] (Figure 49)

**Figure 49.** PFPE Base Oil

Due to the effective shielding of the C-O-C backbone in the structure of PFPE by the trifluor‐ omethyl side chain group PFPE are completely insoluble in water, inert toward alkaline and acids and even oxygen.

PFPE Base oil is used for high temperature purposes and in the presence of aggressive media, mentioned above in junction with PTFE thickener. (Figure 50)

**Figure 50.** PTFE as thickener for PFPE

PFPE sparingly adheres to metal surfaces due to droplet formation. The low adhesion causes creeping across surfaces and mal-lubrication if the surfaces are not cleaned thor‐ oughly. Creeping and low adhesion may cause low friction in certain applications. PFPE is insoluble in most of the common base oils. Use of PFPE hence is restricted to the fluo‐ rine group of base oil.

Inertness of PFPE and PTFE make greases suitable for incidental food contact lubrication.

High temperature combustion of PFPE may cause the emission of hydrogen fluoride and fluoro phosgene which makes PFPE formulations somehow corrosive, especially on steel alloy compositions. Due to this fact, the high temperature corrosiveness should be carefully taken into account in the case of PFPE use.

#### **11. Additives**

Siloxanes with different structures are not generally miscible amongst each other. Miscibility has to be checked carefully. As mentioned, siloxanes are widely used in lubrication technology due their exceptional properties concerning low friction capability, high temperature stability and low toxicity in various applications. Prominent applications are starter components in cars, valves in food industry, slow speed bearings and high temperature applications where arylsiloxanes are in use. Siloxanes creep widely across surfaces and may cause problems in

PFPE Base Oil is created by polymerization of Perfluoroepoxids. Structure of PFPE is similar to polyglycoles but with overall substitution of hydrogen by fluorine [2] [3] (Figure 49)

CF2 O

Due to the effective shielding of the C-O-C backbone in the structure of PFPE by the trifluor‐ omethyl side chain group PFPE are completely insoluble in water, inert toward alkaline and

PFPE Base oil is used for high temperature purposes and in the presence of aggressive media,

PFPE sparingly adheres to metal surfaces due to droplet formation. The low adhesion causes creeping across surfaces and mal-lubrication if the surfaces are not cleaned thor‐ oughly. Creeping and low adhesion may cause low friction in certain applications. PFPE is insoluble in most of the common base oils. Use of PFPE hence is restricted to the fluo‐

Inertness of PFPE and PTFE make greases suitable for incidental food contact lubrication.

CF2

F

CF3

CF2 O

F

CF3

CF2

\*

CF2

n

n

CF2 O

F

CF3

coatings, lacquering and paintings.

32 Tribology - Fundamentals and Advancements

**Figure 49.** PFPE Base Oil

acids and even oxygen.

**Figure 50.** PTFE as thickener for PFPE

rine group of base oil.

**10. Polyfluorinated Polyether (PFPE) base oil**

\* O

CF3

F

mentioned above in junction with PTFE thickener. (Figure 50)

CF2

CF2

Additives in lubricants enhance base oil functionalities. Additive technology is in broad scale based on organic chemistry syntheses. From their origin they are found by chance, less than by a real scientific approach. Nevertheless, literature about their reactions is innumerous from the very beginning. [2, 3, 9]

Modern additive technology commenced in the early 20th century and has progressed contin‐ uously due to advanced organic chemistry syntheses. Upcoming modern spectrometry has been used to clarify their structures and their reaction at different metal sites. [2, 3, 9]

Beyond the basic and industrial reaction mechanism studies, mixtures of additives have been studied extensively by industry and science over the years. Such studies reveal the mechanisms of compatibility and incompatibility of additives acting together at a given application. [9]

For example a functional mismatch is caused by diverse demanding addressed to additives, e.g.: additives acting against corrosion may interfere with additives that have to prevent metal surfaces against fretting or welding.

Modern additive technology is inevitable to reach the "for-life" goal of modern technologies. As "for-life" might be understood in a different manner by users, additive packages are developed during the decades adapted to a given customer demanding. For example, the demanding to get automatic transmission gear oil performances is achieved by additive packages that may not fit for wind turbine or paper mill applications. Hence, additives and their mixtures have to be selected carefully for each purpose. [2, 3, 8, 9]

In general there are no rules up to now to predict additive performances at a given technical application. As a consequence formulations have to be tested in forecast extensively to assure its functionality. Such testing is addressed by international and national regulations.

Additives may cover a distinct structure-property relationship. Since there are no scientific rules declaring on how a chemical structure of an additive causes a function all variations in additives have to be validated by tests.

Additive technologies have been revised many times during their history, either due to a change in demanding or due to their toxicity. Toxicity is a severe problem in additive tech‐ nology, since no one knows their real long term biological and ecological effects. [9]

Since the validation of those different chemical additive structures causes tremendous costs, it is a fact, that additive free technologies or additive technologies with marginal content level are favored as future solutions.

The following chapter addresses additive technologies concerning extreme pressure, anti-wear functions and also corrosion-protecting and antioxidants.

#### **11.1. Extreme Pressure (EP) and Anti-Wear (AW) additives**

#### *11.1.1. General*

Extreme Pressure (EP) and Anti-Wear(AW) Additives are functional chemicals in lubricants with the task to separate metal surfaces in the case of heavy loading and to improve their resistance toward wear in the case of oil film break in the contact [9].

Machinery elements that start to run or stop due to emergency show pronounced loading due to a lack of lubrication, e.g. the oil does not separate the metal surfaces and the protection of the oil film drops down. At that point EP and AW additives are supposed to jump into the arena by causing reaction layers preventing the metal from direct rupture or welding.

Dialkyl-Thiophosphoricacidesters are prominent representatives in sulfur additive chemistry.

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 35

Zincdithiophosphate (ZndtP) represent a prominent group of EP/AW additives. They derive from the neutralization of Thiophosphoric Acids, obtained by ring opening of Phosphorous pentasulfide with alcohols, with Zn-Carbonate or Hydroxides. As a fact, the ZndtP differ strongly by their carbon-chain length. A couple of variants are achieved by choosing different alcohols in the ring opening sequence of Phosphorous (V) sulfide. From the structural perspective, ZndtP may be regarded as chelate complexes rather than a salt (Figure 53).

Molybdenumdithiophosphates contains a Molybdenum [µ-oxo] Core, distinct compared to

Similar to Dithiophosphates, Chelat Complexes from Zinc, Molybdenum but also Bismuth and others may be formed by reaction of Thiocarbamic Acid with the metal precursors. Dithiocar‐ bamic Acid is synthesized via addition of amines to Carbondisulfide. By varying the chain length of the amine different dithiocarbamates are achieved (Figure 55 and Figure 56):

(Figure 52).

**Figure 52.** Thiophosphoric Acid Ester

**Figure 51.** Didodecyltrisulfide as polysulfide representative

**Zinc- and Molybdenum dithio phosphates (ZndtP- ModtP)**

*11.1.3. Dithiophosphates*

ZndtP (Figure 54).

*11.1.4. Dithiocarbamates*

Their chemical structures are found by chance. For example observations during drilling and maching show that tools perform better if lubrication is carried out by use of sulfurized oils derived from vegetables, mixed and heated with sulfur.

Later on intense research the nature and reaction started including modern surface spectrom‐ etry techniques. The transformation of EP/AW additives as a function of the nature of the surfaces, their loading, contact geometry, temperature and their structure shows a clear picture of structure-activity relationship. Also additives perform as a function of their chemical structure, but also as a function of their solubility in base oil and as a function of other additives being present. In that sense, it is shown that additives either may prolong service life but are also capable to shrink life.

#### *11.1.2. Sulfur additives*

Sulfur acts as a powerful extreme pressure additive. The high reactivity, especially toward copper makes it unlike to use sulfur as element in tribology.

Sulfur embedded in organic framework acts as a powerful Extreme Pressure additive. Choosing appropriate organic structures the activity toward copper drops down. However, using sulfurized additives copper deactivation should be present anyhow.

Sulfur is added either by reaction of reactive organic precursors like alkenes and their derivatives by heating up with the element, or by polymerization sequences with activated sulfur precursors such as di-sulfur dichloride. Doing so, all kinds of unsaturated specie gives reaction products leading to sulfurized specie. Prominent representatives are reaction products of Isobutene with Disulfur Dichloride, or reaction products with terpenes (Figure 51) but also unsaturated carboxylic acid esters, like rapseed oil:

Sulfurized Additives (S-Additives) are often used together with phosphoric acid esters, since the synergistic between those additives are known from the past. Doing so, gear oils may contain S-Additives with amine phosphate esters. Also extreme pressure additives containing

**Figure 51.** Didodecyltrisulfide as polysulfide representative

Dialkyl-Thiophosphoricacidesters are prominent representatives in sulfur additive chemistry. (Figure 52).

**Figure 52.** Thiophosphoric Acid Ester

#### *11.1.3. Dithiophosphates*

The following chapter addresses additive technologies concerning extreme pressure, anti-wear

Extreme Pressure (EP) and Anti-Wear(AW) Additives are functional chemicals in lubricants with the task to separate metal surfaces in the case of heavy loading and to improve their

Machinery elements that start to run or stop due to emergency show pronounced loading due to a lack of lubrication, e.g. the oil does not separate the metal surfaces and the protection of the oil film drops down. At that point EP and AW additives are supposed to jump into the arena by causing reaction layers preventing the metal from direct rupture or welding.

Their chemical structures are found by chance. For example observations during drilling and maching show that tools perform better if lubrication is carried out by use of sulfurized oils

Later on intense research the nature and reaction started including modern surface spectrom‐ etry techniques. The transformation of EP/AW additives as a function of the nature of the surfaces, their loading, contact geometry, temperature and their structure shows a clear picture of structure-activity relationship. Also additives perform as a function of their chemical structure, but also as a function of their solubility in base oil and as a function of other additives being present. In that sense, it is shown that additives either may prolong service life but are

Sulfur acts as a powerful extreme pressure additive. The high reactivity, especially toward

Sulfur embedded in organic framework acts as a powerful Extreme Pressure additive. Choosing appropriate organic structures the activity toward copper drops down. However,

Sulfur is added either by reaction of reactive organic precursors like alkenes and their derivatives by heating up with the element, or by polymerization sequences with activated sulfur precursors such as di-sulfur dichloride. Doing so, all kinds of unsaturated specie gives reaction products leading to sulfurized specie. Prominent representatives are reaction products of Isobutene with Disulfur Dichloride, or reaction products with terpenes (Figure

Sulfurized Additives (S-Additives) are often used together with phosphoric acid esters, since the synergistic between those additives are known from the past. Doing so, gear oils may contain S-Additives with amine phosphate esters. Also extreme pressure additives containing

functions and also corrosion-protecting and antioxidants.

**11.1. Extreme Pressure (EP) and Anti-Wear (AW) additives**

derived from vegetables, mixed and heated with sulfur.

copper makes it unlike to use sulfur as element in tribology.

51) but also unsaturated carboxylic acid esters, like rapseed oil:

using sulfurized additives copper deactivation should be present anyhow.

resistance toward wear in the case of oil film break in the contact [9].

*11.1.1. General*

34 Tribology - Fundamentals and Advancements

also capable to shrink life.

*11.1.2. Sulfur additives*

#### **Zinc- and Molybdenum dithio phosphates (ZndtP- ModtP)**

Zincdithiophosphate (ZndtP) represent a prominent group of EP/AW additives. They derive from the neutralization of Thiophosphoric Acids, obtained by ring opening of Phosphorous pentasulfide with alcohols, with Zn-Carbonate or Hydroxides. As a fact, the ZndtP differ strongly by their carbon-chain length. A couple of variants are achieved by choosing different alcohols in the ring opening sequence of Phosphorous (V) sulfide. From the structural perspective, ZndtP may be regarded as chelate complexes rather than a salt (Figure 53).

Molybdenumdithiophosphates contains a Molybdenum [µ-oxo] Core, distinct compared to ZndtP (Figure 54).

#### *11.1.4. Dithiocarbamates*

Similar to Dithiophosphates, Chelat Complexes from Zinc, Molybdenum but also Bismuth and others may be formed by reaction of Thiocarbamic Acid with the metal precursors. Dithiocar‐ bamic Acid is synthesized via addition of amines to Carbondisulfide. By varying the chain length of the amine different dithiocarbamates are achieved (Figure 55 and Figure 56):

**Figure 53.** ZndtP from neutralization of Thiophosphoric Acid with Zn Carbonate

**Figure 54.** Molybdenumdithiophosphate

#### **11.2. Corrosion protection**

#### *11.2.1. General*

Within this chapter only iron as a chief element in technical application is considered.

Generally metal surfaces tend to corrosion if water, oxygen and probably salts, like sodium chloride are present. Corrosion may take place either by cathodic reduction of oxygen or by anodic oxidation of the metal. Charges, either positive (anode) or negative (cathode) pass the

Mo S O

S

ZndtC

S

<sup>H</sup> <sup>N</sup>

S

S C Zn C

ModtC

O

<sup>H</sup> <sup>N</sup>

C C

Mo O

S

S

H

H

S O

N

N

**Figure 56.** Zincdithiocarbamate (ZndtC) and Molybdenumdithiocarbamate (ModtC)

surface layer. [2, 3, 9]

S=C=S

+

**Figure 55.** Syntheses of dithiocarbamates

NH2

Metal-Carbonate

NH S S

Metal

<sup>S</sup> <sup>S</sup>

N H N H

NH

S S

S S R

N H

R-CHO

S

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 37

SH

**Figure 55.** Syntheses of dithiocarbamates

**11.2. Corrosion protection**

**Figure 54.** Molybdenumdithiophosphate

Within this chapter only iron as a chief element in technical application is considered.

Mo S O

<sup>O</sup> <sup>O</sup>

P Mo

O O

<sup>S</sup> <sup>O</sup>

O

S

Generally metal surfaces tend to corrosion if water, oxygen and probably salts, like sodium chloride are present. Corrosion may take place either by cathodic reduction of oxygen or by

*11.2.1. General*

S

S

P

S

P

S

S

P

S

36 Tribology - Fundamentals and Advancements

S

P

Phosphorous (V) Sulfide

Zn Carbonate

S

**Figure 53.** ZndtP from neutralization of Thiophosphoric Acid with Zn Carbonate

OH

Various alcohols

+

P SH

Thiophosphoric Acid

O

O

<sup>S</sup> <sup>O</sup>

O

S

S

ZndtP

O P S

O

O

P P

S

S Zn

S

ModtC

**Figure 56.** Zincdithiocarbamate (ZndtC) and Molybdenumdithiocarbamate (ModtC)

anodic oxidation of the metal. Charges, either positive (anode) or negative (cathode) pass the surface layer. [2, 3, 9]

Charge transport from the metal toward the outer region is hindered by the surface potential (over potential). Thus, corrosion processes have to overcome this potential and start after a certain induction period. Once, if this potential has been overcome the corrosion starts without hindrances by successive material transport. Materials transport ends up in a drastic change of the surface, mainly accompanied by a loss.

may be neutralized. In technical applications mainly alkyl benze sulfonic acids and dodecyl‐ sulfonic acid are neutralized. Production starts from alkenes out of petrol chemistry by

Neutralization with either sodiumhydroxide, Calciumcarbonate, Magnesiumcarbonate or Bariumcarbonate leads to sulfonates: A = Sodiumsulfonate, B = Calciumsulfonate and with excess Carbonate to over based Calciumsulfonate (B'), Magnesiumsulfonate (C) and Barium‐

Dodecylbenzene

Dodecylbenezesulfonic Acid

BaCO3

**D**

SO3-

SO3-

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 39

SO3- Ba++

SO3- Mg++

addition sulfuric acid or SO3.

sulfonate (D) (Figure 57).

+

Catalysts

SO3/H2SO4

**B**

**B'**

**Figure 57.** Sulfonic Acids and their Salts

SO3- Na<sup>+</sup>

**A**

HO3S

CaCO3

NaOH

SO3-

MgCO3

**C**

SO3-

Ca++ X

SO3- Ca++

X = Carbonate, Hydroxide

For iron as metal, the transport of the metal ends up in a flaky layer (Rust) that permits water and oxygen to penetrate. Due to this effect the rust process ends up in a total damage of the metal, especially in an environment that boosts corrosive processes.

Counteracting corrosion, the initial processes of charge transportation have to be blocked. Doing so, the over potential, e.g. the natural barrier of charges passing the surface has to be increased by creating additional layers on the metal surface (Passivation) or by creation of stable, insoluble complexes, formed by interaction of the surface atoms with a complex builder.

Passivation of iron surfaces and enhancing the over potential is achieved by deposition of chromium layers that cause a thin and gas-dense closed layer on the metal. Thus, chromium is a powerful inhibitor toward corrosion processes. As the charge transport phenomena occurring on iron surfaces are cathodic or anodic and vice versa, this process could also be stopped by offering an anodic victim like a zinc coating.

Additives that create a corrosion protection are in general dissolved in a carrier base-oil that spreads over the surface. Due to their adapted functional groups a physical binding toward the surface starts to create a layer. In order to create an appropriate corrosion protection this layer has to be packed dense to avoid the penetration of water and oxygen. This is realized by strong dipolar groups and oil soluble tails with a marginal demand in lateral spacing, e.g. long, - unbranched alkyl chains.

Else, passivation also is achieved by placing insoluble complex builders onto the iron surface, like phosphates are. Iron phosphate builds up a close dense insoluble layer on the surface.

Restriction of iron phosphate is indicated by the fact that, under certain conditions, phosphates start to get reduced forming posphanes. Phosph4anes strongly affect metals due to segregation of phosphorous at grain boundaries and releasing hydrogen into the metal. Hydrogen is detrimental to the microstructure by inducing, e.g. hydrogen enhanced local plasticit (HELP) or hydrogen induced cracking (HIC). Presence of phosphanes by reduction of phosphates takes place in acidic and reducing environment, e.g. presence of hydrogen sulfide, chlorides and others.

The following chapter will show some of the most prominent representatives of corrosion protectors.

#### *11.2.2. Sulfonate-chemistry*

#### **General**

Sulfonates derive from sulfonic acids by neutralization with alkali, earth alkali –metals but also with metals from the transition group, for example zinc. Principally each sulfonic acid may be neutralized. In technical applications mainly alkyl benze sulfonic acids and dodecyl‐ sulfonic acid are neutralized. Production starts from alkenes out of petrol chemistry by addition sulfuric acid or SO3.

Neutralization with either sodiumhydroxide, Calciumcarbonate, Magnesiumcarbonate or Bariumcarbonate leads to sulfonates: A = Sodiumsulfonate, B = Calciumsulfonate and with excess Carbonate to over based Calciumsulfonate (B'), Magnesiumsulfonate (C) and Barium‐ sulfonate (D) (Figure 57).

X = Carbonate, Hydroxide

**Figure 57.** Sulfonic Acids and their Salts

Charge transport from the metal toward the outer region is hindered by the surface potential (over potential). Thus, corrosion processes have to overcome this potential and start after a certain induction period. Once, if this potential has been overcome the corrosion starts without hindrances by successive material transport. Materials transport ends up in a drastic change

For iron as metal, the transport of the metal ends up in a flaky layer (Rust) that permits water and oxygen to penetrate. Due to this effect the rust process ends up in a total damage of the

Counteracting corrosion, the initial processes of charge transportation have to be blocked. Doing so, the over potential, e.g. the natural barrier of charges passing the surface has to be increased by creating additional layers on the metal surface (Passivation) or by creation of stable, insoluble complexes, formed by interaction of the surface atoms with a complex builder.

Passivation of iron surfaces and enhancing the over potential is achieved by deposition of chromium layers that cause a thin and gas-dense closed layer on the metal. Thus, chromium is a powerful inhibitor toward corrosion processes. As the charge transport phenomena occurring on iron surfaces are cathodic or anodic and vice versa, this process could also be

Additives that create a corrosion protection are in general dissolved in a carrier base-oil that spreads over the surface. Due to their adapted functional groups a physical binding toward the surface starts to create a layer. In order to create an appropriate corrosion protection this layer has to be packed dense to avoid the penetration of water and oxygen. This is realized by strong dipolar groups and oil soluble tails with a marginal demand in lateral spacing, e.g. long,

Else, passivation also is achieved by placing insoluble complex builders onto the iron surface, like phosphates are. Iron phosphate builds up a close dense insoluble layer on the surface.

Restriction of iron phosphate is indicated by the fact that, under certain conditions, phosphates start to get reduced forming posphanes. Phosph4anes strongly affect metals due to segregation of phosphorous at grain boundaries and releasing hydrogen into the metal. Hydrogen is detrimental to the microstructure by inducing, e.g. hydrogen enhanced local plasticit (HELP) or hydrogen induced cracking (HIC). Presence of phosphanes by reduction of phosphates takes place in acidic and reducing environment, e.g. presence of hydrogen sulfide, chlorides and

The following chapter will show some of the most prominent representatives of corrosion

Sulfonates derive from sulfonic acids by neutralization with alkali, earth alkali –metals but also with metals from the transition group, for example zinc. Principally each sulfonic acid

of the surface, mainly accompanied by a loss.

38 Tribology - Fundamentals and Advancements

metal, especially in an environment that boosts corrosive processes.

stopped by offering an anodic victim like a zinc coating.


others.

protectors.

**General**

*11.2.2. Sulfonate-chemistry*

#### *11.2.3. Carboxylic acids and derivatives*

Carboxylic Acid and their derivatives, e.g. esters may act as metal corrosion protectors. While carboxylic acids are supposed to cause corrosion, some of them prevent. Rust preventing carboxylic acids are derivatives from α-Aminoacids, like N-Oleylglycine. (Figure 58)

**Figure 58.** N-Oleylglycine

N-Oleylgylcine acts as powerful emulsifier, even at low dosage. Rust protecting is due to the spread of water in the formulation over a big volume. N-Oleylglycine, even at low percentages also counteracts with EP/AW additives, driving their activity down.

Carboxylic Acids, derived from Phenoles such as Nonyl-phenoxiaceticacid is a non emulsify‐ ing corrosion protector but under prohibition, due to its irritating effects (Figure 59).

Nonyl Phenoxy Acetic Acid

**Figure 59.** Nonylphenoxyaceticacid

Succinic Acid Derivatives, such as Succinic Half Ester of Octanole are powerful metal protec‐ tors, but also strong counteracting with EP/AW additives. Synthesis is carried out by reacting succinic acid anhydride with alcoholes (Figure 60).

Carboxylates, derived from neutralizing carboxylic acids with transition metals like Zinc, Lead, Bismuth lead to corrosion protection. Common acids are Napthenic acids or medium chain carboxylic acids like octanoic acid (Figure 61).

O

**Figure 62.** Amine Phosphate Structure

O

+

n COOH

**Figure 61.** Zn (Bi) Carboxylates (Napthenate and Octoate)

OH

O

O

O

ZnCO3

Zinc Naphtenates

O

**Figure 60.** Succinic Half Ester

O

O <sup>P</sup> <sup>O</sup> NH O 2 +

Amine Phosphate

COO-

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 41

Zn++

COO- <sup>n</sup>

OH Succinic Half Esters

#### *11.2.4. Amine phosphate esters*

Amine Phosphate Esters may act as anti-corrosion additives in addition to their anti-wear properties. Due to their synergistic properties and due to the fact, that certain amine phos‐ phates are allowed as additives for incidental food contact, they are often found in all kind of lubricants (Figure 62).

**Figure 60.** Succinic Half Ester

*11.2.3. Carboxylic acids and derivatives*

40 Tribology - Fundamentals and Advancements

**Figure 58.** N-Oleylglycine

**Figure 59.** Nonylphenoxyaceticacid

*11.2.4. Amine phosphate esters*

lubricants (Figure 62).

succinic acid anhydride with alcoholes (Figure 60).

chain carboxylic acids like octanoic acid (Figure 61).

Carboxylic Acid and their derivatives, e.g. esters may act as metal corrosion protectors. While carboxylic acids are supposed to cause corrosion, some of them prevent. Rust preventing

<sup>H</sup> N-Oleyl Glycine

N-Oleylgylcine acts as powerful emulsifier, even at low dosage. Rust protecting is due to the spread of water in the formulation over a big volume. N-Oleylglycine, even at low percentages

Carboxylic Acids, derived from Phenoles such as Nonyl-phenoxiaceticacid is a non emulsify‐

Nonyl Phenoxy Acetic Acid

Succinic Acid Derivatives, such as Succinic Half Ester of Octanole are powerful metal protec‐ tors, but also strong counteracting with EP/AW additives. Synthesis is carried out by reacting

Carboxylates, derived from neutralizing carboxylic acids with transition metals like Zinc, Lead, Bismuth lead to corrosion protection. Common acids are Napthenic acids or medium

Amine Phosphate Esters may act as anti-corrosion additives in addition to their anti-wear properties. Due to their synergistic properties and due to the fact, that certain amine phos‐ phates are allowed as additives for incidental food contact, they are often found in all kind of

O CH2 COOH

ing corrosion protector but under prohibition, due to its irritating effects (Figure 59).

also counteracts with EP/AW additives, driving their activity down.

COOH <sup>N</sup>

carboxylic acids are derivatives from α-Aminoacids, like N-Oleylglycine. (Figure 58)

Zinc Naphtenates

**Figure 62.** Amine Phosphate Structure

Amine Phosphates are powerful activators of copper and zinc and cause leaching of those metals from brass cages in bearings. Adding Amine Phosphates copper deactivators like benzotriazoles have to be present (Figure 63).

**Figure 63.** Benzotriazole and N-Alkylbenzotriazoles as Cooper Passivators

#### **11.3. Antioxidants (AO)**

AO prevent lubricants from oxygen attack. Oxygen is, by nature, a diradical that undergoes several transitions. Electron uptake from metal surfaces by a cathodic transfer, leads to varieties of activated oxygen specie, powerful attacking hydrocarbon sites by abstraction of hydrogen, leading to peroxides, and carbon radicals. The carbon radical itself starts to stabilize by abstraction of hydrogen leaving an alkene as new product [10]. (Figure 64)

Due to radical stabilization the new formed alkene starts to continue the oxidation by sequen‐ tial abstraction of hydrogen, forming di-, tri- and polyalkenes, but also benzene rings. Apart from the hydrogen abstraction, also oxidation takes place by attacking carbon radicals by oxygen. At least the products created by such this procedures are carbonyl compounds, e.g. alcohols, ketones, aldehydes, carboxylic acids and sometimes esters. PAO oxidation at metal surfaces, e.g. iron beyond 120°C results in the formation of lactones (esters that come up by internal reaction between an alcohol group and terminal carboxylic group) (Figure 65).

Hence oxidation sequences dramatically change the original hydrocarbon chain. If once started it is self accelerating till new, different and stable products are reached. Oxidation is unselective and takes place everywhere in the chain. Hence, plenty of products are formed by radical oxygen assisted processes.

The AO radical subsequently stabilizes to form new products like quinones. The quinone structure may form a dark colored charge transfer complex with the original antioxidant. Very often this causes strong discoloration of AO stabilized lubricants since the charge transfer complexes are very intense in color. Sometimes, for example in the case of poly‐ urea greases, such charge transfer complexes may interfere with the grease structure in

**.O-O.**

.O-O-Reactive O-intermediate

Creates Alkenes

Tri-Ene Repetitive Sequnece Poly-Enes

**.**

n

Graphite like Structures

**Figure 64.** Oxygen – Hydrocarbon Attack sequence

Allyl Radical

creates **.**

abstracts hydrogen

Hydrocarbon Radical

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 43

new attack

Di-Ene

electron transfer

attacks

Hydrocarbon

abstracts hydrogen

Persistent radicals formed by AO are dangerous in some cases. In the case of their accumulation in the system they are able to boost oxidation rather than to prevent. Dosage of AO hence

terms of solidification.

Metal

Antioxidants in general prevent the base oil, quenching the oxygen attack by formation of stable radicals. Stabilization of the radicals is realized by a delocalization of the persistent AO radical, created by oxygen attack due to the presence of aromats in the structure (Figure 66).

Graphite like Structures

**Figure 64.** Oxygen – Hydrocarbon Attack sequence

Amine Phosphates are powerful activators of copper and zinc and cause leaching of those metals from brass cages in bearings. Adding Amine Phosphates copper deactivators like

> N N N

AO prevent lubricants from oxygen attack. Oxygen is, by nature, a diradical that undergoes several transitions. Electron uptake from metal surfaces by a cathodic transfer, leads to varieties of activated oxygen specie, powerful attacking hydrocarbon sites by abstraction of hydrogen, leading to peroxides, and carbon radicals. The carbon radical itself starts to stabilize

Due to radical stabilization the new formed alkene starts to continue the oxidation by sequen‐ tial abstraction of hydrogen, forming di-, tri- and polyalkenes, but also benzene rings. Apart from the hydrogen abstraction, also oxidation takes place by attacking carbon radicals by oxygen. At least the products created by such this procedures are carbonyl compounds, e.g. alcohols, ketones, aldehydes, carboxylic acids and sometimes esters. PAO oxidation at metal surfaces, e.g. iron beyond 120°C results in the formation of lactones (esters that come up by internal reaction between an alcohol group and terminal carboxylic group) (Figure 65).

Hence oxidation sequences dramatically change the original hydrocarbon chain. If once started it is self accelerating till new, different and stable products are reached. Oxidation is unselective and takes place everywhere in the chain. Hence, plenty of products are formed by radical

Antioxidants in general prevent the base oil, quenching the oxygen attack by formation of stable radicals. Stabilization of the radicals is realized by a delocalization of the persistent AO radical, created by oxygen attack due to the presence of aromats in the structure (Figure 66).

by abstraction of hydrogen leaving an alkene as new product [10]. (Figure 64)

N-Alkylated Benzotriazoles

benzotriazoles have to be present (Figure 63).

**Figure 63.** Benzotriazole and N-Alkylbenzotriazoles as Cooper Passivators

N H

42 Tribology - Fundamentals and Advancements

Benzotriazole

**11.3. Antioxidants (AO)**

oxygen assisted processes.

N N

> The AO radical subsequently stabilizes to form new products like quinones. The quinone structure may form a dark colored charge transfer complex with the original antioxidant. Very often this causes strong discoloration of AO stabilized lubricants since the charge transfer complexes are very intense in color. Sometimes, for example in the case of poly‐ urea greases, such charge transfer complexes may interfere with the grease structure in terms of solidification.

> Persistent radicals formed by AO are dangerous in some cases. In the case of their accumulation in the system they are able to boost oxidation rather than to prevent. Dosage of AO hence

**Figure 65.** Oxidation sequence of Hydrocarbons toward carbonyl compounds

should be carefully tested. Formation of either charge transfer complexes or oxidation products by the presence of AO may cause increased formation of sludge in the lubricant if the dosage balance is not appropriate.

stirring leads to a raw material where amorphous and crystalline structures are merged. The amount of crystals and amorphous materials depends on the nature of the raw materials on the one side and on the rate of heating and cooling on the other side. Rapid cooling causes homogeneous and amorphous structure, as particles are not able to grow to a large size. The raw grease, as effect of the mixture of solid structures has to be homogenized carefully. Homogenization leads to a smoothened appearance of the grease with a scale distribution of thickener particles as effect of the cooling process. Slow cooling generally leads to material with large sized particles as an effect of nucleation and crystal growth. Oil embedding in such structures is different due to the solid structure of the thickener. Stiffness and flowing capability may change as an effect of the merged structure. Greases, even in the case of identical chemical composition may differ significantly by their manufacturing process. Stiffness of

NH

**B**

**Figure 67.** Structures of AO: (A): BHT, (B) Alkyldiphenylamine, (C) PAN

Octyldiphenylamine

N

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 45

Formation of persistant radicals due to delocalisation

NH

Quinone Structure

**C**

PAN

Phenyl--Naphtylamine

H **.**

H **.**

N=

Activated O

N= O

N H

**Figure 66.** Principal delocalization of radicals created by oxygen attack.

OH

Butylhydroxitoluene

**A**

BHT

Nearly all AO contain aromats as a base principle. Prominent AO candidates are Butylhy‐ droxitoluene (BHT) (A), Alkyldiphenylamine ), Phenyl-α-Naphtylamine (PAN) (C) and various others (Figure 67).

#### **12. Greases**

#### **12.1. General remarks**

Greases are defined apart from their chemical composition by the manufacturing processes. Thickener and oil, getting heated by stirring, start to dissolve. Getting cold, the process of

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 45

**Figure 66.** Principal delocalization of radicals created by oxygen attack.

**Figure 67.** Structures of AO: (A): BHT, (B) Alkyldiphenylamine, (C) PAN

should be carefully tested. Formation of either charge transfer complexes or oxidation products by the presence of AO may cause increased formation of sludge in the lubricant if the dosage

O

Nearly all AO contain aromats as a base principle. Prominent AO candidates are Butylhy‐ droxitoluene (BHT) (A), Alkyldiphenylamine ), Phenyl-α-Naphtylamine (PAN) (C) and

Greases are defined apart from their chemical composition by the manufacturing processes. Thickener and oil, getting heated by stirring, start to dissolve. Getting cold, the process of

balance is not appropriate.

O

H

O O

**.**

44 Tribology - Fundamentals and Advancements

O-OH

Peroxides

Alcoholes Ketones

Lactones

**Figure 65.** Oxidation sequence of Hydrocarbons toward carbonyl compounds

Hydrocarbon Radical

activated oxygen

**.** O-O

<sup>H</sup> <sup>O</sup>

Aldehydes

various others (Figure 67).

**12.1. General remarks**

**12. Greases**

stirring leads to a raw material where amorphous and crystalline structures are merged. The amount of crystals and amorphous materials depends on the nature of the raw materials on the one side and on the rate of heating and cooling on the other side. Rapid cooling causes homogeneous and amorphous structure, as particles are not able to grow to a large size. The raw grease, as effect of the mixture of solid structures has to be homogenized carefully. Homogenization leads to a smoothened appearance of the grease with a scale distribution of thickener particles as effect of the cooling process. Slow cooling generally leads to material with large sized particles as an effect of nucleation and crystal growth. Oil embedding in such structures is different due to the solid structure of the thickener. Stiffness and flowing capability may change as an effect of the merged structure. Greases, even in the case of identical chemical composition may differ significantly by their manufacturing process. Stiffness of greases is defined by the NLGI grade declaration, measured by penetration of standard cone into the grease. The deeper it's penetration the more liquid the grease will be. To get a constant value, the grease is worked by 60 strokes, then tempered to 25 °C and measured by cone penetration. NLGI grades are presented in table 2: [3]

3CH

3CH

Li+ O-

**Figure 68.** Prominent representatives of thickeners for grease production

with dicarboxylic acids like Acelaic or sebacaic acid.

Urea Greases are often called PU-Greases in technical language.

Urea structures are realized by adding amines to isocyanates (Figure 69):

**12.3. Di and Polyurea greases (PU-greases)**

hydoxistearate as thickeners.

**Figure 69.** Urea Formation

O

COO- Li+

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 47

OH

Lithiumstearate

Lithium-12-hydroxystearate

3CH COO- Ca++ -OOC CH3

Thickeners are all substances where gelling in the base oil is achievable. Prominent represen‐ tatives are lithium and calcium salts of carboxylic acids, for example Lithium Stearate, Lithium-12-hydroxistearate, Calciumstearate, Calcium-12-hydroxistearate but also Calciuma‐ cetate. Lithium Complex Greases are created by the co-existence of lithium-12-hydroxistearate

Calcium Complex Greases are composed by calcium acetate, Calcium Stearate and calcium-12-

Salts of magnesium, barium and alumina are used for grease production but to minor extent.


Isocyanate Amine Urea

Calciumacetate

Lithium Acelate

COO- Li+

O- Li+

O


**Table 2.** NLGI Grades of Greases

#### *12.1.1. Oil bleeding*

Within grease the base oil is bound in different states. Some oil is weakly bound to the thickener nuclei and gets easy released. Oil, bound in micelles and large structures with van-der-Waals and dipolar bonding releases less. Oil release takes place due to centrifugal effects in speeding machinery elements, e.g. bearings, creeping across walls e.g. sealings enhanced by tempera‐ ture. Successive loss of oil in grease may lead to its change in performance, accompanied by a malfunction. Oil bleeding is measured with different techniques. Within the most popular one the grease is sat on a sieve and pressed by a static load through it a given temperature. Bleeding is measured as a function of time. For bearings the long term bleeding rate should be less than 5 % per weight in 7 days. [3]

#### *12.1.2. Dropping point*

Greases - if heated - start to get liquid at a certain point. Molten grease will leak out at sealing edges and may cause a malfunction of the grease. For bearings the thumb rule is given by dropping point minus 50 °C as the upper point of applicability. [3]

#### **12.2. Soap based greases**

Greases are soft solids, created by a thickener that gelates in suitable base oils. Gelling takes place by intense mixing of thickeners with the base oil, often accompanied by heating till the gelation is reached [3]. (Figure 68):

**Figure 68.** Prominent representatives of thickeners for grease production

Thickeners are all substances where gelling in the base oil is achievable. Prominent represen‐ tatives are lithium and calcium salts of carboxylic acids, for example Lithium Stearate, Lithium-12-hydroxistearate, Calciumstearate, Calcium-12-hydroxistearate but also Calciuma‐ cetate. Lithium Complex Greases are created by the co-existence of lithium-12-hydroxistearate with dicarboxylic acids like Acelaic or sebacaic acid.

Calciumacetate

Calcium Complex Greases are composed by calcium acetate, Calcium Stearate and calcium-12 hydoxistearate as thickeners.

Salts of magnesium, barium and alumina are used for grease production but to minor extent.

#### **12.3. Di and Polyurea greases (PU-greases)**

Urea Greases are often called PU-Greases in technical language.

Urea structures are realized by adding amines to isocyanates (Figure 69):


**Figure 69.** Urea Formation

greases is defined by the NLGI grade declaration, measured by penetration of standard cone into the grease. The deeper it's penetration the more liquid the grease will be. To get a constant value, the grease is worked by 60 strokes, then tempered to 25 °C and measured by cone

> NLGI Grade Cone Penetration in 1/10 mm OOO 445 - 475 OO 400 - 430 O 355 - 385 1 310 - 340 2 265 - 295 3 220 - 250 4 175 - 205 5 130 - 160 6 85 - 115

Within grease the base oil is bound in different states. Some oil is weakly bound to the thickener nuclei and gets easy released. Oil, bound in micelles and large structures with van-der-Waals and dipolar bonding releases less. Oil release takes place due to centrifugal effects in speeding machinery elements, e.g. bearings, creeping across walls e.g. sealings enhanced by tempera‐ ture. Successive loss of oil in grease may lead to its change in performance, accompanied by a malfunction. Oil bleeding is measured with different techniques. Within the most popular one the grease is sat on a sieve and pressed by a static load through it a given temperature. Bleeding is measured as a function of time. For bearings the long term bleeding rate should be less than

Greases - if heated - start to get liquid at a certain point. Molten grease will leak out at sealing edges and may cause a malfunction of the grease. For bearings the thumb rule is given by

Greases are soft solids, created by a thickener that gelates in suitable base oils. Gelling takes place by intense mixing of thickeners with the base oil, often accompanied by heating till the

dropping point minus 50 °C as the upper point of applicability. [3]

penetration. NLGI grades are presented in table 2: [3]

46 Tribology - Fundamentals and Advancements

**Table 2.** NLGI Grades of Greases

5 % per weight in 7 days. [3]

*12.1.2. Dropping point*

**12.2. Soap based greases**

gelation is reached [3]. (Figure 68):

*12.1.1. Oil bleeding*

Di-Urea grease production take aromatic Isocyanates, like Diphenylmethane Isocyanate (Methylenbisdi-isocyanate, MDI) reacted with various aliphatic amines, like Cyclohexyla‐ mine, Alkylamines from C8 to C18 chain length.

Synthesis of the thickeners and grease formation is carried out simultaneously. Ester Oils, like trimellitic acid esters facilitate the synthesis by solving the precursors before the reaction takes place (Figure 70):

**Figure 70.** Formation of Di-Urea Grease

Tetra- and polyurea Greases are created by mixing Di-Isocyanates like MDI or Toluenediiso‐ cyanates (TDI) with diamines, like ethylene diamine and monoamines, like Octadecylamine in suitable base oils (Figure 71):

white mineral oil or PAO) is used. Also the modern EU REACH regulations are valid for

N=C=O

<sup>2</sup>NH

+

NHCONH

NH2

Diamine

NH2

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 49

NHCONH

N=C=O

Monoamine

Tetra Urea (idealised)

NHCON

NHCONH

Diisocyanate

As the polymeric degree increase, the thickeners may get insoluble and crystalline. Greases are no longer available due to this because a lack of gelling. Due to this fact, variances of PU

MDI and especially TDI are ought to be highly toxic by inhalation. Production of PU greases

Some isocyanates tend to polymerize during production, rather than to react with the amine, especially at the end of the syntheses. Polymeric Isocyanates may remain in the grease and

PU Greases are very sensitive toward ingress of OH – groups (e.g. alkalines, water, polygly‐ coles) as the nitrogen-hydrogen bridging is disturbed. Ingress of such pollutants may cause a change in consistency. Polyglycoles, if heated emit aldehydes that interfere with the NH

have to take care, than none of the precursors are free in air, nor present in the grease.

cause severe toxicitiy, especially if the greases are up -heated.

polymeric structure.

**Figure 71.** Formation of Tetra-and Polyurea Greases

Greases are restricted.

Urea Greases offer plenty nitrogen-hydrogen bridges within their structures. Concordant with the presence of temperature resistant aromatic nuclei and in junction with high temperature resistant base oils, they represent the group of high temperature grease "per se". As to the high variability of taking precursor amines, PU greases offer the possibility to adapt the grease to a given application, much more than soap greases do.

Polyurea Greases that start from tallow amine, tolyenediisocyanate and ethylene di-amine are in accordance with the US FDA regulations H1 (incidental food contact) if H1 base oil (like

**Figure 71.** Formation of Tetra-and Polyurea Greases

Di-Urea grease production take aromatic Isocyanates, like Diphenylmethane Isocyanate (Methylenbisdi-isocyanate, MDI) reacted with various aliphatic amines, like Cyclohexyla‐

Synthesis of the thickeners and grease formation is carried out simultaneously. Ester Oils, like trimellitic acid esters facilitate the synthesis by solving the precursors before the reaction takes

O=C=N N=C=O

<sup>2</sup>NH + NH2

Amine

NHCONH

**Figure 70.** Formation of Di-Urea Grease

in suitable base oils (Figure 71):

a given application, much more than soap greases do.

Diisocyanate

Ester as Solvent

Di-Urea Thickener

Tetra- and polyurea Greases are created by mixing Di-Isocyanates like MDI or Toluenediiso‐ cyanates (TDI) with diamines, like ethylene diamine and monoamines, like Octadecylamine

Urea Greases offer plenty nitrogen-hydrogen bridges within their structures. Concordant with the presence of temperature resistant aromatic nuclei and in junction with high temperature resistant base oils, they represent the group of high temperature grease "per se". As to the high variability of taking precursor amines, PU greases offer the possibility to adapt the grease to

Polyurea Greases that start from tallow amine, tolyenediisocyanate and ethylene di-amine are in accordance with the US FDA regulations H1 (incidental food contact) if H1 base oil (like

NHCONH

mine, Alkylamines from C8 to C18 chain length.

48 Tribology - Fundamentals and Advancements

place (Figure 70):

white mineral oil or PAO) is used. Also the modern EU REACH regulations are valid for polymeric structure.

As the polymeric degree increase, the thickeners may get insoluble and crystalline. Greases are no longer available due to this because a lack of gelling. Due to this fact, variances of PU Greases are restricted.

MDI and especially TDI are ought to be highly toxic by inhalation. Production of PU greases have to take care, than none of the precursors are free in air, nor present in the grease.

Some isocyanates tend to polymerize during production, rather than to react with the amine, especially at the end of the syntheses. Polymeric Isocyanates may remain in the grease and cause severe toxicitiy, especially if the greases are up -heated.

PU Greases are very sensitive toward ingress of OH – groups (e.g. alkalines, water, polygly‐ coles) as the nitrogen-hydrogen bridging is disturbed. Ingress of such pollutants may cause a change in consistency. Polyglycoles, if heated emit aldehydes that interfere with the NH groups in PU greases. This reaction may end up in making the solid PU liquid! PU greases thus should be monitored to those facts (Figure 72):

**Figure 72.** Reaction of PU Grease and Aldehydes

Other incompatibilities of PU Greases arise from mixtures with clay thickeners due to the presence of either OH (Si-OH) or NH functional groups if the clay is modified by organic amines.

#### **12.4. Other thickeners**

#### *12.4.1. Clay greases-structure and use*

Clay Thickeners derive from Alumina-Silicates. Due to their high surface and modification they are suitable for gelling base oils, e.g. Esters, Napthenic Base Oils, sometimes Silicones and Phosphoric Acid Esters. Clay Structure is generated by tetrahedral arrangement of Silica with insertion of alumina (see figure) in layers of approximately 1-2 nm distance. Water and other cations may be inserted in the space in between the two layers. Other cations, e.g. magnesium, may also be inserted in between. [3] (Figure 73).

Gelling takes place by adhesion and insertion of organic molecules in the structure, assisted by polar additives like propylene carbonate. Clay grease is produced by multiple milling the clay with appropriate base oil by addition of water suppliers like glycerol or Propylene carbonate at temperatures below 100°C. If water is lost the structure may break down during the manufacturing. Doing so, the grease produced is a buttery solid with no dropping point.

and conventional greases should be evaluated very carefully. Clay greases are restricted in bearing lubrication strictly due to over rolling speed. In general the speed factor is limited to ndm (Average of outer and inner diameter of the bearing times the speed (revolution per

Silica is in use for thickeners as amorphous material, obtained by flame decomposition of Silica

Silica, due to its powerful surface activity may be used as powerful thickener in low percentage for each kind of base oil. Greases obtained by mixing silica with base oils are transparent. The inorganic structure causes no dropping point for such greases. Silica Thickened greases cause steep and irreversible thickening by heating up due to the increase of internal hydrogen bonding. They never should be in use for high speed and high temperature rotating bearings,

minute)) of 100.000. Only slow moving bearings could bear clay lubrication.

Si

Si

Si

Si

Si

O

O

O

O

O

Si

Si

Si

Si

Si

Si

Si

Si

O

O

OH

OH

O

O

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 51

OH

OH

O

Si

Si

O

Si

O

O

O

O

O

O

O

Si

Si

OH2

O

Si

O

O

O

<sup>O</sup> <sup>O</sup>

OH

OH

Al Al Al Al Al Al

O

O

O

O

<sup>O</sup> <sup>O</sup>

OH

Al Al Al Al Al Al

O

HO

O

Si

Metal Cations

Si

O

Si

O

O

O

O

O

Si

Si

OH2

O

Si

O

O

O

OH

OH

O

O

O

HO

OH

O

O

Si

1-2 nm Distance

Si

O

Si

O

O

O

O

OH2

**13. Silica**

Tetrachloride (Figure 74):

**Figure 73.** Estimated basic structure of clay

#### *12.4.2. Use of clay greases*

Clay greases are used for applications where the grease should not move out and for special high temperature applications, e.g. cement industry in slow motion bearings. Due to the inertness of the inorganic structure toward alkaline and acids, clay greases are preferred in applications where water, alkaline and acids are present, e.g. chain or bearing lubrication with such ingress. Clay is declared as safe for incidental food contact and allowed for lubricants in food industry (USDA H1 regulated) in junction with base oils like white mineral oil, PAO or esters that are allowed for this purpose.

#### *12.4.3. Restrictions in the use of clay greases*

Restrictions for the use of clay greases are the presence of Lithium, - Calcium or Polyurea Greases that may interfere with the hydrogen bonding of the clay structure. Mixtures of clay

**Figure 73.** Estimated basic structure of clay

and conventional greases should be evaluated very carefully. Clay greases are restricted in bearing lubrication strictly due to over rolling speed. In general the speed factor is limited to ndm (Average of outer and inner diameter of the bearing times the speed (revolution per minute)) of 100.000. Only slow moving bearings could bear clay lubrication.

#### **13. Silica**

groups in PU greases. This reaction may end up in making the solid PU liquid! PU greases


OH R'

+

Other incompatibilities of PU Greases arise from mixtures with clay thickeners due to the presence of either OH (Si-OH) or NH functional groups if the clay is modified by organic

Clay Thickeners derive from Alumina-Silicates. Due to their high surface and modification they are suitable for gelling base oils, e.g. Esters, Napthenic Base Oils, sometimes Silicones and Phosphoric Acid Esters. Clay Structure is generated by tetrahedral arrangement of Silica with insertion of alumina (see figure) in layers of approximately 1-2 nm distance. Water and other cations may be inserted in the space in between the two layers. Other cations, e.g. magnesium,

Gelling takes place by adhesion and insertion of organic molecules in the structure, assisted by polar additives like propylene carbonate. Clay grease is produced by multiple milling the clay with appropriate base oil by addition of water suppliers like glycerol or Propylene carbonate at temperatures below 100°C. If water is lost the structure may break down during the manufacturing. Doing so, the grease produced is a buttery solid with no dropping point.

Clay greases are used for applications where the grease should not move out and for special high temperature applications, e.g. cement industry in slow motion bearings. Due to the inertness of the inorganic structure toward alkaline and acids, clay greases are preferred in applications where water, alkaline and acids are present, e.g. chain or bearing lubrication with such ingress. Clay is declared as safe for incidental food contact and allowed for lubricants in food industry (USDA H1 regulated) in junction with base oils like white mineral oil, PAO or

Restrictions for the use of clay greases are the presence of Lithium, - Calcium or Polyurea Greases that may interfere with the hydrogen bonding of the clay structure. Mixtures of clay

R'-CHO

thus should be monitored to those facts (Figure 72):


**Figure 72.** Reaction of PU Grease and Aldehydes

50 Tribology - Fundamentals and Advancements

*12.4.1. Clay greases-structure and use*

may also be inserted in between. [3] (Figure 73).

amines.

**12.4. Other thickeners**

*12.4.2. Use of clay greases*

esters that are allowed for this purpose.

*12.4.3. Restrictions in the use of clay greases*

Silica is in use for thickeners as amorphous material, obtained by flame decomposition of Silica Tetrachloride (Figure 74):

Silica, due to its powerful surface activity may be used as powerful thickener in low percentage for each kind of base oil. Greases obtained by mixing silica with base oils are transparent. The inorganic structure causes no dropping point for such greases. Silica Thickened greases cause steep and irreversible thickening by heating up due to the increase of internal hydrogen bonding. They never should be in use for high speed and high temperature rotating bearings,

tions found here are radical reactions, as a fact of the presence of oxygen and iron. With‐ in such radical reaction sequences hydrogen is abstracted, alkenes and alkynes are formed and their oxidation products (aldehydes, ketones, carboxylic acids and their derivatives).

Fundamentals of Lubricants and Lubrication http://dx.doi.org/10.5772/55731 53

In general, lubrication fundamentals in tribology have overcome the alchemy of the past by

[1] Rudnik L.R., editor. Synthetics, Mineral Oils, and Bio-Based Lubricants. Boca Raton:

[2] Dresel W., Mang T., editors: Lubricants and Lubrication. 2nd Edition. Weinheim: Wi‐

[3] Klamann D. Schmierstoff und verwandte Produkte. Weinheim: VCH-Verlag; 1982.

10.1007/978-1-4020-8662-5/page/1 (accessed 27 December 2012).

[4] Mortier R.M., Fox M.F., Orszullik T.M., editors. Chemistry and Technology of Lubri‐ cants Dordrecht: Springer; 2010. http://link.springer.com/book/

[5] Dowson D., Taylor C., Childs T., Dalmaz G. editors. Lubricants and Lubrication. In: Tribology Series 30 : Proceedings of the 21st Leeds-Lyon Symposium on Tribology.

[6] Bloch, H.P., Practical Lubrication for Industrial Facilities. Lilburn: Fairmont Press;

[8] Lansdown A.R., Lubrication and lubricant selection: a practical guide. 3rd Edition.

[9] Rudnick L. R., editor. Lubricant Additives: Chemistry and Applications, 1st Edition.

[10] March, J., Advanced Organic Chemistry: Reactions, mechanisms, Structure. New

[7] Stepina V., Vesely V. Lubricants and Special Fluids. Amsterdam: Elsevier; 1992.

Additives, in general improve the lubricants by expanding their limits.

Schaeffler Technologies AG & Co.KG, R&D Central Materials, Germany

numerous efforts taken by the scientific community.

**Author details**

Walter Holweger\*

**References**

CRC Press; 2005.

ley-VCH; 2007.

2000.

Amsterdam : Elsevier; 1995.

John Wiley & Sons; 2004.

York: Wiley-VCH; 1992.

New York:Marcel Dekker, 2003.

**Figure 74.** Principal formation of amorphous SiO2 by flame combustion

since they block their motion. The ndm (Average of Bearing Size times revolution per minute) is restricted to 100.000, hence slow motion. Due to the possible entrance of water, silica thickened grease is poorly water stable and should not be in use in applications where water (especially hot water) and alkalines are present. Alkalines react with silica to silicates, starting its degradation.

#### **14. Polytetrafluoroethylene (PTFE)**

PTFE is a convenient thickener in base oils for the purpose of incidental food contact, low friction properties and high temperature. The fluorine entity causes low activity toward oxygen. PTFE Grease is used in oxygen application (valves under oxygen impact), especially with PFPE.

#### **15. Conclusion**

Tribology is highly guided by physics and chemistry of the lubricants. Functionality of lu‐ bricants is given by their physics and their chemical structure. Modern understanding of lubrication hence allows the construction of lubricants appropriate to a given application to a certain extent. Under the conditions of full lubrication their physical properties, e.g. viscosity, viscosity-temperature and viscosity –pressure properties dominate over the chemical structure. Under such circumstances, the lubricant takes away heat (cooling function) from the mating contacts, but also wear and debris (cleaning function). Within a running – in period some reaction layers of lubricant constituents (additives) may be cre‐ ated. Basically those layers stay constant over time and do not change. On the other hand, if lubrication undermines the given roughness's of the mating partners, or over‐ takes the natural temperature limit given by the restrictions of organic chemistry (e.g. temperatures beyond 150°C), chemistry starts to perform reaction scenario highly related to the nature of the chemical structure of the ingredients in the lubricant. The basic reac‐ tions found here are radical reactions, as a fact of the presence of oxygen and iron. With‐ in such radical reaction sequences hydrogen is abstracted, alkenes and alkynes are formed and their oxidation products (aldehydes, ketones, carboxylic acids and their derivatives). Additives, in general improve the lubricants by expanding their limits.

In general, lubrication fundamentals in tribology have overcome the alchemy of the past by numerous efforts taken by the scientific community.

#### **Author details**

Walter Holweger\*

Schaeffler Technologies AG & Co.KG, R&D Central Materials, Germany

#### **References**

since they block their motion. The ndm (Average of Bearing Size times revolution per minute) is restricted to 100.000, hence slow motion. Due to the possible entrance of water, silica thickened grease is poorly water stable and should not be in use in applications where water (especially hot water) and alkalines are present. Alkalines react with silica to silicates, starting

Si O O

SiCl4

O Si O O

Flame

<sup>O</sup> <sup>n</sup>

O

Si O O

Si O O

O

O

**Figure 74.** Principal formation of amorphous SiO2 by flame combustion

PTFE is a convenient thickener in base oils for the purpose of incidental food contact, low friction properties and high temperature. The fluorine entity causes low activity toward oxygen. PTFE Grease is used in oxygen application (valves under oxygen impact), especially

Tribology is highly guided by physics and chemistry of the lubricants. Functionality of lu‐ bricants is given by their physics and their chemical structure. Modern understanding of lubrication hence allows the construction of lubricants appropriate to a given application to a certain extent. Under the conditions of full lubrication their physical properties, e.g. viscosity, viscosity-temperature and viscosity –pressure properties dominate over the chemical structure. Under such circumstances, the lubricant takes away heat (cooling function) from the mating contacts, but also wear and debris (cleaning function). Within a running – in period some reaction layers of lubricant constituents (additives) may be cre‐ ated. Basically those layers stay constant over time and do not change. On the other hand, if lubrication undermines the given roughness's of the mating partners, or over‐ takes the natural temperature limit given by the restrictions of organic chemistry (e.g. temperatures beyond 150°C), chemistry starts to perform reaction scenario highly related to the nature of the chemical structure of the ingredients in the lubricant. The basic reac‐

its degradation.

52 Tribology - Fundamentals and Advancements

with PFPE.

**15. Conclusion**

**14. Polytetrafluoroethylene (PTFE)**


**Chapter 2**

**Lubrication and Lubricants**

Additional information is available at the end of the chapter

The primary purpose of lubrication is to reduce wear and heat between contacting surfaces in relative motion. While wear and heat cannot be completely eliminated, they can be reduced to negligible or acceptable levels. Because heat and wear are associated with friction, both effects can be minimized by reducing the coefficient of friction between the contacting surfaces. Lubrication is also used to reduce oxidation and prevent rust; to provide insulation in transformer applications; to transmit mechanical power in hydraulic fluid power applications;

The modern period of lubrication began with the work of Osborne Reynolds (1842-1912). Reynold's research was concerned with shafts rotating in bearings and cases this show in Fig. 1. When a lubricant was applied to the shaft, Reynolds found that a rotating shaft pulled a converging wedge of lubricant between the shaft and the bearing. He also noted that as the shaft gained velocity, the liquid flowed between the two surfaces at a greater rate. This, because the lubricant is viscous, produces a liquid pressure in the lubricant wedge that is sufficient to keep the two surfaces separated. Under ideal conditions, Reynolds showed that this liquid pressure was great enough to prevent direct contact between the metal surfaces. Fig.2 taking a plain journal bearing as example, Fig.3 which is known as Stribeck curve summarizes the lubrication regimes by describing the relationship between speed, load, oil viscosity, oil film

> © 2013 Ahmed and Nassar; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2013 Ahmed and Nassar; licensee InTech. This is a paper distributed under the terms of the Creative Commons

Nehal S. Ahmed and

http://dx.doi.org/10.5772/56043

and to seal against dust, dirt, and water.

*1.1.1. The lubrication regimes*

thickness, and friction.

Amal M. Nassar

**1. Introduction**

**1.1. Lubrication**

**Chapter 2**

### **Lubrication and Lubricants**

Nehal S. Ahmed and Amal M. Nassar

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/56043

#### **1. Introduction**

#### **1.1. Lubrication**

The primary purpose of lubrication is to reduce wear and heat between contacting surfaces in relative motion. While wear and heat cannot be completely eliminated, they can be reduced to negligible or acceptable levels. Because heat and wear are associated with friction, both effects can be minimized by reducing the coefficient of friction between the contacting surfaces. Lubrication is also used to reduce oxidation and prevent rust; to provide insulation in transformer applications; to transmit mechanical power in hydraulic fluid power applications; and to seal against dust, dirt, and water.

#### *1.1.1. The lubrication regimes*

The modern period of lubrication began with the work of Osborne Reynolds (1842-1912). Reynold's research was concerned with shafts rotating in bearings and cases this show in Fig. 1. When a lubricant was applied to the shaft, Reynolds found that a rotating shaft pulled a converging wedge of lubricant between the shaft and the bearing. He also noted that as the shaft gained velocity, the liquid flowed between the two surfaces at a greater rate. This, because the lubricant is viscous, produces a liquid pressure in the lubricant wedge that is sufficient to keep the two surfaces separated. Under ideal conditions, Reynolds showed that this liquid pressure was great enough to prevent direct contact between the metal surfaces. Fig.2 taking a plain journal bearing as example, Fig.3 which is known as Stribeck curve summarizes the lubrication regimes by describing the relationship between speed, load, oil viscosity, oil film thickness, and friction.

© 2013 Ahmed and Nassar; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Ahmed and Nassar; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

**Figure 1.** Three positions of shaft in a bearing

**Figure 2.** Plain Journal bearing

In this graph, the coefficient of friction is plotted against the expression ZN/P (sometimes referred to as the Hersey number)

$$\text{Where ZN/P} = \frac{\text{oil viscosity} \times \text{shaftspeed}}{\text{bearing pressure}} \tag{1}$$

As shown there are three distinct zones separated by points A and B. At B the oil film is just thick enough to ensure that there is no contact between asperities on the shaft and bearing surfaces. Smoother surfaces shift B to the left, while at point A the oil film thickness reduces virtually to nil. Zone 2, between A and B is known as the zone of mixed lubrication. Mixedfilm lubrication is unstable at which increase in lubrication temperature causes further

 **Mixed-film lubrication** 

Basically, lubrication is governed by one of two principles: hydrodynamic lubrication and boundary lubrication. In the former, a continuous full-fluid film separates the sliding surfaces. In the latter, the oil film is not sufficient to prevent metal-to-metal contact. Hydrodynamic lubrication is the more common, and it is applicable to nearly all types of continuous sliding action where extreme pressures are not involved. Whether the sliding occurs on flat

ZN/P

**Zone 2** 

**A**

 Fig. (2) Stribeck curve ZN/P

**B**

**Zone 3** 

Lubrication and Lubricants http://dx.doi.org/10.5772/56043 57

In this graph, the coefficient of friction is plotted against the expression ZN/P (sometimes referred to as the Hersey

 As shown there are three distinct zones separated by points A and B. At B the oil film is just thick enough to ensure that there is no contact between asperities on the shaft and bearing surfaces. Smoother surfaces shift B to the left, while at point A the oil film thickness reduces virtually to nil. Zone 2, between A and B is known as the zone of mixed lubrication. Mixed-film lubrication is unstable at which increase in lubrication temperature causes

Where ZN/P = **(1)**

**Zone 1** 

further increases in lubrication temperature .

**1.1.1 Hydrodynamic Lubrication.** 

Basically, lubrication is governed by one of two principles: hydrodynamic lubrication and boundary lubrication. In the former, a continuous full-fluid film separates the sliding surfaces. In the latter, the oil film is not sufficient to prevent metal-to-metal contact. Hydrodynamic lubrication is the more common, and it is applicable to nearly all types of continuous sliding 2

increases in lubrication temperature.

**Figure 3. Stribeck curve** 

number)

**Figure 3.** Stribeck curve

*1.1.2. Hydrodynamic lubrication*

**Scheme 1.** Mixed-film lubrication

In this graph, the coefficient of friction is plotted against the expression ZN/P (sometimes referred to as the Hersey

**Figure 3.** Stribeck curve

**Figure 3. Stribeck curve** 

**Figure 1.** Three positions of shaft in a bearing

56 Tribology - Fundamentals and Advancements

**Figure 2.** Plain Journal bearing

referred to as the Hersey number)

In this graph, the coefficient of friction is plotted against the expression ZN/P (sometimes

oilviscosity×shaftspeed Where ZN/P = bearing pressure (1)

#### **Scheme 1.** Mixed-film lubrication

As shown there are three distinct zones separated by points A and B. At B the oil film is just thick enough to ensure that there is no contact between asperities on the shaft and bearing surfaces. Smoother surfaces shift B to the left, while at point A the oil film thickness reduces virtually to nil. Zone 2, between A and B is known as the zone of mixed lubrication. Mixedfilm lubrication is unstable at which increase in lubrication temperature causes further increases in lubrication temperature.  **Mixed-film lubrication 1.1.1 Hydrodynamic Lubrication.**  

Basically, lubrication is governed by one of two principles: hydrodynamic lubrication and boundary lubrication. In

#### *1.1.2. Hydrodynamic lubrication* the former, a continuous full-fluid film separates the sliding surfaces. In the latter, the oil film is not sufficient to prevent metal-to-metal contact. Hydrodynamic lubrication is the more common, and it is applicable to nearly all

Basically, lubrication is governed by one of two principles: hydrodynamic lubrication and boundary lubrication. In the former, a continuous full-fluid film separates the sliding surfaces. In the latter, the oil film is not sufficient to prevent metal-to-metal contact. Hydrodynamic lubrication is the more common, and it is applicable to nearly all types of continuous sliding 2 types of continuous sliding action where extreme pressures are not involved. Whether the sliding occurs on flat

action where extreme pressures are not involved. Whether the sliding occurs on flat surfaces, as it does in most thrust bearings, or whether the surfaces are cylindrical, as in the case of journal (plain or sleeve) bearings, the principle is essentially the same.

It would be reasonable to suppose that, when one part slides on another, the protective oil film between them would be scraped away. Except under some conditions of reciprocating motion, this is not necessarily true at all. With the proper design, in fact, this very sliding motion constitutes the means of creating and maintaining that film.

In zone 3 is the zone of hydrodynamic or fluid film lubrication where there is no wear because there is no contact between the surfaces. Hydrodynamic Lubrication is often referred to as stable lubrication. There are four essential elements in hydrodynamic lubrication, a liquid, relative motion, the viscous properties of the liquid, and the geometry of the surfaces between which the convergent wedge of fluid is produced. Only friction present in a hydrodynamic lubrication system is the friction of the lubricant itself, it would make sense to have a less viscous fluid in order to minimize friction: the less viscous a liquid the lower the friction. Too low of a viscosity jeopardizes our system though. We have to be very careful that the distance between the two surfaces is greater than the largest surface defect. The distance between the two surfaces decreases with higher loads on the bearing, less viscous fluids, and lower speeds. The surface geometry is also very important. The surfaces have to be such that a converging wedge of fluid can develop between the surfaces, allowing the hydrodynamic pressure of the lubricant to support the load of the shaft or moving surface. Hydrodynamic lubrication is an excellent method of lubrication since it is possible to achieve coefficients of friction as low as 0.001, and there is no wear between the moving parts. Special attention must be paid to the heating of the lubricant by the frictional force since viscosity is temperature dependent. One method of accomplishing this is to cycle the lubricant through a cooling reservoir in order to maintain the desired viscosity of the fluid. Another way of handling the heat dissipation is to use commercially available additives to decrease the viscosity's temperature dependence which are known as viscosity index improvers.

*1.1.3. Boundary lubrication*

**Scheme 2.** Hydrodynamic lubrication

reciprocating equipment.

The oil film has become so thin in Zone 1 that there is no hydrodynamic contribution and only boundary lubrication which is defined by Campbell in 1969 as the lubrication by a liquid under conditions where the solid surfaces are so close together that appreciable contact between opposing asperities is possible. The friction and wear in boundary lubrication are determined predominantly by interaction between the solids and between the solids and the liquid. The bulk flow properties of the liquid play little or no part in the friction and wear behavior.

Lubrication and Lubricants http://dx.doi.org/10.5772/56043 59

As mentioned, boundary lubrication is effective when a complete fluid film does not develop between potentially rubbing surfaces, the film thickness may be reduced to permit momentary dry contact between wear surface high points or asperities. Boundary lubrication occurs whenever any of the essential factors that influence formation of a full fluid film are missing. The most common example of boundary lubrication includes bearings, which normally operate with fluid film lubrication but experience boundary lubricating conditions during routine starting and stopping of equipment. Other examples include gear tooth contacts and

A brief explanation of what needs to be added to basic mineral oil in order to create an effective boundary lubricant. Generally, the best additives are active organic compounds with long chain molecules and active end groups. These compounds bind tightly and intricately with each other, forming a film that builds up on the surface of the metal itself. This results in a thin film that is very difficult to penetrate. When two surfaces, each covered with a boundary layer, come in contact with each other they tend to slide along their outermost surfaces, with the actual faces of the surfaces rarely making contact with each other. Liquids are rarely good boundary lubricants. The best boundary lubricants are solids with long chains of high interchain attraction, low shear resistance so as to slip easily, and a high temperature tolerance. The

The formation of fluid film is influenced by the following factors:


#### *1.1.3. Boundary lubrication* **Scheme 2.** Hydrodynamic lubrication

action where extreme pressures are not involved. Whether the sliding occurs on flat surfaces, as it does in most thrust bearings, or whether the surfaces are cylindrical, as in the case of

It would be reasonable to suppose that, when one part slides on another, the protective oil film between them would be scraped away. Except under some conditions of reciprocating motion, this is not necessarily true at all. With the proper design, in fact, this very sliding motion

In zone 3 is the zone of hydrodynamic or fluid film lubrication where there is no wear because there is no contact between the surfaces. Hydrodynamic Lubrication is often referred to as stable lubrication. There are four essential elements in hydrodynamic lubrication, a liquid, relative motion, the viscous properties of the liquid, and the geometry of the surfaces between which the convergent wedge of fluid is produced. Only friction present in a hydrodynamic lubrication system is the friction of the lubricant itself, it would make sense to have a less viscous fluid in order to minimize friction: the less viscous a liquid the lower the friction. Too low of a viscosity jeopardizes our system though. We have to be very careful that the distance between the two surfaces is greater than the largest surface defect. The distance between the two surfaces decreases with higher loads on the bearing, less viscous fluids, and lower speeds. The surface geometry is also very important. The surfaces have to be such that a converging wedge of fluid can develop between the surfaces, allowing the hydrodynamic pressure of the lubricant to support the load of the shaft or moving surface. Hydrodynamic lubrication is an excellent method of lubrication since it is possible to achieve coefficients of friction as low as 0.001, and there is no wear between the moving parts. Special attention must be paid to the heating of the lubricant by the frictional force since viscosity is temperature dependent. One method of accomplishing this is to cycle the lubricant through a cooling reservoir in order to maintain the desired viscosity of the fluid. Another way of handling the heat dissipation is to use commercially available additives to decrease the viscosity's temperature dependence

**•** The contact surfaces must meet at a slight angle to allow formation of the lubricant wedge.

**•** The fluid viscosity must be high to maintain adequate film thickness to separate the

**•** The fluid must be adhering to the contact surfaces for conveyance into the pressure area to

**•** The operating speed must be sufficient to allow formation and maintenance of the fluid film.

**•** The contact surfaces of bearings and journals must be smooth and free from sharp surfaces

**•** The fluid must be distributing itself completely within the bearing clearance area.

journal (plain or sleeve) bearings, the principle is essentially the same.

constitutes the means of creating and maintaining that film.

58 Tribology - Fundamentals and Advancements

which are known as viscosity index improvers.

contacting surfaces at operating speeds.

that will disrupt the fluid film.

support the load.

The formation of fluid film is influenced by the following factors:

The oil film has become so thin in Zone 1 that there is no hydrodynamic contribution and only boundary lubrication which is defined by Campbell in 1969 as the lubrication by a liquid under conditions where the solid surfaces are so close together that appreciable contact between opposing asperities is possible. The friction and wear in boundary lubrication are determined predominantly by interaction between the solids and between the solids and the liquid. The bulk flow properties of the liquid play little or no part in the friction and wear behavior.

As mentioned, boundary lubrication is effective when a complete fluid film does not develop between potentially rubbing surfaces, the film thickness may be reduced to permit momentary dry contact between wear surface high points or asperities. Boundary lubrication occurs whenever any of the essential factors that influence formation of a full fluid film are missing. The most common example of boundary lubrication includes bearings, which normally operate with fluid film lubrication but experience boundary lubricating conditions during routine starting and stopping of equipment. Other examples include gear tooth contacts and reciprocating equipment.

A brief explanation of what needs to be added to basic mineral oil in order to create an effective boundary lubricant. Generally, the best additives are active organic compounds with long chain molecules and active end groups. These compounds bind tightly and intricately with each other, forming a film that builds up on the surface of the metal itself. This results in a thin film that is very difficult to penetrate. When two surfaces, each covered with a boundary layer, come in contact with each other they tend to slide along their outermost surfaces, with the actual faces of the surfaces rarely making contact with each other. Liquids are rarely good boundary lubricants. The best boundary lubricants are solids with long chains of high interchain attraction, low shear resistance so as to slip easily, and a high temperature tolerance. The

**Figure 4.** Schematic View of Crude Oil Distillation

tions are also available [2].

etc.

**1.3. Lubricants**

suitable for different applications. For example:

Base stocks are refined from crude oil to obtain products with the best lubricating properties. Base stocks generally make up 80-95% of a typical engine oil and 5% additives [1]. Base stock is used to describe plain mineral oil. The physical properties of an oil depend on its base stock. In most cases it is chemically inert there are three sources of base stock: biological, mineral and synthetic. The oils manufactured from these sources exhibit different properties and they are

Lubrication and Lubricants http://dx.doi.org/10.5772/56043 61

**a.** Biological oils are suitable in applications where the risk of contamination must be reduced to a minimum, for example, in the food or pharmaceutical industry. They are usually applied to lubricate kilns, bakery ovens, etc. There can be two sources of this type of oil: vegetable and animal. Examples of vegetable oils are: castor, palm and rape-seed oils while the examples of animal oils are: sperm, fish and wool oils from sheep (lanolin).

**b.** Mineral oils are the most commonly used lubricants throughout industry. They are petroleum based and are used in applications where temperature requirements are moderate. Typical applications of mineral oils are to gears, bearings, engines, turbines,

**c.** Synthetic oils are artificially developed substitutes for mineral oils. They are specifical‐ ly developed to provide lubricants with superior properties to mineral oils. For example, temperature resistant synthetic oils are used in high performance machi‐ nery operating at high temperatures. Synthetic oils for very low temperature applica‐

All liquids will provide lubrication of a sort, but some do it a great deal bettor than others. The difference between one lubricating material and another is often the difference between successful operation of a machine and failure. For almost every situation, petroleum products have been found to excel as lubricants. Petroleum lubricants stand high in metal-wetting

#### **Scheme 3.** Boundary lubrication

boundary lubricant should also, obviously, be able to maintain a strong attachment to the surfaces under high temperatures and load pressures.

The most common boundary lubricants are probably greases. Greases are so widely used because they have the most desirable properties of a boundary lubricant. They not only shear easily, they flow. They also dissipate heat easily; form a protective barrier for the surfaces, preventing dust, dirt, and corrosive agents from harming the surfaces

#### **1.2. Base stock**

Petroleum is one of the naturally occurring hydrocarbons that frequently include natural gas, natural bitumen, and natural wax. The name ''petroleum'' is derived from the Latin *petra* (rock) and *oleum* (oil). According to the most generally accepted theory today, petroleum was formed by the decomposition of organic refuse, aided by high temperatures and pressures, over a vast period of geological time.

Although petroleum occurs, as its name indicates, among rocks in the earth, it sometimes seeps to the surface through fissures or is exposed by erosion. The existence of petroleum was known to primitive man, since surface seepage, often sticky and thick, was obvious to anyone passing by prehistoric animals were sometimes mired in it, but few human bones have been recovered from these tar pits. Early man evidently knew enough about the danger of surface seepage to avoid it.

The petroleum remaining from the distillation is thick like pitch; if the distillation has been pushed far, the residuum will flow only languidly in the retort, and in cold weather it becomes a soft solid, resembling much the maltha or mineral pitch Fig. 4 shows that the distillation of crude oil.

**Figure 4.** Schematic View of Crude Oil Distillation

Base stocks are refined from crude oil to obtain products with the best lubricating properties. Base stocks generally make up 80-95% of a typical engine oil and 5% additives [1]. Base stock is used to describe plain mineral oil. The physical properties of an oil depend on its base stock. In most cases it is chemically inert there are three sources of base stock: biological, mineral and synthetic. The oils manufactured from these sources exhibit different properties and they are suitable for different applications. For example:


#### **1.3. Lubricants**

boundary lubricant should also, obviously, be able to maintain a strong attachment to the

The most common boundary lubricants are probably greases. Greases are so widely used because they have the most desirable properties of a boundary lubricant. They not only shear easily, they flow. They also dissipate heat easily; form a protective barrier for the surfaces,

Petroleum is one of the naturally occurring hydrocarbons that frequently include natural gas, natural bitumen, and natural wax. The name ''petroleum'' is derived from the Latin *petra* (rock) and *oleum* (oil). According to the most generally accepted theory today, petroleum was formed by the decomposition of organic refuse, aided by high temperatures and pressures, over a vast

Although petroleum occurs, as its name indicates, among rocks in the earth, it sometimes seeps to the surface through fissures or is exposed by erosion. The existence of petroleum was known to primitive man, since surface seepage, often sticky and thick, was obvious to anyone passing by prehistoric animals were sometimes mired in it, but few human bones have been recovered from these tar pits. Early man evidently knew enough about the danger of surface seepage to

The petroleum remaining from the distillation is thick like pitch; if the distillation has been pushed far, the residuum will flow only languidly in the retort, and in cold weather it becomes a soft solid, resembling much the maltha or mineral pitch Fig. 4 shows that the distillation of

surfaces under high temperatures and load pressures.

**1.2. Base stock**

avoid it.

crude oil.

period of geological time.

**Scheme 3.** Boundary lubrication

60 Tribology - Fundamentals and Advancements

preventing dust, dirt, and corrosive agents from harming the surfaces

All liquids will provide lubrication of a sort, but some do it a great deal bettor than others. The difference between one lubricating material and another is often the difference between successful operation of a machine and failure. For almost every situation, petroleum products have been found to excel as lubricants. Petroleum lubricants stand high in metal-wetting ability, and they possess the body, or viscosity characteristics, that a substantial film requires, these oils have many additional properties that are essential to modern lubrication, such as good water resistance, inherent rust-preventive characteristics, natural adhesiveness, relative‐ ly good thermal stability, and the ability to transfer frictional heat away from lubricated parts. What is more, nearly all of these properties can be modified during manufacture to produce a suitable lubricant for each of a wide variety of applications. Oils have been developed handin- hand with the modern machinery that they lubricate; indeed, the efficiency, if not the existence, of many of today's industries and transportation facilities is dependent upon petroleum lubricants as well as petroleum fuels.

*1.3.1.1. Paraffinic components*

*1.3.1.2. Naphthenic components*

*1.3.1.3. Aromatic components*

*1.3.1.4. Non hydrocarbon components*

[9, 10].

The paraffinic components, show in Fig. 5 (a, b), which determine the pour point, contain not only linear but also branched paraffins. The straight chain paraffins of high molecular weights raise the pour point of oils (waxy compounds) and should be removed by dewaxing processes.

Lubrication and Lubricants http://dx.doi.org/10.5772/56043 63

The branched paraffins are chemically interesting hydrocarbons and they are found in large quantities in lubricating oil fractions from paraffinic crudes. Oil rich in paraffinic hydrocarbons have relatively low density and viscosity for their molecular weight and boiling range. Also, they have good viscosity/ temperature characteristics. In general, paraffinic components are reasonably resistant to oxidation and have particularly good response to oxidation inhibitors

They have rather higher density and viscosity for their molecular weight compared to the paraffinic components. An advantage which naphthenic components have over the paraffinic ones is that they tend to have low pour point and so do not contribute to wax. However, one disadvantage is that they have inferior viscosity/ temperature characteristics. Single ring alicyclics with long paraffinic side chains, however, share many properties with branched paraffins and can in fact be highly desirable components for lubricant base oils. Naphthenic components, Fig. 5 (c), tend to have better solvency power for additives than paraffinic

They have densities and viscosities which are still higher viscosity/ temperature characteristics are in general poor but pour point is low, although they have the best solvency power for additives, their stability to oxidation is poor. As for alicyclics, single ring aromatics with long paraffinic side chain may be very desirable base oil components, Fig. 5 (d). The classifying of hydrocarbon as paraffinic, naphthenic and aromatic groups which are generally used for characterizing the base oil should not be taken as absolute but as an expression of the pre‐

The non hydrocarbons in lubricating oil are analogous in many ways to the hydrocarbons. Sulfur and nitrogen compounds are found almost entirely in ring structures such as sulfides, thiophene, pyridine and pyrrol types. More complex molecules are also thought to exist in lubricating oil in which nitrogen and sulfur atoms are found in the same molecule. As in the case of hydrocarbons, these compounds will probably also have paraffinic side chains and

Although these non hydrocarbons may be present in only trace amounts, they often play a major role in controlling the properties of lubricating oils. In general they are chemically more active than the hydrocarbon, and hence they may markedly affect properties such as oxidation

components but their stability to oxidative processes is inferior [9, 10].

possibly be condensed with naphthenic and aromatic ring structures [11]

dominating chemical tendencies of the base stocks [11].

The basic petroleum lubricant is lubricating oil, which is often referred to simply as "oil." This complex mixture of hydrocarbon molecules represents one of the important classifications of products derived from the refining of crude petroleum oils, and is readily available in a great variety of types and grades.

Any description of lubricating oils would be incomplete without consideration of oils for vehicle engines. These oils are used in greater quantity than all other lubricants combined, and are of interest to more people than any other lubricants. Engine oils are generally recommend‐ ed by automotive builder according to the Society American of Automotive Engineers (SAE) viscosity classification.

Engine oil lubricants make up nearly one half of the lubricant market and therefore attract a lot of interest. The principal function of the engine oil lubricant is to extend the life of moving parts operating under many different conditions of speed, temperature, and pressure. At low temperatures the lubricant is expected to flow sufficiently in order that moving parts are not starved of oil. At higher temperatures they are expected to keep the moving parts apart to minimize wear. The lubricants reduce friction and removing heat from moving parts.

#### *1.3.1. General classification of the lubricating oils*

The term lubricating oil is generally used to include all those classes of lubricating materials that are applied as fluids [3]. Lubricating oils are made from the more viscous portion of the crude oil which remains after removal by distillation of the gas oil and lighter fraction [4-8]. Although crude oils from various parts of the world differ widely in properties and appear‐ ance, there is relatively little difference in their elemental analysis. Thus, crude oil samples will generally show carbon content ranging from 83% to 87 %, and hydrogen content from 11% to 14%. The remainder is composed of elements such as oxygen, nitrogen, and sulfur, and various metallic compounds. An elemental analysis, therefore, gives little indication of the extreme range of physical and chemical properties that actually exists, or of the nature of the lubricating base stocks that can be produced from a particular crude oil.

An idea of the complexity of the lubricating oil-refining problem can obtained from a consid‐ eration of the variations that can exist in a single hydrocarbon molecule with a specific number of carbon atoms. For example, the paraffinic molecule containing 25 carbon atoms has 52 hydrogen atoms. This compound can have about 37,000,000 different molecular arrangements [3]. The hydrocarbons of the crude oils are:

#### *1.3.1.1. Paraffinic components*

ability, and they possess the body, or viscosity characteristics, that a substantial film requires, these oils have many additional properties that are essential to modern lubrication, such as good water resistance, inherent rust-preventive characteristics, natural adhesiveness, relative‐ ly good thermal stability, and the ability to transfer frictional heat away from lubricated parts. What is more, nearly all of these properties can be modified during manufacture to produce a suitable lubricant for each of a wide variety of applications. Oils have been developed handin- hand with the modern machinery that they lubricate; indeed, the efficiency, if not the existence, of many of today's industries and transportation facilities is dependent upon

The basic petroleum lubricant is lubricating oil, which is often referred to simply as "oil." This complex mixture of hydrocarbon molecules represents one of the important classifications of products derived from the refining of crude petroleum oils, and is readily available in a great

Any description of lubricating oils would be incomplete without consideration of oils for vehicle engines. These oils are used in greater quantity than all other lubricants combined, and are of interest to more people than any other lubricants. Engine oils are generally recommend‐ ed by automotive builder according to the Society American of Automotive Engineers (SAE)

Engine oil lubricants make up nearly one half of the lubricant market and therefore attract a lot of interest. The principal function of the engine oil lubricant is to extend the life of moving parts operating under many different conditions of speed, temperature, and pressure. At low temperatures the lubricant is expected to flow sufficiently in order that moving parts are not starved of oil. At higher temperatures they are expected to keep the moving parts apart to

The term lubricating oil is generally used to include all those classes of lubricating materials that are applied as fluids [3]. Lubricating oils are made from the more viscous portion of the crude oil which remains after removal by distillation of the gas oil and lighter fraction [4-8]. Although crude oils from various parts of the world differ widely in properties and appear‐ ance, there is relatively little difference in their elemental analysis. Thus, crude oil samples will generally show carbon content ranging from 83% to 87 %, and hydrogen content from 11% to 14%. The remainder is composed of elements such as oxygen, nitrogen, and sulfur, and various metallic compounds. An elemental analysis, therefore, gives little indication of the extreme range of physical and chemical properties that actually exists, or of the nature of the lubricating

An idea of the complexity of the lubricating oil-refining problem can obtained from a consid‐ eration of the variations that can exist in a single hydrocarbon molecule with a specific number of carbon atoms. For example, the paraffinic molecule containing 25 carbon atoms has 52 hydrogen atoms. This compound can have about 37,000,000 different molecular arrangements

minimize wear. The lubricants reduce friction and removing heat from moving parts.

petroleum lubricants as well as petroleum fuels.

*1.3.1. General classification of the lubricating oils*

[3]. The hydrocarbons of the crude oils are:

base stocks that can be produced from a particular crude oil.

variety of types and grades.

62 Tribology - Fundamentals and Advancements

viscosity classification.

The paraffinic components, show in Fig. 5 (a, b), which determine the pour point, contain not only linear but also branched paraffins. The straight chain paraffins of high molecular weights raise the pour point of oils (waxy compounds) and should be removed by dewaxing processes.

The branched paraffins are chemically interesting hydrocarbons and they are found in large quantities in lubricating oil fractions from paraffinic crudes. Oil rich in paraffinic hydrocarbons have relatively low density and viscosity for their molecular weight and boiling range. Also, they have good viscosity/ temperature characteristics. In general, paraffinic components are reasonably resistant to oxidation and have particularly good response to oxidation inhibitors [9, 10].

#### *1.3.1.2. Naphthenic components*

They have rather higher density and viscosity for their molecular weight compared to the paraffinic components. An advantage which naphthenic components have over the paraffinic ones is that they tend to have low pour point and so do not contribute to wax. However, one disadvantage is that they have inferior viscosity/ temperature characteristics. Single ring alicyclics with long paraffinic side chains, however, share many properties with branched paraffins and can in fact be highly desirable components for lubricant base oils. Naphthenic components, Fig. 5 (c), tend to have better solvency power for additives than paraffinic components but their stability to oxidative processes is inferior [9, 10].

#### *1.3.1.3. Aromatic components*

They have densities and viscosities which are still higher viscosity/ temperature characteristics are in general poor but pour point is low, although they have the best solvency power for additives, their stability to oxidation is poor. As for alicyclics, single ring aromatics with long paraffinic side chain may be very desirable base oil components, Fig. 5 (d). The classifying of hydrocarbon as paraffinic, naphthenic and aromatic groups which are generally used for characterizing the base oil should not be taken as absolute but as an expression of the pre‐ dominating chemical tendencies of the base stocks [11].

#### *1.3.1.4. Non hydrocarbon components*

The non hydrocarbons in lubricating oil are analogous in many ways to the hydrocarbons. Sulfur and nitrogen compounds are found almost entirely in ring structures such as sulfides, thiophene, pyridine and pyrrol types. More complex molecules are also thought to exist in lubricating oil in which nitrogen and sulfur atoms are found in the same molecule. As in the case of hydrocarbons, these compounds will probably also have paraffinic side chains and possibly be condensed with naphthenic and aromatic ring structures [11]

Although these non hydrocarbons may be present in only trace amounts, they often play a major role in controlling the properties of lubricating oils. In general they are chemically more active than the hydrocarbon, and hence they may markedly affect properties such as oxidation

**b. Viscosity index**

distillate feedstock [8].

measured.

**• cloud point**

**• pour point**

point.

teristics of the oil.

**• volatility**

**• flash point**

ization.

**d. High temperature properties.**

**c. Low temperature properties.**

The most frequently used method for comparing the variation of viscosity with temperature between different oils by calculation of dimensionless numbers, known as the viscosity index (VI). The kinematic viscosity of the sample is measured at two different temperatures (40°C, 100°C) and the viscosity compared with an empirical reference scale. VI is used as a convenient measure of the degree of aromatics removal during the base oil manufacturing process, but comparison of VI of different oil samples is only realistic if they are derived from the same

Lubrication and Lubricants http://dx.doi.org/10.5772/56043 65

When a sample of oil is cooled, its viscosity increases in a predictable manner until wax crystals start to form. The matrix of wax crystals becomes sufficiently dense with further cooling to cause an apparent solidification of the oil. Although the solidified oil does not pour under the influence of gravity, it can move if sufficient force is applied. Further decrease in temperature cause more wax to form, increasing the complexity of the wax/oil matrix. Many lubricating oils have to be capable of flow at low temperatures and a number of properties should be

It is the temperature at which the first sign of wax formation can be detected. A sample of oil is warmed sufficiently to be fluid and clear. It is then cooled at a specified rate. The tem‐ perature at which haziness is first observed is recorded as the cloud point, the ASTM D 2500/IP 219 test. The oil sample must be free of water because it interferers with the test.

It is the lowest temperature at which the sample of the sample of oil can make to flow by gravity alone. The oil is warmed and then cooled at a specified rate. The test jar is removed from the cooling bath at intervals to see if the sample is still mobile. The procedure is repeat‐ ed until movement of the oil doesnot occur, ASTM D 97/IP 15. the pour point is the last tem‐ perature before the movement ceases, not the temperature at which solidification occurs. This is an important property of diesel fuels as well as lubricant base oils. High- Viscosity oils may cease to flow at low temperatures because their viscosity becomes too high rather than because of wax formation. In these cases, the pour point will be higher than the cloud

The high temperature properties of oil are governed by distillation or boiling range charac‐

It is important because it is an indication of the tendency of oil to be lost in service by vapor‐

**Figure 5.** Chemical Structure of Lubricating oil

stability, thermal stability and deposit forming tendencies. In refining the general tendency is to reduce the non hydrocarbons content to a minimum.

Naphthenic acid account for most of the oxygenated compounds found in petroleum. These are removed in the refining processes by neutralization and distillation. The naphthenates are retained in the residue from the distillation and can be removed by deasphalting process. Modern refining methods generally remove most of resins, asphaltenes, polycyclic aromatic, di aromatic and their analogous non hydrocarbons, so that the final lubricant consists chiefly of saturated and monocyclic aromatic fraction [12].

#### *1.3.2. Main properties of lubricating oils*

The main properties which a lubricating oil must posses to full performance are :

#### *1.3.2.1. Physical properties of lubricating oil*

#### **a. Viscosity**

Viscosity is the measure of the internal friction within a liquid; the way the molecules interact to resist motion. It is a vital property of a lubricant because it influences the ability of the oil to form a lubricating film or to minimize friction [8]. Newton defined the absolute viscosity of a liquid as the ratio between the applied shear stress and the resulting shear rate.

#### **b. Viscosity index**

The most frequently used method for comparing the variation of viscosity with temperature between different oils by calculation of dimensionless numbers, known as the viscosity index (VI). The kinematic viscosity of the sample is measured at two different temperatures (40°C, 100°C) and the viscosity compared with an empirical reference scale. VI is used as a convenient measure of the degree of aromatics removal during the base oil manufacturing process, but comparison of VI of different oil samples is only realistic if they are derived from the same distillate feedstock [8].

#### **c. Low temperature properties.**

When a sample of oil is cooled, its viscosity increases in a predictable manner until wax crystals start to form. The matrix of wax crystals becomes sufficiently dense with further cooling to cause an apparent solidification of the oil. Although the solidified oil does not pour under the influence of gravity, it can move if sufficient force is applied. Further decrease in temperature cause more wax to form, increasing the complexity of the wax/oil matrix. Many lubricating oils have to be capable of flow at low temperatures and a number of properties should be measured.

#### **• cloud point**

It is the temperature at which the first sign of wax formation can be detected. A sample of oil is warmed sufficiently to be fluid and clear. It is then cooled at a specified rate. The tem‐ perature at which haziness is first observed is recorded as the cloud point, the ASTM D 2500/IP 219 test. The oil sample must be free of water because it interferers with the test.

#### **• pour point**

stability, thermal stability and deposit forming tendencies. In refining the general tendency is

Naphthenic acid account for most of the oxygenated compounds found in petroleum. These are removed in the refining processes by neutralization and distillation. The naphthenates are retained in the residue from the distillation and can be removed by deasphalting process. Modern refining methods generally remove most of resins, asphaltenes, polycyclic aromatic, di aromatic and their analogous non hydrocarbons, so that the final lubricant consists chiefly

Viscosity is the measure of the internal friction within a liquid; the way the molecules interact to resist motion. It is a vital property of a lubricant because it influences the ability of the oil to form a lubricating film or to minimize friction [8]. Newton defined the absolute viscosity of

The main properties which a lubricating oil must posses to full performance are :

a liquid as the ratio between the applied shear stress and the resulting shear rate.

to reduce the non hydrocarbons content to a minimum.

of saturated and monocyclic aromatic fraction [12].

*1.3.2. Main properties of lubricating oils*

**Figure 5.** Chemical Structure of Lubricating oil

64 Tribology - Fundamentals and Advancements

*1.3.2.1. Physical properties of lubricating oil*

**a. Viscosity**

It is the lowest temperature at which the sample of the sample of oil can make to flow by gravity alone. The oil is warmed and then cooled at a specified rate. The test jar is removed from the cooling bath at intervals to see if the sample is still mobile. The procedure is repeat‐ ed until movement of the oil doesnot occur, ASTM D 97/IP 15. the pour point is the last tem‐ perature before the movement ceases, not the temperature at which solidification occurs. This is an important property of diesel fuels as well as lubricant base oils. High- Viscosity oils may cease to flow at low temperatures because their viscosity becomes too high rather than because of wax formation. In these cases, the pour point will be higher than the cloud point.

#### **d. High temperature properties.**

The high temperature properties of oil are governed by distillation or boiling range charac‐ teristics of the oil.

#### **• volatility**

It is important because it is an indication of the tendency of oil to be lost in service by vapor‐ ization.

**• flash point**

It is important for oil from a safety point of view because it is the lowest temperature at which auto-ignition of the vapour occur above the heated oil sample. Different methods are used, ASTM D 92, D93, and it is essential to know which equipment has been used when comparing results.

**b. Low Carbon Forming Tendency.**

**Carbon residue test methods.**

**c. High Oxidation Stability.**

**d. Wear Reduction.**

**i. Abrasive wear**

**ii. Corrosive wear**

wear.

This property is important for high compression ratio petro engine where carbon deposit will adversely affect combustion quality. The extent and also the composition of such formed deposits are causing noisy and rough burning which subjects the engine to high thermal and mechanical stresses resulting in lowering of performance and reduction of engine life. The typical symptoms will be knocking, preignition and surface ignition. These call higher octane fuels which are more expensive and do not eliminate the need for ultimate decarbonizing.

Lubrication and Lubricants http://dx.doi.org/10.5772/56043 67

Provide with some indication about the relative coke forming tendency of the oil in some application and quality-controlled lubricants. So, the test can be helpful in selecting oils for certain industrial applications such as heat treating, lubrication of bearing subjected to high temperature and air compressors. It is claimed that the presence of viscous oil (bright stock)

One of the most important requirement of the lubricant is that its properties are not changed during use [5-10]. The lubricant is often subjected to several oxidizing conditions which are primarily due to the oxidative changes of the oil. While the temperature of the oil, engine parts presence of oxygen, nature by products of fuel composition contribute to the oxidative change the properties of the lubricant during use. Therefore, It's essential that the lubricating oil; when exposed to high temperature; doesn't contribute to the forming of deposits even after a long period of continuous engine running. So, the lubricant resistance to the oxidative depends

Wear occurs in lubricated systems by three mechanisms (abrasion, corrosion and metal-tometal contact. i.e adhesion). The lubricant play an important role in combating each type of

It is caused by solid particles entering into the area between the lubricated surfaces and physically eroding these surfaces and may contaminate wear fragments. To cause wear, the solid particles must be larger than the oil-film thickness and harder than the lubricated surfaces. The flushing action of the lubricant, especially in forced feed or once through systems, severs to remove potentially harmful solid particles from the area of lubricated surfaces.

Corrosive wear is generally caused by the products of oxidation of lubricants. The high sulfur content of the fuel helps the corrosive attack. In other words, corrosion is the principal cause of wear in the internal combustion engines because the products of combustion are highly acidic and contaminate the lubrication oil, lubricants function to minimize corrosive wear is

in the base oils plays an important role in the formation of carbon deposits.

mainly on the nature of the lubricant and the presence of anti-oxidant additives.

#### **e. Other physical properties**

Various other physical properties may be measured, most of them relating to specialized lubricant applications. Some of the more important measurements are:

#### **• density**

Important, because oils may be formulated by weight, but measured by volume.

#### **• demulsification**

Ability of oil and water to separate.

#### **• foam characteristics**

Tendency to foam formation and stability of the foam that results.


Important for heat transfer fluid.

**• electrical properties**

Resistively, dielectric constant.

**• surface properties**

As surface tension, air separation.

#### *1.3.2.2. Chemical properties of lubricating oils*

#### **a. Ease of starting rapidity of warming up.**

The ease of starting depends chiefly on the cranking speed which is influenced by oil viscosity at the temperature of the crankcase. The major factor in the usage of a lubricant is its viscosity. It's not enough that the lubricants should have the proper viscosity but also they should maintain the little viscosity change within the temperature range during and after the apper‐ tain. So, viscosity controls not only frictional and thermal effect but also oil flow as a function of the load speed, temperature and design of the device lubricated. In other words, if the equipment will often have no make a cold start, it's also important that the viscosity at starting temperature is not so high that the machine can not be started. The rapidity with which an engine can be put to work is dependent on the speed of circulation and supply of oil to vital components, all forms of wear and even the safety of the engine are influenced by rapidity of circulation of the lubricants.

#### **b. Low Carbon Forming Tendency.**

It is important for oil from a safety point of view because it is the lowest temperature at which auto-ignition of the vapour occur above the heated oil sample. Different methods are used, ASTM D 92, D93, and it is essential to know which equipment has been used when comparing

Various other physical properties may be measured, most of them relating to specialized

The ease of starting depends chiefly on the cranking speed which is influenced by oil viscosity at the temperature of the crankcase. The major factor in the usage of a lubricant is its viscosity. It's not enough that the lubricants should have the proper viscosity but also they should maintain the little viscosity change within the temperature range during and after the apper‐ tain. So, viscosity controls not only frictional and thermal effect but also oil flow as a function of the load speed, temperature and design of the device lubricated. In other words, if the equipment will often have no make a cold start, it's also important that the viscosity at starting temperature is not so high that the machine can not be started. The rapidity with which an engine can be put to work is dependent on the speed of circulation and supply of oil to vital components, all forms of wear and even the safety of the engine are influenced by rapidity of

lubricant applications. Some of the more important measurements are:

Tendency to foam formation and stability of the foam that results.

Important, because oils may be formulated by weight, but measured by volume.

results.

**• density**

**• demulsification**

**• foam characteristics**

**• thermal conductivity**

**• electrical properties**

**• surface properties**

**e. Other physical properties**

66 Tribology - Fundamentals and Advancements

Ability of oil and water to separate.

**• pressure/viscosity characteristics**

Important for heat transfer fluid.

Resistively, dielectric constant.

As surface tension, air separation.

circulation of the lubricants.

*1.3.2.2. Chemical properties of lubricating oils*

**a. Ease of starting rapidity of warming up.**

This property is important for high compression ratio petro engine where carbon deposit will adversely affect combustion quality. The extent and also the composition of such formed deposits are causing noisy and rough burning which subjects the engine to high thermal and mechanical stresses resulting in lowering of performance and reduction of engine life. The typical symptoms will be knocking, preignition and surface ignition. These call higher octane fuels which are more expensive and do not eliminate the need for ultimate decarbonizing.

#### **Carbon residue test methods.**

Provide with some indication about the relative coke forming tendency of the oil in some application and quality-controlled lubricants. So, the test can be helpful in selecting oils for certain industrial applications such as heat treating, lubrication of bearing subjected to high temperature and air compressors. It is claimed that the presence of viscous oil (bright stock) in the base oils plays an important role in the formation of carbon deposits.

#### **c. High Oxidation Stability.**

One of the most important requirement of the lubricant is that its properties are not changed during use [5-10]. The lubricant is often subjected to several oxidizing conditions which are primarily due to the oxidative changes of the oil. While the temperature of the oil, engine parts presence of oxygen, nature by products of fuel composition contribute to the oxidative change the properties of the lubricant during use. Therefore, It's essential that the lubricating oil; when exposed to high temperature; doesn't contribute to the forming of deposits even after a long period of continuous engine running. So, the lubricant resistance to the oxidative depends mainly on the nature of the lubricant and the presence of anti-oxidant additives.

#### **d. Wear Reduction.**

Wear occurs in lubricated systems by three mechanisms (abrasion, corrosion and metal-tometal contact. i.e adhesion). The lubricant play an important role in combating each type of wear.

#### **i. Abrasive wear**

It is caused by solid particles entering into the area between the lubricated surfaces and physically eroding these surfaces and may contaminate wear fragments. To cause wear, the solid particles must be larger than the oil-film thickness and harder than the lubricated surfaces. The flushing action of the lubricant, especially in forced feed or once through systems, severs to remove potentially harmful solid particles from the area of lubricated surfaces.

#### **ii. Corrosive wear**

Corrosive wear is generally caused by the products of oxidation of lubricants. The high sulfur content of the fuel helps the corrosive attack. In other words, corrosion is the principal cause of wear in the internal combustion engines because the products of combustion are highly acidic and contaminate the lubrication oil, lubricants function to minimize corrosive wear is in two ways: proper refinement plus the use of oxidation inhibitors which reduces lubricant deterioration and keeps the level of corrosive oxidation products low.

is paramount and in another, such as metal cutting, the temperature control may be most important. A lubricating oil performance or requirement for a modern high speed engine

Lubrication and Lubricants http://dx.doi.org/10.5772/56043 69

The reduction of engine resistance to minimum is necessary to ensure maximum mechanical efficiency (running costs of a vehicle or engines are influenced by the lubricant viscosity)

All users wants minimum maintenance costs, longer engine life and increased usefulness.

The reduction of gas and oil leakages in an efficient and lasting manner is necessary to maintain engine performance and to prevent the combustions products from adulterating the oil.

In modern engines, the oil functions and more as a heat exchange medium, dissipating the heat is not converted into work. This is often associated with the first function in this list where the viscous oil give greater frictional resistance and its slow internal circulation leads to a rapid temperature raise of some vital part of the engine to cool efficiency, the oil must be able to

The lubricant give the function of protecting the engine against corrosive and mechanical wear which caused by all injurious impurities. So, the removal of these impurities by lubricants is very important for engine. The function and the corresponding qualities required for engine

Gaseous lubricants belong to the simplest, lowest viscosity lubricants known and include air, nitrogen, oxygen, and helium. They are applied in aerodynamic and aerostatic bearings. Since the chemical properties and the aggregate state of most gases remain unchanged over a wide temperature range, gaseous lubricants offer several advantages over liquid lubricants. First, they can be applied at both very high and very low temperatures. Their chemical stability eliminates any risk of contamination of the bearing by the lubricant, important for the machinery used in many branches of industry, primarily in the food, pharmaceutical and

A useful property of gases is that their viscosities increase with temperature, wheras the opposite is true of liquids, resulting in load – carrying capacity of gas – lubricated bearings

should fulfill the following five important functions:

**2. Protection of the engine against all types of wear:**

**4. Contributing the thermal equilibrium of the engine:**

Modern oil has allowed longer intervals between engines over hauls.

**1. Reduction of the frictional resistance:**

**3. Reduction of gas and oil leakages:**

**5. Removal of all injurious impurities:**

lubricating oils are summarized in Table (1).

circulated quickly.

*1.3.4. Types of lubricants*

*1.3.4.1. Gaseous lubricants*

electronic industries.

#### **iii. Adhesive wear**

This type of wear can significaly affect certain parts of the engine where metal-to-metal contact takes place. Adhesive wear takes place also if power was increased without corresponding modification is design, finishing and composition of the metal parts. Wear of this type also results form breakdown of lubricant film. It can also be the result of excessive surface rough‐ ness or interruption of the lubricant supply. A plentiful supply of the proper viscosity of oil is often the best way to avoid these conditions. The composition of the base oil and addition of certain chemical additives are also the important factors in protection of engine parts compo‐ nents against adhesive wear.

#### **e. Detergency and Dispersancy.**

With the exception of detergency and dispersancy in the combustion chamber, deposit in the oil are controlled by its detergent power.The source of the deposits found in engines are many and their volume depends mainly on the used, the quality of combustion, the temperature of lubricating oil and coolant, and on the gas sealing of the ring in the cylin‐ der. It these deposits are not removed with the oil when it is drained, their accumulation in the engine would drastically shorten the engine life. The role of the detergent addi‐ tives is to reduce the amount of deposits formed and their removal easy. The detergent property imparted to oils by additives seems to perform differently depending upon whether deposits result from high low temperature, low temperature deposits are mainly yielded from the fuel combustion and the detergency function is to keep them in suspen‐ sion or solution in the lubricating oil. However, high temperature deposits are mainly re‐ lated to the oxidized fraction of the oil.

The role of detergency here is not only to maintain these products in suspension, but also to stop the development of those chain reaction which promote the formation of varnishes and lacquers. The physical and functional properties of the lube oil will depend on the properties of carbon atoms in the various ring structures and aliphatic side chain

#### **f. Seal compatibility**

Lubricants are often used in machines where they come into contact with rubber or plastic seal. The strength and degree of swell of these seals may be affected by interaction with the oil. Various tests have been devised to measure the effect of base oils different seals and under different test conditions [13]. The strength and degree of swell of these seals may be affected by interaction with the oil. Various tests measure the effects of base oils on different seals and under different test conditions.

#### *1.3.3. Required performance characteristics for lubricating oils*

Selection and application of lubricating oil are determined by the functions which are expected for performance. In one application, such as delicate instrument bearing, the reduction friction is paramount and in another, such as metal cutting, the temperature control may be most important. A lubricating oil performance or requirement for a modern high speed engine should fulfill the following five important functions:

#### **1. Reduction of the frictional resistance:**

in two ways: proper refinement plus the use of oxidation inhibitors which reduces lubricant

This type of wear can significaly affect certain parts of the engine where metal-to-metal contact takes place. Adhesive wear takes place also if power was increased without corresponding modification is design, finishing and composition of the metal parts. Wear of this type also results form breakdown of lubricant film. It can also be the result of excessive surface rough‐ ness or interruption of the lubricant supply. A plentiful supply of the proper viscosity of oil is often the best way to avoid these conditions. The composition of the base oil and addition of certain chemical additives are also the important factors in protection of engine parts compo‐

With the exception of detergency and dispersancy in the combustion chamber, deposit in the oil are controlled by its detergent power.The source of the deposits found in engines are many and their volume depends mainly on the used, the quality of combustion, the temperature of lubricating oil and coolant, and on the gas sealing of the ring in the cylin‐ der. It these deposits are not removed with the oil when it is drained, their accumulation in the engine would drastically shorten the engine life. The role of the detergent addi‐ tives is to reduce the amount of deposits formed and their removal easy. The detergent property imparted to oils by additives seems to perform differently depending upon whether deposits result from high low temperature, low temperature deposits are mainly yielded from the fuel combustion and the detergency function is to keep them in suspen‐ sion or solution in the lubricating oil. However, high temperature deposits are mainly re‐

The role of detergency here is not only to maintain these products in suspension, but also to stop the development of those chain reaction which promote the formation of varnishes and lacquers. The physical and functional properties of the lube oil will depend on the properties

Lubricants are often used in machines where they come into contact with rubber or plastic seal. The strength and degree of swell of these seals may be affected by interaction with the oil. Various tests have been devised to measure the effect of base oils different seals and under different test conditions [13]. The strength and degree of swell of these seals may be affected by interaction with the oil. Various tests measure the effects of base oils on different seals and

Selection and application of lubricating oil are determined by the functions which are expected for performance. In one application, such as delicate instrument bearing, the reduction friction

of carbon atoms in the various ring structures and aliphatic side chain

*1.3.3. Required performance characteristics for lubricating oils*

deterioration and keeps the level of corrosive oxidation products low.

**iii. Adhesive wear**

68 Tribology - Fundamentals and Advancements

nents against adhesive wear.

**e. Detergency and Dispersancy.**

lated to the oxidized fraction of the oil.

**f. Seal compatibility**

under different test conditions.

The reduction of engine resistance to minimum is necessary to ensure maximum mechanical efficiency (running costs of a vehicle or engines are influenced by the lubricant viscosity)

#### **2. Protection of the engine against all types of wear:**

All users wants minimum maintenance costs, longer engine life and increased usefulness. Modern oil has allowed longer intervals between engines over hauls.

#### **3. Reduction of gas and oil leakages:**

The reduction of gas and oil leakages in an efficient and lasting manner is necessary to maintain engine performance and to prevent the combustions products from adulterating the oil.

#### **4. Contributing the thermal equilibrium of the engine:**

In modern engines, the oil functions and more as a heat exchange medium, dissipating the heat is not converted into work. This is often associated with the first function in this list where the viscous oil give greater frictional resistance and its slow internal circulation leads to a rapid temperature raise of some vital part of the engine to cool efficiency, the oil must be able to circulated quickly.

#### **5. Removal of all injurious impurities:**

The lubricant give the function of protecting the engine against corrosive and mechanical wear which caused by all injurious impurities. So, the removal of these impurities by lubricants is very important for engine. The function and the corresponding qualities required for engine lubricating oils are summarized in Table (1).

#### *1.3.4. Types of lubricants*

#### *1.3.4.1. Gaseous lubricants*

Gaseous lubricants belong to the simplest, lowest viscosity lubricants known and include air, nitrogen, oxygen, and helium. They are applied in aerodynamic and aerostatic bearings. Since the chemical properties and the aggregate state of most gases remain unchanged over a wide temperature range, gaseous lubricants offer several advantages over liquid lubricants. First, they can be applied at both very high and very low temperatures. Their chemical stability eliminates any risk of contamination of the bearing by the lubricant, important for the machinery used in many branches of industry, primarily in the food, pharmaceutical and electronic industries.

A useful property of gases is that their viscosities increase with temperature, wheras the opposite is true of liquids, resulting in load – carrying capacity of gas – lubricated bearings


straight mineral oil raffinates of various viscosity grades. The viscosity grade required is dependent upon bearing speed, oil temperature and load. table (2) provides a general guideline to selecting the correct ISO viscosity grade. The ISO grade number indicated is the preferred grade for the speed and temperature range. ISO 68- and 100- Grade oils are commonly used in indoor, heated applications, with32- grade oils being used for high-speed, 10.000 rpm, units and some outdoor low temperature applications. The higher the bearing speed, the lower the oil viscosity required and also that the higher the unit operating temperature, the higher the oil viscosity required. If vibration or minor shock loading is possible, a higher grade of oil than

Lubrication and Lubricants http://dx.doi.org/10.5772/56043 71

**Bearing Speed (rpm) Bearing / Oil Temperature (oC)** 0-50 60 75 90

300-1,500 - 68 100-150 -

1,800 32 32 - 46 68 -100 100

3,600 32 32 46 - 68 68 -100

10,000 32 32 32 32-46

Other methods for determining the viscosity grade required in an application are to apply minimum and optimum viscosity criteria to a viscosity – temperature plot. A third and more complex method is to calculate the oil viscosity needed to obtain a satisfactory oil film

The lubrication of bearings for machine tools usually requires mineral oils of ISO VG 46 or 68. for fast – running grinding spindles with plain bearings, mineral oils of ISO VG 5 or 7 are required, dependent on bearing clearance and speed. Bearings operating under high loads need lubricants of ISO VG 68 or 100. the service life of the bearing can be increased if the viscosity of the selected liquid lubricant at operating temperature exceeds the calculated

On the other hand, increased viscosity also increases operating temperature. In practice, therefore, the extent to which lubrication can be improved in this way is often limited. The chemical compositions of these oils differ from typical base oils in that they contain somewhat more aromatic hydrocarbons and heterocyclic compounds, which act as natural oxidation inhibitors. An increased viscosity for oils derived from the same crude oil does not significantly change their chemical composition; the difference generally lies with the increasing chain length of the paraffinic hydrocarbons, mostly isoparaffins, and in the aliphatic substituents of naphthenic and aromatic rings, together with a slight increase in the number of naphthenic and aromatic rings. More highly refined mineral oils and oxidation inhibitors are used for

the one indicated in table (2) should be considered.

**Table 2.** Plain Bearing ISO viscosity grade selection

thickness.

optimum viscosity.

**Table 1.** Function and qualities required for engine oils.

increasing with temperature. However, the relatively low viscosity of gases generally limits the load-carrying capacity of self-acting, aerodynamic bearings to 15-20kPa. It is possible to achieve better bearing performance with gaseous lubricants than with liquid lubricants due to the very low viscosity of the gases which results in smaller heat generation by internal friction. In some cases, such as in foil air bearings, sliding contact occurs during stops and starts [14], therefore solid lubricants such as PTFE are used to reduce friction.

#### *1.3.4.2. Liquid lubricants*

**Mineral oils:** As the hydrodynamic behaviour of plain bearings of plain bearings is totally dependent on the viscosity characteristics of the lubricant, typical liquid bearing lubricants are straight mineral oil raffinates of various viscosity grades. The viscosity grade required is dependent upon bearing speed, oil temperature and load. table (2) provides a general guideline to selecting the correct ISO viscosity grade. The ISO grade number indicated is the preferred grade for the speed and temperature range. ISO 68- and 100- Grade oils are commonly used in indoor, heated applications, with32- grade oils being used for high-speed, 10.000 rpm, units and some outdoor low temperature applications. The higher the bearing speed, the lower the oil viscosity required and also that the higher the unit operating temperature, the higher the oil viscosity required. If vibration or minor shock loading is possible, a higher grade of oil than the one indicated in table (2) should be considered.


**Table 2.** Plain Bearing ISO viscosity grade selection

increasing with temperature. However, the relatively low viscosity of gases generally limits the load-carrying capacity of self-acting, aerodynamic bearings to 15-20kPa. It is possible to achieve better bearing performance with gaseous lubricants than with liquid lubricants due to the very low viscosity of the gases which results in smaller heat generation by internal friction. In some cases, such as in foil air bearings, sliding contact occurs during stops and

and physical and chemical condition.

Reduce frictional resistance • Viscosity not too high to provide good pumpability or to cause undue cracking resistance.

Protect against corrosion and wear • Must protect metallic surface against corrosive action of fuel

stability).

outside the hydrodynamic condition.

• Minimum viscosity without risk of metal to metal contact under the

• Sufficiently high viscosity a high temperature; good lubrication property

varying condition of temperatures, speed and load.

decomposition product (wear, So2, HBr, HCl, etec.)

high temperatures, especially on non-ferrous metals.

of unavoidable metal-to-metal contact.

(detergency or dispersancy action).

(dispersancy action). Assist sealing • Must have sufficient viscosity at high temperature and low volatility. • Must limit wear.

> • Must have low volatility. • Viscosity must not be too high.

formation. Contribute to cooling • Must good and thermal stability and oxidation resistance.

• Anti-seizure properties, especially during the run-in period.

• Must resist degradation (resist oxidation and have a good thermal

• Must resist deposit formations which would affect lubrication

• Must contribute to the elimination of dust and other pollutants

• Must not contribute to formation of deposits and fight against such

• Must be able to maintain in fine solid material whatever the temperature

• Must counteract action of fuel and lubricant decomposition product at

• By intervention in the friction mechanism, must reduce the consequences

**Mineral oils:** As the hydrodynamic behaviour of plain bearings of plain bearings is totally dependent on the viscosity characteristics of the lubricant, typical liquid bearing lubricants are

starts [14], therefore solid lubricants such as PTFE are used to reduce friction.

**Main functions required Qualities required**

70 Tribology - Fundamentals and Advancements

*1.3.4.2. Liquid lubricants*

Facilitate the suspension and eliminate undesirable products

**Table 1.** Function and qualities required for engine oils.

Other methods for determining the viscosity grade required in an application are to apply minimum and optimum viscosity criteria to a viscosity – temperature plot. A third and more complex method is to calculate the oil viscosity needed to obtain a satisfactory oil film thickness.

The lubrication of bearings for machine tools usually requires mineral oils of ISO VG 46 or 68. for fast – running grinding spindles with plain bearings, mineral oils of ISO VG 5 or 7 are required, dependent on bearing clearance and speed. Bearings operating under high loads need lubricants of ISO VG 68 or 100. the service life of the bearing can be increased if the viscosity of the selected liquid lubricant at operating temperature exceeds the calculated optimum viscosity.

On the other hand, increased viscosity also increases operating temperature. In practice, therefore, the extent to which lubrication can be improved in this way is often limited. The chemical compositions of these oils differ from typical base oils in that they contain somewhat more aromatic hydrocarbons and heterocyclic compounds, which act as natural oxidation inhibitors. An increased viscosity for oils derived from the same crude oil does not significantly change their chemical composition; the difference generally lies with the increasing chain length of the paraffinic hydrocarbons, mostly isoparaffins, and in the aliphatic substituents of naphthenic and aromatic rings, together with a slight increase in the number of naphthenic and aromatic rings. More highly refined mineral oils and oxidation inhibitors are used for applications where higher temperatures or longer service periods require better ageing stabilites.

materials and allows for the elimination of grease or oil that would evaporate, congeal or solidify, causing premature failure. The graphite matrix can be filled with a variety of embed‐ ded lubricants to enhance chemical, mechanical and tribological properties to give a constant, low friction coefficient rather than just a surface layer, helping to protect against catastrophic failure. Lubrication is maintained during linear motion where lubricant is not frawn out and

Lubrication and Lubricants http://dx.doi.org/10.5772/56043 73

A recent development in solid bearing lubricants is micro –porous polymeric lubricants, MPL, where a polymer containing a continous microporous network has oil contained within the pores, which may include appropriate additives [14]. The oil content in the polymer can be more than 50% by weight and the microporous polymer acts as a sponage, releasing and

Water content (ASTM D95, D1744, D1533, and D96) is the amount of water present in the lubricant. It can be expressed as parts per million, percent by volume or percent by weight. It can be measured by centrifuging, distillation and voltametry. The most popular, although least accurate, method of water content assessment is the centrifuge test. In this method a 50% mixture of oil and solvent is centrifuged at a specified speed until the volumes of water and sediment observed are stable. Apart from water, solids and other solubles are also separated and the results obtained do not correlate well with those obtained by the other two methods. The distillation method is a little more accurate and involves distillation of oil mixed with xylene. Any water present in the sample condenses in a graduated receiver. Voltametry method is the most accurate. It employs electrometric titration, giving the water concentration

Corrosion and oxidation behaviour of lubricants is critically related to water content. An oil mixed with water gives an emulsion. An emulsion has a much lower load carrying capacity than pure oil and lubricant failure followed by damage to the operating surfaces can result. In general, in applications such as turbine oil systems, the limit on water content is 0.2% and for hydraulic systems 0.1%. In dielectric systems excessive water content has a significant effect on dielectric breakdown. Usually the water content in such systems should be kept below 35

Sulphur content (ASTM D1266, D129, D1662) is the amount of sulphur present in an oil. It can have some beneficial, as well as some detrimental, effects on operating machinery. Sulphur is a very good boundary agent, which can effectively operate under extreme conditions of pressure and temperature. On the other hand, it is very corrosive. A commonly used technique for the determination of sulphur content is the bomb oxidation technique. It involves the ignition and combustion of a small oil sample under pressurised oxygen. The sulphur from

dust is not pulled in.

**• Water Content**

in parts per million.

**• Sulphur Content**

the products of combustion is extracted and weighed.

[ppm].

absorbing oil when necessary.

*1.3.5. Lubricant impurities and contaminants*

**Synthetic lubricants:** in practice, every synthetic oil of adequate viscosity and good viscositytemperature behavior can be used as a bearing lubricant, e.g. polyglycols are very good bearing lubricants for mills and calenders in the rubber, plastics, textile and paper industries. However, in most cases the synthetic oils specifically developed for lubricating particular equipment are also used to lubricate its bearings. Although synthetic oils do not form a lubricant film under pressure as well as mineral oils and may not be effective bearing lubricants despite their higher temperature viscosity.

**Biodegradable products:** Biodegradable products of vegetable or animal origin are also considered for liquid lubrication, e.g. the effects of sunflower oil added to base oil on the performance of journal bearings. The use of vegetable oils as lubricants is likely to increase due to environmental and government requirements and is becoming increasingly important.

#### *1.3.4.3. Solid lubricants*

General description: bearings used under vacuum, at very high temperatures or under very high radiation cannot be lubricated by liquid lubricants or greases. For these and many other cases, solid lubricants are used, deemed to be any solid material used to reduce friction and wear between two moving surfaces.

In general, the solid material is interposed as a film between sliding and /or rolling surfaces. Simply stated, an adequate solid material is required for the special lubrication requirements of extreme operating conditions, such as very high or very low temperatures over a wide range, e.g. -200 to 850o C, and corrosive atmospheres. Such materials normally have a layered crystalline structure which ensures low shear strength, thereby minimizing friction. The shear strength between the crystalline layers is weak and sets up a low and sets up a low friction mechanism by slippage of the crystalline layers under low shearing forces. Examples of layerlattice solids are molybdenum disulphide, graphite, boron nitride, cadmium iodide and borax. Solid lubricants are used mainly in the form of powders or as bonded solid films.

A good solid film lubricant has strong adhesion to the bearing substrate material, full surface coverage and good malleability. It should also be chemically stable and prevent corrosion, taking into account operational and environmental conditions. Many solid film lubricants have poor wear resistance, since any breaks in the film are not self-healing, in contrast to the surface coating formed by a liquid lubricant. Advanced solid film lubricants perform reliably in many specific applications and much experience has been gained to better understand their limita‐ tions. The most commonly used disulphide, graphite, polytetrafluroethylene propylene.

Another group of materials, the self-lubricated materials, are related to solid lubricants and are particulary important for bearings. Their self-lubricating characteristics eliminate the need of grease or other lubrication and gove improved performance under high temperature conditions. Graphalloy (Graphite/matal) alloys make use of special properties of graphite, the structure of which can be compared to a deck of cards with individual layers able to easily slide off. This phenomenon gives the material a self lubricating ability matched by few other materials and allows for the elimination of grease or oil that would evaporate, congeal or solidify, causing premature failure. The graphite matrix can be filled with a variety of embed‐ ded lubricants to enhance chemical, mechanical and tribological properties to give a constant, low friction coefficient rather than just a surface layer, helping to protect against catastrophic failure. Lubrication is maintained during linear motion where lubricant is not frawn out and dust is not pulled in.

A recent development in solid bearing lubricants is micro –porous polymeric lubricants, MPL, where a polymer containing a continous microporous network has oil contained within the pores, which may include appropriate additives [14]. The oil content in the polymer can be more than 50% by weight and the microporous polymer acts as a sponage, releasing and absorbing oil when necessary.

#### *1.3.5. Lubricant impurities and contaminants*

#### **• Water Content**

applications where higher temperatures or longer service periods require better ageing

**Synthetic lubricants:** in practice, every synthetic oil of adequate viscosity and good viscositytemperature behavior can be used as a bearing lubricant, e.g. polyglycols are very good bearing lubricants for mills and calenders in the rubber, plastics, textile and paper industries. However, in most cases the synthetic oils specifically developed for lubricating particular equipment are also used to lubricate its bearings. Although synthetic oils do not form a lubricant film under pressure as well as mineral oils and may not be effective bearing lubricants despite their higher

**Biodegradable products:** Biodegradable products of vegetable or animal origin are also considered for liquid lubrication, e.g. the effects of sunflower oil added to base oil on the performance of journal bearings. The use of vegetable oils as lubricants is likely to increase due to environmental and government requirements and is becoming increasingly important.

General description: bearings used under vacuum, at very high temperatures or under very high radiation cannot be lubricated by liquid lubricants or greases. For these and many other cases, solid lubricants are used, deemed to be any solid material used to reduce friction and

In general, the solid material is interposed as a film between sliding and /or rolling surfaces. Simply stated, an adequate solid material is required for the special lubrication requirements of extreme operating conditions, such as very high or very low temperatures over a wide range,

crystalline structure which ensures low shear strength, thereby minimizing friction. The shear strength between the crystalline layers is weak and sets up a low and sets up a low friction mechanism by slippage of the crystalline layers under low shearing forces. Examples of layerlattice solids are molybdenum disulphide, graphite, boron nitride, cadmium iodide and borax.

A good solid film lubricant has strong adhesion to the bearing substrate material, full surface coverage and good malleability. It should also be chemically stable and prevent corrosion, taking into account operational and environmental conditions. Many solid film lubricants have poor wear resistance, since any breaks in the film are not self-healing, in contrast to the surface coating formed by a liquid lubricant. Advanced solid film lubricants perform reliably in many specific applications and much experience has been gained to better understand their limita‐ tions. The most commonly used disulphide, graphite, polytetrafluroethylene propylene.

Another group of materials, the self-lubricated materials, are related to solid lubricants and are particulary important for bearings. Their self-lubricating characteristics eliminate the need of grease or other lubrication and gove improved performance under high temperature conditions. Graphalloy (Graphite/matal) alloys make use of special properties of graphite, the structure of which can be compared to a deck of cards with individual layers able to easily slide off. This phenomenon gives the material a self lubricating ability matched by few other

Solid lubricants are used mainly in the form of powders or as bonded solid films.

C, and corrosive atmospheres. Such materials normally have a layered

stabilites.

temperature viscosity.

72 Tribology - Fundamentals and Advancements

*1.3.4.3. Solid lubricants*

e.g. -200 to 850o

wear between two moving surfaces.

Water content (ASTM D95, D1744, D1533, and D96) is the amount of water present in the lubricant. It can be expressed as parts per million, percent by volume or percent by weight. It can be measured by centrifuging, distillation and voltametry. The most popular, although least accurate, method of water content assessment is the centrifuge test. In this method a 50% mixture of oil and solvent is centrifuged at a specified speed until the volumes of water and sediment observed are stable. Apart from water, solids and other solubles are also separated and the results obtained do not correlate well with those obtained by the other two methods. The distillation method is a little more accurate and involves distillation of oil mixed with xylene. Any water present in the sample condenses in a graduated receiver. Voltametry method is the most accurate. It employs electrometric titration, giving the water concentration in parts per million.

Corrosion and oxidation behaviour of lubricants is critically related to water content. An oil mixed with water gives an emulsion. An emulsion has a much lower load carrying capacity than pure oil and lubricant failure followed by damage to the operating surfaces can result. In general, in applications such as turbine oil systems, the limit on water content is 0.2% and for hydraulic systems 0.1%. In dielectric systems excessive water content has a significant effect on dielectric breakdown. Usually the water content in such systems should be kept below 35 [ppm].

#### **• Sulphur Content**

Sulphur content (ASTM D1266, D129, D1662) is the amount of sulphur present in an oil. It can have some beneficial, as well as some detrimental, effects on operating machinery. Sulphur is a very good boundary agent, which can effectively operate under extreme conditions of pressure and temperature. On the other hand, it is very corrosive. A commonly used technique for the determination of sulphur content is the bomb oxidation technique. It involves the ignition and combustion of a small oil sample under pressurised oxygen. The sulphur from the products of combustion is extracted and weighed.

#### **• Ash Content**

There is some quantity of noncombustible material present in a lubricant which can be determined by measuring the amount of ash remaining after combustion of the oil (ASTM D482, D874). The contaminants may be wear products, solid decomposition products from a fuel or lubricant, atmospheric dust entering through a filter, etc. Some of these contaminants are removed by an oil filter but some settle into the oil. To determine the amount of contami‐ nant, the oil sample is burned in a specially designed vessel. The residue that remains is then ashed in a high temperature muffle furnace and the result displayed as a percentage of the original sample. The ash content is used as a means of monitoring oils for undesirable impurities and sometimes additives. In used oils it can also indicate contaminants such as dirt, wear products, etc.

range of possible applications while gaseous and solid lubricants are recommended in

Lubrication and Lubricants http://dx.doi.org/10.5772/56043 75

**9.** Additives are chemical compounds added to lubricating oils to impart specific properties to the finished oils. Some additives impart new and useful properties to the lubricant; some enhance properties already present, while some act to reduce the rate at which

Additives lab., Department of Petroleum Applications, Egyptian Petroleum Research Insti‐

[1] Rudnick Leslie R., Ewa A. Bardasz, and Gordon D. Lamb; "Lubricant Additives:

[2] Stachowiak Gwidon W.,and Andrew W. Batchelor; Engineering Tribology", third ed‐

[4] Dowson D.; History of Tribology, 2nd Edition, Professional Engineering Publishing,

[5] Pirro D.M. and Wessol A.A.; "Lubrication fundamentals"; Marcel Dekker, Inc. New

[7] Stachowiak G. W. and Batchelor A.w.; Engineering Tribology, 2nd Edition, Butter‐

[9] Avilino S.Jr.,"Lubricant Base Oil and Wax Processing", Morcel Dekker,Inc.,New

[10] Mortier R.M. and Orszulik S.T.; "Chemistry and Technology of Lubricant", Blackie

[8] Pawlak Z., "Tribochemistry of Lubricating Oils"; Elsevier, UK, 45, 17 (2003).

Academic and Professional Publications, Chapter (1), pp.2-12,(1993).

Chemistry and Applications", Marcel Dekker, pages 387-427, (2003).

ition, Amsterdam: Elsevier, pages 2,12,-22,52,62-67,77, (2005).

[3] Geore J.W.; "Lubrication Fundamentals", (1980).

[6] Spikes H.; Tribology International 34, 789 (2001).

special applications.

**Author details**

**References**

Nehal S. Ahmed and Amal M. Nassar

tute, Nasr City, Cairo, Egypt

London (1998).

York and Basel 3, 37 (2001).

worth-Heinemann, Boston,(2001).

York, Chapter (2),pp.17-36,(1994).

**8.** Lubricants do not persist working without additives.

undesirable changes take place in product during its service.

#### **• Chlorine Content**

The amount of chlorine in a lubricant should be at an optimum level. Excess chlorine causes corrosion whereas an insufficient amount of chlorine may cause wear and frictional losses to increase. Chlorine content (ASTM D808, D1317) can be determined either by a bomb test which provides the gravimetric evaluation or by a volumetric test which gives chlorine content, after reacting with sodium metal to produce sodium chloride, then titrating with silver nitride [14].

#### **2. Conclusion**


range of possible applications while gaseous and solid lubricants are recommended in special applications.


#### **Author details**

**• Ash Content**

74 Tribology - Fundamentals and Advancements

wear products, etc. **• Chlorine Content**

**2. Conclusion**

lard.

There is some quantity of noncombustible material present in a lubricant which can be determined by measuring the amount of ash remaining after combustion of the oil (ASTM D482, D874). The contaminants may be wear products, solid decomposition products from a fuel or lubricant, atmospheric dust entering through a filter, etc. Some of these contaminants are removed by an oil filter but some settle into the oil. To determine the amount of contami‐ nant, the oil sample is burned in a specially designed vessel. The residue that remains is then ashed in a high temperature muffle furnace and the result displayed as a percentage of the original sample. The ash content is used as a means of monitoring oils for undesirable impurities and sometimes additives. In used oils it can also indicate contaminants such as dirt,

The amount of chlorine in a lubricant should be at an optimum level. Excess chlorine causes corrosion whereas an insufficient amount of chlorine may cause wear and frictional losses to increase. Chlorine content (ASTM D808, D1317) can be determined either by a bomb test which provides the gravimetric evaluation or by a volumetric test which gives chlorine content, after reacting with sodium metal to produce sodium chloride, then titrating with silver nitride [14].

**1.** The technology of lubrication has been used from the ancient times, from the pyramid

**2.** The main purpose of lubrication is to reduce friction and wear in bearings or sliding

**3.** Adequate lubrication also helps to prevent foreign material from entering the bearings and guards against corrosion and rusting. Satisfactory bearing performance can be achieved by adopting the lubricating method that is most suitable for the particular

**4.** A lubricant prevents the direct contact of rubbing surfaces and thus reduces wear. It keeps the surface of metals clean.Lubricants can also act as coolants by removing heat effects

**5.** lubricant is consisting of either oil or grease. Most grease is from animal fats or vegetable

**6.** Lubricating oils are made from the more viscous portion of the crude oil which remains

**7.** There are three major types of lubricants: Gaseous lubricants e.g. air, helium, Liquid lubricants e.g. oils, water and Solid lubricants e.g. graphite, grease, teflon, molybdenum disulphide etc. Liquid lubricant is the most commonly used lubricant because of its wide

and also prevent rusting and deposition of solids on close fitting parts.

after removal by distillation of the gas oil and lighter fraction

building where massive rock slabs are moved, up to present modern times.

components to prevent premature failure.

application and operating conditions

Nehal S. Ahmed and Amal M. Nassar

Additives lab., Department of Petroleum Applications, Egyptian Petroleum Research Insti‐ tute, Nasr City, Cairo, Egypt

#### **References**


[11] O'Connar J.J., Boyd J. and Auallane E.A.; "Standard Hand Book of Lubrication Engi‐ neering", McGrow Hill, New York, 14-2 (1968).

**Chapter 3**

**Some Aspects of Grease Flow in**

Additional information is available at the end of the chapter

Maciej Paszkowski

**1. Introduction**

http://dx.doi.org/10.5772/55929

**Lubrication Systems and Friction Nodes**

lead to a considerable increase in the opearting costs of the machine.

An optimally designed lubrication system should reliably distribute a lubricant to particular reception points. The distribution must be precise and preferably fully automated. It is particularly difficult to design such a system if it is to supply a lubricant to heavily loaded machines, featuring a considerable (even up to a few hundred) number of different kinds of friction nodes distributed in a non-linear way, with long distances between each other. An additional problem might be hard conditions of the environment in which the system is expected to work (high or low temperature, high air humidity, etc.). All the conditions make the task of the designing of a reliable central lubrication system a real challenge, even for an experienced designer specialising in that particular field. While building a lubrication system it is important to define fundamental parameters which will determine its reliable operation. One of the crucial things is to select an appropriate grease that ought to reduce friction resistance and wear of the friction nodes, to protect them against the influence of the environ‐ ment, as well as to guarantee the lowest possible flow resistance during its distribution (mainly through the lubricating conduits). An inappropriately selected grease, in terms of its rheolog‐ ical and tribological properties, and also the dynamically changing working conditions, may

The chapter contains the most important information on the structure of greases along with the discussion of the influence of the thickener's microstructure on the behaviour of the lubrication formula mainly in lubrication systems, but also in reception points, namely in the friction nodes (including roller bearings). It also refers to the problems of the influence of the mechanical stability of a thickener's microstructure on the quality of lubricating the roller bearings and their service life. Additionally, the chapter presents works on generating the lubrication film on the friction nodes working surface. Also, the fundamental problems

> © 2013 Paszkowski; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,

© 2013 Paszkowski; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

distribution, and reproduction in any medium, provided the original work is properly cited.


### **Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes**

Maciej Paszkowski

[11] O'Connar J.J., Boyd J. and Auallane E.A.; "Standard Hand Book of Lubrication Engi‐

[12] Allyson M., Keith D., Vincent R. and Thibon A.; Tribology International 34, 389-395

[14] Roy M. M., Malcolm F. F., and Stefan T. Orszulik; Chemistry and Technology of Lu‐

neering", McGrow Hill, New York, 14-2 (1968).

[13] Anon, Machinery and Production, 19 July 24 (1996).

bricants, 3rd Edition, 12-13,( 2010).

(2001).

76 Tribology - Fundamentals and Advancements

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/55929

#### **1. Introduction**

An optimally designed lubrication system should reliably distribute a lubricant to particular reception points. The distribution must be precise and preferably fully automated. It is particularly difficult to design such a system if it is to supply a lubricant to heavily loaded machines, featuring a considerable (even up to a few hundred) number of different kinds of friction nodes distributed in a non-linear way, with long distances between each other. An additional problem might be hard conditions of the environment in which the system is expected to work (high or low temperature, high air humidity, etc.). All the conditions make the task of the designing of a reliable central lubrication system a real challenge, even for an experienced designer specialising in that particular field. While building a lubrication system it is important to define fundamental parameters which will determine its reliable operation. One of the crucial things is to select an appropriate grease that ought to reduce friction resistance and wear of the friction nodes, to protect them against the influence of the environ‐ ment, as well as to guarantee the lowest possible flow resistance during its distribution (mainly through the lubricating conduits). An inappropriately selected grease, in terms of its rheolog‐ ical and tribological properties, and also the dynamically changing working conditions, may lead to a considerable increase in the opearting costs of the machine.

The chapter contains the most important information on the structure of greases along with the discussion of the influence of the thickener's microstructure on the behaviour of the lubrication formula mainly in lubrication systems, but also in reception points, namely in the friction nodes (including roller bearings). It also refers to the problems of the influence of the mechanical stability of a thickener's microstructure on the quality of lubricating the roller bearings and their service life. Additionally, the chapter presents works on generating the lubrication film on the friction nodes working surface. Also, the fundamental problems

© 2013 Paszkowski; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Paszkowski; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

connected with the grease rheology (the grease performance in the fixed flow rate conditions and the notion of the linear and the non-linear viscoelasticity) have been discussed in the text. The state of the art knowledge on the fundamental phenomena observed in the lubrication systems, including the mechanisms of the thixotropic changes in the grease microstructure at shearing and relaxation, as well as the forming of the boundary layer at the grease flowing through the conduits in the main line of the lubrication system have been presented.

exceed the size of 1 µm, their structure is rough and resembles twisted ropes. The crystallites of the colloidal size are definitely smoother and less twisted [3]. The microscope photographs also show a clear difference in the surface structure of the lithium, sodium and calcium thickener floccules [7,9]. The calcium soap particles are rougher than the particles of the lithium and sodium soaps, independently of their length [5]. In the case of the non-organic thickeners, the grease microstructure is made of numerous individual aggregates popularly called *open card-house* [10], for instance bentonite (Figure 2), mica or vermiculite. Such particles have a skeleton structure which resembles heterogeneous, fuzzy-edged, curled flakes or straight

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

79

The shape of the soap particles and other organic or non-organic thickeners, their anisometry (in reference to their length and lateral dimensions), as well as their dispersivity and the percentage in the full volume of the grease, greatly influence the physical properties of the ultimate lubricating compound [7,12-14]. The last two factors are critical for the easiness of the making of different energy connection types between the elements of the microstructure. Apart from providing the appropriate consistency, thickeners influence the way the lubricating grease flows, changes its shape or the type of flow resistance (pumpability) it presents. It is particularly important in the central lubrication systems where the lubricant is often trans‐

**Figure 1.** Floccules of the lithium thickener with particles of polytetrafluoroethylene (PTFE) improving the tribological properties of the lubricant. The microphotograph was taken by means of a scanning electron microscope (SEM) in sec‐

plates piled one on the top of the other [11].

ported in long conduits to particular receiving points.

ondary electron imaging (SEI) mode at the magnification of 12 500 times.

#### **2. Microstructure of the lubricating greases**

Lubricating greases are rheologically complex two-phase non-Newtonian fluids. They are chemically and physically heterogeneous. The dispersive phase is normally a mineral oil, a synthetic oil or a vegetable oil, whereas the dispersed phase is a thickener and, depending on the needs, solid additives. The particles of the thickeners can vary in their dimensions. Soaps, for instance, do not normally exceed 100 µm in length, and their diameter is not shorter than 0.1 µm and not longer than 0.5 µm [1-3]. The lithium and calcium soaps particles are definitely bigger than the sodium soaps particles. The isometric aggregates of bentonite clay and mica are approximately 0.5 µm in width and 0.1 µm in thickness [1]. The solid additives have similar dimensions. Due to the size of the thickener particles, the greases acquire the characteristics of the mechanically dispersed (suspension) or the colloidal system. The thickener particles size depends primarily on the process of the grease production [4,5], as well as on the conditions in the friction node. During the shearing of a grease in the friction node, the size of the particles can change considerably. According to Boner [5], in order to achieve the optimum tribological and rheological properties of a lubricating compound, greases containing variously dispersed thickener particles ought to be mixed. It concerns particularly the lubricants thickened with metal soaps. Too high particles dispersion of the thickener in the grease can negatively influence its lubricating properties. Such particles do not present enough capability of making spatial, three-dimensional structures resulting from the physicochemical interactions. Too long particles of the thickener cause too high an increase of the consistency of the lubricating grease and lead to its easy breaking both in the lubrication systems and the tribological pairs.

Microstructure of the lubricating greases with soap thickeners can be compared to a sponge with a lubricating oil (Figure 1). It makes a three-dimensional, coherent network of intercon‐ nected particles (flow units), in the literature referred to as *floccules*. The oil is locked in the free spaces of the microstructure through the mechanical occlusion, the capillary phenomena as well as the molecular attraction between the thickener and the polar components of oil [6]. It is estimated that the amount of the oil locked in the microstructure of the soap greases can amount even to 75%. The soap particles making the microstructure, from the chemical viewpoint, are associated molecules [7], namely groups of identical molecules generated as a result of the dipol-dipol type of interaction or the hydrogen bonds [8]. The final skeleton of the microstructure – shaped in the dispersion center – is made in situ in the process of crystallization of the soap particles and/or through nucleation, namely making crystallite nucleuses and their further growth [5]. The shape of the thickener particles and their surface topography can be different, depending on the kind of soap used and the particles' size. If they exceed the size of 1 µm, their structure is rough and resembles twisted ropes. The crystallites of the colloidal size are definitely smoother and less twisted [3]. The microscope photographs also show a clear difference in the surface structure of the lithium, sodium and calcium thickener floccules [7,9]. The calcium soap particles are rougher than the particles of the lithium and sodium soaps, independently of their length [5]. In the case of the non-organic thickeners, the grease microstructure is made of numerous individual aggregates popularly called *open card-house* [10], for instance bentonite (Figure 2), mica or vermiculite. Such particles have a skeleton structure which resembles heterogeneous, fuzzy-edged, curled flakes or straight plates piled one on the top of the other [11].

connected with the grease rheology (the grease performance in the fixed flow rate conditions and the notion of the linear and the non-linear viscoelasticity) have been discussed in the text. The state of the art knowledge on the fundamental phenomena observed in the lubrication systems, including the mechanisms of the thixotropic changes in the grease microstructure at shearing and relaxation, as well as the forming of the boundary layer at the grease flowing

Lubricating greases are rheologically complex two-phase non-Newtonian fluids. They are chemically and physically heterogeneous. The dispersive phase is normally a mineral oil, a synthetic oil or a vegetable oil, whereas the dispersed phase is a thickener and, depending on the needs, solid additives. The particles of the thickeners can vary in their dimensions. Soaps, for instance, do not normally exceed 100 µm in length, and their diameter is not shorter than 0.1 µm and not longer than 0.5 µm [1-3]. The lithium and calcium soaps particles are definitely bigger than the sodium soaps particles. The isometric aggregates of bentonite clay and mica are approximately 0.5 µm in width and 0.1 µm in thickness [1]. The solid additives have similar dimensions. Due to the size of the thickener particles, the greases acquire the characteristics of the mechanically dispersed (suspension) or the colloidal system. The thickener particles size depends primarily on the process of the grease production [4,5], as well as on the conditions in the friction node. During the shearing of a grease in the friction node, the size of the particles can change considerably. According to Boner [5], in order to achieve the optimum tribological and rheological properties of a lubricating compound, greases containing variously dispersed thickener particles ought to be mixed. It concerns particularly the lubricants thickened with metal soaps. Too high particles dispersion of the thickener in the grease can negatively influence its lubricating properties. Such particles do not present enough capability of making spatial, three-dimensional structures resulting from the physicochemical interactions. Too long particles of the thickener cause too high an increase of the consistency of the lubricating grease and lead to its easy breaking both in the lubrication systems and the tribological pairs. Microstructure of the lubricating greases with soap thickeners can be compared to a sponge with a lubricating oil (Figure 1). It makes a three-dimensional, coherent network of intercon‐ nected particles (flow units), in the literature referred to as *floccules*. The oil is locked in the free spaces of the microstructure through the mechanical occlusion, the capillary phenomena as well as the molecular attraction between the thickener and the polar components of oil [6]. It is estimated that the amount of the oil locked in the microstructure of the soap greases can amount even to 75%. The soap particles making the microstructure, from the chemical viewpoint, are associated molecules [7], namely groups of identical molecules generated as a result of the dipol-dipol type of interaction or the hydrogen bonds [8]. The final skeleton of the microstructure – shaped in the dispersion center – is made in situ in the process of crystallization of the soap particles and/or through nucleation, namely making crystallite nucleuses and their further growth [5]. The shape of the thickener particles and their surface topography can be different, depending on the kind of soap used and the particles' size. If they

through the conduits in the main line of the lubrication system have been presented.

**2. Microstructure of the lubricating greases**

78 Tribology - Fundamentals and Advancements

The shape of the soap particles and other organic or non-organic thickeners, their anisometry (in reference to their length and lateral dimensions), as well as their dispersivity and the percentage in the full volume of the grease, greatly influence the physical properties of the ultimate lubricating compound [7,12-14]. The last two factors are critical for the easiness of the making of different energy connection types between the elements of the microstructure. Apart from providing the appropriate consistency, thickeners influence the way the lubricating grease flows, changes its shape or the type of flow resistance (pumpability) it presents. It is particularly important in the central lubrication systems where the lubricant is often trans‐ ported in long conduits to particular receiving points.

**Figure 1.** Floccules of the lithium thickener with particles of polytetrafluoroethylene (PTFE) improving the tribological properties of the lubricant. The microphotograph was taken by means of a scanning electron microscope (SEM) in sec‐ ondary electron imaging (SEI) mode at the magnification of 12 500 times.

nantly by the degradation of the thickener's microstructure and worsening of the grease's properties. Vinogradov et al. proved that interactions between the thickener particles and the microstructure state may influence the increase of the friction moment in the roller bearing [18]. The thickener in the lubricating grease is responsible also for the thickness of the lubrication film on the working surfaces of the friction joints. Spikes, Cann, Wiliamson and Kendal in their works [19,20] presented results of the research on the interaction of the thickener's particles with the treadmill sprinkled with titanium oxide. They observed an increase in the thickness of the lubricating film in the elastohydrodynamic conditions of the collaboration of a steel ball with the treadmill. The measurement was done by means of Alström's method. In the case of the lubricants thickened with metal soaps, the biggest capability of making the lubrication film was recorded in the lithium grease. The thickness of the generated grease layer, apart from the kind of thickener, was also determined by the flow speed of the lubricant between the treadmill of the roller bearing and the rolling element. The thickness of the observed thickener's layer was 100 up to approximately 250 nm, depending on the kind of the lubricating grease. Longlasting tests showed that the lubrication film's thickness generated on the bearing's treadmill,

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

81

at a constant load, falls with time up to the stabilization at the level from 20 to 50 nm.

compounds [21,22].

generated lubrication compound.

Apart from the thickener, the lubricating grease microstructure can also contain solid addi‐ tives. Their main function is to improve the tribological properties of the lubricant. The solid additives are normally graphite, molybdenum disulfide (MoS2), polytetrafluoroethylene (PTFE) as well as metal powders. The percentage of the solid additives in the grease usually does not exceed 5%. Research results indicate the benefits of the increase of the percentage of the solid additives in the lubricating greases. Of course, there are also works which show the advantages of the synergism of the powdered PTFE as well as tin and copper in the lubricating

Solid additives significantly influence the rheological properties of the lubrication compound [23,24]. The lubricants thickened with the polar lithium soap without the solid additives show quite high values of the structural viscosity and of the shear stress. After the enriching the lubricants with the graphite thickeners and MoS2 the lowering of the parameters' values occurs. The additives reduce the shear stress both in the bulk of the grease as well as in the boundary layer. Research show that PTFE does not influence the change of the values of the parameters in the lithium soap. In the case of the lubricants thickened with the bentonite clay, solid additives cause the increase of the yield stress both close to the wall and in the bulk of the grease. The most influential determinant of the values increase is the graphite thickener. Adding the PTFE in this case does not significantly determine the structural changes thus

Greases, depending on their characteristic structure, are shear-thinning (pseudoplastic) and rheounstable fluids having a yield point. They show very complex rheological properties.

**3. The basic rheological properties of the lubricating greases**

**3.1. Behaviour of the lubricating greases in the steady flow conditions**

**Figure 2.** Bentonite clay aggregates making the *open card-house* structure.The microphotograph was taken by the scanning electron microscope (SEM) in secondary electron imaging (SEI) mode at the magnification of 12 000 times.

The mechanical durability and the stability of the microstructure of the thickener in a lubri‐ cating grease determines the friction reduction, the protection of the lubricated surfaces as well as the grease performance during the mechanical loading in the friction node and work at ultimately high temperatures. Cann [15] analyzed thin layers of the lithium lubricants without the solid additives on treadmill roller bearings with the use of attenuated total reflectance – Fourier transform infrared (ATR-FTIR) spectroscopy. The research showed that the oil gets out of the lubricated roller bearing as a result of damage in the thickener's microstructure. In the case of the untouched microstructure, the oil is connected with the thickener by means of the capillary forces. When the microstructure is damaged, it gets out of the bearing, washing out individual soap crystallites. Such an oil undergoes oxidation and evaporation easier, which lowers its lubrication quality. Additionally, its rheological properties are much worse than the properties of the base oil, and its kinematic viscosity is significantly reduced [16]. Similar conclusions were drawn by Farcas and Gafitanu [17]. Based on experimental research, Farcas and Gafitanu presented a mathematical model correlating the degree of the thickener micro‐ structure destruction with the service life of the roller bearing. Their SEM photographs of the lubricating greases subjected to shearing in the roller bearing showed a damaged microstruc‐ ture of the grease with the base oil bled out of it. According to the researchers, a crucial role in the process of the microstructure destruction is played by temperature. Above 60-70 o C, along with its increase by each 10-15 o C, the service life of the bearing falls by half. Above 70 o C, the bearing's failures are most frequently caused not by the contact fatigue, but predomi‐ nantly by the degradation of the thickener's microstructure and worsening of the grease's properties. Vinogradov et al. proved that interactions between the thickener particles and the microstructure state may influence the increase of the friction moment in the roller bearing [18].

The thickener in the lubricating grease is responsible also for the thickness of the lubrication film on the working surfaces of the friction joints. Spikes, Cann, Wiliamson and Kendal in their works [19,20] presented results of the research on the interaction of the thickener's particles with the treadmill sprinkled with titanium oxide. They observed an increase in the thickness of the lubricating film in the elastohydrodynamic conditions of the collaboration of a steel ball with the treadmill. The measurement was done by means of Alström's method. In the case of the lubricants thickened with metal soaps, the biggest capability of making the lubrication film was recorded in the lithium grease. The thickness of the generated grease layer, apart from the kind of thickener, was also determined by the flow speed of the lubricant between the treadmill of the roller bearing and the rolling element. The thickness of the observed thickener's layer was 100 up to approximately 250 nm, depending on the kind of the lubricating grease. Longlasting tests showed that the lubrication film's thickness generated on the bearing's treadmill, at a constant load, falls with time up to the stabilization at the level from 20 to 50 nm.

Apart from the thickener, the lubricating grease microstructure can also contain solid addi‐ tives. Their main function is to improve the tribological properties of the lubricant. The solid additives are normally graphite, molybdenum disulfide (MoS2), polytetrafluoroethylene (PTFE) as well as metal powders. The percentage of the solid additives in the grease usually does not exceed 5%. Research results indicate the benefits of the increase of the percentage of the solid additives in the lubricating greases. Of course, there are also works which show the advantages of the synergism of the powdered PTFE as well as tin and copper in the lubricating compounds [21,22].

**Figure 2.** Bentonite clay aggregates making the *open card-house* structure.The microphotograph was taken by the scanning electron microscope (SEM) in secondary electron imaging (SEI) mode at the magnification of 12 000 times.

The mechanical durability and the stability of the microstructure of the thickener in a lubri‐ cating grease determines the friction reduction, the protection of the lubricated surfaces as well as the grease performance during the mechanical loading in the friction node and work at ultimately high temperatures. Cann [15] analyzed thin layers of the lithium lubricants without the solid additives on treadmill roller bearings with the use of attenuated total reflectance – Fourier transform infrared (ATR-FTIR) spectroscopy. The research showed that the oil gets out of the lubricated roller bearing as a result of damage in the thickener's microstructure. In the case of the untouched microstructure, the oil is connected with the thickener by means of the capillary forces. When the microstructure is damaged, it gets out of the bearing, washing out individual soap crystallites. Such an oil undergoes oxidation and evaporation easier, which lowers its lubrication quality. Additionally, its rheological properties are much worse than the properties of the base oil, and its kinematic viscosity is significantly reduced [16]. Similar conclusions were drawn by Farcas and Gafitanu [17]. Based on experimental research, Farcas and Gafitanu presented a mathematical model correlating the degree of the thickener micro‐ structure destruction with the service life of the roller bearing. Their SEM photographs of the lubricating greases subjected to shearing in the roller bearing showed a damaged microstruc‐ ture of the grease with the base oil bled out of it. According to the researchers, a crucial role in the process of the microstructure destruction is played by temperature. Above 60-70 o

C, the bearing's failures are most frequently caused not by the contact fatigue, but predomi‐

along with its increase by each 10-15 o

80 Tribology - Fundamentals and Advancements

o

C,

C, the service life of the bearing falls by half. Above 70

Solid additives significantly influence the rheological properties of the lubrication compound [23,24]. The lubricants thickened with the polar lithium soap without the solid additives show quite high values of the structural viscosity and of the shear stress. After the enriching the lubricants with the graphite thickeners and MoS2 the lowering of the parameters' values occurs. The additives reduce the shear stress both in the bulk of the grease as well as in the boundary layer. Research show that PTFE does not influence the change of the values of the parameters in the lithium soap. In the case of the lubricants thickened with the bentonite clay, solid additives cause the increase of the yield stress both close to the wall and in the bulk of the grease. The most influential determinant of the values increase is the graphite thickener. Adding the PTFE in this case does not significantly determine the structural changes thus generated lubrication compound.

#### **3. The basic rheological properties of the lubricating greases**

#### **3.1. Behaviour of the lubricating greases in the steady flow conditions**

Greases, depending on their characteristic structure, are shear-thinning (pseudoplastic) and rheounstable fluids having a yield point. They show very complex rheological properties. Lubricating greases, at very low shear rates behave like a Newtonian fluid of a constant viscosity *η*0, being a slope of a tangent to the flow curve, equal to the shear rate going to zero (*η*<sup>0</sup> =lim*γ*˙ <sup>→</sup>0*η*(*γ*˙)). At the shear rate going to infinity, lubricating greases also behave like Newtonian fluids, but of viscosity *η∞*close to the dispersion phase, namely of the base oil (*η∞* =lim*γ*˙ <sup>→</sup>*∞η*(*γ*˙)). The viscosity *η∞* is a slope of a straight line being a flow curve asymptote. In the bracket of the indirect shear rates, the curve of the lubricating grease flow curve is described with the following equation:

$$
\eta = \frac{\pi}{\dot{\eta}} \,\tag{1}
$$

The Herschel-Bulkley (2) dependence takes into account the existence of a yield stress *τ*0 value, namely such a value of the shear stress above which the lubricating grease starts flowing:

**Figure 3.** Dependences of the shear stress (a) and the structural viscosity (b) in a function of the shear rate for the lithium greases based on the mineral oil, with a various thickener percentage. The diagrams, in order to achieve a

<sup>0</sup> ( ), *<sup>m</sup> <sup>C</sup>*

 g

tt

a)

b)

greater clarity, do not contain the confidence intervals.

1

=+ × *k* & (2)

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

83

where: *η* - dynamic viscosity, *τ* - shear stress, *γ*˙ - shear rate.

Figure 3 illustrates results of the rheological measurements made on the lithium greases of a various thickener percentage, showing the shear stress changes and of the structural viscosity as a function of the shear rate. Additionally, the theoretical curves generated from the Herschel-Bulkley's [25] and Carreau-Yasuda's [26] dependences have been presented. Values of the parameters defining the theoretical curves traced according to the abovementioned depend‐ ences for the greases of a various thickener percentage have been presented in Table 1. The greases were thickened with lithium 12-hydroxystearate, and the oil base was ORLEN OIL SN-400 (ORLEN OIL, Cracow, Poland) mineral oil. The flow curves were defined at temper‐ ature of 25 o C. The tests were repeated five times, and their results were statistically elaborated at the confidence level *p =* 0.95, using t-Student test. Additionally, correlation coefficient values were calculated. The experiment was carried out by means of Physica Anton-Paar MCR 101 rotational rheometer. The rheometer was working in the cone-and-plate system (CP-25-1, 25 mm, 1o ) at a constant measuring gap height of 49 µm.


**Table 1.** Values of the parameters determining the theoretical curves traces, defined according to dependences (1) and (2) for the greases of the thickener content 4-9%.

Lubricating greases, at very low shear rates behave like a Newtonian fluid of a constant viscosity *η*0, being a slope of a tangent to the flow curve, equal to the shear rate going to zero (*η*<sup>0</sup> =lim*γ*˙ <sup>→</sup>0*η*(*γ*˙)). At the shear rate going to infinity, lubricating greases also behave like Newtonian fluids, but of viscosity *η∞*close to the dispersion phase, namely of the base oil (*η∞* =lim*γ*˙ <sup>→</sup>*∞η*(*γ*˙)). The viscosity *η∞* is a slope of a straight line being a flow curve asymptote. In the bracket of the indirect shear rates, the curve of the lubricating grease flow curve is

> , t

Figure 3 illustrates results of the rheological measurements made on the lithium greases of a various thickener percentage, showing the shear stress changes and of the structural viscosity as a function of the shear rate. Additionally, the theoretical curves generated from the Herschel-Bulkley's [25] and Carreau-Yasuda's [26] dependences have been presented. Values of the parameters defining the theoretical curves traced according to the abovementioned depend‐ ences for the greases of a various thickener percentage have been presented in Table 1. The greases were thickened with lithium 12-hydroxystearate, and the oil base was ORLEN OIL SN-400 (ORLEN OIL, Cracow, Poland) mineral oil. The flow curves were defined at temper‐

at the confidence level *p =* 0.95, using t-Student test. Additionally, correlation coefficient values were calculated. The experiment was carried out by means of Physica Anton-Paar MCR 101 rotational rheometer. The rheometer was working in the cone-and-plate system (CP-25-1, 25

**Herschel-Bulkley model Carreau-Yasuda model**

4.0 120.27 5.826 0.66964 0.99983 1852.4 0.38843 47.456 2 0.20073 0.99600 5.0 190.52 83.155 0.37458 0.99958 4189.5 0.20014 47.457 2 0.23465 0.99668 6.0 216.00 83.867 0.38633 0.99973 4742.5 0.26093 47.939 2 0.22985 0.99811 6.5 241.47 212.430 0.29112 0.99982 9090.5 0.18576 47.940 2 0.20100 0.99919 7.0 672.60 148.930 0.35189 0.99750 15225.0 0.26328 47.455 2 0.17468 0.99696 8.0 712.54 285.550 0.28687 0.99592 21453.0 0.25487 47.423 2 0.16048 0.99944 9.0 785.29 230.630 0.37532 0.99956 16023.0 0.54032 47.458 2 0.21285 0.99743

**Table 1.** Values of the parameters determining the theoretical curves traces, defined according to dependences (1)

η0– η∞ (Pas)

η∞ (Pas)

λ (s)

*an Rxy*

C. The tests were repeated five times, and their results were statistically elaborated

<sup>=</sup> & (1)

g

h

where: *η* - dynamic viscosity, *τ* - shear stress, *γ*˙ - shear rate.

) at a constant measuring gap height of 49 µm.

*m Rxy*

described with the following equation:

82 Tribology - Fundamentals and Advancements

ature of 25 o

mm, 1o

**Thickener content (%)**

τ (Pa)

kC (Pas)

and (2) for the greases of the thickener content 4-9%.

**Figure 3.** Dependences of the shear stress (a) and the structural viscosity (b) in a function of the shear rate for the lithium greases based on the mineral oil, with a various thickener percentage. The diagrams, in order to achieve a greater clarity, do not contain the confidence intervals.

The Herschel-Bulkley (2) dependence takes into account the existence of a yield stress *τ*0 value, namely such a value of the shear stress above which the lubricating grease starts flowing:

$$
\tau = \tau\_0 + (k\_\mathcal{C} \cdot \dot{\mathcal{Y}})^m,\tag{2}
$$

where: *τ*0 –yield stress, *kC* – consistency factor of the thickener in the grease, *m* – nondimen‐ sional index exponent. Carreau-Yasuda (3) dependence enables approximation of the course of changes in the structural viscosity as a function of the shear rate with viscosity *η∞* and *η*<sup>0</sup> included:

$$
\eta(\dot{\boldsymbol{\gamma}}) = \eta\_{\boldsymbol{\alpha}} + (\eta\_0 - \eta\_{\boldsymbol{\alpha}}) \cdot \left[1 + (\boldsymbol{\lambda} \cdot \dot{\boldsymbol{\gamma}})^{\boldsymbol{a}}\right]^{\frac{n-1}{\boldsymbol{a}}} \boldsymbol{\tag{3}
$$

where: *G* ′

Storage modulus *G* ′

where:


0 0 *G* cos , s

0 0 *G* sin . s

g

tan . *G G* d

> 0 0

0 0

*G*

g w

g w

sin , *<sup>G</sup>*

 d

cos .

dynamic viscosity, whereas imaginary part *η* ″ is called the elastic component. The measure‐

Figure 4 illustrates example results of a strain sweep test of Nanolubricant 2010 (Orapi, Saint-Vulbas, France) commercialgrease, thickened with the lithium 12-hydroxystearate, based on

 d

viscosity, similarly to the complex modulus can be formulated as:

h hh

h

h

, *η* ″

The real part of complex viscosity *η* ′

terms of their viscoelastic properties.

, *G* ″

and *η* ′

ment of quantities *G* ′

g

d

d

cycle, whereas loss modulus *G* ″ is a measure of the energy lost per a sinusoidal deformation cycle. What results from equations (5) and (6) is that phase angle shift *δ* can be defined as:

The dispersion effects can also be described by means of complex viscosity *η* <sup>∗</sup>*.* The complex

is a measure of the energy stored and recovered per each deformation

(viscous component shifted in the phase in relation to the stress). Components *G* ′


are equal:

85

*, G* ″

http://dx.doi.org/10.5772/55929

¢ <sup>=</sup> (5)

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

¢¢ <sup>=</sup> (6)

¢¢ <sup>=</sup> ¢ (7)

*i* , \* = -¢ ¢¢ (8)

¢ <sup>=</sup> (9)

¢¢ <sup>=</sup> (10)

is called the viscous component or, less frequently, the

enables the rheological characteristics of the greases in

where: *η∞* =lim*γ*˙ <sup>→</sup>*∞η*(*γ*˙), *η*<sup>0</sup> =lim*γ*˙ <sup>→</sup>0*η*(*γ*˙), *λ* – time-constant, *a* and *n* – nondimensionalparame‐ ters. For the shear-thinning fluids, *a* parameter normally equals 2 [27].

The flow curves clearly illustrate the fact that the greases have the pseudoplastic proper‐ ties. It results from the degradation of the lubricating greases' microstructure as well as from the further orientation of the dispersed particles of the thickener. During relaxation of the stress in the lubricating grease, a considerable entanglement of the thickener particles is observed. However, during shearing of the grease, one can record straightening and untangling of the particles which are directed along the current line. With the increase of the shear rate, the effect is more and more visible. The reduction of the internal friction resulting from a smaller size of the particles and a limited activity between them is observed. At very high shear rates, the total orientation of the particles in the grease is achieved. The internal friction remains constant at a low level. The structural viscosity of the grease then gets close to the viscosity of the base oil.

Figure 3 shows that at a large enough percentage of the dispersion, the interaction between the thickener particles may cause formation of the spatial microstructure resistant to the shear stress which does not exceed certain value. Below the value, the lubricating grease behaves like an elastic solid. When yield stress *τ*<sup>0</sup> (yield point) is reached, the microstructure gets damaged and the grease starts behaving as a viscous liquid. When the shear stress in lubri‐ cating grease becomes lower than the limit value, the immediate reconstruction of the microstructure takes place.The determination of the yield stress value in the lubricating grease is very important because it enables the determining of its usefulness for the application in the central lubrication systems of machines [9,28] or in the friction nodes [7].

#### **3.2. Linear and non-linear viscoelasticity**

When a lubricating grease undergoes the periodic stress oscillation, its response in a form of a relative deformation will be shifted in the phase by *δ* angle. The angle, called the phase angle shift, falls in the range from 0o to 90o . The response of the lubricating grease determined by the extortion is influenced primarily by: complex modulus |*G* <sup>∗</sup> | and complex viscosity | *η* <sup>∗</sup> |. The complex modulus is defined as a sum of the real and the imaginary part:

$$\left| \mathbf{G}^\* \right| = \mathbf{G}' + i\mathbf{G}'' \,, \tag{4}$$

where: *G* ′ - storage modulus (elastic component in the phase with the stress), *G* ″ - loss modulus (viscous component shifted in the phase in relation to the stress). Components *G* ′ *, G* ″ are equal:

$$G' = \frac{\sigma\_0}{\mathcal{V}\_0} \cos \delta\_\prime \tag{5}$$

$$G'' = \frac{\sigma\_0}{\mathcal{Y}\_0} \sin \delta. \tag{6}$$

Storage modulus *G* ′ is a measure of the energy stored and recovered per each deformation cycle, whereas loss modulus *G* ″ is a measure of the energy lost per a sinusoidal deformation cycle. What results from equations (5) and (6) is that phase angle shift *δ* can be defined as:

$$\tan \delta = \frac{G''}{G'}.\tag{7}$$

The dispersion effects can also be described by means of complex viscosity *η* <sup>∗</sup>*.* The complex viscosity, similarly to the complex modulus can be formulated as:

$$\left| \eta^\* \right| = \eta' - i\eta'',\tag{8}$$

where:

where: *τ*0 –yield stress, *kC* – consistency factor of the thickener in the grease, *m* – nondimen‐ sional index exponent. Carreau-Yasuda (3) dependence enables approximation of the course of changes in the structural viscosity as a function of the shear rate with viscosity *η∞* and *η*<sup>0</sup>

<sup>0</sup> ( ) ( ) [1 ( ) ] ,

where: *η∞* =lim*γ*˙ <sup>→</sup>*∞η*(*γ*˙), *η*<sup>0</sup> =lim*γ*˙ <sup>→</sup>0*η*(*γ*˙), *λ* – time-constant, *a* and *n* – nondimensionalparame‐

The flow curves clearly illustrate the fact that the greases have the pseudoplastic proper‐ ties. It results from the degradation of the lubricating greases' microstructure as well as from the further orientation of the dispersed particles of the thickener. During relaxation of the stress in the lubricating grease, a considerable entanglement of the thickener particles is observed. However, during shearing of the grease, one can record straightening and untangling of the particles which are directed along the current line. With the increase of the shear rate, the effect is more and more visible. The reduction of the internal friction resulting from a smaller size of the particles and a limited activity between them is observed. At very high shear rates, the total orientation of the particles in the grease is achieved. The internal friction remains constant at a low level. The structural viscosity of

Figure 3 shows that at a large enough percentage of the dispersion, the interaction between the thickener particles may cause formation of the spatial microstructure resistant to the shear stress which does not exceed certain value. Below the value, the lubricating grease behaves like an elastic solid. When yield stress *τ*<sup>0</sup> (yield point) is reached, the microstructure gets damaged and the grease starts behaving as a viscous liquid. When the shear stress in lubri‐ cating grease becomes lower than the limit value, the immediate reconstruction of the microstructure takes place.The determination of the yield stress value in the lubricating grease is very important because it enables the determining of its usefulness for the application in the

When a lubricating grease undergoes the periodic stress oscillation, its response in a form of a relative deformation will be shifted in the phase by *δ* angle. The angle, called the phase angle

extortion is influenced primarily by: complex modulus |*G* <sup>∗</sup> | and complex viscosity | *η* <sup>∗</sup> |.

. The response of the lubricating grease determined by the

*G G iG* , \* = +¢ ¢¢ (4)

 l g *a* - ¥ ¥ & & = + - ×+ × (3)

hg

 h

the grease then gets close to the viscosity of the base oil.

**3.2. Linear and non-linear viscoelasticity**

shift, falls in the range from 0o

central lubrication systems of machines [9,28] or in the friction nodes [7].

The complex modulus is defined as a sum of the real and the imaginary part:

to 90o

 h h

ters. For the shear-thinning fluids, *a* parameter normally equals 2 [27].

1

*n a*

included:

84 Tribology - Fundamentals and Advancements

$$
\eta' = \frac{G\_0}{\gamma\_0 a} \sin \delta \,, \tag{9}
$$

$$
\eta'' = \frac{G\_0}{\chi\_0 a} \cos \delta.\tag{10}
$$

The real part of complex viscosity *η* ′ is called the viscous component or, less frequently, the dynamic viscosity, whereas imaginary part *η* ″ is called the elastic component. The measure‐ ment of quantities *G* ′ , *G* ″ and *η* ′ , *η* ″ enables the rheological characteristics of the greases in terms of their viscoelastic properties.

Figure 4 illustrates example results of a strain sweep test of Nanolubricant 2010 (Orapi, Saint-Vulbas, France) commercialgrease, thickened with the lithium 12-hydroxystearate, based on a synthetic oil with the addition of tungsten disulfide nanoparticles. The deformation range of the grease was 0 to 100% at 1 Hz oscillation frequency. The measurements were done at 25 o C. The oscillation measurements made it possible to determine the limit of the linear viscoe‐ lasticity (marked in the diagram with the vertical line). The reasearch showed that plateau modulus *G0 <sup>N</sup>* for Nanolubricant 2010 was 75 743 Pa at deformation *γC* equal 0.336%. The review of the methods used for the determining of the linear viscoelasticity limit (the critical point) on the basis of the dynamic-oscillation tests is widely discussed in [29].

**Figure 4.** The storage and loss moduli and the phase angle shiftas a function ofNanolubricant 2010 grease strain.

Figure 5 illustrates the influence of the soap thickener (lithium 12-hydroxystearate) percentage in the lubricating grease based on the mineral oil on (a) value of critical strain *γC* and (b) of shear stress *τ* in the critical point, as a function of the percentage of the thickener. The diagrams show that the range of the linear viscoelasticity depends on the percentage of the thickener in the grease – the smaller percentage of the thickener, the wider range of the linear viscoelasticity. In the case of the shear stress in the critical point, there is an adverse situation. The research was carried out by means of a stress/strain controlled rotational rheometerPhysica Anton-Paar MCR 101. The rheometer was working in the cone-and-plate system (CP-25-1, 25 mm, 1o ) at a constant measuring gap of 49 µm.

influence of oscillation. Carrying out such research is important for the evaluation of the lubricating greases' behaviour at the start-up phase of the friction nodes and of the flow in the lubrication systems conduits. The dynamic-oscillation research on the lubricating greases has been, among others, conducted by Yeong et al. [14] as well as Delgado [12], Martín-Alfonso et

**Figure 5.** Critical strain γ*C* (a) and shear stress τ in the critical point (b),as a function of the percentage of the lithium

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

87

The term 'thixotropy' comes from the Greek words *thixis* (mix, shake) and *trepo* (revolve, change), and it was first suggested by Péterfi [32] (the original name – *thixitropy)* [33]. The

**4. Mechanism of thixotropic changes in the microstructure of the**

**lubricating grease during its shearing and relaxation**

al. [30,31].

thickener in the lubricating grease

a)

b)

The rheological dynamic-oscillation research (which is a combination of the creep and relaxation experiments) are currently the basic tool at evaluating the structure of dispersion and its mechanical stability as well as the behaviour of the lubricating grease in the start phase of the flow. At high frequencies and amplitudes, the dynamic-oscillation tests can be used for the determining of the speed at which the microstructure is created and damaged under the

a synthetic oil with the addition of tungsten disulfide nanoparticles. The deformation range of the grease was 0 to 100% at 1 Hz oscillation frequency. The measurements were done at 25

C. The oscillation measurements made it possible to determine the limit of the linear viscoe‐ lasticity (marked in the diagram with the vertical line). The reasearch showed that plateau

of the methods used for the determining of the linear viscoelasticity limit (the critical point)

**Figure 4.** The storage and loss moduli and the phase angle shiftas a function ofNanolubricant 2010 grease strain.

Figure 5 illustrates the influence of the soap thickener (lithium 12-hydroxystearate) percentage in the lubricating grease based on the mineral oil on (a) value of critical strain *γC* and (b) of shear stress *τ* in the critical point, as a function of the percentage of the thickener. The diagrams show that the range of the linear viscoelasticity depends on the percentage of the thickener in the grease – the smaller percentage of the thickener, the wider range of the linear viscoelasticity. In the case of the shear stress in the critical point, there is an adverse situation. The research was carried out by means of a stress/strain controlled rotational rheometerPhysica Anton-Paar MCR 101. The rheometer was working in the cone-and-plate system (CP-25-1, 25 mm, 1o

The rheological dynamic-oscillation research (which is a combination of the creep and relaxation experiments) are currently the basic tool at evaluating the structure of dispersion and its mechanical stability as well as the behaviour of the lubricating grease in the start phase of the flow. At high frequencies and amplitudes, the dynamic-oscillation tests can be used for the determining of the speed at which the microstructure is created and damaged under the

) at a

on the basis of the dynamic-oscillation tests is widely discussed in [29].

*<sup>N</sup>* for Nanolubricant 2010 was 75 743 Pa at deformation *γC* equal 0.336%. The review

o

modulus *G0*

86 Tribology - Fundamentals and Advancements

constant measuring gap of 49 µm.

**Figure 5.** Critical strain γ*C* (a) and shear stress τ in the critical point (b),as a function of the percentage of the lithium thickener in the lubricating grease

influence of oscillation. Carrying out such research is important for the evaluation of the lubricating greases' behaviour at the start-up phase of the friction nodes and of the flow in the lubrication systems conduits. The dynamic-oscillation research on the lubricating greases has been, among others, conducted by Yeong et al. [14] as well as Delgado [12], Martín-Alfonso et al. [30,31].

#### **4. Mechanism of thixotropic changes in the microstructure of the lubricating grease during its shearing and relaxation**

The term 'thixotropy' comes from the Greek words *thixis* (mix, shake) and *trepo* (revolve, change), and it was first suggested by Péterfi [32] (the original name – *thixitropy)* [33]. The phenomenon is defined as an isothermal decrease in structural viscosity during shearing followed by an increase in the viscosity and the re-solidification of the substance once shearing ends [34]. The phenomenon of thixotropy concerns suspensions of the internal microstructure made of particles which undergo aggregation or flocculation. In such systems, there are physical interactions between particles, which, during relaxation of a substance, determine the creation of a spatial, cross-linked microstructure. They are primarily the Van der Waals-London attractive forces, the electrostatic repulsive forces (particularly important when the dispersed phase particles are anisometric in shape) as well as the steric forces. Also, the Brownian motion may have an important part to play in the process, but their appearance is limited only to colloids. A wide description of the intermolecular interactions which are common in thixotropic systems can be found in the publications of Efremov et al. [35,36] as well as of Sonntag's [37] and Scheludko's [38]. During shearing, the microstructure made of the dispersed phase particles, as a consequence of their mutual interaction, gets disintegrated forming bigger aggregates and individual particles, floccules, suspended freely in the disper‐ sion medium. The review of the references concerning the thixotropy of suspensions can be found, for instance, in the publications of Mewis and Barnes [10,39].

The lubricating greases also show thixotropy. The structural viscosity of the lubricating grease decreases systematically during its shearing in the tribological node or during its flow through lubrication system conduits. The standstill of the grease leads to its re-solidification and a little increase in its structural viscosity. The decrease of viscosity during flowing of the grease is reduced to a certain minimum value, primarily depending on the viscosity of the base oil, a kind and amount of the thickener as well as the shear rate. At high shear rates, the structural viscosity of the grease reaches the value close to the base oil viscosity. According to Czarny [1] "Both in the slide bearings and in the roller bearings, the high-speed ones in particular, the rheological properties of the lubricant, so its hydrodynamic lift force, are determined mainly by the base oil. The thickener influences the force at the start-up and at the stopping of the bearing.". Due to the complexity of the processes of degradation and reconstruction of the thickener's microstructure in the lubricating greases, scientific works which would thoroughly explain the processes have not been published so far.

The first publications on thixotropy of the lubricating greases appeared at the beginning of the 1950s. In 1951, Moore and Cravathpusblished their work [13] on the mechanical process of the soap fibre disruption in the lubricating greases caused by the shearing force and on the influence of the disruption on the change of the consistency of the greases. The research concerned the lubricants thickened with the barium, sodium, lithium and calcium soaps. The degradation of the soap fibre consisted in the working of the samples in a penetration appa‐ ratus, as well as on their long-lasting disintegration in a rolling apparatus. During shearing, at different intervals, the measuring of the greases micropenetration level was performed. The tests showed that the biggest differences in the consistency of the greases were observed in the commercial grease thickened with the barium soap, whereas the smallest differences were recorded for the lithium grease. Additionally, a series of photographs of single soap fibres by means of an electron microscope was taken. The photographs showed a considerable differ‐ ence in the fibre length before and after shearing for most of the tested samples. Moore and

Cravath proposed a soap fibre disruption intensity model at different times of the external

**Figure 6.** Microstructure of the lithium thickener in the paraffin oil [40]: (a) fresh grease, (b) grease sheared at 120 oC for 10 hours at ˙γ = 0.01 s-1 shear rate. The microphotograph taken by means of the atomic force microscope (AFM).

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

89

A similar analysis to the one conducted by Moore and Cravath was carried out by Renshaw [41]. He was particularly interested in the lithium thickened greases based on the ester oil, which showed high shear resistance. The measuring of the micropenetration was done directly before and after the 9-hour long shearing of the grease in a bearing at the speed of 10 000 rev/ min and after 100 000 double strokes of the piston in the penetration apparatus. In the first case, Renshaw observed in the electron microscope, a considerable shortening of the lithium

shearing force activity.

Scanning area 20 µm x 20 µm.

a)

b)

phenomenon is defined as an isothermal decrease in structural viscosity during shearing followed by an increase in the viscosity and the re-solidification of the substance once shearing ends [34]. The phenomenon of thixotropy concerns suspensions of the internal microstructure made of particles which undergo aggregation or flocculation. In such systems, there are physical interactions between particles, which, during relaxation of a substance, determine the creation of a spatial, cross-linked microstructure. They are primarily the Van der Waals-London attractive forces, the electrostatic repulsive forces (particularly important when the dispersed phase particles are anisometric in shape) as well as the steric forces. Also, the Brownian motion may have an important part to play in the process, but their appearance is limited only to colloids. A wide description of the intermolecular interactions which are common in thixotropic systems can be found in the publications of Efremov et al. [35,36] as well as of Sonntag's [37] and Scheludko's [38]. During shearing, the microstructure made of the dispersed phase particles, as a consequence of their mutual interaction, gets disintegrated forming bigger aggregates and individual particles, floccules, suspended freely in the disper‐ sion medium. The review of the references concerning the thixotropy of suspensions can be

The lubricating greases also show thixotropy. The structural viscosity of the lubricating grease decreases systematically during its shearing in the tribological node or during its flow through lubrication system conduits. The standstill of the grease leads to its re-solidification and a little increase in its structural viscosity. The decrease of viscosity during flowing of the grease is reduced to a certain minimum value, primarily depending on the viscosity of the base oil, a kind and amount of the thickener as well as the shear rate. At high shear rates, the structural viscosity of the grease reaches the value close to the base oil viscosity. According to Czarny [1] "Both in the slide bearings and in the roller bearings, the high-speed ones in particular, the rheological properties of the lubricant, so its hydrodynamic lift force, are determined mainly by the base oil. The thickener influences the force at the start-up and at the stopping of the bearing.". Due to the complexity of the processes of degradation and reconstruction of the thickener's microstructure in the lubricating greases, scientific works which would thoroughly

The first publications on thixotropy of the lubricating greases appeared at the beginning of the 1950s. In 1951, Moore and Cravathpusblished their work [13] on the mechanical process of the soap fibre disruption in the lubricating greases caused by the shearing force and on the influence of the disruption on the change of the consistency of the greases. The research concerned the lubricants thickened with the barium, sodium, lithium and calcium soaps. The degradation of the soap fibre consisted in the working of the samples in a penetration appa‐ ratus, as well as on their long-lasting disintegration in a rolling apparatus. During shearing, at different intervals, the measuring of the greases micropenetration level was performed. The tests showed that the biggest differences in the consistency of the greases were observed in the commercial grease thickened with the barium soap, whereas the smallest differences were recorded for the lithium grease. Additionally, a series of photographs of single soap fibres by means of an electron microscope was taken. The photographs showed a considerable differ‐ ence in the fibre length before and after shearing for most of the tested samples. Moore and

found, for instance, in the publications of Mewis and Barnes [10,39].

88 Tribology - Fundamentals and Advancements

explain the processes have not been published so far.

**Figure 6.** Microstructure of the lithium thickener in the paraffin oil [40]: (a) fresh grease, (b) grease sheared at 120 oC for 10 hours at ˙γ = 0.01 s-1 shear rate. The microphotograph taken by means of the atomic force microscope (AFM). Scanning area 20 µm x 20 µm.

Cravath proposed a soap fibre disruption intensity model at different times of the external shearing force activity.

A similar analysis to the one conducted by Moore and Cravath was carried out by Renshaw [41]. He was particularly interested in the lithium thickened greases based on the ester oil, which showed high shear resistance. The measuring of the micropenetration was done directly before and after the 9-hour long shearing of the grease in a bearing at the speed of 10 000 rev/ min and after 100 000 double strokes of the piston in the penetration apparatus. In the first case, Renshaw observed in the electron microscope, a considerable shortening of the lithium thickener's fibre and slight changes in the micropenetration values. After the working of the grease in the penetration apparatus, the situation was exactly reverse, namely the changes in the micropenetration were remarkable but the particles were shortened only to a minor extent. The research showed that the intensity of degradation in the thickener's microstructure depends not only on the mechanical strength of individual fibres, but mainly on the shearing way and on the direction of the fibres at the flow of the grease.

Sirianni et al. [42] investigated the question of the influence of size of the calcium and barium soaps and their concentration on the change of the lubricating greases *thixotropy coefficient*. For the experiment, they used a theory proposed earlier by Goodeveand Whitfield [43] which was aimed at the determination of the thixotropy of the black suspension in the mineral oil [44]. According to the theory, the speed of the microstructure reconstruction (proportional to the thickener's concentration) equals the speed of its degradation. The thickener's concentration, however, is proportional to the difference between apparent viscosity *η* and grease equillibri‐ um viscosity *η*0 following the total degradation of the cross-linked microstructure of the thickener at the infinitely high shear rate.

$$
\eta - \eta\_0 = \frac{\theta}{\dot{\mathcal{Y}}} = \frac{F\_T}{\dot{\mathcal{Y}}} \,\,\,\tag{11}
$$

dimensional network and make the solution acquire the non-Newtonian fluid properties) and the not entangled particles (single floccules which make the solution acquire the Newtonian fluid properties). The number of the former and the latter depends on the temporary shear

where: *χ*<sup>1</sup> – share of the not entangled particles, *χ*2– share of the entangled particles, *1/α1* – the not entangled particles shear stress, *1/α2* – the entangled particles shear stress, *1/β1* – shear intensity for the not entangled particles, *1/β2* – shear intensity for the entangled particles, *γ*˙ –

Hahn and the collaborators assumed that at the increase, and next at the decrease of the shear rate, the free energy of the soap particles changes, and it is directly dependent on their strain energy. The cross-linked particles aggregates featuring the non-Newtonian characteristics get untangled making individual floccules, freely suspended in the dispersion medium, featuring the Newtonian characteristics. The process is reversible. Two years later, the scientists

> 2 2 0 0 (1 )( )

 h



2 1 , *w w kT kT*

where: *kf* – parameter determining the time of the entangled particles → not entangled particles reaction; *kb* – parameter determining the time of the not entangled → entangled particles; *w* – elastic strain energy of the entangled particles, *w* ′ – elastic strain energy of the not entangled particles, *η*0 – viscosity of the entire thixotropy system at zero shear rate, *k* – Boltzman constant,

Thus, Hahn et al. elaborated theoretical flow curves for the lubricating greases during the disruption and reinforcenment of the thickener's structure. The models have been widely applied for the determination of the thixotropic properties of, for instance, motor oils [48]. Validity of the assumptions made at the making of the mathematical dependences was proven by the experimentally developed curves. The mathematical model proposed by Hahn was also used by Utsugi et al. [49]. They analyzed the thixotropy of the calcium greases and the silica gel thickened greases. At the determining of the thixotropic hysteresis loop, Utsugi used the rotational rheometer. Bair [50] proved that the models proposed in the 1960s by Hahn and Eyring were still true. During the verification of the models, Bair used more technologically advanced rotational rheometers. He drew particular attention to the possibility of the depend‐ ences application for describing the thixotropy phenomenon of the lubricants working at the

*f b <sup>d</sup> ke ke*

 c

h

2


Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

91

b g

sinh ( ),

11 2 1

 c

 

1 2

 g

cb

stress. The flow curve equation used by Hahn et al. was:

presented the equation of the changes kinetics [47]:

2

c

*dt*

friction contact, in the elastohydrodynamic conditions.

c

shear rate.

*T* – temperature.

t

where: *η* – apparent viscosity, *η*0 – equillibrium viscosity, *γ*˙ – shear rate, *θ* – thixotropy coefficient, *FT* – momentum generated by the intermolecular forces transmitted by the unit of surface in the unit of time.

The theory assumes that at the increase of 1 / *γ*˙ expression, apparent viscosity *η* rises until it reaches the plateau. According to Goodeve and Whitfield, the thixotropy coefficient is a source of the non-Newtonian effects, which is directly correlated with the intermolecular bonds disruption effects, the average life span of the bonds and the change of size of the interacting molecules in the thixotropic process. In order to calculate the thixotropy coefficient, Goodeve assumed that the bonds between the molecules of the dispersed phase are subject to Hooke's law.

Moses and Puddington [45], using the same theory, determined the influence of the thickener's percentage (10%, 14% and 19.7%), the shear rate (from 1 400 s-1 to 250 000 s-1), temperature (from 35 o C up to the value at which a visible sedimentation occurs) as well as the size-reduction of the thickener particles, on the change of the *thixotropy coefficient* of the lithium, sodium and aluminium greases. Different sizes of the soap fibre were generated by means of the colloid mill. For determining equillibrium viscosity *η*0 a metal capillary was used.

Hahn, Ree and Eyring [46], while observing behaviour of the macromolecular polymers suspensions, formulated a theory (later called *sinh law*) which ultimately was widely used for the quantity description of the phenomena of the lubricating greases microstructure degrada‐ tion and reconstruction, where the soap is a thickener. They took into account the fact that the thickener's structure contains the so called entangled particles (aggregates which create a threedimensional network and make the solution acquire the non-Newtonian fluid properties) and the not entangled particles (single floccules which make the solution acquire the Newtonian fluid properties). The number of the former and the latter depends on the temporary shear stress. The flow curve equation used by Hahn et al. was:

thickener's fibre and slight changes in the micropenetration values. After the working of the grease in the penetration apparatus, the situation was exactly reverse, namely the changes in the micropenetration were remarkable but the particles were shortened only to a minor extent. The research showed that the intensity of degradation in the thickener's microstructure depends not only on the mechanical strength of individual fibres, but mainly on the shearing

Sirianni et al. [42] investigated the question of the influence of size of the calcium and barium soaps and their concentration on the change of the lubricating greases *thixotropy coefficient*. For the experiment, they used a theory proposed earlier by Goodeveand Whitfield [43] which was aimed at the determination of the thixotropy of the black suspension in the mineral oil [44]. According to the theory, the speed of the microstructure reconstruction (proportional to the thickener's concentration) equals the speed of its degradation. The thickener's concentration, however, is proportional to the difference between apparent viscosity *η* and grease equillibri‐ um viscosity *η*0 following the total degradation of the cross-linked microstructure of the

> <sup>0</sup> , *<sup>T</sup>* q*F*

where: *η* – apparent viscosity, *η*0 – equillibrium viscosity, *γ*˙ – shear rate, *θ* – thixotropy coefficient, *FT* – momentum generated by the intermolecular forces transmitted by the unit of

The theory assumes that at the increase of 1 / *γ*˙ expression, apparent viscosity *η* rises until it reaches the plateau. According to Goodeve and Whitfield, the thixotropy coefficient is a source of the non-Newtonian effects, which is directly correlated with the intermolecular bonds disruption effects, the average life span of the bonds and the change of size of the interacting molecules in the thixotropic process. In order to calculate the thixotropy coefficient, Goodeve assumed that the bonds between the molecules of the dispersed phase are subject to Hooke's

Moses and Puddington [45], using the same theory, determined the influence of the thickener's percentage (10%, 14% and 19.7%), the shear rate (from 1 400 s-1 to 250 000 s-1), temperature

of the thickener particles, on the change of the *thixotropy coefficient* of the lithium, sodium and aluminium greases. Different sizes of the soap fibre were generated by means of the colloid

Hahn, Ree and Eyring [46], while observing behaviour of the macromolecular polymers suspensions, formulated a theory (later called *sinh law*) which ultimately was widely used for the quantity description of the phenomena of the lubricating greases microstructure degrada‐ tion and reconstruction, where the soap is a thickener. They took into account the fact that the thickener's structure contains the so called entangled particles (aggregates which create a three-

mill. For determining equillibrium viscosity *η*0 a metal capillary was used.

C up to the value at which a visible sedimentation occurs) as well as the size-reduction


g g

h h

way and on the direction of the fibres at the flow of the grease.

thickener at the infinitely high shear rate.

90 Tribology - Fundamentals and Advancements

surface in the unit of time.

law.

(from 35 o

$$
\tau = \frac{\varkappa\_1 \beta\_1}{a\_1} \dot{\gamma} + \frac{\varkappa\_2}{a\_2} \sinh^{-1}(\beta\_2 \dot{\gamma}),
\tag{12}
$$

where: *χ*<sup>1</sup> – share of the not entangled particles, *χ*2– share of the entangled particles, *1/α1* – the not entangled particles shear stress, *1/α2* – the entangled particles shear stress, *1/β1* – shear intensity for the not entangled particles, *1/β2* – shear intensity for the entangled particles, *γ*˙ – shear rate.

Hahn and the collaborators assumed that at the increase, and next at the decrease of the shear rate, the free energy of the soap particles changes, and it is directly dependent on their strain energy. The cross-linked particles aggregates featuring the non-Newtonian characteristics get untangled making individual floccules, freely suspended in the dispersion medium, featuring the Newtonian characteristics. The process is reversible. Two years later, the scientists presented the equation of the changes kinetics [47]:

$$-\frac{d\mathcal{X}\_2}{dt} = \mathcal{X}\_2 k\_f e^{\frac{\eta\_0 w^2}{kT}} - \mathcal{X}\_1 k\_b e^{\frac{-(1-\eta\_0)(w')^2}{kT}},\tag{13}$$

where: *kf* – parameter determining the time of the entangled particles → not entangled particles reaction; *kb* – parameter determining the time of the not entangled → entangled particles; *w* – elastic strain energy of the entangled particles, *w* ′ – elastic strain energy of the not entangled particles, *η*0 – viscosity of the entire thixotropy system at zero shear rate, *k* – Boltzman constant, *T* – temperature.

Thus, Hahn et al. elaborated theoretical flow curves for the lubricating greases during the disruption and reinforcenment of the thickener's structure. The models have been widely applied for the determination of the thixotropic properties of, for instance, motor oils [48]. Validity of the assumptions made at the making of the mathematical dependences was proven by the experimentally developed curves. The mathematical model proposed by Hahn was also used by Utsugi et al. [49]. They analyzed the thixotropy of the calcium greases and the silica gel thickened greases. At the determining of the thixotropic hysteresis loop, Utsugi used the rotational rheometer. Bair [50] proved that the models proposed in the 1960s by Hahn and Eyring were still true. During the verification of the models, Bair used more technologically advanced rotational rheometers. He drew particular attention to the possibility of the depend‐ ences application for describing the thixotropy phenomenon of the lubricants working at the friction contact, in the elastohydrodynamic conditions.

Vinogradov, Deinega and Verbitsky [18] investigated the correlation between the changes of the friction momentum in the ball bearings and the thixotropic properties of the calcium and the lithium greases. At the research, they observed that the shape of the dispersed phase particles considerably influences the behaviour of the lubricating greases during their further deformation and shear-thinning. On the turn of the 1960s and 1970s, Vinogradov also inves‐ tigated the lubricating greases using the polarimetry and other optical methods. It enabled the quantity evaluation of the level of the thickener microstructure reconstruction. He also conducted research on the thixotropic hysteresis loop for the calcium and lithium greases with the use of the rotational coaxial cylinder rheometers. His works are widely discussed by Froischteter in [7].

Sacchettini, Magnin, Piau and Pierrard [51] published in the Journal of Theoretical and Applied Mechanics a paper on the thixotropic properties of commercial greases: Shell Alvania, ELF Multi and ELF EL series (8305-8308 and 9309) of NLGI (National Lubricating Grease Institute) 2 consistency class. The analyzed lubricating greases showed a different fibre length of the soap thickener (lithium 12-hydroxystearate). The oil base in the greases was the mineral oil. Sacchettini et al. focused mainly on the research of change in the shear stress as a function of shearing time. They analyzed changes in the lithium thickener microstructure at the cyclic shearing lasting a few seconds at different shear rates (from 0.018 s-1 to 18 s-1) and different grease relaxation time (from 2 minutes to 38 hours). They did not analyzed the structural viscosity increase of the greases at their relaxation, however. Apart from the thixotropy, Sacchettini et al. investigated the wall effects.

Czarny, in his paper [52], presented results of the research on the process of degradation and reconstruction of the thickener microstructure in the commercial calcium greases (STP) and the lithium greases (ŁT-4S2). He used the method of cyclic shearing of the lubricating grease at specified time intervals, at a constant shear rate. The research on the grease was carried out without removing it from the rheometer head. He thus simulated the conditions observed in the conduits of lubrication system. In 1990, Czarny investigated the influence of temperature on the thixotropy of the lubricating greases [53]. He proposed a mathematical description of the energetic changes which take place in the microstructure of the thickener during its shearing at various temperatures. Czarny observed that along with the temperature increase, the structural viscosity of the lubricating grease is reduced, and the activity of the thickener particles increases. According to Czarny, the motion of the particles along with the increase in the lubricant temperature, a certain energetic barrier, which Czarny called *the activation energy*, must be overcome. Czarny also carried out the research on the degradation of the thickener microstructure by means of the Couette rotational rheometer, and on the influence of temperature on the process. He investigated the bentonite greases, the calcium greases (STP, Kalton EP1), the lithium greases (Shell Alvania EP2, ŁT4-S2) as well as the lithium-calcium greases. In accordance with the experimental research and the theory proposed by Czarny, when temperature increases, the thickener intermolecular bonds get weakened, making a cross-linked, compact microstructure, and the activity of single crystallites freely suspended in the base oil increases. As a result, the lubricating greases, in the process of long-lasting shearing, take on the Newtonian fluid characteristics.

**Figure 7.** Dependence of the shear stress as a function of shearing time at constant shear rate ˙γ = 8.1 s-1 (*p* = 0.95) for the greases with the thickener percentage: (a) 6%, (b) 7%, (c) 8%. 1 – the first stage of shearing, 2 – the second stage

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

93

of shearing, 3 – resolidification process of the grease in its relaxation phase.

Vinogradov, Deinega and Verbitsky [18] investigated the correlation between the changes of the friction momentum in the ball bearings and the thixotropic properties of the calcium and the lithium greases. At the research, they observed that the shape of the dispersed phase particles considerably influences the behaviour of the lubricating greases during their further deformation and shear-thinning. On the turn of the 1960s and 1970s, Vinogradov also inves‐ tigated the lubricating greases using the polarimetry and other optical methods. It enabled the quantity evaluation of the level of the thickener microstructure reconstruction. He also conducted research on the thixotropic hysteresis loop for the calcium and lithium greases with the use of the rotational coaxial cylinder rheometers. His works are widely discussed by

Sacchettini, Magnin, Piau and Pierrard [51] published in the Journal of Theoretical and Applied Mechanics a paper on the thixotropic properties of commercial greases: Shell Alvania, ELF Multi and ELF EL series (8305-8308 and 9309) of NLGI (National Lubricating Grease Institute) 2 consistency class. The analyzed lubricating greases showed a different fibre length of the soap thickener (lithium 12-hydroxystearate). The oil base in the greases was the mineral oil. Sacchettini et al. focused mainly on the research of change in the shear stress as a function of shearing time. They analyzed changes in the lithium thickener microstructure at the cyclic shearing lasting a few seconds at different shear rates (from 0.018 s-1 to 18 s-1) and different grease relaxation time (from 2 minutes to 38 hours). They did not analyzed the structural viscosity increase of the greases at their relaxation, however. Apart from the thixotropy,

Czarny, in his paper [52], presented results of the research on the process of degradation and reconstruction of the thickener microstructure in the commercial calcium greases (STP) and the lithium greases (ŁT-4S2). He used the method of cyclic shearing of the lubricating grease at specified time intervals, at a constant shear rate. The research on the grease was carried out without removing it from the rheometer head. He thus simulated the conditions observed in the conduits of lubrication system. In 1990, Czarny investigated the influence of temperature on the thixotropy of the lubricating greases [53]. He proposed a mathematical description of the energetic changes which take place in the microstructure of the thickener during its shearing at various temperatures. Czarny observed that along with the temperature increase, the structural viscosity of the lubricating grease is reduced, and the activity of the thickener particles increases. According to Czarny, the motion of the particles along with the increase in the lubricant temperature, a certain energetic barrier, which Czarny called *the activation energy*, must be overcome. Czarny also carried out the research on the degradation of the thickener microstructure by means of the Couette rotational rheometer, and on the influence of temperature on the process. He investigated the bentonite greases, the calcium greases (STP, Kalton EP1), the lithium greases (Shell Alvania EP2, ŁT4-S2) as well as the lithium-calcium greases. In accordance with the experimental research and the theory proposed by Czarny, when temperature increases, the thickener intermolecular bonds get weakened, making a cross-linked, compact microstructure, and the activity of single crystallites freely suspended in the base oil increases. As a result, the lubricating greases, in the process of long-lasting

Froischteter in [7].

92 Tribology - Fundamentals and Advancements

Sacchettini et al. investigated the wall effects.

shearing, take on the Newtonian fluid characteristics.

**Figure 7.** Dependence of the shear stress as a function of shearing time at constant shear rate ˙γ = 8.1 s-1 (*p* = 0.95) for the greases with the thickener percentage: (a) 6%, (b) 7%, (c) 8%. 1 – the first stage of shearing, 2 – the second stage of shearing, 3 – resolidification process of the grease in its relaxation phase.

Figure 7 illustrates the thixotropy curves for the lubricating greases thickened with the lithi‐ um 12-hydroxystearate based on the mineral oil, received by means of the measuring meth‐ od proposed by Czarny. The grease shearing time in two cycles was 120 minutes in total. The grease relaxation time between the shearing cycles was 24 hours. The research was repe‐ aed five times. The experiment was carried out with the use of Rheotest 2.1 rotational co‐ axial cylinder rheometer working at 2 mm measuring gap. For the measurements, a right handed helix shaped grooved polytetrafluoroethylene internal cylinder, with 95 0.5 mmdeep grooves was used. Such a surface structure enabled a visible reduction of the Weissen‐ berg effect and of the slip effect at the cylinder wall, which could have negatively influenced the measurements.

The energetic interpretation of the lubricating greases thixotropy was presented by Kuhn [54]. He focused primarily on those cases of thixotropy of the lubricating greases which occur during the hydrodynamic lubrication of the slide bearing. According to Kuhn, the energy loss during the friction shearing of the grease is a function of the lubricant's qualities and the conditions it is sheared in. If the lubricant's qualities change in the process of the fluid friction, *the energy density* (energy/volume) also changes:

$$
\sigma\_{r\hbar\_e} = \eta \dot{\gamma} \left(\frac{\overline{d}\_e}{\hbar\_0^\*}\right) \tag{14}
$$

Kuhn's research proved that in the quantification of the tribological processes, the energy loss resulting from the structural changes which take place in the frictional contact region must be

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

95

**5. The rheological properties of the boundary layer in the lubricating**

the conduit diameter, the boundary layer influence considerably decreases [61].

repulsion of the soap particles suspended in the base oil from the walls.

The first mentions of the boundary layer in the lubricating greases appeared in 1960. Bramhal and Hutton [62] carried out the research on the influence of the wall types on the boundary layer formation. The research was conducted with the use of a plunger viscometer. While doing the research, they recorded a slip in the grease, which occurred in the area of the walls. Bramhal and Hutton observed inconsiderable influence of the wall's material and the wall surface topography on the grease slip. They explained the observed effect as a consequence of

An attempt of explaining the phenomenon of the boundary layer formation in the lubricating greases within the area of the lubricating grease contact with the lubrication system conduit wall and the working surface of the sliding pair was made by Czarny. In [52,63] he proves that the boundary layer formation in the lubricating greases depends mainly on the surface material which the lubricating grease reacts with. He observed the changes of the shear stress in the commercial greases in the area of the wall made, among others, of methyl polymethacrylate, polytetrafluoroethylene, polyamide, brass, bronze, duralumin and cast iron. The phenomenon of the boundary layer formation was particularly visible at the low shear rates. According to Czarny, the boundary layer formation is an outcome of the thickener's particles adsorption on the surface of the material. Thus, during the flow of the grease, a toroidal space depleted of

Lubricating greases are, on the one hand, required the best possible slide and anticorrosion properties, and on the other hand, the lowest possible and independent of conditions flow resistance at the supplying of the friction nodes. The problem of the flow resistance of the lubricating greases is currently particularly important in the context of lubrication systems. It results primarily from the fact of a higher and higher automation of the systems. The flow resistance have negative influence on the proper functioning of the central lubrication systems. An important factor reducing the resistance in the lubrication systems is the appearance in the region of the lubricating grease contact with the conduit wall, of the so called *the boundary layer* (or *the wall effect*) which has different rheological properties than the rest of the grease's volume. The phenomenon of the boundary layer formation while the suspension flows through the axisymmetric conduits is frequently referred to as *Segr*é*-Silberberg's effect* or the *sigma effect*. For the first time, the phenomenon was thoroughly described in 1962 [60]. The influence of the boundary layer formation on the fall of the suspension flow resistance is particularly visible in the small diameter axisymmetric conduits where the suspension flow is laminar. It leads to the decrease in the pressure gradient along the conduit. Simultaneously with the increase in

taken into account.

**greases**

where: *erh <sup>e</sup>* – rheological density of energy in the microcontact region (J/mm3 ), *η* –dynamic viscosity (Pas), *γ*˙ – shear rate (s-1), *de* – average diameter of the microcontact (mm), *ho* \* – average lubrication film thickness (mm).

Using the dependence presented above, Kuhn estimated friction coefficient *μ* which was:

$$
\mu = \frac{e\_{rh}}{p\_r} \mathbf{i}\_{rh'} \tag{15}
$$

where: *erh* – sum of the rheological energy densities in the region of all microcontacts (J/mm3 ), *i rh* – friction intensity (non-dimensional value), *pr* – pressure (MPa).

Kuhn also observed a similarity between the curve of changes in the rheological energy density as a function of time at different shear rates and the curves received by Czarny [55] on the experimental way. Later in his work, Kuhn developed the energetic model of the lubricating greases thixotropy [56,57], and together with Balan suggested the experimental ways of estimating the rheological density of energy with the use of the rotational cone-and-plate rheometer [58]. In [59], he defined the tribological properties of the lithium and calcium greases based on the mineral oil by means of the energetic parameters, and also presented the relations between the viscoelastic properties of the lubricating grease, the friction in the frictional contact and the elastic strain energy accumulation and the shear stress at the shearing of the grease. Kuhn's research proved that in the quantification of the tribological processes, the energy loss resulting from the structural changes which take place in the frictional contact region must be taken into account.

Figure 7 illustrates the thixotropy curves for the lubricating greases thickened with the lithi‐ um 12-hydroxystearate based on the mineral oil, received by means of the measuring meth‐ od proposed by Czarny. The grease shearing time in two cycles was 120 minutes in total. The grease relaxation time between the shearing cycles was 24 hours. The research was repe‐ aed five times. The experiment was carried out with the use of Rheotest 2.1 rotational co‐ axial cylinder rheometer working at 2 mm measuring gap. For the measurements, a right handed helix shaped grooved polytetrafluoroethylene internal cylinder, with 95 0.5 mmdeep grooves was used. Such a surface structure enabled a visible reduction of the Weissen‐ berg effect and of the slip effect at the cylinder wall, which could have negatively influenced

The energetic interpretation of the lubricating greases thixotropy was presented by Kuhn [54]. He focused primarily on those cases of thixotropy of the lubricating greases which occur during the hydrodynamic lubrication of the slide bearing. According to Kuhn, the energy loss during the friction shearing of the grease is a function of the lubricant's qualities and the conditions it is sheared in. If the lubricant's qualities change in the process of the fluid friction, *the energy*

> \* 0 , *<sup>e</sup> e*

& (14)

= (15)

), *η* –dynamic

\* – average

),

*d*

æ ö <sup>=</sup> ç ÷ ç ÷ è ø

*h* hg

– rheological density of energy in the microcontact region (J/mm3

Using the dependence presented above, Kuhn estimated friction coefficient *μ* which was:

, *rh rh r e i p* m

where: *erh* – sum of the rheological energy densities in the region of all microcontacts (J/mm3

Kuhn also observed a similarity between the curve of changes in the rheological energy density as a function of time at different shear rates and the curves received by Czarny [55] on the experimental way. Later in his work, Kuhn developed the energetic model of the lubricating greases thixotropy [56,57], and together with Balan suggested the experimental ways of estimating the rheological density of energy with the use of the rotational cone-and-plate rheometer [58]. In [59], he defined the tribological properties of the lithium and calcium greases based on the mineral oil by means of the energetic parameters, and also presented the relations between the viscoelastic properties of the lubricating grease, the friction in the frictional contact and the elastic strain energy accumulation and the shear stress at the shearing of the grease.

viscosity (Pas), *γ*˙ – shear rate (s-1), *de* – average diameter of the microcontact (mm), *ho*

*rh*

*e*

*rh* – friction intensity (non-dimensional value), *pr* – pressure (MPa).

the measurements.

94 Tribology - Fundamentals and Advancements

where: *erh <sup>e</sup>*

*i*

*density* (energy/volume) also changes:

lubrication film thickness (mm).

### **5. The rheological properties of the boundary layer in the lubricating greases**

Lubricating greases are, on the one hand, required the best possible slide and anticorrosion properties, and on the other hand, the lowest possible and independent of conditions flow resistance at the supplying of the friction nodes. The problem of the flow resistance of the lubricating greases is currently particularly important in the context of lubrication systems. It results primarily from the fact of a higher and higher automation of the systems. The flow resistance have negative influence on the proper functioning of the central lubrication systems. An important factor reducing the resistance in the lubrication systems is the appearance in the region of the lubricating grease contact with the conduit wall, of the so called *the boundary layer* (or *the wall effect*) which has different rheological properties than the rest of the grease's volume. The phenomenon of the boundary layer formation while the suspension flows through the axisymmetric conduits is frequently referred to as *Segr*é*-Silberberg's effect* or the *sigma effect*. For the first time, the phenomenon was thoroughly described in 1962 [60]. The influence of the boundary layer formation on the fall of the suspension flow resistance is particularly visible in the small diameter axisymmetric conduits where the suspension flow is laminar. It leads to the decrease in the pressure gradient along the conduit. Simultaneously with the increase in the conduit diameter, the boundary layer influence considerably decreases [61].

The first mentions of the boundary layer in the lubricating greases appeared in 1960. Bramhal and Hutton [62] carried out the research on the influence of the wall types on the boundary layer formation. The research was conducted with the use of a plunger viscometer. While doing the research, they recorded a slip in the grease, which occurred in the area of the walls. Bramhal and Hutton observed inconsiderable influence of the wall's material and the wall surface topography on the grease slip. They explained the observed effect as a consequence of repulsion of the soap particles suspended in the base oil from the walls.

An attempt of explaining the phenomenon of the boundary layer formation in the lubricating greases within the area of the lubricating grease contact with the lubrication system conduit wall and the working surface of the sliding pair was made by Czarny. In [52,63] he proves that the boundary layer formation in the lubricating greases depends mainly on the surface material which the lubricating grease reacts with. He observed the changes of the shear stress in the commercial greases in the area of the wall made, among others, of methyl polymethacrylate, polytetrafluoroethylene, polyamide, brass, bronze, duralumin and cast iron. The phenomenon of the boundary layer formation was particularly visible at the low shear rates. According to Czarny, the boundary layer formation is an outcome of the thickener's particles adsorption on the surface of the material. Thus, during the flow of the grease, a toroidal space depleted of the thickener is created. The number of the adsorbed thickener particles depends on the kind of grease and on the collaborating materials, and is inversely proportional to the thickener percentage.

Czarny and Moes [28] proposed a dependence describing the change of the shear stress in the lubricating grease as a function of the distance from the wall:

$$\tau = \tau\_0 \left\{ 1 - e^{-(\frac{z+d}{s})} \right\} \,\,\,\,\tag{16}$$

the increase of the grease consistency level. Simultaneously, the increase determines the reduction in the thickness of boundary layer *s* and surface layer *d*. The observed boundary layer thickness *s* and surface layer *d* equaled, in the case of Alvania EP2, respectively 3.0-9.5 and 0.6-1.5 µm. According to Czarny, the boundary layer formation in the lubricating greases is catalyzed by different thickener particles size as well as the lubricating grease temperature. Along with the temperature increase, the thickness of the absorbed layer *d* increases. Czarny explains the increase by an easier adsorption of the active soap particles on the material surface at a higher temperature. The thickest, 25 µm boundary layer of Alvania EP2 was observed at temperature of 373 K. It might be interesting to know that, for instance at temperature of 273

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

97

Vinogradov et al. [64,65] investigated the shear stress decrease in the lubricating grease at the wall. The research was carried out on the lithium and calcium soap as well as ceresine thickened greases. All the researched lubricating greases formed the boundary layer. The experiments were conducted with the use of the rotational cone-and-plate rheometer and of the capillary rheometer. They found that the thickness of the boundary layer in the grease is determined by the stress which is present there. Based on the Herschel-Bulkley model, they defined dependences which enable the determination of the grease flow parameters in the gap

The research of the lubricating grease flow in the conduits was conducted also by Swartz and Hardy [66]. They recorded a decrease in resistance of the lubricating grease flow through the steel axisymmetric conduits. According to their hypothesis, it was caused by bleeding of the base oil on the conduit wall and by repulsing the thickener particles from the wall. The

Biernacki in [67-69] investigated the problem of influence of the wall material and of the grease temperature on the boundary layer formation. He studied the phenomenon of the boundary layer formation in the commercial greases, in the area of the walls made of plastics (polyamide, polyvinyl chloride, methyl polymethacrylate, polytetrafluoroethylene) as well as the walls made of metal (steel, copper, mazak, duralumin, aluminium) of different surface topography. The research was carried out mainly with the use of the rotational coaxial cylinder rheometer. In the lithium and calcium greases he recorded lower shear stress values at the walls of the cylinders made of metal than at those made of plastics. The differences in the shear stress values were particularly visible in the low shear rates range and at the low temperature. With the increase of temperature and the shear stress, the differences in the values of the shear stress in the greases depending on the wall material disappear. In the polymer and aluminium greases Biernacki did not observed considerable differences in the shear stress values recorded at the

Biernacki also investigated the problem of the boundary layer thickness in the complex lithium and bentonite greases in the area of two copper and polypropylene plates which approach parallel to each other. The experiment showed that the lithium grease at the copper plate forms a 65 µm thick boundary layer. The depleted part of the lithium thickener was formed 27 µm from the plate surface. Biernacki did not observe the boundary layer formation in the lithium

K, the boundary layer was approximately 2 µm thick.

and in the axisymmetric conduit, including the wall effect.

hypothesis, however, was not experimentally verified.

surface of the walls made of different materials.

where: *τ*<sup>0</sup> – yield stress, *z* – normal wall coordinate, *d* – surface layer thickness, *s* – thickness of the boundary layer with the exponentially decreasing yield stress, down to its minimum value on the wall *τ* =*τw* (Figure 8).

**Figure 8.** Stress distribution τ in the grease boundary layer as a function of the distance to wall *z*, according to Czarny and Moes [28].

According to Czarny, the model can be used for practical purposes at specific design solutions of lubrication systems. Parameter *s* and *d* values, needed for the determination of the stress distribution, which depend on the kind of grease, temperature of the tribological pair, the kind of the pair's wall material and its surface topography can be experimentally defined.

Czarny and Moes, during their research on the lithium soap thickened grease – Alvania EP2, found that the value of the yield stress in the grease mass *τ*<sup>0</sup> and at the wall *τ<sup>w</sup>* rise along with the increase of the grease consistency level. Simultaneously, the increase determines the reduction in the thickness of boundary layer *s* and surface layer *d*. The observed boundary layer thickness *s* and surface layer *d* equaled, in the case of Alvania EP2, respectively 3.0-9.5 and 0.6-1.5 µm. According to Czarny, the boundary layer formation in the lubricating greases is catalyzed by different thickener particles size as well as the lubricating grease temperature. Along with the temperature increase, the thickness of the absorbed layer *d* increases. Czarny explains the increase by an easier adsorption of the active soap particles on the material surface at a higher temperature. The thickest, 25 µm boundary layer of Alvania EP2 was observed at temperature of 373 K. It might be interesting to know that, for instance at temperature of 273 K, the boundary layer was approximately 2 µm thick.

the thickener is created. The number of the adsorbed thickener particles depends on the kind of grease and on the collaborating materials, and is inversely proportional to the thickener

Czarny and Moes [28] proposed a dependence describing the change of the shear stress in the

( ) <sup>0</sup> 1 , *z d s*

where: *τ*<sup>0</sup> – yield stress, *z* – normal wall coordinate, *d* – surface layer thickness, *s* – thickness of the boundary layer with the exponentially decreasing yield stress, down to its minimum

**Figure 8.** Stress distribution τ in the grease boundary layer as a function of the distance to wall *z*, according to Czarny

According to Czarny, the model can be used for practical purposes at specific design solutions of lubrication systems. Parameter *s* and *d* values, needed for the determination of the stress distribution, which depend on the kind of grease, temperature of the tribological pair, the kind

Czarny and Moes, during their research on the lithium soap thickened grease – Alvania EP2, found that the value of the yield stress in the grease mass *τ*<sup>0</sup> and at the wall *τ<sup>w</sup>* rise along with

of the pair's wall material and its surface topography can be experimentally defined.

(16)

lubricating grease as a function of the distance from the wall:

t t *e* <sup>+</sup> ì ü - ï ï = - í ý ï ï î þ

percentage.

and Moes [28].

value on the wall *τ* =*τw* (Figure 8).

96 Tribology - Fundamentals and Advancements

Vinogradov et al. [64,65] investigated the shear stress decrease in the lubricating grease at the wall. The research was carried out on the lithium and calcium soap as well as ceresine thickened greases. All the researched lubricating greases formed the boundary layer. The experiments were conducted with the use of the rotational cone-and-plate rheometer and of the capillary rheometer. They found that the thickness of the boundary layer in the grease is determined by the stress which is present there. Based on the Herschel-Bulkley model, they defined dependences which enable the determination of the grease flow parameters in the gap and in the axisymmetric conduit, including the wall effect.

The research of the lubricating grease flow in the conduits was conducted also by Swartz and Hardy [66]. They recorded a decrease in resistance of the lubricating grease flow through the steel axisymmetric conduits. According to their hypothesis, it was caused by bleeding of the base oil on the conduit wall and by repulsing the thickener particles from the wall. The hypothesis, however, was not experimentally verified.

Biernacki in [67-69] investigated the problem of influence of the wall material and of the grease temperature on the boundary layer formation. He studied the phenomenon of the boundary layer formation in the commercial greases, in the area of the walls made of plastics (polyamide, polyvinyl chloride, methyl polymethacrylate, polytetrafluoroethylene) as well as the walls made of metal (steel, copper, mazak, duralumin, aluminium) of different surface topography. The research was carried out mainly with the use of the rotational coaxial cylinder rheometer. In the lithium and calcium greases he recorded lower shear stress values at the walls of the cylinders made of metal than at those made of plastics. The differences in the shear stress values were particularly visible in the low shear rates range and at the low temperature. With the increase of temperature and the shear stress, the differences in the values of the shear stress in the greases depending on the wall material disappear. In the polymer and aluminium greases Biernacki did not observed considerable differences in the shear stress values recorded at the surface of the walls made of different materials.

Biernacki also investigated the problem of the boundary layer thickness in the complex lithium and bentonite greases in the area of two copper and polypropylene plates which approach parallel to each other. The experiment showed that the lithium grease at the copper plate forms a 65 µm thick boundary layer. The depleted part of the lithium thickener was formed 27 µm from the plate surface. Biernacki did not observe the boundary layer formation in the lithium grease at the polypropylene plate. In the bentonite grease, the boundary layer was not formed in any of the cases.

Critical for the question of the flow resistance is also the material used for the making of the lubrication conduits as well as their surface topography. Some materials (copper, steel and cast iron in particular) show the increased adsorption of the particles of the thickener on their surface. It results in the generation of certain thickener depleted zones, in a shape of a ring, presenting a lower structural viscosity. The flow resistance at the start of the pump dosing the grease depending on the material of which the lubrication conduit is made, can be reduced even by half. In the case when the roughness of the internal surface finish of the conduit is higher or equal to the thickness of the thickener depleted layer, then the grease flow resistance is considerably higher than the one which would be observed at the smooth surface finish. While designing the lubrication systems, it is critical to pay attention to whether the roughness of the lubrication conduit surface are as low as possible. It is important to mention here that the pressure drops connected with the resistance, in the majority of the lubrication system elements are local in their character, whereas in the lubrication conduits, there are constant pressure drops. In the case of the particularly long conduits, the flow resistance for the entire lubrication system is predominant there. Other elements which can increase the lubricant flow resistance as well as the pressure in the system are various types of gaps in dividers. Also in this case, their shape, section area and the material of which they are made are important. Similarly complex is a problem of determining the resistance generated by the very friction nodes and defining the pressure necessary to deliver the required grease portion to the bearing. The most important design characteristics of the delivered grease flow resistance reduction bearings are primarily their shape, dimensions of the bearings' working side surfaces, the distribution and way the lubrication gaps are cut, as well as the circumferential backlash.

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

99

**Nomenclature**

*m* – index exponent

*γC* – critical strain, %

*γ*˙ – shear rate, s-1

*η* ′

*α*1 – the not entangled particles shear stress, Pa

*β*1*–* shear intensity for the not entangled particles, s-1

*β*2 – shear intensity for the entangled particles, s-1

*η* – apparent viscosity (dynamic viscosity), Pas

*η*0 – viscosity at zero shear rate, Pas

– viscous component, Pas

*η∞* – viscosity at infinite shear rate, Pas

*α*2 – the entangled particles shear stress, Pa

Delgado et al. in [70-72] did research on the influence of air, roughness of the steel conduit wall surface, as well as the conduit diameter on the pressure drop at the horizontal flow of the grease. They observed that the roughness of the conduit wall influences the formation of the boundary layer in the lubricating grease. The reduction of the surface roughness considerably reduced the pressure gradient. The differences grew with the conduit diameter reduction or with the grease flow rise. Delgado et al. also found out that the air present in the lubricating grease causes the reduction of the wall slip. According to their hypothesis, it is caused by the fact that the air in the lubricating grease generates considerable differences in the moistening of the internal walls of the conduit. The research was carried out mainly on the complex soap lubricants.

#### **6. Conclusions**

A conclusion of the presented research is that the flow of the plastic greases in the lubrication systems depends not only on the design of the systems but also on the kind of grease. Greases are non-Newtonian fluids of a highly complex structure, and therefore the to carry out the evaluation process of their behaviour in particular conditions is multi-stage and very difficult. The flow resistance of the grease in particular elements of the lubrication system depend, among others, on the structural viscosity of the grease. All greases get thinned, to a larger or smaller extent, in the shearing process. At the distribution of the grease to the friction nodes in a particular period of time, the flow resistance gets reduced as a result of the grease thickener's microstructure degradation. The decrease in the grease flow resistance is strictly correlated with the decrease of the shear stress in the grease. In the first seconds of supplying a new portion of the grease into the main conduit of the lubrication system, the pump must in the first place overcome the flow resistance generated by the grease, which is connected with the occurrence of the yield stress in the grease. The value of the stress, as well as the charac‐ teristics of the shear thinning for a particular grease, among other factors, depend on the flow rate generated by the lubrication pump, but also on the physical and chemical activity of the environment. An important determinant on the generating the flow resistance in the lubrica‐ tion systems are the physical-chemical phenomena connected with the interaction of the grease thickener particles suspended in the base oil. One of them is a phenomenon of the grease thixotropy, consisting in its gel→sol→gel transition, namely the reduction in the grease structural viscosity at its flow and a partial thickener microstructure reconstruction in the phase of the grease relaxation, between the consecutive cycles of its distribution to the friction nodes. The phenomenon is crucial in the case of the greases which stay in the conduits and dividers of the lubrication system. It can be dangerous for the complex systems, working cyclically (for instance, the central lubrication systems of big mining machines). A short-term, uncontrolled increase in the flow resistance in such systems may lead to the excessive energy consumption by the pump, and in the extreme cases, to damage of the charging unit, and consequently, to the system failure.

Critical for the question of the flow resistance is also the material used for the making of the lubrication conduits as well as their surface topography. Some materials (copper, steel and cast iron in particular) show the increased adsorption of the particles of the thickener on their surface. It results in the generation of certain thickener depleted zones, in a shape of a ring, presenting a lower structural viscosity. The flow resistance at the start of the pump dosing the grease depending on the material of which the lubrication conduit is made, can be reduced even by half. In the case when the roughness of the internal surface finish of the conduit is higher or equal to the thickness of the thickener depleted layer, then the grease flow resistance is considerably higher than the one which would be observed at the smooth surface finish. While designing the lubrication systems, it is critical to pay attention to whether the roughness of the lubrication conduit surface are as low as possible. It is important to mention here that the pressure drops connected with the resistance, in the majority of the lubrication system elements are local in their character, whereas in the lubrication conduits, there are constant pressure drops. In the case of the particularly long conduits, the flow resistance for the entire lubrication system is predominant there. Other elements which can increase the lubricant flow resistance as well as the pressure in the system are various types of gaps in dividers. Also in this case, their shape, section area and the material of which they are made are important. Similarly complex is a problem of determining the resistance generated by the very friction nodes and defining the pressure necessary to deliver the required grease portion to the bearing. The most important design characteristics of the delivered grease flow resistance reduction bearings are primarily their shape, dimensions of the bearings' working side surfaces, the distribution and way the lubrication gaps are cut, as well as the circumferential backlash.

#### **Nomenclature**

grease at the polypropylene plate. In the bentonite grease, the boundary layer was not formed

Delgado et al. in [70-72] did research on the influence of air, roughness of the steel conduit wall surface, as well as the conduit diameter on the pressure drop at the horizontal flow of the grease. They observed that the roughness of the conduit wall influences the formation of the boundary layer in the lubricating grease. The reduction of the surface roughness considerably reduced the pressure gradient. The differences grew with the conduit diameter reduction or with the grease flow rise. Delgado et al. also found out that the air present in the lubricating grease causes the reduction of the wall slip. According to their hypothesis, it is caused by the fact that the air in the lubricating grease generates considerable differences in the moistening of the internal walls of the conduit. The research was carried out mainly on the complex soap

A conclusion of the presented research is that the flow of the plastic greases in the lubrication systems depends not only on the design of the systems but also on the kind of grease. Greases are non-Newtonian fluids of a highly complex structure, and therefore the to carry out the evaluation process of their behaviour in particular conditions is multi-stage and very difficult. The flow resistance of the grease in particular elements of the lubrication system depend, among others, on the structural viscosity of the grease. All greases get thinned, to a larger or smaller extent, in the shearing process. At the distribution of the grease to the friction nodes in a particular period of time, the flow resistance gets reduced as a result of the grease thickener's microstructure degradation. The decrease in the grease flow resistance is strictly correlated with the decrease of the shear stress in the grease. In the first seconds of supplying a new portion of the grease into the main conduit of the lubrication system, the pump must in the first place overcome the flow resistance generated by the grease, which is connected with the occurrence of the yield stress in the grease. The value of the stress, as well as the charac‐ teristics of the shear thinning for a particular grease, among other factors, depend on the flow rate generated by the lubrication pump, but also on the physical and chemical activity of the environment. An important determinant on the generating the flow resistance in the lubrica‐ tion systems are the physical-chemical phenomena connected with the interaction of the grease thickener particles suspended in the base oil. One of them is a phenomenon of the grease thixotropy, consisting in its gel→sol→gel transition, namely the reduction in the grease structural viscosity at its flow and a partial thickener microstructure reconstruction in the phase of the grease relaxation, between the consecutive cycles of its distribution to the friction nodes. The phenomenon is crucial in the case of the greases which stay in the conduits and dividers of the lubrication system. It can be dangerous for the complex systems, working cyclically (for instance, the central lubrication systems of big mining machines). A short-term, uncontrolled increase in the flow resistance in such systems may lead to the excessive energy consumption by the pump, and in the extreme cases, to damage of the charging unit, and

in any of the cases.

98 Tribology - Fundamentals and Advancements

lubricants.

**6. Conclusions**

consequently, to the system failure.


*T* – temperature, K

**Author details**

Maciej Paszkowski

Poland

**References**

1956;60(4) 439-442.

ers; 1975 (in Polish).

nology; 1992 (in Polish).

ing Chemistry 1936;28(4) 414-416.

ternational Limited Publishers, 2005.

cation and Tribology 1995;47 3-7.

*z* – normal wall coordinate

*w* ′

*w* – elastic strain energy of the entangled particles

– elastic strain energy of the not entangled particles

Address all correspondence to: maciej.paszkowski@pwr.wroc.pl

Wroclaw University of Technology, Institute of Machines Design and Operation, Wroclaw,

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

101

[1] Czarny, R. Lubricating Greases. Warsaw: WNT Publishers, (2004) (in Polish).

[5] Boner CJ. Modern Lubricating Greases, Broseley: Scientific Publication; 1976.

[2] Vold JM. Colloidal Structure in Lithium Stearate Greases. Physical Chemistry

[3] Farrington BB, Davis WN. Structure of Lubricating Greases. Industrial and Engineer‐

[4] IshchukYuL.Lubricating Grease Manufacturing Technology.New Delhi: New Age In‐

[6] Czarny R. Effect of Changes in Grease Structure on Sliding Friction, Industrial Lubri‐

[7] Froischteter GB, Trilisky KK, IshchukYuL, Stupak PM. Rheological and Thermophys‐ ical Properties of Greases. London: Gordon & Breach Science Publishers, 1989.

[8] Basiński A et al. Polish Dictionary of Chemical Terminology. Warsaw: WNT Publish‐

[9] Czarny R. Investigation into Phenomena Accompanying the Flow of Greases in Lu‐ brication Systems. Wroclaw: Publishing House of the Wroclaw University of Tech‐


*kf* – parameter determining the time of the entangled particles → not entangled particles reaction *m* – index exponent


*s* – thickness of the boundary layer with the exponentially decreasing yield stress, µm

*T* – temperature, K

*η* ″

*erh e*

*G* ′

*G* ″

*G* <sup>0</sup>

*ho*

*i*

– elastic component, Pas

100 Tribology - Fundamentals and Advancements


*χ*1 – share of the not entangled particles

*de* – average diameter of the microcontact, mm

\* – average lubrication film thickness, mm

*kc* – consistency factor of the thickener in the grease, Pas

*erh* – sum of the rheological energy densities in the region of all microcontacts, J/mm3

– rheological density of energy in the microcontact region, J/mm3

*kb* – parameter determining the time of the not entangled → entangled particles

*kf* – parameter determining the time of the entangled particles → not entangled particles

*s* – thickness of the boundary layer with the exponentially decreasing yield stress, µm

*χ*2 – share of the entangled particles

*a* – rheological parameter

*d* – surface layer thickness, µm

– storagemodulus, Pa

*<sup>N</sup>* – plateau modulus, Pa


– lossmodulus, Pa

*rh* – friction intensity

*k* – Boltzmann constant, J/K

reaction *m* – index exponent

*n* – index exponent

*pr* – pressure, MPa

*θ* – thixotropy coefficient

*λ* – time-constant, s *μ* – friction coefficient

*τ* – shear stress, Pa

*τ*0 – yield stress, Pa


#### **Author details**

Maciej Paszkowski

Address all correspondence to: maciej.paszkowski@pwr.wroc.pl

Wroclaw University of Technology, Institute of Machines Design and Operation, Wroclaw, Poland

#### **References**


[10] Mewis J. Thixotropy – A General Review, Journal of Non-Newtonian Fluid Mechan‐ ics 1979;6 1-20.

[25] Herschel WH, Bulkley R. Measurement of Consistency as Applied to Rubber – Ben‐ zene Solutions. Proceedings of American Society of Testing Materials 1926;26,

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

103

[26] Yasuda K, Armstrong R, Cohen RE. Shear Flow Properties of Concentrated Solutions of Linear and Star Branched Polystyrenes. RheologicaActa 1981;20 163-178.

[27] Bird R, Armstrong R, Hassager O. Dynamics of Polymer Liquids: Volume 1. Fluid

[28] Czarny R, Moes H. Some Aspects of Lubricating Grease Flow, 3rd International Tri‐

[29] Liu Ch, He J, Ruymbeke E, Keunings R, Bailly C. Evaluation of Different Methods for the Determination of the Plateau Modulus and the Entanglement Molecular Weight.

[30] Martín-Alfonso JE, Valencia C, Sánchez MC, Franco JM, Gallegos C. Recycled and Virgin LDPE as Rheology Modifiers of Lithium Lubricating Greases: A Comparative

[31] Martín-Alfonso JE, Valencia C, Sánchez MC, Franco JM, Gallegos C. Rheological Modification of Lubricating Greases with Recycled Polymers from Different Plastics

[32] Péterfi T. Die Abhebung der Befruchtungsmembran bei Seeigeleiern. Archiv für En‐

[33] Lagaly G, Beneke K. Eighty years of colloid science in Hungary and Germany. Uni‐

[35] Efremov IF, Us'yarov OG. The Long-range Interaction Between Colloid and Other Particles and the Formation of Periodic Colloid Structures. Russian Chemical Re‐

[36] Efremov IF. The Dilatancy of Colloidal Structures and Polymer Solutions, Russian

[37] Sonntag H. Lehrbuch der Kolloidwissenschaft. Berlin: VEB Deutscher Verlag der

[38] Scheludko A. Colloid Chemistry. Warsaw: WNT Publishers, 1969. Translated from

[39] Barnes HA. Thixotropy – a Review. Journal of Non-Newtonian Fluid Mechanics

Waste. Industrial & Engineering Chemistry Research 2009;48 4136–4144.

Mechanics. New York: John Wiley & Sons; 1987.

Polymer 2006;47 4461–4479.

versity Kiel, 2002.

views 1976;45(5) 435-453.

Wissenschaften, 1977.

1997;70 1-33.

bology Congress – Eurotrib '81, 1981, Warsaw, Poland.

Study. Polymer Engineering and Science 2008;48 1112–1119.

twicklungsmechanik der Organismen 1927;112 660-695.

[34] Freundlich H. Thixotropie. Paris: Hermann, 1935.

Chemical Reviews 1982;51(2) 160-177.

Russian by Siedlecka Z (in Polish).

621-633.


[10] Mewis J. Thixotropy – A General Review, Journal of Non-Newtonian Fluid Mechan‐

[11] Luckham PF, Rossi S. The Colloidal and Rheological Properties of Bentonite Suspen‐

[12] Delgado MA, Valencia C, Sanchez MC, Franco JM, Gallegos C. Influence of Soap Concentration and Oil Viscosity on the Rheology and Microstructure of Lubricating

Greases. Industrial and Engineering Chemistry Research 2006;45(6) 1902-1910.

[13] Moore RJ, Cravath AM. Mechanical Breakdown of Soap-base Greases, Industrial and

[14] Yeong SK, Luckham PF, TadrosThF. Steady Flow and Viscoelastic Properties of Lu‐ bricating Grease Containing Various Thickener Concentrations. Journal of Colloid

[15] Cann PM. Grease Degradation in Bearing Simulation Device. Tribology International

[16] Cousseau T, Graca BM, Campos AV, Seabra JHO. Influence of Grease Rheology on Thrust Ball Bearings Friction Torque, Tribology International 2012;46 106-113.

[17] Farcas F, Gafitanu MD. Some Influence Parameters on Grease Lubricated Rolling

[18] Vinogradov GV, DeinegaYuF, VerbitskyYaA. Thixotropy of Greases and Low-speed Operation of Rotational Plastoviscometers and Rolling Bearings.RheologicaActa

[19] CannPM, Spikes HA. Film Thickness Measurements of Lubricating Grease Under

[20] Wiliamson BP, Kendal DR, Cann PM. The Influence of Grease Composition on Film

[21] Krawiec S. On the Mechanism of the Synergistic Effect of PTFE and Copper in a Lith‐ ium Grease Lubricant. Industrial Lubrication and Tribology 2011;63(3) 171-177.

[22] Krawiec S. The Synergistic Effect of Copper Powder with PTFE in a Grease Lubricant Under Mixed Friction Conditions. Archives of Civil and Mechanical Engineering

[23] Czarny R, Paszkowski M. Einfluss von Zusatzen auf die rheologischen Eigenschaften der Schmierfette, 15th International Colloquium Tribology. Automotive and Indus‐

[24] Czarny R, Paszkowski M. The Influence of Graphite Solid Additives, MoS2 and PTFE on Changes in Shear Stress Values in Lubricating Greases. Journal of Synthetic Lubri‐

Normally Starved Conditions. NLGI Spokesman 1992;56 (2) 21-26.

Thickness in EHD Contacts. NLGI Spokesmen 1993;57(8) 13-18.

trial Lubrication, 2006, Stuttgart-Ostfildern, Germany.

sions. Advances in Colloid and Interface Science 1999;82 43-92.

Engineering Chemistry 1951;43(12) 2892-2897.

Contacts Service Life. Wear 1999;225-229 1004-1010.

and Interface Science 2004;274 285-293.

ics 1979;6 1-20.

102 Tribology - Fundamentals and Advancements

2006;39 1698-1706.

1967; 6(3) 252-259.

2011;11(2) 379-390.

cation 2007;24(1) 19-29.


[40] Franco JM, Delgado MA, Valencia C. Combined Oxidative-shear Resistance of Cas‐ trol Oil-based Lubricating Greases, 3rd Arnold Tross Colloquium, 2007, Hamburg, Germany.

[55] Czarny R. Einfluss der Thixotropie auf die rheologischen Eigenschaften der

Some Aspects of Grease Flow in Lubrication Systems and Friction Nodes

http://dx.doi.org/10.5772/55929

105

[56] Kuhn E. Energy Investigation of the Rheological Wear of Lubricant Greases. Applied

[57] Kuhn E. Description of the Energy Level of Tribologically Stressed Greases. Wear

[58] Kuhn E, Balan C. Experimental Procedure for the Evaluation of the Friction Energy

[59] Kuhn E. Experimental Grease Investigation From an Energy Point of View. Industrial

[60] Segré G, Silberberg A. Behavior of Macroscopic Rigid Spheres in Poiseuille Flow. Part II. Experimental Results and Interpretation.Journal of fluid mechanics 1962;14

[61] Heywood NI, Cheng DC-H. Flow in Pipes. Part II. Multiphase flow.Physics in Tech‐

[62] Bramhall AD, Hutton JF. Wall Effect in the Flow of Lubricating Greases in Plunger

[63] Czarny R. Influence of the Wall Type on Formation of the Grease Wall Effect, Confer‐ ence materials, XXI Autumn Tribological School – Machine and appliance friction pairs lubrication. Contemporary tendencies in the research theory development,

[64] Vinogradov GV, Froishteter GB, Trilsky KK. The Generalized Theory of Flow of Plas‐ tic Disperse Systems with Account of the Wall Effect.RheologicaActa 1978;17(2)

[65] Vinogradov GV, Froishteter GB, Trilsky KK, Smorodinsky EL. The Flow of Plastic Disperse Systems in the Presence of the Wall Effect.RheologicaActa 1975;14(9)

[66] Swartz CJ, Hardy B. Mathematical Model of Grease Flow in Pipes. NLGI Spokesman

[67] Biernacki K. Influence of Surface Roughness and Temperature on the Lubricating

[68] Biernacki K, Czarny R. Influence of the Wall Material and Temperature on the Rheo‐ logical Properties of the Boundary Layer in the Lithium and Calcium Greases. Hy‐

[69] Biernacki K. Analysis of the Wall Effect in Lubricating Greases.Tribologia 2006;5

Grease Flow in the Boundary Layer.Tribologia 2000;2 187-198 (in Polish).

Viscometers. British Journal of Applied Physics 1960;11 363-369.

Schmierfette. Tribologie und Schmierungstechnik 1989;3 (36) 134-140.

of Lubricating Greases. Wear 1997;209(1997) 237-240.

Lubricating and Tribology 1999;51(5) 246-251.

1996, Lodz – Arturowek, Poland (in Polish).

draulika i Pneumatyka 2002;4 24-25 (in Polish).

Rheology 1992;2(4) 252-257.

1995;188(1995) 138-141.

136-157.

156-165.

765-775.

1991;55(3) 14 –17.

111-129 (in Polish).

nology 1984;15.


[55] Czarny R. Einfluss der Thixotropie auf die rheologischen Eigenschaften der Schmierfette. Tribologie und Schmierungstechnik 1989;3 (36) 134-140.

[40] Franco JM, Delgado MA, Valencia C. Combined Oxidative-shear Resistance of Cas‐ trol Oil-based Lubricating Greases, 3rd Arnold Tross Colloquium, 2007, Hamburg,

[41] Renshaw TA. Effect of Shear on Lithium Greases and Their Soap Phase. Industrial

[42] Sirianni AF, Moses GB, Puddington IE. Particle Shape in Thixotropic Suspensions.

[43] Goodeve CF, Whitfield GW, The Measurement of Thixotropy in Absolute Units.

[44] [44] Arnold JE, Goodeve CF. The Coefficient of Thixotropy of Suspensions of Carbon

[45] Moses GB, Puddington IE. Rheological Properties of Some Soap-Oil Systems, Cana‐

[46] Hahn SJ, Ree T, Eyring H. A Theory of Thixotropy and its Application to

[47] Houser EA, Reed CE. Studies in Thixotropy II – The Thixotropic Behavior and Struc‐

[48] Hahn SJ, Eyring H, Higuchi I, Ree T. Flow Properties of Lubricating Oils Under Pres‐

[49] Utsugi H, Kim K, Ree T, Eyring H. Thixotropy Property of Lubricating Grease, NLGI

[50] Bair S. Actual Eyring Models for Thixotropy and Shear-thinning: Experimental Vali‐ dation and Application to EHD. ASME Journal of Tribology 2004;126 728-732.

[51] Sacchettini M, Magnin A, Piau JM, Pierrard JM. Charactérisation d'une graisse lubri‐ fiante en écoulements viscosimétriques transitoires. Journal of Theoretical and Ap‐

[52] Czarny R. Influence of the Lubricating Grease Structure on its Rheological Proper‐ ties.Scientific Works of the Institute of Machine Design and Operation of Wroclaw University of Technology.Problems of Friction, Lubrication and Machine System Computerization. Wroclaw: Publishing House of Wroclaw University of Technology;

[53] Czarny R. Effect of Temperature on the Thixotropy Phenomenon in Lubricating Greases, Proceedings of the Japan International Tribology Conference, 1990, Nagoya,

[54] Kuhn E. An Energy Interpretation of Thixotropic Effect. Wear 1991;142(1991) 203-205.

Black in Mineral Oil. Journal of Physical Chemistry 1940;44(5) 652-670.

ture of Bentonite. Journal of Physical Chemistry 1937;41(7) 911-934.

Spokesman, NLGI 28th Annual Meeting, 1960, Chicago, USA.

and Engineering Chemistry 1955;47(4) 834-838.

Canadian Journal of Chemistry 1951;29 166-172.

Transaction of Faraday Society 1938;34 511-520.

dian Journal of Chemistry 1951;29 996-1009.

Grease.NLGI Spokesman 1957;21(3) 12-20.

sure. NLGI Spokesman 1958;21(3) 123-128.

plied Mechanics, Numérospécial 1985; 165-199.

1989 (in Polish).

Japan.

Germany.

104 Tribology - Fundamentals and Advancements


[70] Delgado MA, Franco JM, Partal P, Gallegos C. Experimental Study of Grease Flow in Pipelines: Wall Slip and Air Entrainment Effects. Chemical Engineering and Process‐ ing 2005;44 805–817.

**Section 2**

**Boundary Lubrication Applications**


**Boundary Lubrication Applications**

[70] Delgado MA, Franco JM, Partal P, Gallegos C. Experimental Study of Grease Flow in Pipelines: Wall Slip and Air Entrainment Effects. Chemical Engineering and Process‐

[71] Ruiz-Viera MJ, Delgado MA, Franco JM, Gallegos C. Evaluation of Wall Slip Effects in the Lubricating Grease/Air Two-phase Flow Along Pipelines. Journal of Non-

[72] Ruiz-Viera MJ, Delgado MA, Franco JM, Sanchez MC, Gallegos C. On the Drag Re‐ duction for the Two-phase Horizontal Pipe Flow of Highly Viscous Non-Newtonian Liquid/Air Mixtures: Case of Lubricating Grease. International Journal of Multiphase

ing 2005;44 805–817.

106 Tribology - Fundamentals and Advancements

Flow 2006;32 232–247.

Newtonian Fluid Mechanics 2005;139 190–196.

**Chapter 4**

**Cryogenic Tribology in High-Speed Bearings and**

In recent years, as a rule, improvement of the reliability of liquid propellant rockets becomes an international technical problem for built-up of safe space transport systems. The high performance, liquid propellant rocket engines require high-pressured turbopumps to deliver extremely low temperature propellants of liquid oxygen (LO2, boiling point 90 K) and liquid hydrogen (LH2, boiling point 20 K) to a combustion chamber in engine [1]. In LO2/LH2 turbopumps, cryogenic high-speed bearings and rotating-shaft seals are very important parts to sustain high reliability of the high-rotating-shaft systems. The turbopump bearings are directly equipped in cryogenic propellants in pump side [2]. The shaft seal systems are also set up between the cryogenic pumps and the hot turbines to restrain the leakage of cryogenic

These bearing and shaft seals have to operate under poor lubricating conditions due to extremely small viscosity at cryogenic temperatures. Furthermore, the turbopump bearings and shaft seals have to overcome a severe high-speed operation that has several critical speeds demonstrating self-induced severe vibration of the rotating shaft. In order to develop turbo‐ pump bearings and shaft seals, many inexperienced technical and tribological problems must be solved for extremely low temperature and high speed of operational conditions. Such cryogenic tribological technology has been playing a key role in cryogenic turbopumps to

This chapter presents a topical review of cryogenic tribological studies (for about 30 years in Japan) on the research and development of the cryogenic high-speed bearings and shaft seals of rocket turbopumps [4, 5]. The high-speed bearings and shaft seals were continually studied for the LE-5 engine used in the Japanese H-I rocket (developed in 1986) and the LE-7 engine used in the H-II rocket (developed in 1994). The bearings and shaft seals used in LO2/LH2

> © 2013 Nosaka and Kato; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

> © 2013 Nosaka and Kato; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

**Shaft Seals of Rocket Turbopumps**

Additional information is available at the end of the chapter

Masataka Nosaka and Takahisa Kato

http://dx.doi.org/10.5772/55733

propellants and hot turbine gas [3].

achieve high reliability.

**1. Introduction**

### **Cryogenic Tribology in High-Speed Bearings and Shaft Seals of Rocket Turbopumps**

Masataka Nosaka and Takahisa Kato

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/55733

#### **1. Introduction**

In recent years, as a rule, improvement of the reliability of liquid propellant rockets becomes an international technical problem for built-up of safe space transport systems. The high performance, liquid propellant rocket engines require high-pressured turbopumps to deliver extremely low temperature propellants of liquid oxygen (LO2, boiling point 90 K) and liquid hydrogen (LH2, boiling point 20 K) to a combustion chamber in engine [1]. In LO2/LH2 turbopumps, cryogenic high-speed bearings and rotating-shaft seals are very important parts to sustain high reliability of the high-rotating-shaft systems. The turbopump bearings are directly equipped in cryogenic propellants in pump side [2]. The shaft seal systems are also set up between the cryogenic pumps and the hot turbines to restrain the leakage of cryogenic propellants and hot turbine gas [3].

These bearing and shaft seals have to operate under poor lubricating conditions due to extremely small viscosity at cryogenic temperatures. Furthermore, the turbopump bearings and shaft seals have to overcome a severe high-speed operation that has several critical speeds demonstrating self-induced severe vibration of the rotating shaft. In order to develop turbo‐ pump bearings and shaft seals, many inexperienced technical and tribological problems must be solved for extremely low temperature and high speed of operational conditions. Such cryogenic tribological technology has been playing a key role in cryogenic turbopumps to achieve high reliability.

This chapter presents a topical review of cryogenic tribological studies (for about 30 years in Japan) on the research and development of the cryogenic high-speed bearings and shaft seals of rocket turbopumps [4, 5]. The high-speed bearings and shaft seals were continually studied for the LE-5 engine used in the Japanese H-I rocket (developed in 1986) and the LE-7 engine used in the H-II rocket (developed in 1994). The bearings and shaft seals used in LO2/LH2

© 2013 Nosaka and Kato; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Nosaka and Kato; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

turbopumps of the LE-5 and LE-7 had a rotational speed level of 50,000 rpm and had been studied and developed from the mid-1970 to the mid-1990. Specially, the all-steel bearings (made of AISI 440C) of the LH2 turbopump of the LE-7 demonstrated high performance with high reliability at high-speed level at 2 million DN (40 mm x 50,000 rpm). The shaft seal systems in the LE-5/LE-7 turbopumps that used a mechanical seal, a floating ring seal (annular seal) and a segmented seal are also reviewed.

reviews are based on previous studies carried out by Japan Aerospace Exploration Agency (JAXA) at Kakuda Space Center. All materials used in this chapter have been published by

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

111

**Figure 1.** Typical tribo-components and solid lubricants used in turbopumps

The LO2/LH2 turbopumps as well as the tribo-components, such as high-speed bearings and rotating shaft-seals, were studied and developed to use in the LE-5 and LE-7. In reference to the structure of the LH2 turbopump of the LE-7, the tribo-components and solid lubricants used in the LE-5 and LE-7 turbopumps are typically indicated in Fig. 1 [4]. In addition, main design parameters of the turbopumps and DN values of bearings for the LE-5 and LE-7 are listed in Table 1 [5]. Here, the DN value that represents high-speed level of bearing is defined as the product of the inner-race bore diameter D (in mm) and the pump rotational speed N (in

rpm). The rotor speed is typically restricted by the DN limits of the bearing.

**2. Bearings and shaft seals of turbopumps**

**2.1. Turbopumps and tribo–components**

papers.

Furthermore, for future space transport systems to reduce launch cost and to increase effi‐ ciency, advanced rocket engines which are characterized by high durability (long life) and high performance (light weight) are required in recent years. Advanced bearing and shaft seal that have high durability, i.e., a long life of 7.5 hours for the turbopump bearings used in reusable space shuttle main engine (the SSME). Its required life is 15 times longer than that (30 minutes) of the turbopump bearings used in the LE-7. At the first time, the SSME turbopump bearings experienced a serious wear problem in LO2 due to poor self-lubrication of the retainer [6]. In order to extend bearing life, the hybrid ceramic bearing with Si3N4 balls was used to reduce serious wear in the conventional all-steel bearing. A new type of the retainer having PTFE/ bronze-powder insert was also developed to obtain sufficient self-lubrication of the hybrid ceramic bearing. Consequently, the improvement of the SSME turbopump bearings needed a long time of about 20 years [7].

Today, ultra-high speed level above 100,000 rpm is required to make a small and light turbopump for advanced second-stage engine. These advanced research and development are actively underway. In Japan, a new type of hybrid ceramic bearing having Si3N4 balls with a single guided retainer demonstrated excellent performance at an ultra-high speed of 120,000 rpm (3 million DN) in LH2 and recorded the world's top speed (in 2001) [8]. The result of this bearing was applied to the LH2 turbopump (rotational speed, 90,000 rpm) of the RL60 demonstrator engine (in 2003). The RL60 demonstrator engine was developed in the USA with international collaboration (USA, Japan, Russia and Sweden) and the LH2 turbopump was developed by a Japanese company [9]. In Europe, for the VINCI engine under development, high-DN hybrid ceramic bearing was tested in LH2 at a speed of 70,000 rpm (2.8 million DN) and continuous studies on a high-DN bearing was conducted at DN up to 3.3 million (120,000 rpm) in LH2 (in 2005) [10]. Furthermore, in Russia, for the developed RD0146 engine, its rotational speed of the main LH2 turbopump was 123,000 rpm (3.08 million DN), but detail of its bearing was unknown (in 2003) [11].

This chapter also reviews advanced bearings and shaft seals which were studied from the mid-1990 to the mid-2000 after the development of turbopump bearings and shaft seals of the LE-7 [4,5]. It is typical that a long-life bearing with single-guided retainer demonstrat‐ ed a long operation for 12 hours under 50,000 rpm. A hybrid ceramic bearing having singleguided retainer and Si3N4 balls was able to demonstrate ultra-high-speed performance at speeds up to 120,000 rpm and show excellent performance under 3 million DN. An annular seal made of an Ag plated steel ring also presented two-phase seal performance at speeds up to 120,000 rpm.

These historical reviews are intended to help the technical succession to next young generation who challenges research and development of the future space transportation system. These reviews are based on previous studies carried out by Japan Aerospace Exploration Agency (JAXA) at Kakuda Space Center. All materials used in this chapter have been published by papers.

**Figure 1.** Typical tribo-components and solid lubricants used in turbopumps

#### **2. Bearings and shaft seals of turbopumps**

#### **2.1. Turbopumps and tribo–components**

turbopumps of the LE-5 and LE-7 had a rotational speed level of 50,000 rpm and had been studied and developed from the mid-1970 to the mid-1990. Specially, the all-steel bearings (made of AISI 440C) of the LH2 turbopump of the LE-7 demonstrated high performance with high reliability at high-speed level at 2 million DN (40 mm x 50,000 rpm). The shaft seal systems in the LE-5/LE-7 turbopumps that used a mechanical seal, a floating ring seal (annular seal)

Furthermore, for future space transport systems to reduce launch cost and to increase effi‐ ciency, advanced rocket engines which are characterized by high durability (long life) and high performance (light weight) are required in recent years. Advanced bearing and shaft seal that have high durability, i.e., a long life of 7.5 hours for the turbopump bearings used in reusable space shuttle main engine (the SSME). Its required life is 15 times longer than that (30 minutes) of the turbopump bearings used in the LE-7. At the first time, the SSME turbopump bearings experienced a serious wear problem in LO2 due to poor self-lubrication of the retainer [6]. In order to extend bearing life, the hybrid ceramic bearing with Si3N4 balls was used to reduce serious wear in the conventional all-steel bearing. A new type of the retainer having PTFE/ bronze-powder insert was also developed to obtain sufficient self-lubrication of the hybrid ceramic bearing. Consequently, the improvement of the SSME turbopump bearings needed a

Today, ultra-high speed level above 100,000 rpm is required to make a small and light turbopump for advanced second-stage engine. These advanced research and development are actively underway. In Japan, a new type of hybrid ceramic bearing having Si3N4 balls with a single guided retainer demonstrated excellent performance at an ultra-high speed of 120,000 rpm (3 million DN) in LH2 and recorded the world's top speed (in 2001) [8]. The result of this bearing was applied to the LH2 turbopump (rotational speed, 90,000 rpm) of the RL60 demonstrator engine (in 2003). The RL60 demonstrator engine was developed in the USA with international collaboration (USA, Japan, Russia and Sweden) and the LH2 turbopump was developed by a Japanese company [9]. In Europe, for the VINCI engine under development, high-DN hybrid ceramic bearing was tested in LH2 at a speed of 70,000 rpm (2.8 million DN) and continuous studies on a high-DN bearing was conducted at DN up to 3.3 million (120,000 rpm) in LH2 (in 2005) [10]. Furthermore, in Russia, for the developed RD0146 engine, its rotational speed of the main LH2 turbopump was 123,000 rpm (3.08 million DN), but detail of

This chapter also reviews advanced bearings and shaft seals which were studied from the mid-1990 to the mid-2000 after the development of turbopump bearings and shaft seals of the LE-7 [4,5]. It is typical that a long-life bearing with single-guided retainer demonstrat‐ ed a long operation for 12 hours under 50,000 rpm. A hybrid ceramic bearing having singleguided retainer and Si3N4 balls was able to demonstrate ultra-high-speed performance at speeds up to 120,000 rpm and show excellent performance under 3 million DN. An annular seal made of an Ag plated steel ring also presented two-phase seal performance at speeds

These historical reviews are intended to help the technical succession to next young generation who challenges research and development of the future space transportation system. These

and a segmented seal are also reviewed.

110 Tribology - Fundamentals and Advancements

long time of about 20 years [7].

its bearing was unknown (in 2003) [11].

up to 120,000 rpm.

The LO2/LH2 turbopumps as well as the tribo-components, such as high-speed bearings and rotating shaft-seals, were studied and developed to use in the LE-5 and LE-7. In reference to the structure of the LH2 turbopump of the LE-7, the tribo-components and solid lubricants used in the LE-5 and LE-7 turbopumps are typically indicated in Fig. 1 [4]. In addition, main design parameters of the turbopumps and DN values of bearings for the LE-5 and LE-7 are listed in Table 1 [5]. Here, the DN value that represents high-speed level of bearing is defined as the product of the inner-race bore diameter D (in mm) and the pump rotational speed N (in rpm). The rotor speed is typically restricted by the DN limits of the bearing.


turbine pressure was 19.1 MPa. The paired bearings of 32-mm bore were located at the inducer side and the paired bearings of 45-mm bore were at the turbine side. These bearings operated

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

113

As shown in Fig. 1, it is important to prohibit severe friction and wear in cryogenic environment that various solid lubricants are applied to the frictional parts in static and dynamic tribo-components. Since the turbopums are operated under large power condi‐ tions connecting with high fluid and mechanical vibration, it must pay attention that many components in contact are sure to generate relative motion and resulted in severe adhe‐ sive conditions. It needs proper lubrication to avoid severe frictional adhesion of assem‐

The rotor of turbopump is directly supported by two sets of self-lubricated ball bearings in cryogenic pump fluid. The shaft seal of turbopump is installed between the cryogenic pump and the hot gas turbine. The shaft seal system must seal the cryogenic propellants and the combustion gases (steam with rich hydrogen gas) safely and securely. High-speed compo‐ nents, such as bearings, shaft seals, Labyrinth seals, wear rings and balance pistons, used the proper solid lubricants to protect them from severe friction and wear in the reduction (LH2) or oxidation (LO2) environment of the cryogenic propellants. It is noted that these high-speed

The turbopump bearings are all-steel (AISI 440C) bearings that are self-lubricated by the PTFE transfer film as a lubricant from the reinforced PTFE (polytetra fluoroethylene) retainer. AISI 440C is martensitic stainless steel (with 16-18%Cr) and is one of the most widely used bearing materials in space systems because such high-Cr steel has a high corrosion resistance due to a superficial surface layer of Cr2O3. The resin PTFE retainer is reinforced with glass fiber, carbon fiber and laminated glass cloth to reduce wear as well as thermal contraction of the retainer. Although PTFE material has poor mechanical strength at room temperature, it has the best lubricant for use at cryogenic temperature because its mechanical tensile stress drastically increases and reaches to 80 MPa in LO2 and 130 MPa in LH2, respectively. In order to reduce wear of the PTFE composite retainer with poor thermal conductivity, sufficient cooling of the retainer is need to eliminate heat generation detrimental to successful bearing operation at

Since LH2 and LO2 are particularly poor as lubricants because of their low viscosity under conditions of reduction or oxidation, hydrodynamic fluid lubrication is less effective. It is noted that the cryogenic pump fluids works to remove severe frictional heat and to prevent the temperature rise in the bearing. At low temperatures, the PTFE transfer film as a lubricant is kept to be hard and to sustain the bearing load, so that softening and rupturing of the transfer film due to a rise in temperature have to be eliminated. Under poor cooling conditions, it

at a speed of 18,000 rpm and sustained the shaft power of 4,700 kW [14,15].

tribo-components are important life-controlling parts in engines [4].

**c.** *Tribo-components in turbopumps*

bled parts used in cryogenic environment.

**2.2. Self–lubricating bearings**

*2.2.1. Self–lubrication of retainer*

high speeds [16].

**Table 1.** Design parameters of turbopumps and DN values of bearings for LE-5 and LE-7

#### **a.** *LE-5 turbopumps*

For the upper stage of the H-I rocket, the LE-5 had a gas-generator cycle with 10-ton thrust and its chamber pressure of 3.4 MPa was relatively low. Its engine cycle is not able to achieve a high engine performance due to an open cycle. For the LH2 turbopump of the LE-5, the pump discharge pressure was relatively low at 5.5 MPa and the discharge flow rate was 51 liters/s. The turbine pressure was 2.4 MPa. The paired bearings of 25-mm bore operated at a speed of 50,000 rpm (1.25 million DN) and sustained the shaft power of 490 kW [12].

For the LO2 turbopumps, the discharge pressure was 5.2 MPa and the discharge flow rate was 18 liters/s. The turbine pressure was 0.48 MPa. The paired bearings of 30-mm bore operated at a speed of 16,500 rpm and sustained the shaft power of 130 kW. Basic design and technology of the cryogenic tribo-components used in the small turbopumps was experimentally estab‐ lished under the development of the LE-5.

#### **b.** *LE-7 turbopumps*

For next technical challenge in the first stage engine of the H-II rocket, the LE-7 had a stagedcombustion cycle (similar to that of the SSME) with 100-ton thrust and a high chamber pressure of 13 MPa. Its engine cycle can obtain high performance due to a closed engine cycle. For the high-pressure, large LH2 turbopump of the LE-7, the pump discharge pressure was increased to 27 MPa, and the discharge LH2 flow rate was 510 liters/s. The turbine pressure was relatively high at 20.6 MPa. The paired bearings of 35-mm bore were at the inducer side, and the paired bearings of 40-mm bore were at the turbine side. These bearings operated at a speed of 42,000 rpm (1.68 million DN) and sustained the shaft power of 19,700 kW [13,14].

For the LO2 turbopumps, the discharge pressure was 18 MPa for the main pump and 26 MPa for the preburner pump, respectively. The total discharge LO2 flow rate was 240 liters/s. The turbine pressure was 19.1 MPa. The paired bearings of 32-mm bore were located at the inducer side and the paired bearings of 45-mm bore were at the turbine side. These bearings operated at a speed of 18,000 rpm and sustained the shaft power of 4,700 kW [14,15].

#### **c.** *Tribo-components in turbopumps*

**Engine (thrust) Rocket Engine cycle**

112 Tribology - Fundamentals and Advancements

Pump pressure [MPa] Pump flow rate [kg/s] Shaft rotational speed [rpm] Bearing DN [mm x rpm] Turbine pressure [MPa] Turbine temperature [K] Turbine gas flow rate [kg/s]

Shaft power [kW] Weight [kg]

For pre-burner in bracket; ( )\*

**a.** *LE-5 turbopumps*

**b.** *LE-7 turbopumps*

lished under the development of the LE-5.

**LE-5 (10 tons) Second stage of H-1 Gas-generator cycle**

Turbopump LO2 LH2 LO2 LH2

5.5 3.6 50,000 125 x 104 2.4 840 0.42 490 25

5.2 20 16,500 49.5 x 104 0.48 690 0.39 130 23

For the upper stage of the H-I rocket, the LE-5 had a gas-generator cycle with 10-ton thrust and its chamber pressure of 3.4 MPa was relatively low. Its engine cycle is not able to achieve a high engine performance due to an open cycle. For the LH2 turbopump of the LE-5, the pump discharge pressure was relatively low at 5.5 MPa and the discharge flow rate was 51 liters/s. The turbine pressure was 2.4 MPa. The paired bearings of 25-mm bore operated at a speed of

For the LO2 turbopumps, the discharge pressure was 5.2 MPa and the discharge flow rate was 18 liters/s. The turbine pressure was 0.48 MPa. The paired bearings of 30-mm bore operated at a speed of 16,500 rpm and sustained the shaft power of 130 kW. Basic design and technology of the cryogenic tribo-components used in the small turbopumps was experimentally estab‐

For next technical challenge in the first stage engine of the H-II rocket, the LE-7 had a stagedcombustion cycle (similar to that of the SSME) with 100-ton thrust and a high chamber pressure of 13 MPa. Its engine cycle can obtain high performance due to a closed engine cycle. For the high-pressure, large LH2 turbopump of the LE-7, the pump discharge pressure was increased to 27 MPa, and the discharge LH2 flow rate was 510 liters/s. The turbine pressure was relatively high at 20.6 MPa. The paired bearings of 35-mm bore were at the inducer side, and the paired bearings of 40-mm bore were at the turbine side. These bearings operated at a speed of 42,000

For the LO2 turbopumps, the discharge pressure was 18 MPa for the main pump and 26 MPa for the preburner pump, respectively. The total discharge LO2 flow rate was 240 liters/s. The

**Table 1.** Design parameters of turbopumps and DN values of bearings for LE-5 and LE-7

50,000 rpm (1.25 million DN) and sustained the shaft power of 490 kW [12].

rpm (1.68 million DN) and sustained the shaft power of 19,700 kW [13,14].

**LE-7 (86 tons) First stage of H-2 Staged-combustion cycle**

> 27.0 36 42,000 168 x 104 20.6 830 33.6 19,700 200

17.4 (25.8)\* 212 (46)\* 18,000 81 x 104 19.1 810 14.9 4,700 160

As shown in Fig. 1, it is important to prohibit severe friction and wear in cryogenic environment that various solid lubricants are applied to the frictional parts in static and dynamic tribo-components. Since the turbopums are operated under large power condi‐ tions connecting with high fluid and mechanical vibration, it must pay attention that many components in contact are sure to generate relative motion and resulted in severe adhe‐ sive conditions. It needs proper lubrication to avoid severe frictional adhesion of assem‐ bled parts used in cryogenic environment.

The rotor of turbopump is directly supported by two sets of self-lubricated ball bearings in cryogenic pump fluid. The shaft seal of turbopump is installed between the cryogenic pump and the hot gas turbine. The shaft seal system must seal the cryogenic propellants and the combustion gases (steam with rich hydrogen gas) safely and securely. High-speed compo‐ nents, such as bearings, shaft seals, Labyrinth seals, wear rings and balance pistons, used the proper solid lubricants to protect them from severe friction and wear in the reduction (LH2) or oxidation (LO2) environment of the cryogenic propellants. It is noted that these high-speed tribo-components are important life-controlling parts in engines [4].

#### **2.2. Self–lubricating bearings**

#### *2.2.1. Self–lubrication of retainer*

The turbopump bearings are all-steel (AISI 440C) bearings that are self-lubricated by the PTFE transfer film as a lubricant from the reinforced PTFE (polytetra fluoroethylene) retainer. AISI 440C is martensitic stainless steel (with 16-18%Cr) and is one of the most widely used bearing materials in space systems because such high-Cr steel has a high corrosion resistance due to a superficial surface layer of Cr2O3. The resin PTFE retainer is reinforced with glass fiber, carbon fiber and laminated glass cloth to reduce wear as well as thermal contraction of the retainer. Although PTFE material has poor mechanical strength at room temperature, it has the best lubricant for use at cryogenic temperature because its mechanical tensile stress drastically increases and reaches to 80 MPa in LO2 and 130 MPa in LH2, respectively. In order to reduce wear of the PTFE composite retainer with poor thermal conductivity, sufficient cooling of the retainer is need to eliminate heat generation detrimental to successful bearing operation at high speeds [16].

Since LH2 and LO2 are particularly poor as lubricants because of their low viscosity under conditions of reduction or oxidation, hydrodynamic fluid lubrication is less effective. It is noted that the cryogenic pump fluids works to remove severe frictional heat and to prevent the temperature rise in the bearing. At low temperatures, the PTFE transfer film as a lubricant is kept to be hard and to sustain the bearing load, so that softening and rupturing of the transfer film due to a rise in temperature have to be eliminated. Under poor cooling conditions, it appears that the blackened transfer film due to thermal decomposition of PTFE should occur at a high temperature above about 500 K, and the degraded transfer film was not able to sustain the bearing load. Therefore, sufficient cooling by cryogenic fluids, as well as reduction of frictional heat generation, is very important to produce a durable lubricant film transferred from the retainer even in cryogenic fluid [14].

#### *2.2.2. High–speed and load conditions of bearing*

For the turbopump bearings, angular-contact bearings are usually used in pairs in duplex mounts (back to back). For example, Table 2 shows main design parameters and internal load conditions for the bearings used in the LH2 turbopumps of the LE-5 and LE-7 [17,18]. In this table, the *SVmax* value (=*Smax* x *Vmax*/2) that represents the maximum product of stress times spinning velocity in the contact ellipse zone at the inner race are shown. Here, *Smax* is the maximum contact stress and *Vmax* is the maximum spinning velocity. The *SVmax* value is an important factor related to lubrication and wear at the inner race with ball spinning [13,19]. High *SVmax* value leads to high frictional heating and to wear of the PTFE transfer film due to spin wear. Under poor cooling condition and large tilted misalignment, the turbopump bearings have an initial contact angel of 15-25 deg. with a large radial clearance to prevent a loss of operating clearance from bearing seizer. As mention later, high-speed bearing has the outer-race ball control that produces high ball spinning at the inner race. In order to reduce the stress level within the spinning contact zone, race curvatures were controlled to be 0.54-0.56 for inner race and 0.52 for the outer race, respectively. The inner race has a counter-bore type to gain sufficient cooling within the bearing.

small. Furthermore, for the retainer, the sliding velocity is 50 m/s at the ball pocket and 45 m/

**Table 2.** Design parameters and internal load conditions for LH2 turbopump bearings of LE-5 and LE-7

**Parameters LE-5 LE-7**

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

25 x 52 x 15 38.5 7.938 11 20 57

> 50,000 784 125 x 104

157 / 343 1.58 / 1.49 2.4 x 103

115

http://dx.doi.org/10.5772/55733

46,000 1,176 184 x 104

176 / 637 1.54 / 1.63 3.1 x 103

For system design of the turbopump high-speed rotor, the thrust load applied on the rotor due to unbalanced fluid pressures is balanced automatically by a balance piston mecha‐ nism during operation [17]. As a result, the turbopump bearings can operate only with a spring thrust load to remove internal clearance and control radial stiffness. However, the shaft vibration as well as the fluid action around the impeller should add high dynamic radial load to the thrust load on the bearing. For example, the LH2 turbopump bearings of the LE-7 had to operate at a speed of 42,000 rpm that was beyond the third critical speed of 32,000 rpm and must support the high shaft-power under high shaft-vibration. Therefore, the bearings must have high combined radial and thrust load capacity at all

 unbalanced fluid pressures is balanced automatically by a balance piston mechanism during operation [17]. As a result, the turbopump bearings can operate only with a spring thrust load to remove internal clearance and control radial stiffness. However, the shaft vibration as well as the fluid action around the impeller should add high dynamic radial load to the thrust load on the bearing. For example, the LH2 5 turbopump bearings of the LE-7 had to operate at a speed of 42,000 rpm that was beyond the third critical speed of 32,000 rpm and must support the high shaft-power under high shaft-vibration. Therefore, the bearings must have high combined radial and thrust load capacity at all extremes of the

10 11 Figure 2. Sliding frictional conditions at inner and outer raceways for LH2 bearing (25-mm bore,

**Figure 2.** Sliding frictional conditions at inner and outer raceways for LH2 bearing (25-mm bore, 50,000 rpm, 980 N)

Differential slip sliding

 The required functions for shaft seal systems vary for different engine cycles. Similar to the SSME, the LE-7 has a two-stage combustion cycle. It requires a high pressure seal since the pressure in the pump and turbine is extremely high. To contrast, the pressure of the pump and turbine in the LE-5 with a gas generation cycle is comparatively low. Design parameters (the seal diameter, rubbing speed and seal pressure) for the seal elements used in the LE-5 and LE-7 turbopumps are listed in Table 3 [21]. The seal elements are the LO2 seal, LH2 seal, gas helium (GHe) purge seal and turbine gas seal. These shaft seals prevent or minimize the leakage of LO2 and LH2 21 for pump side and hot turbine gas (steam with rich hydrogen gas) for turbine side. In order to make a short length of the shaft, the shaft seals have to be compactly installed between the cryogenic pump and hot turbine.

Parameters Seal diameter [mm]-Rubbing velocity [m/s]

Engine LE-5 LE-7

46.6-40 (16,500)-0.98-(a) 43.2-113 (50,000)-1.4-(a)

40-35 (16,500)-0.3-(b) 70-61 (16,500)-0.3-(b)

For the LO2 28 turbopumps, when the leakage of LO2 and hot turbine gas are mixed, an explosion will occur. In order to separate the leakage of LO2 and hot turbine gas in safety, the system is complicated and requires three types of seal elements (the LO2 seal, GHe purge seal and turbine gas seal). The GHe purge seal installed between the LO2 seal and turbine gas seal supplies GHe as a barrier gas. To contrast, for the LH2 turbopumps, the LH2 32 leakage can be discharged to the turbine side so that the seal system is relatively simple. However, the rubbing speed of the seal face becomes considerably

25 (a) Mechanical seal, (b) Segmented seal, (c) Floating ring seal, (d) Lift-off seal 26 Table 3. Design parameters for seal elements used in LE-5 and LE-7 turbopumps

(Rotating speed [rpm])-Seal pressure [MPa]-Seal type


*SVmax* 176 N/mm2 x m/s

*Smax*  1,470 N/mm2 *Vmax* 0.12 m/s

6 Book Title

s at the outer guide land at a speed of 50,000 rpm, respectively.

extremes of the tutbopump operating conditions [14].

*Vmax* 2.8 m/s

*Smax*  1,480 N/mm2 *SVmax* 4,140 N/mm2 x m/s

LO2 seal LH2 seal GHe purge seal Turbine gas seal

8 tutbopump operating conditions [14].

12 50,000 rpm, 980 N)

Spin sliding at inner raceway

14 2.3 Shaft seal systems

9

*Bearing*

Dimension [mm] Pitch diameter [mm] Ball diameter [mm] Number of balls

Initial contact angle [deg.] Initial radial clearance [μm]

*Operating condition* Rotational speed [rpm] Thrust pre-load [N] Bearing DN [mm x rpm]

*Internal load condition*

Normal load at inner / outer races [N]

Maximum contact stress at inner / outer races (*Smax*) [GPa] Maximum SV at inner race (*SVmax*) [N/mm2 x m/s]

13

24

27

As the centrifugal force developed on the balls increases at high speeds, the operational contact angle at the inner and the outer races are changed to be different each other. The operational contact angle at the inner race increases rather than the initial contact angle and decreases to near zero at the outer race. This divergence of contact angles tends to increase ball spinning in addition to rolling at the inner race. Its spin velocity due to ball spinning becomes high and results in an occurrence of frictional heat generation. To contrast, rolling contact at the outer race generates differential slip due to curvature of contact ellipse [20].

Under the outer-race control connected with ball spinning at the inner race, heat generation due to ball spin is significantly higher than that of differentia slip, so that sufficient cooling is necessary at the inner race side. Furthermore, sliding velocity of the rolling balls in contact with the outer guide land and the ball pocket is high and resulted in a generation of frictional heating of the retainer. The bearings were effectively cooled by the pump cryogenic fluids circulating in the turbopumps. For example, Fig. 2 shows sliding frictional conditions of the inner and outer raceways for the 25-mm-bore bearing that is at a speed of 50,000 rpm under a thrust load of 980 N [16]. This bearing was used in the LH2 turbopump bearing for the LE-5. In this figure, the distribution of the contact stress, the spinning velocity and the *SV* value with spin at the inner race are shown. Pattern of spin wear generated by ball spinning becomes similar to the distribution of the *SV* value. To contrast, for the outer race, the differential slip velocity and the *SV* value with differential slip are light so that wear due to differential slip is


**Table 2.** Design parameters and internal load conditions for LH2 turbopump bearings of LE-5 and LE-7

8 tutbopump operating conditions [14].

14 2.3 Shaft seal systems

LO2 seal LH2 seal GHe purge seal Turbine gas seal

9

13

24

27

appears that the blackened transfer film due to thermal decomposition of PTFE should occur at a high temperature above about 500 K, and the degraded transfer film was not able to sustain the bearing load. Therefore, sufficient cooling by cryogenic fluids, as well as reduction of frictional heat generation, is very important to produce a durable lubricant film transferred

For the turbopump bearings, angular-contact bearings are usually used in pairs in duplex mounts (back to back). For example, Table 2 shows main design parameters and internal load conditions for the bearings used in the LH2 turbopumps of the LE-5 and LE-7 [17,18]. In this table, the *SVmax* value (=*Smax* x *Vmax*/2) that represents the maximum product of stress times spinning velocity in the contact ellipse zone at the inner race are shown. Here, *Smax* is the maximum contact stress and *Vmax* is the maximum spinning velocity. The *SVmax* value is an important factor related to lubrication and wear at the inner race with ball spinning [13,19]. High *SVmax* value leads to high frictional heating and to wear of the PTFE transfer film due to spin wear. Under poor cooling condition and large tilted misalignment, the turbopump bearings have an initial contact angel of 15-25 deg. with a large radial clearance to prevent a loss of operating clearance from bearing seizer. As mention later, high-speed bearing has the outer-race ball control that produces high ball spinning at the inner race. In order to reduce the stress level within the spinning contact zone, race curvatures were controlled to be 0.54-0.56 for inner race and 0.52 for the outer race, respectively. The inner race has a counter-bore type

As the centrifugal force developed on the balls increases at high speeds, the operational contact angle at the inner and the outer races are changed to be different each other. The operational contact angle at the inner race increases rather than the initial contact angle and decreases to near zero at the outer race. This divergence of contact angles tends to increase ball spinning in addition to rolling at the inner race. Its spin velocity due to ball spinning becomes high and results in an occurrence of frictional heat generation. To contrast, rolling contact at the outer

Under the outer-race control connected with ball spinning at the inner race, heat generation due to ball spin is significantly higher than that of differentia slip, so that sufficient cooling is necessary at the inner race side. Furthermore, sliding velocity of the rolling balls in contact with the outer guide land and the ball pocket is high and resulted in a generation of frictional heating of the retainer. The bearings were effectively cooled by the pump cryogenic fluids circulating in the turbopumps. For example, Fig. 2 shows sliding frictional conditions of the inner and outer raceways for the 25-mm-bore bearing that is at a speed of 50,000 rpm under a thrust load of 980 N [16]. This bearing was used in the LH2 turbopump bearing for the LE-5. In this figure, the distribution of the contact stress, the spinning velocity and the *SV* value with spin at the inner race are shown. Pattern of spin wear generated by ball spinning becomes similar to the distribution of the *SV* value. To contrast, for the outer race, the differential slip velocity and the *SV* value with differential slip are light so that wear due to differential slip is

race generates differential slip due to curvature of contact ellipse [20].

from the retainer even in cryogenic fluid [14].

114 Tribology - Fundamentals and Advancements

*2.2.2. High–speed and load conditions of bearing*

to gain sufficient cooling within the bearing.

small. Furthermore, for the retainer, the sliding velocity is 50 m/s at the ball pocket and 45 m/ s at the outer guide land at a speed of 50,000 rpm, respectively.

For system design of the turbopump high-speed rotor, the thrust load applied on the rotor due to unbalanced fluid pressures is balanced automatically by a balance piston mecha‐ nism during operation [17]. As a result, the turbopump bearings can operate only with a spring thrust load to remove internal clearance and control radial stiffness. However, the shaft vibration as well as the fluid action around the impeller should add high dynamic radial load to the thrust load on the bearing. For example, the LH2 turbopump bearings of the LE-7 had to operate at a speed of 42,000 rpm that was beyond the third critical speed of 32,000 rpm and must support the high shaft-power under high shaft-vibration. Therefore, the bearings must have high combined radial and thrust load capacity at all extremes of the tutbopump operating conditions [14]. 6 Book Title 1 unbalanced fluid pressures is balanced automatically by a balance piston mechanism during operation 2 [17]. As a result, the turbopump bearings can operate only with a spring thrust load to remove internal 3 clearance and control radial stiffness. However, the shaft vibration as well as the fluid action around 4 the impeller should add high dynamic radial load to the thrust load on the bearing. For example, the LH2 5 turbopump bearings of the LE-7 had to operate at a speed of 42,000 rpm that was beyond the 6 third critical speed of 32,000 rpm and must support the high shaft-power under high shaft-vibration. 7 Therefore, the bearings must have high combined radial and thrust load capacity at all extremes of the

12 50,000 rpm, 980 N) **Figure 2.** Sliding frictional conditions at inner and outer raceways for LH2 bearing (25-mm bore, 50,000 rpm, 980 N)

 The required functions for shaft seal systems vary for different engine cycles. Similar to the SSME, the LE-7 has a two-stage combustion cycle. It requires a high pressure seal since the pressure in the pump and turbine is extremely high. To contrast, the pressure of the pump and turbine in the LE-5 with a gas generation cycle is comparatively low. Design parameters (the seal diameter, rubbing speed and seal pressure) for the seal elements used in the LE-5 and LE-7 turbopumps are listed in Table 3 [21]. The seal elements are the LO2 seal, LH2 seal, gas helium (GHe) purge seal and turbine gas seal. These shaft seals prevent or minimize the leakage of LO2 and LH2 21 for pump side and hot turbine gas (steam with rich hydrogen gas) for turbine side. In order to make a short length of the shaft, the shaft seals have to be compactly installed between the cryogenic pump and hot turbine.

Parameters Seal diameter [mm]-Rubbing velocity [m/s]

Engine LE-5 LE-7

46.6-40 (16,500)-0.98-(a) 43.2-113 (50,000)-1.4-(a)

40-35 (16,500)-0.3-(b) 70-61 (16,500)-0.3-(b)

For the LO2 28 turbopumps, when the leakage of LO2 and hot turbine gas are mixed, an explosion will occur. In order to separate the leakage of LO2 and hot turbine gas in safety, the system is complicated and requires three types of seal elements (the LO2 seal, GHe purge seal and turbine gas seal). The GHe purge seal installed between the LO2 seal and turbine gas seal supplies GHe as a barrier gas. To contrast, for the LH2 turbopumps, the LH2 32 leakage can be discharged to the turbine side so that the seal system is relatively simple. However, the rubbing speed of the seal face becomes considerably

25 (a) Mechanical seal, (b) Segmented seal, (c) Floating ring seal, (d) Lift-off seal 26 Table 3. Design parameters for seal elements used in LE-5 and LE-7 turbopumps

(Rotating speed [rpm])-Seal pressure [MPa]-Seal type


#### **2.3. Shaft seal systems**

The required functions for shaft seal systems vary for different engine cycles. Similar to the SSME, the LE-7 has a two-stage combustion cycle. It requires a high pressure seal since the pressure in the pump and turbine is extremely high. To contrast, the pressure of the pump and turbine in the LE-5 with a gas generation cycle is comparatively low. Design parameters (the seal diameter, rubbing speed and seal pressure) for the seal elements used in the LE-5 and LE-7 turbopumps are listed in Table 3 [21]. The seal elements are the LO2 seal, LH2 seal, gas helium (GHe) purge seal and turbine gas seal. These shaft seals prevent or minimize the leakage of LO2 and LH2 for pump side and hot turbine gas (steam with rich hydrogen gas) for turbine side. In order to make a short length of the shaft, the shaft seals have to be compactly installed between the cryogenic pump and hot turbine.

was kept at a lower temperature against the hot turbine section and the reliability of the shaftseal system was further increased. Between the LO2 seal and the turbine gas seal, the segmented circumferential seal (GHe purge seal), that had shrouded Rayleigh step hydrodynamic liftpads to increase opening force, was paired and purged with GHe to prevent mixing of the

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

117

The LH2 seal system of the high pressured LH2 turbopump was assembled with the floatingring seal and lift-off seal. The lift-off seal is similar to a face-contact mechanical seal and is in contact with the mating ring (rotating seal-ring) and its leakage is small when the seal pressure is low. As the rotational speed of the turbopump increases and the seal pressure becomes high, the seal faces are automatically disengaged from contacting and changed to be non-contact

leakage of LO2 and GH2.

**Figure 3.** Shaft seal system for high-pressure LO2 turbopump of LE-7

**3. Cryogenic tribological problems [4, 16, 20, 23]**

condition resulted in severe adhesive (welding) wear in LH2.

LH2 is a particularly poor lubricant due to its extremely low viscosity (approximately equal to that of room-temperature air) and chemical reducing effect to remove native oxide film and to make fresh frictional surface, resulting in a severe lubricating condition at the frictional interfaces. Furthermore, at extremely low temperatures in LH2, the specific heats and thermal conductivities of tribo-materials drop off rapidly rather than those at the liquid nitrogen (LN2, boiling point 77 K) temperature. At a high temperature in LN2, the specific heats and thermal conductivities are less changed and same as those at a room temperature. In addition with vaporization of LH2, it is easy to produce local hot spots at frictional interfaces, so that frictional

seal [21].


(a) Mechanical seal, (b) Segmented seal, (c) Floating ring seal, (d) Lift-off seal

**Table 3.** Design parameters for seal elements used in LE-5 and LE-7 turbopumps

For the LO2 turbopumps, when the leakage of LO2 and hot turbine gas are mixed, an explosion will occur. In order to separate the leakage of LO2 and hot turbine gas in safety, the system is complicated and requires three types of seal elements (the LO2 seal, GHe purge seal and turbine gas seal). The GHe purge seal installed between the LO2 seal and turbine gas seal supplies GHe as a barrier gas. To contrast, for the LH2 turbopumps, the LH2 leakage can be discharged to the turbine side so that the seal system is relatively simple. However, the rubbing speed of the seal face becomes considerably high and the contacting seal face is opposite severe tribological condition.

For the low-pressure turbopumps of the LE-5, the LO2 and LH2 seals used face-contact mechanical seals to gain small leakage. The GHe purge seal and turbine gas seal used contacttype segmented seal. For the high-pressure turbopumps of the LE-7, the LO2 seal, LH2 seal and turbine gas seal used non-contact type, floating-ring seal (annular seal) due to high seal pressure. For example, the shaft seal system of the high- pressure LO2 turbopump of the LE-7 is shown in Fig. 3 [22]. The shaft seal system was set up between the cryogenic pumps and the hot turbine and prevented the mixing of the leakage of LO2 and hot turbine gas. The LO2 seal was composed of a floating-ring seal. The turbine gas seal used two floating-ring seals to seal the low temperature GH2 that made a barrier to the turbine hot gas. So that the turbine gas seal was kept at a lower temperature against the hot turbine section and the reliability of the shaftseal system was further increased. Between the LO2 seal and the turbine gas seal, the segmented circumferential seal (GHe purge seal), that had shrouded Rayleigh step hydrodynamic liftpads to increase opening force, was paired and purged with GHe to prevent mixing of the leakage of LO2 and GH2.

The LH2 seal system of the high pressured LH2 turbopump was assembled with the floatingring seal and lift-off seal. The lift-off seal is similar to a face-contact mechanical seal and is in contact with the mating ring (rotating seal-ring) and its leakage is small when the seal pressure is low. As the rotational speed of the turbopump increases and the seal pressure becomes high, the seal faces are automatically disengaged from contacting and changed to be non-contact seal [21].

**Figure 3.** Shaft seal system for high-pressure LO2 turbopump of LE-7

**2.3. Shaft seal systems**

116 Tribology - Fundamentals and Advancements

LO2 seal LH2 seal GHe purge seal Turbine gas seal

condition.

between the cryogenic pump and hot turbine.

The required functions for shaft seal systems vary for different engine cycles. Similar to the SSME, the LE-7 has a two-stage combustion cycle. It requires a high pressure seal since the pressure in the pump and turbine is extremely high. To contrast, the pressure of the pump and turbine in the LE-5 with a gas generation cycle is comparatively low. Design parameters (the seal diameter, rubbing speed and seal pressure) for the seal elements used in the LE-5 and LE-7 turbopumps are listed in Table 3 [21]. The seal elements are the LO2 seal, LH2 seal, gas helium (GHe) purge seal and turbine gas seal. These shaft seals prevent or minimize the leakage of LO2 and LH2 for pump side and hot turbine gas (steam with rich hydrogen gas) for turbine side. In order to make a short length of the shaft, the shaft seals have to be compactly installed

**Parameters Seal diameter [mm]—Rubbing velocity [m/s]**

46.6—40 (16,500)—0.98—(a) 43.2—113 (50,000)—1.4—(a) 40—35 (16,500)—0.3—(b) 70—61 (16,500)—0.3—(b)

(a) Mechanical seal, (b) Segmented seal, (c) Floating ring seal, (d) Lift-off seal

**Table 3.** Design parameters for seal elements used in LE-5 and LE-7 turbopumps

Engine LE-5 LE-7

For the LO2 turbopumps, when the leakage of LO2 and hot turbine gas are mixed, an explosion will occur. In order to separate the leakage of LO2 and hot turbine gas in safety, the system is complicated and requires three types of seal elements (the LO2 seal, GHe purge seal and turbine gas seal). The GHe purge seal installed between the LO2 seal and turbine gas seal supplies GHe as a barrier gas. To contrast, for the LH2 turbopumps, the LH2 leakage can be discharged to the turbine side so that the seal system is relatively simple. However, the rubbing speed of the seal face becomes considerably high and the contacting seal face is opposite severe tribological

For the low-pressure turbopumps of the LE-5, the LO2 and LH2 seals used face-contact mechanical seals to gain small leakage. The GHe purge seal and turbine gas seal used contacttype segmented seal. For the high-pressure turbopumps of the LE-7, the LO2 seal, LH2 seal and turbine gas seal used non-contact type, floating-ring seal (annular seal) due to high seal pressure. For example, the shaft seal system of the high- pressure LO2 turbopump of the LE-7 is shown in Fig. 3 [22]. The shaft seal system was set up between the cryogenic pumps and the hot turbine and prevented the mixing of the leakage of LO2 and hot turbine gas. The LO2 seal was composed of a floating-ring seal. The turbine gas seal used two floating-ring seals to seal the low temperature GH2 that made a barrier to the turbine hot gas. So that the turbine gas seal

**(Rotating speed [rpm])— Seal pressure [MPa]— Seal type**

55—58 (18,000)—4.9—(b) 50—120 (42,000)—7.1—(c) 69—173 (42,000)—0.6—(d) 100—105 (18,000)—0.6—(b) 55—58 (18,000)—16.7—(c)

#### **3. Cryogenic tribological problems [4, 16, 20, 23]**

LH2 is a particularly poor lubricant due to its extremely low viscosity (approximately equal to that of room-temperature air) and chemical reducing effect to remove native oxide film and to make fresh frictional surface, resulting in a severe lubricating condition at the frictional interfaces. Furthermore, at extremely low temperatures in LH2, the specific heats and thermal conductivities of tribo-materials drop off rapidly rather than those at the liquid nitrogen (LN2, boiling point 77 K) temperature. At a high temperature in LN2, the specific heats and thermal conductivities are less changed and same as those at a room temperature. In addition with vaporization of LH2, it is easy to produce local hot spots at frictional interfaces, so that frictional condition resulted in severe adhesive (welding) wear in LH2.

LO2 has high oxidization power and forms oxide film at frictional surfaces, so that oxide film produces lower friction compared with that in LH2; however, in boiling of LO2, oxide wear should increase due to high oxidization power. Active cooling is important to prohibit boiling of LO2 at frictional interfaces. Furthermore, violent frictional heating in LO2 can lead to the ignition of tribo-elements due to burn-out phenomenon occurring in nucleate boiling, that is defined by engineering heat transfer. Under burn-out phenomenon in boiling, an extreme rise in surface temperature was experienced because a marked reduction occurred in heat transfer. For example, in boiling of LN2, the sliding surface of Ag-10%Cu alloy (melting point 1,155 K) against Ti alloy (Ti-5Al-2.5Sn) melted due to burn-out wear during friction test [24]. The surface coating of TiN or TiO2 had a high resistance to adhesive welding to the Ti alloy disk was able to protect from burn-out wear. The results were applied to the balance-piston system in the LH2 turbopump of the LE-7.

**Figure 5.** Friction and wear of PTFE pin against oxidized 440C disk in cryogenic GO2 as a function of pin temperature

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

119

It is noted that the tribo-characteristics at cryogenic temperatures tend to change complexly. For example, Fig. 4 shows the change of friction and wear of a PTFE pin against a 440C steel disk in cryogenic gaseous oxygen (GO2) as a function of pin temperature [23,25]. This figure denotes the glass transition temperature of PTFE, about 170 K, 230 K and 260 K, those are defined by relaxation of its amorphous layer in the PTFE band structure. When the frictional environment changed from the liquid phase to the gas phase at boiling, the friction coefficient increased drastically and wear began. To the glass transition temperature of 170 K (amorphous layer begins to relax), the friction coefficient remains at a low constant value, but the specific wear drastically decreased at 170 K. In an inert gaseous nitrogen (GN2), there was not such drastically decrease in the specific wear at 170 K. After that, friction and wear begin to increase gradually up to 230 K. The increase of friction and wear above 170 K surely depends on the

However, when the surface of 440C steel was oxidized, the characteristic curve of friction and wear depended on cryogenic temperatures was changed drastically. Figure 5 shows the change

fact that the strength property of PTFE begins to decrease rapidly above 170 K.

**Figure 4.** Friction and wear of PTFE pin against 440C disk in cryogenic GO2 as a function of pin temperature

LO2 has high oxidization power and forms oxide film at frictional surfaces, so that oxide film produces lower friction compared with that in LH2; however, in boiling of LO2, oxide wear should increase due to high oxidization power. Active cooling is important to prohibit boiling of LO2 at frictional interfaces. Furthermore, violent frictional heating in LO2 can lead to the ignition of tribo-elements due to burn-out phenomenon occurring in nucleate boiling, that is defined by engineering heat transfer. Under burn-out phenomenon in boiling, an extreme rise in surface temperature was experienced because a marked reduction occurred in heat transfer. For example, in boiling of LN2, the sliding surface of Ag-10%Cu alloy (melting point 1,155 K) against Ti alloy (Ti-5Al-2.5Sn) melted due to burn-out wear during friction test [24]. The surface coating of TiN or TiO2 had a high resistance to adhesive welding to the Ti alloy disk was able to protect from burn-out wear. The results were applied to the balance-piston system in the

**Figure 4.** Friction and wear of PTFE pin against 440C disk in cryogenic GO2 as a function of pin temperature

LH2 turbopump of the LE-7.

118 Tribology - Fundamentals and Advancements

**Figure 5.** Friction and wear of PTFE pin against oxidized 440C disk in cryogenic GO2 as a function of pin temperature

It is noted that the tribo-characteristics at cryogenic temperatures tend to change complexly. For example, Fig. 4 shows the change of friction and wear of a PTFE pin against a 440C steel disk in cryogenic gaseous oxygen (GO2) as a function of pin temperature [23,25]. This figure denotes the glass transition temperature of PTFE, about 170 K, 230 K and 260 K, those are defined by relaxation of its amorphous layer in the PTFE band structure. When the frictional environment changed from the liquid phase to the gas phase at boiling, the friction coefficient increased drastically and wear began. To the glass transition temperature of 170 K (amorphous layer begins to relax), the friction coefficient remains at a low constant value, but the specific wear drastically decreased at 170 K. In an inert gaseous nitrogen (GN2), there was not such drastically decrease in the specific wear at 170 K. After that, friction and wear begin to increase gradually up to 230 K. The increase of friction and wear above 170 K surely depends on the fact that the strength property of PTFE begins to decrease rapidly above 170 K.

However, when the surface of 440C steel was oxidized, the characteristic curve of friction and wear depended on cryogenic temperatures was changed drastically. Figure 5 shows the change of friction and wear of a PTFE pin in case of using an oxidized 440C steel disk [23,25]. At the pin temperatures above boiling point of LO2 (90 K), the friction and wear of PTFE pin showed relatively high values as compared with that showed in Fig. 4. As the pin temperature increased from 90 K to near 170 K, the friction and wear of PTFE drastically decreased to low values. The oxidized 440C steel disk was obtained by heating in air at about 623 K for 3 hours. The surface of the oxidized 440C showed an increase of FeO/Fe2O3 film in comparison with Cr2O3 film. It is noted that the oxidization of 440C steel should result in an increase of friction and wear of PTFE. It seems that PTFE transfer film was less formed due to poor adhesion of PTFE against FeO/Fe2O3, and frictional condition became to be severe. Thus, it is very interesting that the friction and wear properties of PTFE changed characteristically at its glass transition temper‐ ature, depending on the oxidization of 440C steel.

At cryogenic temperatures, it is noted that sufficient cooling and the restriction of frictional heat generation are essential to prohibit severe tribological conditions. In order to solve these cryogenic tribological problems, it is important that (1) understanding the complex character‐ istics of tribology at low temperatures, (2) selection of the proper solid-lubricants against the oxidation or reduction power, and (3) active cooling to remove severe frictional heat at local

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

121

**Figure 6.** Wear of five kinds of the ceramic balls against 440C disk in LO2 and LN2

**Figure 7.** Friction of five kinds of the ceramic balls against 440C disk in LO2 and LN2

hot spots [4].

For other friction tests, wear of PTFE in cryogenic GO2 was increased as surface roughness of 440C disk was increased; however, in cryogenic GN2, surface roughness had less effect on wear increase of PTFE. Furthermore, friction and wear of PTFE against Si3N4 disk was determined in cryogenic GO2 and GN2. In both cryogenic environments, friction coefficient was higher than that of 440C disk. It was noted that wear of PTFE in GO2 was drastically high compared with that in GN2. It was assumed that poor formation of PTFE transfer film on the SI3N4 disk resulted in an increase of friction and wear in GO2. This result indicated that the hybrid ceramic bearing with Si3N4 ball showed poor self-lubrication in LO2.

It is interesting to use ceramic material as tribo-materials in cryogenic environments. Friction and wear behavior of typical fine ceramics against 440C disk were evaluated in LO2 and LN2. Figure 6 and 7 show wear and friction of five kinds of the ceramic balls in comparison with those in LO2 and LN2 [23], respectively. In all the cases of friction tests, the sliding contact surface of ceramic pin was covered by the transfer film of wear debris of 440C steel. The metallic transfer film prevented direct contact between metal and ceramic. As a result, the metal-tometal contact should control the friction and wear behavior of the sliding pair, and the order of friction seemed to be less affected in the wear resistance of ceramic pins.

In LO2, Al2O3 indicated the lowest wear rate and was followed by SiC, Si3N4, Sialon and ZrO2 in order of the wear resistance. For Al2O3 pin, the metallic oxide film of 440C seemed to be strongly adhered onto the ceramic pin and resulted in an increase of protection of the pin wear; however, wear of the 440C disk was prolonged. For SiC, Si3N4 and Sialon, sliding friction in oxidized environment made the glassy formation of SiO2 film due to tribo-chemical reaction. The hardness of SiO2 is much less than that of ceramic substrate and resulted in an increase in the wear of ceramic pins. It was noted that the wear rate of ZrO2 was considerably high as similar to that of self-mated 440C steels. Since ZrO2 has the lowest hardness compared with other ceramics, the hard oxide film of 440C should increase wear of ZrO2 pin.

To the contrary, in LN2, Zr2O3 indicated the lowest wear rate and was followed by Si3N4, Sialon, Al2O3 and SiC in order of the wear resistance. The high wear of Al2O3 and SiC pins was seemed to be induced by lack of protective film of 440C steel due to weak adhesion to ceramic pin. It is found that the order of wear resistance of ceramics against 440C steel in LO2 was opposed to that in LN2 [23].

At cryogenic temperatures, it is noted that sufficient cooling and the restriction of frictional heat generation are essential to prohibit severe tribological conditions. In order to solve these cryogenic tribological problems, it is important that (1) understanding the complex character‐ istics of tribology at low temperatures, (2) selection of the proper solid-lubricants against the oxidation or reduction power, and (3) active cooling to remove severe frictional heat at local hot spots [4].

**Figure 6.** Wear of five kinds of the ceramic balls against 440C disk in LO2 and LN2

of friction and wear of a PTFE pin in case of using an oxidized 440C steel disk [23,25]. At the pin temperatures above boiling point of LO2 (90 K), the friction and wear of PTFE pin showed relatively high values as compared with that showed in Fig. 4. As the pin temperature increased from 90 K to near 170 K, the friction and wear of PTFE drastically decreased to low values. The oxidized 440C steel disk was obtained by heating in air at about 623 K for 3 hours. The surface of the oxidized 440C showed an increase of FeO/Fe2O3 film in comparison with Cr2O3 film. It is noted that the oxidization of 440C steel should result in an increase of friction and wear of PTFE. It seems that PTFE transfer film was less formed due to poor adhesion of PTFE against FeO/Fe2O3, and frictional condition became to be severe. Thus, it is very interesting that the friction and wear properties of PTFE changed characteristically at its glass transition temper‐

For other friction tests, wear of PTFE in cryogenic GO2 was increased as surface roughness of 440C disk was increased; however, in cryogenic GN2, surface roughness had less effect on wear increase of PTFE. Furthermore, friction and wear of PTFE against Si3N4 disk was determined in cryogenic GO2 and GN2. In both cryogenic environments, friction coefficient was higher than that of 440C disk. It was noted that wear of PTFE in GO2 was drastically high compared with that in GN2. It was assumed that poor formation of PTFE transfer film on the SI3N4 disk resulted in an increase of friction and wear in GO2. This result indicated that the hybrid ceramic

It is interesting to use ceramic material as tribo-materials in cryogenic environments. Friction and wear behavior of typical fine ceramics against 440C disk were evaluated in LO2 and LN2. Figure 6 and 7 show wear and friction of five kinds of the ceramic balls in comparison with those in LO2 and LN2 [23], respectively. In all the cases of friction tests, the sliding contact surface of ceramic pin was covered by the transfer film of wear debris of 440C steel. The metallic transfer film prevented direct contact between metal and ceramic. As a result, the metal-tometal contact should control the friction and wear behavior of the sliding pair, and the order

In LO2, Al2O3 indicated the lowest wear rate and was followed by SiC, Si3N4, Sialon and ZrO2 in order of the wear resistance. For Al2O3 pin, the metallic oxide film of 440C seemed to be strongly adhered onto the ceramic pin and resulted in an increase of protection of the pin wear; however, wear of the 440C disk was prolonged. For SiC, Si3N4 and Sialon, sliding friction in oxidized environment made the glassy formation of SiO2 film due to tribo-chemical reaction. The hardness of SiO2 is much less than that of ceramic substrate and resulted in an increase in the wear of ceramic pins. It was noted that the wear rate of ZrO2 was considerably high as similar to that of self-mated 440C steels. Since ZrO2 has the lowest hardness compared with

To the contrary, in LN2, Zr2O3 indicated the lowest wear rate and was followed by Si3N4, Sialon, Al2O3 and SiC in order of the wear resistance. The high wear of Al2O3 and SiC pins was seemed to be induced by lack of protective film of 440C steel due to weak adhesion to ceramic pin. It is found that the order of wear resistance of ceramics against 440C steel in LO2 was opposed

ature, depending on the oxidization of 440C steel.

120 Tribology - Fundamentals and Advancements

bearing with Si3N4 ball showed poor self-lubrication in LO2.

of friction seemed to be less affected in the wear resistance of ceramic pins.

other ceramics, the hard oxide film of 440C should increase wear of ZrO2 pin.

to that in LN2 [23].

**Figure 7.** Friction of five kinds of the ceramic balls against 440C disk in LO2 and LN2

### **4. High–speed bearings**

#### **4.1. Improvement of self–lubrication of retainer [16, 17, 18, 26, 27]**

In the beginning of the development of the turbopump bearing for the LE-5, the bearing had used the composite PTFE retainer reinforced with glass fiber or carbon fiber. The bearing tested in LH2 by using a bearing tester showed that the glass fiber-reinforced PTFE retainer (24 wt.% glass fiber and additive) could demonstrate stable bearing-torque performance as compared with that of the carbon fiber-reinforced retainer (15 wt.% carbon fiber). From inspection of the ball-pocket surface of the carbon fiber-reinforced retainer, it was found that pile-up of the wear debris of carbon fiber might reduce supply of PTFE transfer film to ball surface. As a result, the LH2 turbopump bearing selected the glass fiber-reinforced PTFE retainer; however, the real turbopump test showed severe wear of the retainer when the turbopump was operated under poor cooling conditions. This fact indicated low wear resistance of the glass fiberreinforced PTFE retainer under severe operation of turbopump [16,17].

For the rocket-turbopump bearings, a laminated glass cloth with PTFE binder (laminated glass cloth of 45 wt.% and PTFE of 55 wt.%) was currently used because of its great strength to protect against dangerous retainer rupture [4,17]. This retainer showed poor self-lubrication resulting from abrasion by glass cloth layers exposed on the ball-pocket surface. During the develop‐ ment of the LH2 turbopumps for the LE-5, the bearing showed unstable high-temperature rise and poor lubrication was observed, resulting in severe wear of the balls. In case of the reusable turbopumps used in the SSME, the bearings similarly experienced a serious wear problem [6]. In order to improve the self-lubricating performance of the retainer, special surface treatment of the retainer was developed [12,18]. The abrasive retainer surface with the exposed glass cloth was chemically etched with hydrofluoric acid (HF) to a depth of 0.10-0.15 mm. Following this treatment, smooth surface for the retainer was obtained. The sliding friction and wear between the ball and ball-pocket surface was reduced, resulting in a sufficient supply of PTFE transfer film from the retainer to the rolling balls.

such as Ca and Mg were able to react easily with F by severe dry sliding friction and resulted in the formation of CaF2 and MgF2 within the transfer film [4]. The tribo-chemical formation of CaF2 and MgF2 might enhance adhesion of transfer film. When CaF2 and MgF2 added as fillers to PTFE, there was no tribo-chemical reaction, resulting in poor wear resistance. Furthermore, oxidation of the Mo filler in GO2 seemed to be extremely effective except in GN2.

**Figure 8.** Wear of PTFE composite pins with various fillers against 440C disk in cryogenic GO2 (123 K) under high-slid‐

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

123

During testing of the LH2 turbopump for the LE-7, the conventional bearings using a retainer with circular pockets showed a significant temperature rise under high shaft vibration. Since high shaft vibration increases the radial load applied to the bearings, ball excursion occurring in the ball pockets of the retainer due to ball-speed-variation (BSV) becomes significantly large. Figure 9 shows the ball excursion due to the BSV *vs.* the radial load for the 40-mm-bore bearing at a speed of 42,000 rpm [13]. The ball excursion tends to increase with increasing of the radial load. At a radial load of about 1.5 times thrust load, the ball excursion reaches a maximum value. When the pocket clearance of the retainer is smaller than the maximum ball excursion,

For the 40-mm-bore bearing, a retainer having elliptical pockets with a large pocket clearance was developed. As shown in Fig. 10, this retainer with elliptical pockets is able to allow maximum ball excursion due to BSV in the circumferential direction and to stabilize wobbling of the retainer due to a narrow clearance in the axial direction [13]. The pocket clearance of 1.8 mm was twice as large as that of the conventional circular pocket. Consequently, the LE-7

**4.2. Development of elliptical ball–pockets of retainer [13, 14, 26]**

ing speed (10 m/s)

severe contact occurs between the ball and the retainer pocket.

For the HF etched retainer tested in LH2, detailed examination of the transfer film on the sound ball surface with hardly any wear was conducted by electron probe microanalysis (EPMA) [12]. The result indicated that F of PTFE of the retainer strongly depended on the Ca concen‐ tration on the map and resulted in the tribo-chemical formation of CaF2 transfer film. The reacted oxide material (49 wt.% of glass fiber) consisted mainly of an oxide of Ca (CaO) remained on the HF etched retainer surface. Therefore, it seems that the formation of CaF2 transfer film was conducted by tribo-chemical reaction between F of PTFE and CaO remained on the retainer surface in chemical reduction environment in LH2.

In order to determine the effect of tribo-chemical formation of CaF2 in transfer film, additional friction tests were conducted. Figure 8 shows the wear of PTFE composite pins with 15 wt.% of various fillers against the 440C disk in cryogenic oxygen gas (GO2, 123 K) under a high sliding speed (10 m/s) [15]. The PTFE composites with CaO and MgO fillers showed excellent wear resistance (progression of the pin-wear was stopped) due to the formation of good transfer film even in both cryogenic GO2 and GN2 (123 K). It seems that alkali-earth-metals

**Figure 8.** Wear of PTFE composite pins with various fillers against 440C disk in cryogenic GO2 (123 K) under high-slid‐ ing speed (10 m/s)

such as Ca and Mg were able to react easily with F by severe dry sliding friction and resulted in the formation of CaF2 and MgF2 within the transfer film [4]. The tribo-chemical formation of CaF2 and MgF2 might enhance adhesion of transfer film. When CaF2 and MgF2 added as fillers to PTFE, there was no tribo-chemical reaction, resulting in poor wear resistance. Furthermore, oxidation of the Mo filler in GO2 seemed to be extremely effective except in GN2.

#### **4.2. Development of elliptical ball–pockets of retainer [13, 14, 26]**

**4. High–speed bearings**

122 Tribology - Fundamentals and Advancements

**4.1. Improvement of self–lubrication of retainer [16, 17, 18, 26, 27]**

reinforced PTFE retainer under severe operation of turbopump [16,17].

transfer film from the retainer to the rolling balls.

on the retainer surface in chemical reduction environment in LH2.

In the beginning of the development of the turbopump bearing for the LE-5, the bearing had used the composite PTFE retainer reinforced with glass fiber or carbon fiber. The bearing tested in LH2 by using a bearing tester showed that the glass fiber-reinforced PTFE retainer (24 wt.% glass fiber and additive) could demonstrate stable bearing-torque performance as compared with that of the carbon fiber-reinforced retainer (15 wt.% carbon fiber). From inspection of the ball-pocket surface of the carbon fiber-reinforced retainer, it was found that pile-up of the wear debris of carbon fiber might reduce supply of PTFE transfer film to ball surface. As a result, the LH2 turbopump bearing selected the glass fiber-reinforced PTFE retainer; however, the real turbopump test showed severe wear of the retainer when the turbopump was operated under poor cooling conditions. This fact indicated low wear resistance of the glass fiber-

For the rocket-turbopump bearings, a laminated glass cloth with PTFE binder (laminated glass cloth of 45 wt.% and PTFE of 55 wt.%) was currently used because of its great strength to protect against dangerous retainer rupture [4,17]. This retainer showed poor self-lubrication resulting from abrasion by glass cloth layers exposed on the ball-pocket surface. During the develop‐ ment of the LH2 turbopumps for the LE-5, the bearing showed unstable high-temperature rise and poor lubrication was observed, resulting in severe wear of the balls. In case of the reusable turbopumps used in the SSME, the bearings similarly experienced a serious wear problem [6]. In order to improve the self-lubricating performance of the retainer, special surface treatment of the retainer was developed [12,18]. The abrasive retainer surface with the exposed glass cloth was chemically etched with hydrofluoric acid (HF) to a depth of 0.10-0.15 mm. Following this treatment, smooth surface for the retainer was obtained. The sliding friction and wear between the ball and ball-pocket surface was reduced, resulting in a sufficient supply of PTFE

For the HF etched retainer tested in LH2, detailed examination of the transfer film on the sound ball surface with hardly any wear was conducted by electron probe microanalysis (EPMA) [12]. The result indicated that F of PTFE of the retainer strongly depended on the Ca concen‐ tration on the map and resulted in the tribo-chemical formation of CaF2 transfer film. The reacted oxide material (49 wt.% of glass fiber) consisted mainly of an oxide of Ca (CaO) remained on the HF etched retainer surface. Therefore, it seems that the formation of CaF2 transfer film was conducted by tribo-chemical reaction between F of PTFE and CaO remained

In order to determine the effect of tribo-chemical formation of CaF2 in transfer film, additional friction tests were conducted. Figure 8 shows the wear of PTFE composite pins with 15 wt.% of various fillers against the 440C disk in cryogenic oxygen gas (GO2, 123 K) under a high sliding speed (10 m/s) [15]. The PTFE composites with CaO and MgO fillers showed excellent wear resistance (progression of the pin-wear was stopped) due to the formation of good transfer film even in both cryogenic GO2 and GN2 (123 K). It seems that alkali-earth-metals

During testing of the LH2 turbopump for the LE-7, the conventional bearings using a retainer with circular pockets showed a significant temperature rise under high shaft vibration. Since high shaft vibration increases the radial load applied to the bearings, ball excursion occurring in the ball pockets of the retainer due to ball-speed-variation (BSV) becomes significantly large. Figure 9 shows the ball excursion due to the BSV *vs.* the radial load for the 40-mm-bore bearing at a speed of 42,000 rpm [13]. The ball excursion tends to increase with increasing of the radial load. At a radial load of about 1.5 times thrust load, the ball excursion reaches a maximum value. When the pocket clearance of the retainer is smaller than the maximum ball excursion, severe contact occurs between the ball and the retainer pocket.

For the 40-mm-bore bearing, a retainer having elliptical pockets with a large pocket clearance was developed. As shown in Fig. 10, this retainer with elliptical pockets is able to allow maximum ball excursion due to BSV in the circumferential direction and to stabilize wobbling of the retainer due to a narrow clearance in the axial direction [13]. The pocket clearance of 1.8 mm was twice as large as that of the conventional circular pocket. Consequently, the LE-7

**Figure 9.** Ball excursion due to BSV *vs.* radial load for LE-7 LH2 bearing at 42,000 rpm (40-mm-bore bearing)

turbopump bearings with the elliptical-pocket retainer exhibited excellent performance by reducing severe frictional heating and high wear of bearing components at a high-speed level of 50,000 rpm (2 million DN). Basic study of the elliptical pocket of the retainer was conducted

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

125

During the development of the LE-7A, the LH2 turbopump experienced severe operation with high vibration of the rotating shaft. As a result, high vibration of the rotating heavy turbine-disk increased radial load at the turbine-side bearings (40-mm bore) and broke the retainer due to large BSV [26]. It was considered that the ball-retainer contact force due to BSV bent the retainer and hoop stress occurred on the retainer inside, resulting in fracture of the thin (weak) web section of the ball pocket. In order to gain high reliability of the LH2 turbopump, the retainer using elliptical ball pocket was improved by increasing the

**Figure 11.** Maximum ball excursion vs. tilted misalignment under various thrust loads at 50,000 rpm (40-mm-bore

in the development of the LE-5 turbopump bearing [12,17].

pocket clearance to 2.2 mm.

bearing)


**Figure 10.** Circular and elliptical pockets of retainer and ball pocket clearances for 40-mm-bore bearing

turbopump bearings with the elliptical-pocket retainer exhibited excellent performance by reducing severe frictional heating and high wear of bearing components at a high-speed level of 50,000 rpm (2 million DN). Basic study of the elliptical pocket of the retainer was conducted in the development of the LE-5 turbopump bearing [12,17].

During the development of the LE-7A, the LH2 turbopump experienced severe operation with high vibration of the rotating shaft. As a result, high vibration of the rotating heavy turbine-disk increased radial load at the turbine-side bearings (40-mm bore) and broke the retainer due to large BSV [26]. It was considered that the ball-retainer contact force due to BSV bent the retainer and hoop stress occurred on the retainer inside, resulting in fracture of the thin (weak) web section of the ball pocket. In order to gain high reliability of the LH2 turbopump, the retainer using elliptical ball pocket was improved by increasing the pocket clearance to 2.2 mm.

**Figure 9.** Ball excursion due to BSV *vs.* radial load for LE-7 LH2 bearing at 42,000 rpm (40-mm-bore bearing)

124 Tribology - Fundamentals and Advancements

**Figure 10.** Circular and elliptical pockets of retainer and ball pocket clearances for 40-mm-bore bearing

**Figure 11.** Maximum ball excursion vs. tilted misalignment under various thrust loads at 50,000 rpm (40-mm-bore bearing)

**Figure 12.** Maximum ball excursion and tilted misalignment *vs.* thrust load at 50,000 rpm (40-mm-bore bearing)

**Figure 13.** Load capacity of transfer film under inner race ball-spinning in LH2

Performance of self-lubricating bearing coated with PTFE or MoS2 films was evaluated for the LH2 turbopump bearing of the LE-5. The PTFE and MoS2 films were coated with rf-sputtering. Bearing test was conducted for about 2 hours at a speed of 50,000 rpm in LH2. Frictional heating was estimated from the temperature rise of cooling flow through the test bearing [12]. The coated films are hoped to induce smooth running in the initial operation when the amount of the PTFE transfer film is insufficient. The high self-lubricating performance and durability were experimentally confirmed with the PTFE coated bearing indicating frictional heating of 170-250 W. For the MoS2 coated bearing, the frictional heating was 250-330 W and relatively high. The retainer of the PTFE coated bearing showed less ball-pocket wear than that of the

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

127

For high-speed bearings, since the bearing was under the outer-race ball control at high speed, the transfer film of the inner raceway was damaged due to the spinning of the ball. In order to evaluate stable operating condition without bearing damage, the load capacity of the transfer film under inner race ball-spinning in LH2 was determined as shown in Fig. 13[13].

**4.3. Performance of LH2 bearing [12, 13]**

MoS2 coated bearing.

Such BSV was also caused by inclination of the outer race to the shaft (tilted misalignment). The effect of tilted misalignment in a level of 1.9-3.5 x 10-3 mm/mm on the tribo-characteristics of 40-mm-bore ball bearing was determined. The bearing used a retainer having various elliptical ball pockets to restrain the ball-retainer contact due to high BSV. The elliptical ball pocket changed the pocket clearance (1.75mm, 1.95 mm and 2.15 mm). Figure 11 shows the relationship of the tilted misalignment and the maximum ball excursion under various thrust loads at a speed of 50,000 rpm [26]. It is understood that maximum ball excursion increased with an enlargement of tilted misalignment.

Figure 12 shows the relationship of the maximum ball excursion and the tilted misalign‐ ment *vs.* the thrust load at a speed of 50,000 rpm [26]. The relationship of the maximum ball excursion *vs.* the thrust load was calculated by assuming that the tilted misalign‐ ment linearly increased with an increase of the thrust load. As the thrust load increased, the calculated maximum ball excursion tended to increase in a parabolic pattern. It was found that, in case of the pocket clearance of 1.95 mm, ball-retainer contact due to ball excursion possibly occurred within a limited range of thrust loads, resulting in high increase of bearing torque and bearing temperature.

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps http://dx.doi.org/10.5772/55733 127

**Figure 13.** Load capacity of transfer film under inner race ball-spinning in LH2

#### **4.3. Performance of LH2 bearing [12, 13]**

**Figure 12.** Maximum ball excursion and tilted misalignment *vs.* thrust load at 50,000 rpm (40-mm-bore bearing)

with an enlargement of tilted misalignment.

126 Tribology - Fundamentals and Advancements

increase of bearing torque and bearing temperature.

Such BSV was also caused by inclination of the outer race to the shaft (tilted misalignment). The effect of tilted misalignment in a level of 1.9-3.5 x 10-3 mm/mm on the tribo-characteristics of 40-mm-bore ball bearing was determined. The bearing used a retainer having various elliptical ball pockets to restrain the ball-retainer contact due to high BSV. The elliptical ball pocket changed the pocket clearance (1.75mm, 1.95 mm and 2.15 mm). Figure 11 shows the relationship of the tilted misalignment and the maximum ball excursion under various thrust loads at a speed of 50,000 rpm [26]. It is understood that maximum ball excursion increased

Figure 12 shows the relationship of the maximum ball excursion and the tilted misalign‐ ment *vs.* the thrust load at a speed of 50,000 rpm [26]. The relationship of the maximum ball excursion *vs.* the thrust load was calculated by assuming that the tilted misalign‐ ment linearly increased with an increase of the thrust load. As the thrust load increased, the calculated maximum ball excursion tended to increase in a parabolic pattern. It was found that, in case of the pocket clearance of 1.95 mm, ball-retainer contact due to ball excursion possibly occurred within a limited range of thrust loads, resulting in high Performance of self-lubricating bearing coated with PTFE or MoS2 films was evaluated for the LH2 turbopump bearing of the LE-5. The PTFE and MoS2 films were coated with rf-sputtering. Bearing test was conducted for about 2 hours at a speed of 50,000 rpm in LH2. Frictional heating was estimated from the temperature rise of cooling flow through the test bearing [12]. The coated films are hoped to induce smooth running in the initial operation when the amount of the PTFE transfer film is insufficient. The high self-lubricating performance and durability were experimentally confirmed with the PTFE coated bearing indicating frictional heating of 170-250 W. For the MoS2 coated bearing, the frictional heating was 250-330 W and relatively high. The retainer of the PTFE coated bearing showed less ball-pocket wear than that of the MoS2 coated bearing.

For high-speed bearings, since the bearing was under the outer-race ball control at high speed, the transfer film of the inner raceway was damaged due to the spinning of the ball. In order to evaluate stable operating condition without bearing damage, the load capacity of the transfer film under inner race ball-spinning in LH2 was determined as shown in Fig. 13[13]. This figure shows the critical load capacity, that is, maximum Herze stress (*Smax*) *vs.* maximum spinning speed (*Vmax*). Under high thrust loads, an increasing of the bearing torque and bearing temperature (at limit A) was determined by the bearing tester which could measure the bearing torque in LH2. The film local rupture (at limit B) was also defined by the electrical resistance monitoring between the inner race and outer race. Up to a *Vmax* of 5 m/s at 50,000 rpm, the transfer film was able to sustain a *Smax* up to 2 GPa. It was determined that the load capacity of the transfer film depended more on *Smax* than on *Vmax*. So, in order to increase durability of the bearing, it is important to limit the stress level to a *Smax* of 2 GPa to prevent transfer-film rupture and sufficiently to cool the frictional heat due to high *Vmax*.

raceway. For the retainer with elliptical pockets, the wear depths in the pockets were smaller than the depth (0.10-0.15 mm) of chemical etching of the glass cloth. The PTFE layer without

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

129

To contrary, the all inner raceways of the LO2 turbopump bearings showed typical spin wear with light oxidative wear [14]. These turbopump bearings tested for a total time of 34.6 minutes with 23 engine start-stops. The surface profiles included the thickness of the initial film coatings of sputtered PTFE film (1 µm) on Ion-plated Au film (0.4 µm). The wear depths of raceways seemed to be relatively high; however, smooth surface roughness demonstrated mild wear without severe adhesion due to metal-to-metal. For bearing D that was affected by turbine whirling with radial overload, heavy spin wear with a wear depth of 7 µm was generated on the inner raceway. Furthermore, slight flaking was observed on the inner and outer raceways. This flaking was characterized by a very shallow depth and by fractures on the surface.

For the retainer with conventional circular pockets, the wear depths in the pockets were relatively light compared with those of the LH2 bearing. The contact area in the retainer pocket and on the ball surfaces was blackened by the thermally degraded transfer film. The degra‐ dation of the transfer film seemed to occur at a temperature above about 500 K. This was confirmed by a heating test of the retainer. These facts indicated that the transfer film was severely heated even in cryogenic fluid and the LO2 turbopump bearings were operated under poor cooling conditions. Thus, to increase the durability of the bearings, it is apparent that

In order to evaluate the excellent lubricating conditions without severe wear, XPS depth analysis of a transfer film on a ball used in the LH2 turbopump bearing of the LE-7 was conducted. Inspected ball that showed excellent wear condition was from the turbine-side bearing tested for 31.4 minutes in engine tests. The XPS depth analysis with an etching depth of 30 nm (SiO2 rate) indicated that F(1s) and Fe(2p) spectra show the significant formation of thick CaF2 and FeF2 film as shown in Fig. 14 [4]. It seemed that, due to the reduction power of LH2, the reacted CaO (remained on the retainer surface chemically etched with HF) was tribochemically changed to CaF2 with the F of PTFE retainer during bearing operation. In addition, due to removing of native oxide film by the LH2 reducing power, a FeF2 film was formed by a chemical reaction between the F of PTFE retainer and the Fe of 440C steel. It is noted that the formation of FeF2 film at the stressed contact area resulted in demonstrating high resistance to

Thus, the LH2 turbopump bearings used in the engine firing tests demonstrated excellent performance due to the formation of thick CaF2 and FeF2 film. The tribo-chemical formation of CaF2/FeF2 film possibly reduced wear at frictional interfaces within the bearings used in

( ) ( ) <sup>2</sup> n n 2 22 -CF - + CaO + Fe -CF -CO- + CaF + FeF ® (1)

sufficient cooling is essential.

**b.** *XPS analysis of transfer films*

metal-to-metal adhesion and in leading to less wear [27].

LH2. The basic tribo-chemical reaction was determined as follows [4]:

the abrasive glass cloth sufficiently remained at the bottom of the pocket wear scar.

#### **4.4. Durability of LO2 bearing [15]**

It is noted that violent frictional heating in LO2 can lead to the ignition of tribo-elements due to burn-out phenomenon. Burn out is overheat occurring in a transition from nucleate boiling to film boiling at critical heat flux that is defined by engineering heat transfer. For the LO2 turbopump bearings (32-mm and 45-mm bore) of the LE-7, the durability and fatigue life were evaluated by applying heavy radial loads at a speed of 20,000 rpm in LO2 or LN2. During testing, the bearing-cartridge-acceleration (BCA), i.e., *Gpk* (peak value) and *Grms* (rot-meansquare value), was monitored to detect bearing damage. Testing in LO2 for about 2.2 hours under a system radial load of 5,880 N showed that excellent lubricating conditions without abnormal BCA were obtained for all bearings.

Durability test in LN2 (to keep safety in the experience) under a heavy system radial load of 11,760 N was conducted at a speed of 20,000 rpm for about 5.1 hours [15]. The result detected that the fatigue life of the bearing was about the same as the calculated B10 fatigue life. The bearings were operated at steady conditions for 5.1 hours with 20 start-stops. For BCA on bearings A/B, *Gpk* and *Grms* on the chart were abnormally separated from each other in a pattern of abnormal BCA showing an increase of surface roughness due to an occurrence of slight flaking. Then, at a total test time of 3.8 hours, the loaded and unloaded BCA abnormally began to increase concomitantly. Examination of tested bearing B indicted that slight flaking with very shallow depth (about 8.5 µm) was observed on the inner raceway.

#### **4.5. Evaluation of turbopump bearings [14]**

The durability of the bearings of the LO2/LH2 turbopumps used in the firing tests of the LE-7 was evaluated based of findings of wear inspection and X-ray photoelectron spectroscopic (XPS) analysis of PTFE transfer film. Inspection of the turbopump bearings used in the engine firing tests is essential for evaluation of their durability under engine operation.

#### **a.** *Bearing wear*

After the engine firing test, surface profiles of the raceways of the LH2 turbopump bearings was evaluated [14]. The engine test was conducted for a total time of 31.4 minutes with 20 engine start-stops. The surface profiles included the thickness (1µm) of the initial film coatings of sputtered PTFE film. It is obvious that the wear scars on the raceways of all bearings were flat and spin wear was not observed despite conditions of higher ball spinning on the inner raceway. For the retainer with elliptical pockets, the wear depths in the pockets were smaller than the depth (0.10-0.15 mm) of chemical etching of the glass cloth. The PTFE layer without the abrasive glass cloth sufficiently remained at the bottom of the pocket wear scar.

To contrary, the all inner raceways of the LO2 turbopump bearings showed typical spin wear with light oxidative wear [14]. These turbopump bearings tested for a total time of 34.6 minutes with 23 engine start-stops. The surface profiles included the thickness of the initial film coatings of sputtered PTFE film (1 µm) on Ion-plated Au film (0.4 µm). The wear depths of raceways seemed to be relatively high; however, smooth surface roughness demonstrated mild wear without severe adhesion due to metal-to-metal. For bearing D that was affected by turbine whirling with radial overload, heavy spin wear with a wear depth of 7 µm was generated on the inner raceway. Furthermore, slight flaking was observed on the inner and outer raceways. This flaking was characterized by a very shallow depth and by fractures on the surface.

For the retainer with conventional circular pockets, the wear depths in the pockets were relatively light compared with those of the LH2 bearing. The contact area in the retainer pocket and on the ball surfaces was blackened by the thermally degraded transfer film. The degra‐ dation of the transfer film seemed to occur at a temperature above about 500 K. This was confirmed by a heating test of the retainer. These facts indicated that the transfer film was severely heated even in cryogenic fluid and the LO2 turbopump bearings were operated under poor cooling conditions. Thus, to increase the durability of the bearings, it is apparent that sufficient cooling is essential.

#### **b.** *XPS analysis of transfer films*

This figure shows the critical load capacity, that is, maximum Herze stress (*Smax*) *vs.* maximum spinning speed (*Vmax*). Under high thrust loads, an increasing of the bearing torque and bearing temperature (at limit A) was determined by the bearing tester which could measure the bearing torque in LH2. The film local rupture (at limit B) was also defined by the electrical resistance monitoring between the inner race and outer race. Up to a *Vmax* of 5 m/s at 50,000 rpm, the transfer film was able to sustain a *Smax* up to 2 GPa. It was determined that the load capacity of the transfer film depended more on *Smax* than on *Vmax*. So, in order to increase durability of the bearing, it is important to limit the stress level to a *Smax* of 2 GPa to prevent

It is noted that violent frictional heating in LO2 can lead to the ignition of tribo-elements due to burn-out phenomenon. Burn out is overheat occurring in a transition from nucleate boiling to film boiling at critical heat flux that is defined by engineering heat transfer. For the LO2 turbopump bearings (32-mm and 45-mm bore) of the LE-7, the durability and fatigue life were evaluated by applying heavy radial loads at a speed of 20,000 rpm in LO2 or LN2. During testing, the bearing-cartridge-acceleration (BCA), i.e., *Gpk* (peak value) and *Grms* (rot-meansquare value), was monitored to detect bearing damage. Testing in LO2 for about 2.2 hours under a system radial load of 5,880 N showed that excellent lubricating conditions without

Durability test in LN2 (to keep safety in the experience) under a heavy system radial load of 11,760 N was conducted at a speed of 20,000 rpm for about 5.1 hours [15]. The result detected that the fatigue life of the bearing was about the same as the calculated B10 fatigue life. The bearings were operated at steady conditions for 5.1 hours with 20 start-stops. For BCA on bearings A/B, *Gpk* and *Grms* on the chart were abnormally separated from each other in a pattern of abnormal BCA showing an increase of surface roughness due to an occurrence of slight flaking. Then, at a total test time of 3.8 hours, the loaded and unloaded BCA abnormally began to increase concomitantly. Examination of tested bearing B indicted that slight flaking

The durability of the bearings of the LO2/LH2 turbopumps used in the firing tests of the LE-7 was evaluated based of findings of wear inspection and X-ray photoelectron spectroscopic (XPS) analysis of PTFE transfer film. Inspection of the turbopump bearings used in the engine

After the engine firing test, surface profiles of the raceways of the LH2 turbopump bearings was evaluated [14]. The engine test was conducted for a total time of 31.4 minutes with 20 engine start-stops. The surface profiles included the thickness (1µm) of the initial film coatings of sputtered PTFE film. It is obvious that the wear scars on the raceways of all bearings were flat and spin wear was not observed despite conditions of higher ball spinning on the inner

with very shallow depth (about 8.5 µm) was observed on the inner raceway.

firing tests is essential for evaluation of their durability under engine operation.

transfer-film rupture and sufficiently to cool the frictional heat due to high *Vmax*.

**4.4. Durability of LO2 bearing [15]**

128 Tribology - Fundamentals and Advancements

abnormal BCA were obtained for all bearings.

**4.5. Evaluation of turbopump bearings [14]**

**a.** *Bearing wear*

In order to evaluate the excellent lubricating conditions without severe wear, XPS depth analysis of a transfer film on a ball used in the LH2 turbopump bearing of the LE-7 was conducted. Inspected ball that showed excellent wear condition was from the turbine-side bearing tested for 31.4 minutes in engine tests. The XPS depth analysis with an etching depth of 30 nm (SiO2 rate) indicated that F(1s) and Fe(2p) spectra show the significant formation of thick CaF2 and FeF2 film as shown in Fig. 14 [4]. It seemed that, due to the reduction power of LH2, the reacted CaO (remained on the retainer surface chemically etched with HF) was tribochemically changed to CaF2 with the F of PTFE retainer during bearing operation. In addition, due to removing of native oxide film by the LH2 reducing power, a FeF2 film was formed by a chemical reaction between the F of PTFE retainer and the Fe of 440C steel. It is noted that the formation of FeF2 film at the stressed contact area resulted in demonstrating high resistance to metal-to-metal adhesion and in leading to less wear [27].

Thus, the LH2 turbopump bearings used in the engine firing tests demonstrated excellent performance due to the formation of thick CaF2 and FeF2 film. The tribo-chemical formation of CaF2/FeF2 film possibly reduced wear at frictional interfaces within the bearings used in LH2. The basic tribo-chemical reaction was determined as follows [4]:

$$\text{-}\left(\text{-CF}\_{2}\text{-}\right)\_{\text{n}} + \text{CaO} + \text{Fe} \rightarrow \left(\text{-CF}\_{2}\text{-}\text{CO-}\right)\_{\text{n}} + \text{CaF}\_{2} + \text{FeF}\_{2} \tag{1}$$

Electron counts (cps)

Etching

http://dx.doi.org/10.5772/55733

131

**CaF2**

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

Etching depth; 30 nm (SiO2 rate)

**Fe2O3 Fe**

Binding energy (eV)

**Figure 15.** XPS depth analysis of ball for LO2 turbopump bearing (turbine side)

**Figure 16.** Face-contact mechanical seal for LH2 turbopump of LE-5

direction

**Figure 14.** XPS depth analysis of ball for LH2 turbopump bearing (turbine side)

On the contrary, for the LO2 turbopump bearings of the LE-7, the inspected ball was from the turbine-side bearing that was tested for 34.6 minutes in engine tests and showed heavy spin wear. Figure 15 shows the XPS depth analysis with an etching depth of 30 nm (SiO2 rate) for the worn ball due to spin wear. It indicated that the oxidization power of LO2 prohibited the tribo-chemical formation of CaF2 /FeF2 transfer film. This bearing was operated under poor cooling conditions, so that the bearing wear was relatively increased and shallow flaking was formed on the raceways. From the F spectrum, it was shown that very thin PTFE/CaF2 transfer film was formed compared with the thick PTFE/CaF2 transfer film in the LH2 bearing. Fur‐ thermore, from the Fe spectrum, formation of Fe2O3 oxide film was typically shown. Fe2O3 oxide film was apt to form at elevated temperature, so that the oxidative mild wear in the bearing was increased due to poor cooling conditions in LO2 [5]. As mention later (in 6.1.1), for the bearing tested under sufficient cooling condition, the intense formation of Cr2O3 film without Fe2O3 film was found beneath an extremely thin PTFE film, resulting in high resistance to metal-to-metal adhesion and in a decrease of the bearing wear [28].

**Figure 15.** XPS depth analysis of ball for LO2 turbopump bearing (turbine side)

**Figure 14.** XPS depth analysis of ball for LH2 turbopump bearing (turbine side)

130 Tribology - Fundamentals and Advancements

to metal-to-metal adhesion and in a decrease of the bearing wear [28].

On the contrary, for the LO2 turbopump bearings of the LE-7, the inspected ball was from the turbine-side bearing that was tested for 34.6 minutes in engine tests and showed heavy spin wear. Figure 15 shows the XPS depth analysis with an etching depth of 30 nm (SiO2 rate) for the worn ball due to spin wear. It indicated that the oxidization power of LO2 prohibited the tribo-chemical formation of CaF2 /FeF2 transfer film. This bearing was operated under poor cooling conditions, so that the bearing wear was relatively increased and shallow flaking was formed on the raceways. From the F spectrum, it was shown that very thin PTFE/CaF2 transfer film was formed compared with the thick PTFE/CaF2 transfer film in the LH2 bearing. Fur‐ thermore, from the Fe spectrum, formation of Fe2O3 oxide film was typically shown. Fe2O3 oxide film was apt to form at elevated temperature, so that the oxidative mild wear in the bearing was increased due to poor cooling conditions in LO2 [5]. As mention later (in 6.1.1), for the bearing tested under sufficient cooling condition, the intense formation of Cr2O3 film without Fe2O3 film was found beneath an extremely thin PTFE film, resulting in high resistance

**Figure 16.** Face-contact mechanical seal for LH2 turbopump of LE-5

#### **5. Turbopump shaft seals**

#### **5.1. Mechanical seal [29-34]**

For the LE-5 turbopumps operating under the gas generator cycle, the contact-type mechanical seal was able to use for the propellant seals because the pump and turbine pressures were relatively low. Specially, for the LH2 turbopump, a high-speed mechani‐ cal seal was required to withstand high rubbing speed (113 m/s) at a speed of 50,000 rpm in LH2. Figure 16 shows the face-contact mechanical seal with a seal diameter of 43.2 mm developed for the LH2 turbopump of the LE-5 [29,30]. In order to reduce seal leakage of LH2, it has a modified seal nose that could reduce the seal face distortion and control the direction of its distortion (to contact at outside of the seal face) under low temperature and high pressure. Furthermore, a modified vibration damper made of PTFE sheets is attach‐ ed around the seal nose to prevent fluttering during rapid start or stop of the turbopump.

When the closing force to contact seal faces is increased to make seal leakage smaller, wear rate of the seal faces is increased due to the poor lubrication of LH2. If the closing force is set to be smaller than the fluid opening force separating seal face each other, the leakage is considered to be quite large because of the extremely low viscosity and density of LH2. Therefore, to obtain the stable seal performance and the long wear life, it is important that the proper balance between the closing force and the opening force is retained.

Critical value of the seal balance ratio that obtained stable seal performance and reduce wear of the seal faces was experimentally and analytically evaluated [32,33]. In this study, the experimental and analytical study on the friction power loss and seal performance was conducted. It was indicated that the friction power loss fell to a small value after the seal faces were sufficiently run-in. The seal balance ratio *[B]* that stabilized seal performance was in a range of 0.77-0.82. The seal balance ratio *[B]* is determined by the following equation;

$$
\begin{bmatrix} \mathbf{B} \end{bmatrix} = \mathbf{B} + F\mathbf{s}p \text{ /} \left( \mathbf{A}\mathbf{s}\Delta P \right) \tag{2}
$$

constant *<sup>m</sup> Pv* = (3)

http://dx.doi.org/10.5772/55733

133

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

[*K Fo As P* = D /] ( ) (4)

where, *P* is the pressure, *v* is the specific volume and *m* is the ausfluss exponent. As *m* decreases with the temperature rise of gas due to viscous friction, the pressure of leakage flow increases particularly in the gas region within gas-liquid phase, and it resulted in the increase of the opening force. The analysis of phase change model of leakage was conducted using the flow

Figure 17 shows the calculated and experimental results of the relationship between the seal clearance and the opening force ratio *[K]* at a speed of 50,000 rpm in LH2. The opening force

where, *Fo* is the opening force. It was also shown that the opening force within seal clearance increases linearly as the seal clearance decreases. After the seal faces were sufficiently run-in and the seal clearance was maintained in an average of 0.6 µm, the opening force ratio *[K]* ap‐ proaches the critical balance ratio *[B]c* (= 0.77) that showed critical seal performance. As a result, the difference of *[K]* and *[B]c* was decreased and it resulted in the reduction of the load on the seal faces. The frictional loss power was decreased to a small value, resulting in a restrain of wear rate of seal faces. If the seal clearance increases, the leakage becomes large; however, the load on the seal faces is increased with the decrease of the opening force and the seal clearance would become small enough to reduce leakage. Furthermore, the starting torque and static seal performance were markedly affected by the change of the seal face distortion due to wear [31].

Durability of the mechanical seal was evaluated by the long-run test [29]. The long-run test was conducted at a speed of 50,000 rpm with a seal pressure of 1.37 MPaG for 83 minutes. The experimental results showed that the leakage gradually increased until total test time was 50 minutes. During its step, wear of the seal faces was running-in, then the leakage was stabilized. It is noted that an extremely small LH2 leakage (8-19 cc/min) was kept during test. The seal after the durability test indicated an excellent condition that maximum wear of carbon-ring

Temperature on the rubbing seal faces was estimated from the reduction rate of the hardness of hard Cr plating on the rotating mating ring [34]. The estimated temperature of rubbing seal face was possibly reached to be about 773 K at a rubbing speed of 113 m/s in LH2. In an initial stage of running-in, extremely high temperature of the seal faces caused thermal cracks in wear surface of the Cr plating, so that it is necessary to cool the contacting seal faces sufficiently. When the cooling of the sealing unit is insufficient, the surface of the carbon seal ring showed abnormal wear. Furthermore, the Cr plating showed better wear results than the tungsten carbide (WC) coating, because the Cr plating easily forms thin transfer films of graphite contained in the carbon. In the case of the WC coating, the transfer film of graphite was hardly

formed in LH2, resulting in an occurrence of severe seal wear.

and energy equations of liquid and gas leakages.

ratio *[K]* is expressed by the following equation;

was 8 µm.

where, *B* is the fluid balance ratio, *Fsp* is the spring force of bellows, *As* is the seal area and *△P* is the seal pressure. *[B]* is determined by the initial spring force of the bellows.

When the seal balance ratio was below 0.77, the leakage was apt to increase due to lack of the closing force. In this case, the critical balance ratio *[B]c* that gains stable seal performance showing small leakage was 0.77. To contrast, its balance ratio above 0.82 increased wear of the seal face by rise of the closing force. This high value of critical balance ratio was due to large opening force that could be explained with leakage flow model, assuming the phase change of leakage (from liquid phase to gas-liquid phase and gas phase) due to viscous frictional heating at high rubbing speed. In this phase change model, a state change of gas was assumed to be irreversibly adiabatic and a curve of gas expansion expressed by the following equation;

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps http://dx.doi.org/10.5772/55733 133

$$Pv^{\prime\prime} = \text{constant} \tag{3}$$

where, *P* is the pressure, *v* is the specific volume and *m* is the ausfluss exponent. As *m* decreases with the temperature rise of gas due to viscous friction, the pressure of leakage flow increases particularly in the gas region within gas-liquid phase, and it resulted in the increase of the opening force. The analysis of phase change model of leakage was conducted using the flow and energy equations of liquid and gas leakages.

**5. Turbopump shaft seals**

132 Tribology - Fundamentals and Advancements

For the LE-5 turbopumps operating under the gas generator cycle, the contact-type mechanical seal was able to use for the propellant seals because the pump and turbine pressures were relatively low. Specially, for the LH2 turbopump, a high-speed mechani‐ cal seal was required to withstand high rubbing speed (113 m/s) at a speed of 50,000 rpm in LH2. Figure 16 shows the face-contact mechanical seal with a seal diameter of 43.2 mm developed for the LH2 turbopump of the LE-5 [29,30]. In order to reduce seal leakage of LH2, it has a modified seal nose that could reduce the seal face distortion and control the direction of its distortion (to contact at outside of the seal face) under low temperature and high pressure. Furthermore, a modified vibration damper made of PTFE sheets is attach‐ ed around the seal nose to prevent fluttering during rapid start or stop of the turbopump.

When the closing force to contact seal faces is increased to make seal leakage smaller, wear rate of the seal faces is increased due to the poor lubrication of LH2. If the closing force is set to be smaller than the fluid opening force separating seal face each other, the leakage is considered to be quite large because of the extremely low viscosity and density of LH2. Therefore, to obtain the stable seal performance and the long wear life, it is important that the

Critical value of the seal balance ratio that obtained stable seal performance and reduce wear of the seal faces was experimentally and analytically evaluated [32,33]. In this study, the experimental and analytical study on the friction power loss and seal performance was conducted. It was indicated that the friction power loss fell to a small value after the seal faces were sufficiently run-in. The seal balance ratio *[B]* that stabilized seal performance was in a

where, *B* is the fluid balance ratio, *Fsp* is the spring force of bellows, *As* is the seal area and *△P*

When the seal balance ratio was below 0.77, the leakage was apt to increase due to lack of the closing force. In this case, the critical balance ratio *[B]c* that gains stable seal performance showing small leakage was 0.77. To contrast, its balance ratio above 0.82 increased wear of the seal face by rise of the closing force. This high value of critical balance ratio was due to large opening force that could be explained with leakage flow model, assuming the phase change of leakage (from liquid phase to gas-liquid phase and gas phase) due to viscous frictional heating at high rubbing speed. In this phase change model, a state change of gas was assumed to be irreversibly adiabatic and a curve of gas expansion

é ù *B B Fsp As P* / ( ) û = ë + D (2)

range of 0.77-0.82. The seal balance ratio *[B]* is determined by the following equation;

is the seal pressure. *[B]* is determined by the initial spring force of the bellows.

expressed by the following equation;

proper balance between the closing force and the opening force is retained.

**5.1. Mechanical seal [29-34]**

Figure 17 shows the calculated and experimental results of the relationship between the seal clearance and the opening force ratio *[K]* at a speed of 50,000 rpm in LH2. The opening force ratio *[K]* is expressed by the following equation;

$$\text{E}\begin{bmatrix}\text{K}\end{bmatrix} = \text{Fo} \begin{pmatrix}\text{(As $\Delta P$ )} & \text{(As $\Delta P$ )} \end{pmatrix} \tag{4}$$

where, *Fo* is the opening force. It was also shown that the opening force within seal clearance increases linearly as the seal clearance decreases. After the seal faces were sufficiently run-in and the seal clearance was maintained in an average of 0.6 µm, the opening force ratio *[K]* ap‐ proaches the critical balance ratio *[B]c* (= 0.77) that showed critical seal performance. As a result, the difference of *[K]* and *[B]c* was decreased and it resulted in the reduction of the load on the seal faces. The frictional loss power was decreased to a small value, resulting in a restrain of wear rate of seal faces. If the seal clearance increases, the leakage becomes large; however, the load on the seal faces is increased with the decrease of the opening force and the seal clearance would become small enough to reduce leakage. Furthermore, the starting torque and static seal performance were markedly affected by the change of the seal face distortion due to wear [31].

Durability of the mechanical seal was evaluated by the long-run test [29]. The long-run test was conducted at a speed of 50,000 rpm with a seal pressure of 1.37 MPaG for 83 minutes. The experimental results showed that the leakage gradually increased until total test time was 50 minutes. During its step, wear of the seal faces was running-in, then the leakage was stabilized. It is noted that an extremely small LH2 leakage (8-19 cc/min) was kept during test. The seal after the durability test indicated an excellent condition that maximum wear of carbon-ring was 8 µm.

Temperature on the rubbing seal faces was estimated from the reduction rate of the hardness of hard Cr plating on the rotating mating ring [34]. The estimated temperature of rubbing seal face was possibly reached to be about 773 K at a rubbing speed of 113 m/s in LH2. In an initial stage of running-in, extremely high temperature of the seal faces caused thermal cracks in wear surface of the Cr plating, so that it is necessary to cool the contacting seal faces sufficiently. When the cooling of the sealing unit is insufficient, the surface of the carbon seal ring showed abnormal wear. Furthermore, the Cr plating showed better wear results than the tungsten carbide (WC) coating, because the Cr plating easily forms thin transfer films of graphite contained in the carbon. In the case of the WC coating, the transfer film of graphite was hardly formed in LH2, resulting in an occurrence of severe seal wear.

**Figure 18.** Floating ring seal for LO2 turbopump of LE-7

the seal gap.

phase flow.

The leakage from the floating-ring seal for the LH2 and LO2 seal can be calculated from the equation of the incompressible fluid flow in the rotating double cylinders when the leakage is liquid phase flow and the mass flow flux (mass flow/seal area in the flow direction) is large [29,35]. When the seal gap is narrow and the seal pressure is low, the mass flow flux of leakage is reduced, and vaporization of leakage occurred by viscous frictional heating and pressure

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

135

Comparison between the experimental and calculated leakage of LH2 was evaluated by the mass flow flux of leakage for the floating-ring seal with one seal ring or two seal rings [29]. In this study, the LH2 seal with a seal diameter of 32 mm and various seal gap of 30-86 µm was tested at rotating speeds to 50,000 rpm. It is shown that the leakage of LH2 is less than the calculated value from incompressible fluid flow equation because the leakage is changed to be tow-phase flow. When the mass flow flux is large, most of leakage flows out in liquid phase. This means that there is not sufficient time to vaporize the leakage to be tow-phase flow within

A flow visualization study of floating-ring seal was conducted to identify the two-phase flow area induced by viscous frictional heating and pressure drop [36]. In order to visualize the two-phase flow in seal gap, the floating ring made of transparent hard plastic (polycarbonate) was tested in a seal fluid of LN2. It was confirmed that the two-phase flow seemed to be homogeneous mixture of liquid and vapor flow and the two-phase flow area increases with increasing rotational speed and decreases leakage flow rate. When the two-phase flow area was fully prolonged within the seal gap, the leakage rate contrary increased with instability because the inlet flow resistance at the high-pressure side of the seal ring was reduced by two-

drop changes liquid phase flow to gas-liquid phase flow (two-phase flow).

**Figure 17.** Opening force ratio *[K]vs.* seal clearance at 50,000 rpm in LH2

#### **5.2. Floating ring seal [22, 29, 35, 37]**

A floating-ring seal is a type of no-contact annular seal without a rubbing seal surface. It has a simple structure and is able to seal high-pressure fluids, restraining leakage through a small clearance (gap) between the seal ring and the runner. Its gap is in an order of several dozens of µm. The seal ring is free to move in the radial direction, and thus severe contact with the rotating runner can be prevented. Leakage of floating-ring seal is much larger than that of facecontact mechanical seal, but the floating-ring seal shows a high resistance to pressure and a high reliability when used as high-pressure seal. A multi-seal system consisting of several seal rings arranged in series is employed for the high-pressure turbopumps. The floating-ring seals were developed and used in the LE-5 and LE-7.

Figure 18 shows the floating-ring seal with a seal diameter of 50 mm developed for the LO2 turbopump of the LE-7 [22,35]. The carbon seal ring is enclosed with a retainer of the same material as the seal runner. Since the retainer contracts thermally nearly as much as the seal runner at low temperature, the seal gap hardly changes. The seal gap was 50-60µm. When the seal pressure increases, the floating ring is pressed against the secondary seal by the fluid force and its movement in the radial direction is restrained. In order to smooth the radial movement of the floating ring, on the secondary seal of the housing, the PTFE film was coated for the LO2 seal and the MoS2 film was coated for the turbine gas seal (to seal the low temperature GH2). For the GH2 leakage of the floating-ring seal used in the turbine gas seal, leakage rate calculated by the quasi-one-dimensional compressible flow equation agreed quite well with experimental value.

**Figure 18.** Floating ring seal for LO2 turbopump of LE-7

**Figure 17.** Opening force ratio *[K]vs.* seal clearance at 50,000 rpm in LH2

A floating-ring seal is a type of no-contact annular seal without a rubbing seal surface. It has a simple structure and is able to seal high-pressure fluids, restraining leakage through a small clearance (gap) between the seal ring and the runner. Its gap is in an order of several dozens of µm. The seal ring is free to move in the radial direction, and thus severe contact with the rotating runner can be prevented. Leakage of floating-ring seal is much larger than that of facecontact mechanical seal, but the floating-ring seal shows a high resistance to pressure and a high reliability when used as high-pressure seal. A multi-seal system consisting of several seal rings arranged in series is employed for the high-pressure turbopumps. The floating-ring seals

Figure 18 shows the floating-ring seal with a seal diameter of 50 mm developed for the LO2 turbopump of the LE-7 [22,35]. The carbon seal ring is enclosed with a retainer of the same material as the seal runner. Since the retainer contracts thermally nearly as much as the seal runner at low temperature, the seal gap hardly changes. The seal gap was 50-60µm. When the seal pressure increases, the floating ring is pressed against the secondary seal by the fluid force and its movement in the radial direction is restrained. In order to smooth the radial movement of the floating ring, on the secondary seal of the housing, the PTFE film was coated for the LO2 seal and the MoS2 film was coated for the turbine gas seal (to seal the low temperature GH2). For the GH2 leakage of the floating-ring seal used in the turbine gas seal, leakage rate calculated by the quasi-one-dimensional compressible flow equation agreed quite well with

**5.2. Floating ring seal [22, 29, 35, 37]**

134 Tribology - Fundamentals and Advancements

were developed and used in the LE-5 and LE-7.

experimental value.

The leakage from the floating-ring seal for the LH2 and LO2 seal can be calculated from the equation of the incompressible fluid flow in the rotating double cylinders when the leakage is liquid phase flow and the mass flow flux (mass flow/seal area in the flow direction) is large [29,35]. When the seal gap is narrow and the seal pressure is low, the mass flow flux of leakage is reduced, and vaporization of leakage occurred by viscous frictional heating and pressure drop changes liquid phase flow to gas-liquid phase flow (two-phase flow).

Comparison between the experimental and calculated leakage of LH2 was evaluated by the mass flow flux of leakage for the floating-ring seal with one seal ring or two seal rings [29]. In this study, the LH2 seal with a seal diameter of 32 mm and various seal gap of 30-86 µm was tested at rotating speeds to 50,000 rpm. It is shown that the leakage of LH2 is less than the calculated value from incompressible fluid flow equation because the leakage is changed to be tow-phase flow. When the mass flow flux is large, most of leakage flows out in liquid phase. This means that there is not sufficient time to vaporize the leakage to be tow-phase flow within the seal gap.

A flow visualization study of floating-ring seal was conducted to identify the two-phase flow area induced by viscous frictional heating and pressure drop [36]. In order to visualize the two-phase flow in seal gap, the floating ring made of transparent hard plastic (polycarbonate) was tested in a seal fluid of LN2. It was confirmed that the two-phase flow seemed to be homogeneous mixture of liquid and vapor flow and the two-phase flow area increases with increasing rotational speed and decreases leakage flow rate. When the two-phase flow area was fully prolonged within the seal gap, the leakage rate contrary increased with instability because the inlet flow resistance at the high-pressure side of the seal ring was reduced by twophase flow.

#### **5.3. Segmented seal [22, 35, 37, 38]**

Contact-type segmented seal were used in the GHe purge seals and the low pressured turbine gas seals. The GHe purge seal used in the LO2 turbopump of the LE-7 is shown in Fig. 19 [22]. Segmented seal has a carbon seal ring cut into three segments. The segmented annular seal ring is pressed on the seal runner with a coil spring and maintains high purge-pressure of GHe as a barrier gas. Wear of the carbon seal ring is reduced by using the shrouded Rayleigh step lift-pads to increase the opening force within the seal clearance. As the rubbing speed increases, the opening force in the Rayleigh step increases, so that the rubbing speed is increased by enlarging the seal diameter using a T-type runner.

R

unning Title

Without MoS2 coati

**Figure 20.** Comparison of wear of MoS2 coated and uncoated seal surfaces

of the carbon seal ring at low temperature, as shown in Fig. 20.

balance between the opening and closing forces.

**6. Advanced bearings and shaft seals**

 ng

> With MoS2 coating

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

Change of the friction and wear of the carbon pin as a function of the pin temperature was determined in the cryogenic GHe environment [23]. Friction test was conducted against the Cr-plated steel disk at a sliding speed of 12 m/s and load of 9.8 N. When the pin temperature is below the solidification temperature of CO2 (216 K), it is noted that lubricating property of the carbon pin suddenly disappeared and friction and wear became intensive. When absorbed CO2 gas was changed to be solid phase, lubricity of carbon was lost. This phenomenon resembles that when phase of moisture is transfer to solid phase (ice) below 273 K, lubricity decreases; be well known. From this fact, it seemed that severe wear of the GHe purge seal was generated because the environmental temperature around the seal was lower than 216 K. Spray MoS2 coating on the carbon seal face was drastically able to prohibit progression of wear

After a total operating time of 29 minutes for the engine firing test, the GHe purge seal used in the LE-7 indicated that the seal surfaces coated by MoS2 were found to be in excellent condition and wear depth of the carbon seal ring was about 7 µm. It assumes that high opening force produced by the Rayleigh step was kept by prohibit of wear of the Rayleigh step and the GHe purge seal was operated under conditions of nearly no load on the seal surfaces due to

Future space transport systems require reusable launch vehicles to reduce launch cost and to increase efficiency. The durability of reusable turbopump bearings must be greater than that 25

137

1 2

Relationship between the purge pressure and the leakage rate of GHe purge seal was evaluated at a steady speed of 20,000 rpm [22]. When the purge pressure is low, the seal face is kept to be non-contact because the Rayleigh step increases the seal opening force. As the purge pressure is set to be high, the seal face condition is changed from the non-contact state to the contact state, it resulted that the dynamic leakage almost equals that of the resting state. Furthermore, for the GHe purge seal combined with the LO2 floating-ring seal, the environ‐ mental temperature around the GHe purge seal was equal to that of LO2 leakage, so that the carbon seal ring showed severe wear with an appearance of worn-out of the Rayleigh step.

**Figure 19.** GHe purge seal for LO2 turbopump of LE-7

25

**Figure 20.** Comparison of wear of MoS2 coated and uncoated seal surfaces

R

unning Title

1 2

**5.3. Segmented seal [22, 35, 37, 38]**

136 Tribology - Fundamentals and Advancements

enlarging the seal diameter using a T-type runner.

**Figure 19.** GHe purge seal for LO2 turbopump of LE-7

Contact-type segmented seal were used in the GHe purge seals and the low pressured turbine gas seals. The GHe purge seal used in the LO2 turbopump of the LE-7 is shown in Fig. 19 [22]. Segmented seal has a carbon seal ring cut into three segments. The segmented annular seal ring is pressed on the seal runner with a coil spring and maintains high purge-pressure of GHe as a barrier gas. Wear of the carbon seal ring is reduced by using the shrouded Rayleigh step lift-pads to increase the opening force within the seal clearance. As the rubbing speed increases, the opening force in the Rayleigh step increases, so that the rubbing speed is increased by

Relationship between the purge pressure and the leakage rate of GHe purge seal was evaluated at a steady speed of 20,000 rpm [22]. When the purge pressure is low, the seal face is kept to be non-contact because the Rayleigh step increases the seal opening force. As the purge pressure is set to be high, the seal face condition is changed from the non-contact state to the contact state, it resulted that the dynamic leakage almost equals that of the resting state. Furthermore, for the GHe purge seal combined with the LO2 floating-ring seal, the environ‐ mental temperature around the GHe purge seal was equal to that of LO2 leakage, so that the carbon seal ring showed severe wear with an appearance of worn-out of the Rayleigh step.

> Change of the friction and wear of the carbon pin as a function of the pin temperature was determined in the cryogenic GHe environment [23]. Friction test was conducted against the Cr-plated steel disk at a sliding speed of 12 m/s and load of 9.8 N. When the pin temperature is below the solidification temperature of CO2 (216 K), it is noted that lubricating property of the carbon pin suddenly disappeared and friction and wear became intensive. When absorbed CO2 gas was changed to be solid phase, lubricity of carbon was lost. This phenomenon resembles that when phase of moisture is transfer to solid phase (ice) below 273 K, lubricity decreases; be well known. From this fact, it seemed that severe wear of the GHe purge seal was generated because the environmental temperature around the seal was lower than 216 K. Spray MoS2 coating on the carbon seal face was drastically able to prohibit progression of wear of the carbon seal ring at low temperature, as shown in Fig. 20.

> After a total operating time of 29 minutes for the engine firing test, the GHe purge seal used in the LE-7 indicated that the seal surfaces coated by MoS2 were found to be in excellent condition and wear depth of the carbon seal ring was about 7 µm. It assumes that high opening force produced by the Rayleigh step was kept by prohibit of wear of the Rayleigh step and the GHe purge seal was operated under conditions of nearly no load on the seal surfaces due to balance between the opening and closing forces.

#### **6. Advanced bearings and shaft seals**

Future space transport systems require reusable launch vehicles to reduce launch cost and to increase efficiency. The durability of reusable turbopump bearings must be greater than that of currently available (expendable) turbopumps. For the improved high-pressure LO2 turbopump of the SSME that reduced serious wear of the all-steel bearing, the hybrid ceramic bearing with Si3N4 balls was developed and accomplished the required life of 7.5 hours. In this case, to improve self-lubrication of the abrasive retainer made of glass cloth-reinforced PTFE, a new type of the retainer that had PTFE/bronze-powder insert fitted on the ball pocket was developed [7].

loads in LO2 [28]. These bearings used the glass cloth-reinforced PTFE retainer which was

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

139

chemically treated with HF to improve self-lubrication.

**Figure 21.** Advanced bearing having single-guided retainer with elliptical ball pocket

**Figure 22.** Bearing torque of single-guided bearings and double guided bearing to 50,000 rpm in LH2

It is noted that, at high speeds, the hybrid ceramic bearing that consists of hard, light weight ceramic balls as well as steel rings shows a lower centrifugal force on the ceramic ball. The centrifugal force of the Si3N4 ball makes about 60 % lighter than that of the 440C steel ball. This leads to a reduction of bearing load and a smaller contact area with a lower spinning speed, resulting in a low level of heat generation due to ball spin. Additionally, good tribological combinations of the ceramic balls against the steel rings result in a decrease in bearing wear and in instances of seizure, even under insufficient lubricating conditions. Thus, the hybrid ceramic bearing enables higher speed operation rather than the all-steel bearing.

On the other hand, advanced rocket engines that are characterized by high performance (light weight) and high durability (long life) are required today. Ultra-high speed turbopump having a rotational speed level of 100,000 rpm needs to make engine smaller and lighter. Hybrid ceramic bearing is suitable to ultra-high speed turbopump because of lower centrifugal force. In recent years, these advanced research and development on the hybrid ceramic bearing are actively underway.

#### **6.1. Single–guided bearing [27, 28, 39]**

In order to increase the durability of self-lubricated bearing, it is apparent that sufficient cooling and restriction of the frictional heat generation in the bearing are essential. Its notifi‐ cation is experimentally identified by a series of studies on the turbopump bearing. In order to improve internal coolant flow through the bearing and to reduce bearing frictional torque, a new type of bearing having a single-guided retainer was developed. Figure 21 shows the 25 mm-bore bearing having a single-guided retainer with elliptical ball pockets [39]. The singleguided retainer is guided only by one side of the outer-ring bore (land) to reduce land friction and to increase the cooling ability within the bearing. However, reduce of retainer guiding is apt to generate unstable wobbling at high speed, so that the elliptical ball pockets with narrow axial clearance is needed to reduce wobbling of the retainer. For the elliptical ball pocket of the single-guided retainer, its circumferential clearance of 1.3 mm was twice as large as that of the conventional circular pocket to reduce ball-to-pocket interaction under high BSV. Further‐ more, the axial clearance of 0.1 mm was narrow to stabilize wobbling of the single-guided retainer at high speeds.

Self-lubricating performance, bearing wear and transfer film of two-types of the single-guided bearing, i.e., a hybrid ceramic bearing with Si3N4 and all-steel bearing, was evaluated under high thrust loads at speeds up to 50,000 rpm in LH2, LO2 and LN2 [27,39]. Furthermore, to evaluate the durability of the single-guided bearing for long-life bearing, the all steel bearing was tested for total operation times up to 11.7 hours at a speed of 50,000 rpm with high thrust loads in LO2 [28]. These bearings used the glass cloth-reinforced PTFE retainer which was chemically treated with HF to improve self-lubrication.

**Figure 21.** Advanced bearing having single-guided retainer with elliptical ball pocket

of currently available (expendable) turbopumps. For the improved high-pressure LO2 turbopump of the SSME that reduced serious wear of the all-steel bearing, the hybrid ceramic bearing with Si3N4 balls was developed and accomplished the required life of 7.5 hours. In this case, to improve self-lubrication of the abrasive retainer made of glass cloth-reinforced PTFE, a new type of the retainer that had PTFE/bronze-powder insert fitted on the ball pocket was

It is noted that, at high speeds, the hybrid ceramic bearing that consists of hard, light weight ceramic balls as well as steel rings shows a lower centrifugal force on the ceramic ball. The centrifugal force of the Si3N4 ball makes about 60 % lighter than that of the 440C steel ball. This leads to a reduction of bearing load and a smaller contact area with a lower spinning speed, resulting in a low level of heat generation due to ball spin. Additionally, good tribological combinations of the ceramic balls against the steel rings result in a decrease in bearing wear and in instances of seizure, even under insufficient lubricating conditions. Thus, the hybrid

On the other hand, advanced rocket engines that are characterized by high performance (light weight) and high durability (long life) are required today. Ultra-high speed turbopump having a rotational speed level of 100,000 rpm needs to make engine smaller and lighter. Hybrid ceramic bearing is suitable to ultra-high speed turbopump because of lower centrifugal force. In recent years, these advanced research and development on the hybrid ceramic bearing are

In order to increase the durability of self-lubricated bearing, it is apparent that sufficient cooling and restriction of the frictional heat generation in the bearing are essential. Its notifi‐ cation is experimentally identified by a series of studies on the turbopump bearing. In order to improve internal coolant flow through the bearing and to reduce bearing frictional torque, a new type of bearing having a single-guided retainer was developed. Figure 21 shows the 25 mm-bore bearing having a single-guided retainer with elliptical ball pockets [39]. The singleguided retainer is guided only by one side of the outer-ring bore (land) to reduce land friction and to increase the cooling ability within the bearing. However, reduce of retainer guiding is apt to generate unstable wobbling at high speed, so that the elliptical ball pockets with narrow axial clearance is needed to reduce wobbling of the retainer. For the elliptical ball pocket of the single-guided retainer, its circumferential clearance of 1.3 mm was twice as large as that of the conventional circular pocket to reduce ball-to-pocket interaction under high BSV. Further‐ more, the axial clearance of 0.1 mm was narrow to stabilize wobbling of the single-guided

Self-lubricating performance, bearing wear and transfer film of two-types of the single-guided bearing, i.e., a hybrid ceramic bearing with Si3N4 and all-steel bearing, was evaluated under high thrust loads at speeds up to 50,000 rpm in LH2, LO2 and LN2 [27,39]. Furthermore, to evaluate the durability of the single-guided bearing for long-life bearing, the all steel bearing was tested for total operation times up to 11.7 hours at a speed of 50,000 rpm with high thrust

ceramic bearing enables higher speed operation rather than the all-steel bearing.

developed [7].

138 Tribology - Fundamentals and Advancements

actively underway.

retainer at high speeds.

**6.1. Single–guided bearing [27, 28, 39]**

**Figure 22.** Bearing torque of single-guided bearings and double guided bearing to 50,000 rpm in LH2

#### *6.1.1. Self–lubricating performance and transfer film [27,39]*

#### **a.** *In LH2*

Figure 22 shows the bearing torque of the single-guided bearings (hybrid ceramic bearing and all-steel bearing) and the conventional double-guided bearing at speeds to 50,000 rpm in LH2 [39]. It was observed that the bearing torque of the single-guided bearing effectively decreased to about one-half of that of the double-guided bearing. Its result identified that bearing torque induced by high-speed sliding of the outer land guide of the retainer almost accounted for an overall bearing torque generated at high speeds. In addition, the hybrid ceramic bearing showed lower bearing torque than the all-steel bearing at high speeds.

Figure 23 shows the XPS depth analysis of a Si3N4 ball taken from the hybrid ceramic bearing tested in LH2 [27]. Its etching depth was 120 nm (SiO2 rate). It was found that a considerably thick transfer film consisting of CaF2/FeF2 was formed on the ceramic balls. CaF2 and FeF2 seemed to be tribo-chemically formed by the reducing power of LH2. The considerably thick transfer film of CaF2 and FeF2 led to exhibit high load capacity. For the all-steel bearing tested in LH2, a thick CaF2 film was formed beneath an extremely thin PTFE overlay, but its thickness

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

141

In LO2, the hybrid ceramic bearing exhibited poor self-lubricating performance even at a low speed of 10,000 rpm. To the contrary, the all-steel bearing indicated excellent load capacity accompanied by a stable bearing and enabled to sustain a thrust load of 2,650 N (*Smax*, 2.7

For the hybrid ceramic bearing, an extremely thin, weakly adhesive PTFE film was formed on ceramic balls and resulted in a poor load capacity of the bearing. For the all-steel bearing, the intense formation of a Cr2O3 film was beneath an extremely thin PTFE film. It is noted that the tribo-chemical formation of Cr2O3 film due to high oxidation power of LO2 could exhibit high

The hybrid ceramic bearing exhibited better load capacity than that of the all-steel bearing in LN2. The hybrid ceramic bearing enabled to sustain a thrust load of 2,700 N (*Smax*, 3.1 GPa) at a speed of 50,000 rpm without seizure. To the contrary, the all-steel bearing showed unstable change of bearing torque and seized at a relatively light-thrust load of 1,470 N (*Smax*, 2.2 GPa)

For the hybrid ceramic bearing, the thick transfer film consisting of FeF2/iron oxide formed on the ceramic balls. To the contrary, the seized all-steel bearing was lubricated by only thin PTFE transfer film, without the tribo-chemical formation of CaF2/FeF2/Cr2O3 films because of its inert environment of LN2. This fact was determined by that the all-steel bearing once tested in LH2 or LO2, whose bearing formed the CaF2/FeF2/Cr2O3 films, showed stable change of the bearing torque without seizure even under high thrust loads above 1,470 N in LN2 [27].

The single-guided all steel bearing was tested for a total operation time to 11.7 hours at a speed of 50,000 rpm with high thrust loads to 2,400 N in LO2. During long-run test, one-hour operation at a speed of 50,000 rpm was repeated nine times. The test bearing was effectively cooled by the jet-cooling with using nozzles. During the long-run test, the bearing exhibited stable variation of the bearing torque in a range of 93-95 N-mm [28]. The bearing exhibited excellent self-lubrication performance that there was no abnormal change of the bearing torque

From the examination of the bearing tested for the long-run test in LO2, it was observed that sound surface conditions with hardly any wear were determined. The XPS depth

of CaF2 transfer film was thinner than that of the hybrid ceramic bearing.

GPa) at a speed of 50,000 rpm without seizure in LO2 [39].

resistance to metal-to-metal adhesion leading to seizure [27].

**b.** *In LO2*

**c.** *In LN2*

at a speed of 50,000 rpm [39].

*6.1.2. Long–life bearing [28]*

and bearing temperature.

Critical load capacity of the single-guided bearing without a significant rise of the bearing torque and bearing temperature was evaluated. For the single-guided hybrid ceramic bearing tested in LH2, the critical thrust load was 1,960 N (*Smax* of inner race, 2.7 GPa) at 50,000 rpm and was two times higher than that of the double-guided all-steel bearing. Furthermore, even when bear‐ ing torque increased with a rise of bearing temperature, the hybrid ceramic bearing was able to sustain a thrust load of 2,840 N (*Smax*, 3.2 GPa) at 50,000 rpm without seizure in LH2. High critical load capacity of the single-guided hybrid ceramic bearing was demonstrated [39].

**Figure 23.** XPS depth analysis of Si3N4 ball of hybrid ceramic bearing tested in LH2

Figure 23 shows the XPS depth analysis of a Si3N4 ball taken from the hybrid ceramic bearing tested in LH2 [27]. Its etching depth was 120 nm (SiO2 rate). It was found that a considerably thick transfer film consisting of CaF2/FeF2 was formed on the ceramic balls. CaF2 and FeF2 seemed to be tribo-chemically formed by the reducing power of LH2. The considerably thick transfer film of CaF2 and FeF2 led to exhibit high load capacity. For the all-steel bearing tested in LH2, a thick CaF2 film was formed beneath an extremely thin PTFE overlay, but its thickness of CaF2 transfer film was thinner than that of the hybrid ceramic bearing.

**b.** *In LO2*

*6.1.1. Self–lubricating performance and transfer film [27,39]*

Figure 22 shows the bearing torque of the single-guided bearings (hybrid ceramic bearing and all-steel bearing) and the conventional double-guided bearing at speeds to 50,000 rpm in LH2 [39]. It was observed that the bearing torque of the single-guided bearing effectively decreased to about one-half of that of the double-guided bearing. Its result identified that bearing torque induced by high-speed sliding of the outer land guide of the retainer almost accounted for an overall bearing torque generated at high speeds. In addition, the hybrid ceramic bearing showed lower bearing torque than the all-steel bearing at high speeds.

Critical load capacity of the single-guided bearing without a significant rise of the bearing torque and bearing temperature was evaluated. For the single-guided hybrid ceramic bearing tested in LH2, the critical thrust load was 1,960 N (*Smax* of inner race, 2.7 GPa) at 50,000 rpm and was two times higher than that of the double-guided all-steel bearing. Furthermore, even when bear‐ ing torque increased with a rise of bearing temperature, the hybrid ceramic bearing was able to sustain a thrust load of 2,840 N (*Smax*, 3.2 GPa) at 50,000 rpm without seizure in LH2. High critical

load capacity of the single-guided hybrid ceramic bearing was demonstrated [39].

**Figure 23.** XPS depth analysis of Si3N4 ball of hybrid ceramic bearing tested in LH2

**a.** *In LH2*

140 Tribology - Fundamentals and Advancements

In LO2, the hybrid ceramic bearing exhibited poor self-lubricating performance even at a low speed of 10,000 rpm. To the contrary, the all-steel bearing indicated excellent load capacity accompanied by a stable bearing and enabled to sustain a thrust load of 2,650 N (*Smax*, 2.7 GPa) at a speed of 50,000 rpm without seizure in LO2 [39].

For the hybrid ceramic bearing, an extremely thin, weakly adhesive PTFE film was formed on ceramic balls and resulted in a poor load capacity of the bearing. For the all-steel bearing, the intense formation of a Cr2O3 film was beneath an extremely thin PTFE film. It is noted that the tribo-chemical formation of Cr2O3 film due to high oxidation power of LO2 could exhibit high resistance to metal-to-metal adhesion leading to seizure [27].

**c.** *In LN2*

The hybrid ceramic bearing exhibited better load capacity than that of the all-steel bearing in LN2. The hybrid ceramic bearing enabled to sustain a thrust load of 2,700 N (*Smax*, 3.1 GPa) at a speed of 50,000 rpm without seizure. To the contrary, the all-steel bearing showed unstable change of bearing torque and seized at a relatively light-thrust load of 1,470 N (*Smax*, 2.2 GPa) at a speed of 50,000 rpm [39].

For the hybrid ceramic bearing, the thick transfer film consisting of FeF2/iron oxide formed on the ceramic balls. To the contrary, the seized all-steel bearing was lubricated by only thin PTFE transfer film, without the tribo-chemical formation of CaF2/FeF2/Cr2O3 films because of its inert environment of LN2. This fact was determined by that the all-steel bearing once tested in LH2 or LO2, whose bearing formed the CaF2/FeF2/Cr2O3 films, showed stable change of the bearing torque without seizure even under high thrust loads above 1,470 N in LN2 [27].

#### *6.1.2. Long–life bearing [28]*

The single-guided all steel bearing was tested for a total operation time to 11.7 hours at a speed of 50,000 rpm with high thrust loads to 2,400 N in LO2. During long-run test, one-hour operation at a speed of 50,000 rpm was repeated nine times. The test bearing was effectively cooled by the jet-cooling with using nozzles. During the long-run test, the bearing exhibited stable variation of the bearing torque in a range of 93-95 N-mm [28]. The bearing exhibited excellent self-lubrication performance that there was no abnormal change of the bearing torque and bearing temperature.

From the examination of the bearing tested for the long-run test in LO2, it was observed that sound surface conditions with hardly any wear were determined. The XPS depth

tion in hot pure F2 gas. The fluorine-passivated bearings coated with FeF2 film was tested by long run for 11.7 hours at a speed of 50,000 rpm under high thrust loads in LH2, LO2 and LN2. The fluorine-passivated bearings showed excellent self-lubrication in both LH2 and LN2 [28].

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

143

In a reduce environment of LH2, even under poor cooling conditions controlled by reducing of the coolant flow, the fluorine-passivated bearing exhibited superior durability for a total test time to 4.4 hours, as compared with signs of seizure for the untreated bearing. The XPS analysis of the transfer film indicated that the fluorine-passivated bearing was tribo-chemically

In an inert environment of LN2, the fluorine-passivated bearing showed excellent selflubrication and wear conditions for the long-run test up to 11.7 hours at a speed of 50,000 rpm. Stable change of the bearing torque (75-80 N-mm) was shown for the passivated bearing during the long-run test in LN2 [28]. The bearing test was repeated seven times at a speed of 50,000 rpm and a thrust load of 2,600 N in LN2. From the examination of the fluorine-passivated bearing tested in LN2, sound surface conditions with hardly any wear were determined. It was found that a thick CaF2 film was tribo-chemically formed on thick FeF2/Cr2O3 films of the bearing. On the other hand, the untreated bearing was seized at a low thrust load of 1,470 N due to less tribo-chemical reaction in LN2, as mentioned before. In such inert environment in LN2, there was less formation of CaF2/FeF2/Cr2O3 films, so that poor self-lubrication and load

To the contrary, in an oxide environment of LO2, the fluorine-passivated bearing indicated a higher bearing torque with greater unstable change than that of the untreated bearing [28]. The bearing tests were repeated seven times of the bearing test at a speed of 50,000 rpm and a thrust load of 2,450 N in LO2. Its total test time was 11.7 hours. During long-run test, high bearing torque continued to vary erratically with the variation in a range of 75-120 N-mm. The fluorine-passivated bearing tested in LO2 showed somewhat high wear. To the contrary, the untreated bearing demonstrated excellent self-lubrication with hardly any wear during the long-run test as mentioned before. It was clearly showed that the FeF2 film in LO2 made a

Inspection of the fluorine-passivated bearing tested in LO2 indicated that the initial coated film of FeF2 was worm away. Its result also indicated that oxide power of LO2 restricted the tribochemical formation of FeF2 film. Such reduction in self-lubrication possibly resulted from that the coated FeF2 film restricted the tribo-chemical formation of Cr2O3 film in LO2, resulting in an increase of metal-to-metal adhesion. These results indicated that excellent lubrication depended on the tribo-chemical formation of CaF2/FeF2 films in LH2 or Cr2O3 film in LO2, respectively. In order to obtain high self-lubrication and durability of the bearing, it is noted that tribo-chemical reaction is necessary at the frictional interfaces within the bearing [4].

Based on previous bearing tests at high speeds up to 50,000 rpm, the hybrid ceramic bearing (25-mm bore) was tested at ultra-high-speeds up to 120,000 rpm, and results were compared with the all-steel bearing in LH2. At a ultra-high speed of 120,000 rpm, the inner-race growth

lubricated by a thick CaF2 film overlaid on a thick FeF2/Cr2O3 films.

capacity of the bearing were shown.

typical reduction in self-lubrication.

**6.3. Ultra–high–speed hybrid ceramic bearing [8,40]**

**Figure 24.** XPS depth analysis of SUS440C ball tested for long run in LO2 and new ball

analysis of a ball taken from the tested bearing is shown in Fig. 24 [28]. Its etching depth was 30 nm (SiO2 rate). It is noted that the intense formation of a Cr2O3 film was detected and its thickness was thicker than that of the native Cr2O3 film on the new ball. Under sufficient cooling conditions in LO2, the thick Cr2O3 film formed by tribo-chemical reaction could provide an extremely high resistance to metal-to-metal adhesion beneath an extremely thin CaF2 film. To the contrary, under poor cooling conditions in LO2, the intense formation of oxide film (Fe2O3) was mainly produced and led to large mild wear, as discussed in the LO2 turbopump bearing. Furthermore, the formation of Fe2O3 might reduce adhesion of PTFE transfer film, resulting in less lubricant within the bearing. The results indicated that thick formation of a Cr2O3 film due to tribo-chemical reaction in LO2 is important to reduce the bearing wear. Its effect needs sufficient cooling with jet within the bearing components to eliminate the formation of Fe2O3 [28].

#### **6.2. Fluorine–passivated bearing [28]**

It is experimentally found that the FeF2 film formed by a tribo-chemical reaction between the F of PTFE and Fe of 440C steel was facilitated by the high reduction power of LH2 and enhanced to reduce the bearing wear in LH2. This may suggest that the FeF2 film has a good solidlubricant performance to improve the tribological performance of the bearing. Effect of the coated FeF2 film on the self-lubrication and durability of the all-steel bearing was evaluated. An FeF2 film was chemically formed by means of a passivating surface treatment of fluorida‐ tion in hot pure F2 gas. The fluorine-passivated bearings coated with FeF2 film was tested by long run for 11.7 hours at a speed of 50,000 rpm under high thrust loads in LH2, LO2 and LN2. The fluorine-passivated bearings showed excellent self-lubrication in both LH2 and LN2 [28].

In a reduce environment of LH2, even under poor cooling conditions controlled by reducing of the coolant flow, the fluorine-passivated bearing exhibited superior durability for a total test time to 4.4 hours, as compared with signs of seizure for the untreated bearing. The XPS analysis of the transfer film indicated that the fluorine-passivated bearing was tribo-chemically lubricated by a thick CaF2 film overlaid on a thick FeF2/Cr2O3 films.

In an inert environment of LN2, the fluorine-passivated bearing showed excellent selflubrication and wear conditions for the long-run test up to 11.7 hours at a speed of 50,000 rpm. Stable change of the bearing torque (75-80 N-mm) was shown for the passivated bearing during the long-run test in LN2 [28]. The bearing test was repeated seven times at a speed of 50,000 rpm and a thrust load of 2,600 N in LN2. From the examination of the fluorine-passivated bearing tested in LN2, sound surface conditions with hardly any wear were determined. It was found that a thick CaF2 film was tribo-chemically formed on thick FeF2/Cr2O3 films of the bearing. On the other hand, the untreated bearing was seized at a low thrust load of 1,470 N due to less tribo-chemical reaction in LN2, as mentioned before. In such inert environment in LN2, there was less formation of CaF2/FeF2/Cr2O3 films, so that poor self-lubrication and load capacity of the bearing were shown.

To the contrary, in an oxide environment of LO2, the fluorine-passivated bearing indicated a higher bearing torque with greater unstable change than that of the untreated bearing [28]. The bearing tests were repeated seven times of the bearing test at a speed of 50,000 rpm and a thrust load of 2,450 N in LO2. Its total test time was 11.7 hours. During long-run test, high bearing torque continued to vary erratically with the variation in a range of 75-120 N-mm. The fluorine-passivated bearing tested in LO2 showed somewhat high wear. To the contrary, the untreated bearing demonstrated excellent self-lubrication with hardly any wear during the long-run test as mentioned before. It was clearly showed that the FeF2 film in LO2 made a typical reduction in self-lubrication.

Inspection of the fluorine-passivated bearing tested in LO2 indicated that the initial coated film of FeF2 was worm away. Its result also indicated that oxide power of LO2 restricted the tribochemical formation of FeF2 film. Such reduction in self-lubrication possibly resulted from that the coated FeF2 film restricted the tribo-chemical formation of Cr2O3 film in LO2, resulting in an increase of metal-to-metal adhesion. These results indicated that excellent lubrication depended on the tribo-chemical formation of CaF2/FeF2 films in LH2 or Cr2O3 film in LO2, respectively. In order to obtain high self-lubrication and durability of the bearing, it is noted that tribo-chemical reaction is necessary at the frictional interfaces within the bearing [4].

#### **6.3. Ultra–high–speed hybrid ceramic bearing [8,40]**

analysis of a ball taken from the tested bearing is shown in Fig. 24 [28]. Its etching depth was 30 nm (SiO2 rate). It is noted that the intense formation of a Cr2O3 film was detected and its thickness was thicker than that of the native Cr2O3 film on the new ball. Under sufficient cooling conditions in LO2, the thick Cr2O3 film formed by tribo-chemical reaction could provide an extremely high resistance to metal-to-metal adhesion beneath an extremely thin CaF2 film. To the contrary, under poor cooling conditions in LO2, the intense formation of oxide film (Fe2O3) was mainly produced and led to large mild wear, as discussed in the LO2 turbopump bearing. Furthermore, the formation of Fe2O3 might reduce adhesion of PTFE transfer film, resulting in less lubricant within the bearing. The results indicated that thick formation of a Cr2O3 film due to tribo-chemical reaction in LO2 is important to reduce the bearing wear. Its effect needs sufficient cooling with jet within the

It is experimentally found that the FeF2 film formed by a tribo-chemical reaction between the F of PTFE and Fe of 440C steel was facilitated by the high reduction power of LH2 and enhanced to reduce the bearing wear in LH2. This may suggest that the FeF2 film has a good solidlubricant performance to improve the tribological performance of the bearing. Effect of the coated FeF2 film on the self-lubrication and durability of the all-steel bearing was evaluated. An FeF2 film was chemically formed by means of a passivating surface treatment of fluorida‐

bearing components to eliminate the formation of Fe2O3 [28].

**Figure 24.** XPS depth analysis of SUS440C ball tested for long run in LO2 and new ball

**6.2. Fluorine–passivated bearing [28]**

142 Tribology - Fundamentals and Advancements

Based on previous bearing tests at high speeds up to 50,000 rpm, the hybrid ceramic bearing (25-mm bore) was tested at ultra-high-speeds up to 120,000 rpm, and results were compared with the all-steel bearing in LH2. At a ultra-high speed of 120,000 rpm, the inner-race growth of 34µm due to centrifugal force results in a reduction of the radial clearance within the bearing. Table 4 summarizes comparison of the bearing load and speed conditions for the hybrid ceramic bearing and all-steel bearing at a speed of 120,000 rpm with a thrust load of 980 N [8]. At 120,000 rpm, the initial radial clearance of 77 µm was decreased to 43µm. For the hybrid ceramic bearing, the maximum contact stress *Smax* at the inner race is apt to increase rather than that of the all-steel bearing due to a high elastic modulus. However, the maximum spinning velocity *Vmax* is reduced and resulted in a lower *SVmax* value that leads to a reduction of the bearing temperature and spin wear. The maximum contact stress at the outer race becomes higher due to centrifugal force. For sliding conditions of the retainer, the sliding velocity at the outer land and ball pocket reaches to a high level of 110 m/s and the frictional heat generation of the retainer is to be severe. For the cooling system to remove the bearing heat generation at 120,000 rpm, effective jet cooling with nozzles needs to obtain sufficient coolant flow within the bearing. The nozzles were directed to cool the single outer land-guiding side of the retainer where high frictional heat is generated.


The power loss around the bearing was estimated based on the heat absorbed by the cooling flow [8]. Figure 26 shows the power loss of the hybrid ceramic and all-steel bearings as a function of rotational speed up to 120,000 rpm in LH2 under different cooling conditions at a thrust load of 980 N. It was found that the power loss of the bearing significantly increased above 80,000 rpm with increasing cooling flow rate. At 120,000 rpm, the power loss of the bearing that contained the viscous power loss of 2.2 kW at the shaft side was estimated. The power loss was 6.0 kW for the hybrid ceramic bearing and 6.4 kW for the all-steel bearing, respectively. There was not typical difference of the power loss of the bearing because viscous power loss within the bearing almost accounted for an overall power loss generated at ultrahigh speeds. It seems that the power loss around the bearing was mainly induced by viscous

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

145

**Figure 25.** Change of bearing temperature of hybrid ceramic and all-steel bearings at 120,000 rpm with 2,160 N

The components of the hybrid ceramic bearing were in excellent condition with regard to wear at a speed of 120,000 rpm with a thrust load of 3,140 N in LH2 [40]. On the contrary, the seized all-steel bearing exhibited severe adhesive wear. It was found that the ceramic balls formed superficial micro-cracks on the contact track. Superficial micro-cracks visually extended in a mesh-like pattern on the Si3N4 ball tested. It was shown that network of hair crack was propagated along wide-ditch crack. A marked feature of these superficial micro-cracks was that they were very shallow to about 3 µm at maximum and did not extend deeply into the ball. From detailed observation with a scanning electron microscope (SEM), such wide-ditch cracks seemed to be formed by removal of fragments fractured due to contact stress repeated by the rolling balls as shown in Fig. 27. Thus, when the Si3N4 balls had lower mechanical strength and fracture toughness, it was clear that wide-ditch cracks were apt to be formed.

drag and churning of the cooling flow passing through the bearing.

**Table 4.** Bearing load and speed conditions for hybrid ceramic and all-steel bearings at 120,000 rpm with 980 N (25 mm bore)

Figure 25 shows the change of the bearing temperature at a steady speed of 120,000 rpm with a thrust load of 2,160 N [8]. The hybrid ceramic bearing showed excellent performance with a stable condition of the bearing temperature, compared to the seized all-steel bearing showing an irregular change of high bearing temperature. When the thrust load was increased to 3,140 N, the hybrid bearing showed slight damage with a spiky rise of the bearing temperature. It was found that the critical load capacity *Smax* without seizure at a speed of 120,000 rpm was reached to 3.0 GPa (at a thrust load of 2,160 N) for the hybrid ceramic bearing and 2.0 GPa (980 N) for the all-steel bearing, respectively.

of 34µm due to centrifugal force results in a reduction of the radial clearance within the bearing. Table 4 summarizes comparison of the bearing load and speed conditions for the hybrid ceramic bearing and all-steel bearing at a speed of 120,000 rpm with a thrust load of 980 N [8]. At 120,000 rpm, the initial radial clearance of 77 µm was decreased to 43µm. For the hybrid ceramic bearing, the maximum contact stress *Smax* at the inner race is apt to increase rather than that of the all-steel bearing due to a high elastic modulus. However, the maximum spinning velocity *Vmax* is reduced and resulted in a lower *SVmax* value that leads to a reduction of the bearing temperature and spin wear. The maximum contact stress at the outer race becomes higher due to centrifugal force. For sliding conditions of the retainer, the sliding velocity at the outer land and ball pocket reaches to a high level of 110 m/s and the frictional heat generation of the retainer is to be severe. For the cooling system to remove the bearing heat generation at 120,000 rpm, effective jet cooling with nozzles needs to obtain sufficient coolant flow within the bearing. The nozzles were directed to cool the single outer land-guiding

**Parameters Hybrid ceramic bearing All-steel bearing**

Centrifugal force on ball [N] 454 1,120

**Table 4.** Bearing load and speed conditions for hybrid ceramic and all-steel bearings at 120,000 rpm with 980 N (25-

Figure 25 shows the change of the bearing temperature at a steady speed of 120,000 rpm with a thrust load of 2,160 N [8]. The hybrid ceramic bearing showed excellent performance with a stable condition of the bearing temperature, compared to the seized all-steel bearing showing an irregular change of high bearing temperature. When the thrust load was increased to 3,140 N, the hybrid bearing showed slight damage with a spiky rise of the bearing temperature. It was found that the critical load capacity *Smax* without seizure at a speed of 120,000 rpm was reached to 3.0 GPa (at a thrust load of 2,160 N) for the hybrid ceramic bearing and 2.0 GPa (980

2.31 / 2.14 2.00 / 2.35

5.8 7.5

108 116

side of the retainer where high frictional heat is generated.

*Bearing* Rotational speed [rpm] Thrust load [N] Initial contact angle [deg.] Initial radial clearance [μm] Operational radial clearance [μm]

144 Tribology - Fundamentals and Advancements

Maximum contact stress at inner/outer races (*Smax*) [GPa]

> Maximum spinning velocity at inner race (*Vmax*) [m/s]

*Retainer* Sliding velocity at outer land [m/s] Sliding velocity at ball pocket [m/s]

N) for the all-steel bearing, respectively.

mm bore)

**Figure 25.** Change of bearing temperature of hybrid ceramic and all-steel bearings at 120,000 rpm with 2,160 N

The power loss around the bearing was estimated based on the heat absorbed by the cooling flow [8]. Figure 26 shows the power loss of the hybrid ceramic and all-steel bearings as a function of rotational speed up to 120,000 rpm in LH2 under different cooling conditions at a thrust load of 980 N. It was found that the power loss of the bearing significantly increased above 80,000 rpm with increasing cooling flow rate. At 120,000 rpm, the power loss of the bearing that contained the viscous power loss of 2.2 kW at the shaft side was estimated. The power loss was 6.0 kW for the hybrid ceramic bearing and 6.4 kW for the all-steel bearing, respectively. There was not typical difference of the power loss of the bearing because viscous power loss within the bearing almost accounted for an overall power loss generated at ultrahigh speeds. It seems that the power loss around the bearing was mainly induced by viscous drag and churning of the cooling flow passing through the bearing.

The components of the hybrid ceramic bearing were in excellent condition with regard to wear at a speed of 120,000 rpm with a thrust load of 3,140 N in LH2 [40]. On the contrary, the seized all-steel bearing exhibited severe adhesive wear. It was found that the ceramic balls formed superficial micro-cracks on the contact track. Superficial micro-cracks visually extended in a mesh-like pattern on the Si3N4 ball tested. It was shown that network of hair crack was propagated along wide-ditch crack. A marked feature of these superficial micro-cracks was that they were very shallow to about 3 µm at maximum and did not extend deeply into the ball. From detailed observation with a scanning electron microscope (SEM), such wide-ditch cracks seemed to be formed by removal of fragments fractured due to contact stress repeated by the rolling balls as shown in Fig. 27. Thus, when the Si3N4 balls had lower mechanical strength and fracture toughness, it was clear that wide-ditch cracks were apt to be formed.

0.01mm (a) Hair crack (b) Wide-ditch crack

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

147

The floating-ring seal due to noncontact-type is suitable for high-pressure turbopumps; however, conventional seals using carbon seal-rings were weak under high speed and high pressure conditions. Since metal seal-rings have higher mechanical strength and durability, advanced floating-ring seal (with one-seal and two-seal rings) that used Ag-plated metal sealrings with a seal diameter of 30 mm [8]. This metal seal was studied at ultra-high speeds up to 120,000 rpm in LH2. Calculated runner growth due to centrifugal force at 120,000 rpm was 29µm, so that the initial seal clearance (gap) was decreased as the rotational speed increased. The test seal had an Ag-plated seal ring made of Inconel 718 that was the same material used for the runner. The runner was coated with a Cr2O3 plasma spray, and this coating exhibited excellent friction and wear without adhesion to Ag in LN2. In order to obtain smooth radial movement of the seal ring, the static seal surface of the housing was coated with a sprayed

Figure 28 shows the seal performance of the one-ring seal *vs.* the two-ring seal up to a speed of 100,000 rpm in LH2 [8]. These seals had a straight bore with a seal gap of 110-120 µm. Figure 29 also shows the phase change models of leakage flow within the seal gap [4]. Seal perform‐ ance depended on the two-phase flow (gas/liquid phase) of leakage, because the vaporization of leakage was generated by the viscous friction heat and by the seal pressure drop. At lower speeds, the leakage of the one-ring seal was relatively greater than that of the two-ring seal; however, with increasing speed, the leakage of the one-ring seal was drastically decreased and approached the same level of the two-ring seal due to enlargement of the two-phase flow.

Removing of fragments Wide-ditch crack

Hair crack

**Figure 27.** Process model of wide-ditch crack formation on Si3N4 ball

**6.4. Ultra–high–speed two–phase seal [8]**

MoS2 film.

**Figure 26.** Power loss of hybrid ceramic and all-steel bearings as a function of rotational speed up to 120,000 rpm in LH2

An advanced study was conducted to select a tough Si3N4 ball capable of restraining crack propagation as well as to evaluate the efficient bearing cooling with nozzles. A Si3N4 ball having higher thermal-shock resistance, as well as higher fracture toughness, was found to reduce the propagation of superficial micro-cracks, resulting in a decrease of ball wear. Furthermore, it was observed that the cooling ability of the LH2 jet-flow aimed at the retainer was superior to that aimed at the inner raceway, further reducing the propagation of thermal micro-cracks on the Si3N4 balls. This result also indicated that micro-cracks on the balls were possibly generated at the trace contacting the outer raceway due to a higher centrifugal force under insufficient cooling conditions. Furthermore, under the same cooling rate, the four nozzles achieved a higher cooling ability than the two nozzles with increasing jet speed above 208 m/s. The jet-speed of nozzles reached to the twice of the sliding speed of 108 m/s at the retainer outer-land [40].

In order to prevent the propagation of superficial thermal micro-cracks on the balls, the outer race contact stress was reduced by decreasing the outer race curvature to a limited value of 0.51. Furthermore, sufficient cooling at the outer raceway was gained by a proper clearance of the outer land of the retainer. Decreasing the maximum outer-race stress to 2.0 GPa (thrust load, 1,960 N) in conjunction with sufficient cooling through a narrow outer land clearance could prevent the propagation of superficial micro-cracks even under insufficient cooling conditions [40].

**Figure 27.** Process model of wide-ditch crack formation on Si3N4 ball

#### **6.4. Ultra–high–speed two–phase seal [8]**

An advanced study was conducted to select a tough Si3N4 ball capable of restraining crack propagation as well as to evaluate the efficient bearing cooling with nozzles. A Si3N4 ball having higher thermal-shock resistance, as well as higher fracture toughness, was found to reduce the propagation of superficial micro-cracks, resulting in a decrease of ball wear. Furthermore, it was observed that the cooling ability of the LH2 jet-flow aimed at the retainer was superior to that aimed at the inner raceway, further reducing the propagation of thermal micro-cracks on the Si3N4 balls. This result also indicated that micro-cracks on the balls were possibly generated at the trace contacting the outer raceway due to a higher centrifugal force under insufficient cooling conditions. Furthermore, under the same cooling rate, the four nozzles achieved a higher cooling ability than the two nozzles with increasing jet speed above 208 m/s. The jet-speed of nozzles reached to the twice of the sliding speed of 108 m/s at the

**Figure 26.** Power loss of hybrid ceramic and all-steel bearings as a function of rotational speed up to 120,000 rpm in

In order to prevent the propagation of superficial thermal micro-cracks on the balls, the outer race contact stress was reduced by decreasing the outer race curvature to a limited value of 0.51. Furthermore, sufficient cooling at the outer raceway was gained by a proper clearance of the outer land of the retainer. Decreasing the maximum outer-race stress to 2.0 GPa (thrust load, 1,960 N) in conjunction with sufficient cooling through a narrow outer land clearance could prevent the propagation of superficial micro-cracks even under insufficient cooling

retainer outer-land [40].

146 Tribology - Fundamentals and Advancements

LH2

conditions [40].

The floating-ring seal due to noncontact-type is suitable for high-pressure turbopumps; however, conventional seals using carbon seal-rings were weak under high speed and high pressure conditions. Since metal seal-rings have higher mechanical strength and durability, advanced floating-ring seal (with one-seal and two-seal rings) that used Ag-plated metal sealrings with a seal diameter of 30 mm [8]. This metal seal was studied at ultra-high speeds up to 120,000 rpm in LH2. Calculated runner growth due to centrifugal force at 120,000 rpm was 29µm, so that the initial seal clearance (gap) was decreased as the rotational speed increased. The test seal had an Ag-plated seal ring made of Inconel 718 that was the same material used for the runner. The runner was coated with a Cr2O3 plasma spray, and this coating exhibited excellent friction and wear without adhesion to Ag in LN2. In order to obtain smooth radial movement of the seal ring, the static seal surface of the housing was coated with a sprayed MoS2 film.

Figure 28 shows the seal performance of the one-ring seal *vs.* the two-ring seal up to a speed of 100,000 rpm in LH2 [8]. These seals had a straight bore with a seal gap of 110-120 µm. Figure 29 also shows the phase change models of leakage flow within the seal gap [4]. Seal perform‐ ance depended on the two-phase flow (gas/liquid phase) of leakage, because the vaporization of leakage was generated by the viscous friction heat and by the seal pressure drop. At lower speeds, the leakage of the one-ring seal was relatively greater than that of the two-ring seal; however, with increasing speed, the leakage of the one-ring seal was drastically decreased and approached the same level of the two-ring seal due to enlargement of the two-phase flow.

For the two-ring seal, the two-phase flow was fully enlarged within the secondary seal ring that was at the downstream of the primary seal ring. Seal leakage was reduced within limits; however, the hydrodynamic force of the liquid phase flow that sustained the seal ring was lost and resulted in seal-ring seizure at a relatively lower speed of 98,700 rpm. Also, shaft vibration for the two-ring seal was likely produced by wobbling of the seal ring under severe rubbing conditions and abruptly increased at speeds of more than 92,000 rpm before resulting in sealring seizure at a speed of 98,700 rpm. Furthermore, in the two-ring seal with a seal gap of 70-80 µm, the primary seal-ring seized a speed of 108,600 rpm, because the hydrostatic force decreased due to a low differential pressure.

**Figure 29.** Phase change models of leakage flow within seal gap at ultra-high speed

of two-phase flow within the sealing clearance.

**7. Concluding remarks**

In contrast, the one-ring seal successfully functioned with no abnormal signs of seizure during tests, because the liquid-phase flow remained within a seal clearance even though the twophase flow increased. As a result, the hydrodynamic force in the liquid-phase flow as well as the hydrostatic force due to high differential pressure possibly helped to prevent seal-ring seizure. At a steady speed of 120,000 rpm, the one-ring seal exhibited a stable leakage in a range of 0.21-0.24 liters/s that is similar to leakage in the two-ring seal as shown in Fig. 28. Thus, the one-ring seal was superior to the two-ring seal, preventing seal-ring seizure due to an increase

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

149

For built-up of safe space transport system to achieve high reliability, cryogenic high-speed bearing and shaft seal used in the rocket turbopumps are reviewed historically. These tribocomponents have specific lubrication, materials and design requirements in pumping cryo‐ genic liquid propellants in rocket engines. Nowadays, as earth scale issues of energy conservation and environment preservation, a breakaway from the conventional fossil-fuel society becomes a big problem. Clean hydrogen energy is attractive due to its energy efficiency and its smaller impact on the environment, and it is expected to be a key technology in the 21st century. It is proposed that, to build hydrogen infrastructure for LH2 storage and distribution, development of an industrial tribo-system with long durability and high reliability is essential

and advances by supporting of cryogenic tribology studied for LH2 rocket system.

**Figure 28.** Seal performance of one-ring seal *vs.* two-ring seal up to 100,000 rpm

**Figure 29.** Phase change models of leakage flow within seal gap at ultra-high speed

In contrast, the one-ring seal successfully functioned with no abnormal signs of seizure during tests, because the liquid-phase flow remained within a seal clearance even though the twophase flow increased. As a result, the hydrodynamic force in the liquid-phase flow as well as the hydrostatic force due to high differential pressure possibly helped to prevent seal-ring seizure. At a steady speed of 120,000 rpm, the one-ring seal exhibited a stable leakage in a range of 0.21-0.24 liters/s that is similar to leakage in the two-ring seal as shown in Fig. 28. Thus, the one-ring seal was superior to the two-ring seal, preventing seal-ring seizure due to an increase of two-phase flow within the sealing clearance.

#### **7. Concluding remarks**

For the two-ring seal, the two-phase flow was fully enlarged within the secondary seal ring that was at the downstream of the primary seal ring. Seal leakage was reduced within limits; however, the hydrodynamic force of the liquid phase flow that sustained the seal ring was lost and resulted in seal-ring seizure at a relatively lower speed of 98,700 rpm. Also, shaft vibration for the two-ring seal was likely produced by wobbling of the seal ring under severe rubbing conditions and abruptly increased at speeds of more than 92,000 rpm before resulting in sealring seizure at a speed of 98,700 rpm. Furthermore, in the two-ring seal with a seal gap of 70-80 µm, the primary seal-ring seized a speed of 108,600 rpm, because the hydrostatic force

decreased due to a low differential pressure.

148 Tribology - Fundamentals and Advancements

**Figure 28.** Seal performance of one-ring seal *vs.* two-ring seal up to 100,000 rpm

For built-up of safe space transport system to achieve high reliability, cryogenic high-speed bearing and shaft seal used in the rocket turbopumps are reviewed historically. These tribocomponents have specific lubrication, materials and design requirements in pumping cryo‐ genic liquid propellants in rocket engines. Nowadays, as earth scale issues of energy conservation and environment preservation, a breakaway from the conventional fossil-fuel society becomes a big problem. Clean hydrogen energy is attractive due to its energy efficiency and its smaller impact on the environment, and it is expected to be a key technology in the 21st century. It is proposed that, to build hydrogen infrastructure for LH2 storage and distribution, development of an industrial tribo-system with long durability and high reliability is essential and advances by supporting of cryogenic tribology studied for LH2 rocket system.

#### **Acknowledgements**

This paper is based on previous cryogenic tribology studies carried out by Japan Aerospace Exploration Agency (JAXA) at Kakuda Space Center. These studies were also supported by IHI Corporation for turbopumps, by NTN corporation for bearings and by Eagle Industry Co., LTD. for shaft seals, respectively. The author is indebted to researchers engaged for their valuable support, to organizations for their enthusiastic cooperation. At last, the author has to thank late Prof. Miyakawa, Y. of Hhosei University, as a pioneer in space tribology in Japan, for his guidance to cryogenic tribology with profound appreciation.

[10] Collongeat L, Edeline E, Frocot M & Dehouve J. Development status of high DN LH2

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

151

[11] Rachuk V & Titkov N. The First Russian LOX-LH2 Expander Cycle LRV: RD0146,

[12] Nosaka M, Oike M, Kamijo K, Kikuchi M & Katsuta H. Experimental Study on Lubri‐ cating Performance of Self-Lubricating Ball Bearings for Liquid Hydrogen Turbo‐

[13] Nosaka M, Oike M, Kikuchi M, Kamijo K & Tajiri M. Tribo-Characteristics of Self-Lu‐ bricating Ball Bearings for the LE-7 Liquid Hydrogen Rocket-Turbopump, *Trib.*

[14] Nosaka M, Oike M, Kikuchi M, Nagao R & Mayumi T. Evaluation of Durability for Cryogenic High-Speed Ball Bearings for LE-7 Rocket Turbopumps, *Lubr. Eng.*, 52, 3,

[15] Nosaka M, Oike M, Kikuchi M, Kamijo K & Tajiri M. Self-Lubricating Performance and Durability of Ball Bearings for the LE-7 Liquid Oxygen Rocket-Turbopump,

[16] Nosaka M, Oike M & Kikuchi M. Tribology at Low and High Temperatures, Lubrica‐ tion in Rocket-Turbopumps, *J. of Japan Society of Lubrication Engineers*, 33, 2, (1988),

[17] Nosaka M. Self-Lubricating Performance of High-Speed Ball Bearing for Liquid Hy‐ drogen (1), Design Problems, *J. of Japan Society of Lubrication Engineers*, 32, 10, (1987),

[18] Nosaka M. Self-Lubricating Performance of High-Speed Ball Bearing for Liquid Hy‐ drogen (2), Self-Lubricating Performance Improvements. *J. of Japan Society of Lubrica‐*

[19] Winn L W, Eusepi M W & Smalley A J. Small, High-Speed Bearing Technology for

[21] Nosaka M & Oike M. Rotating-Shaft Seals in Rocket-Turbopumps, *J. of Japanese Soci‐*

[22] Oike M, Nosaka M, Watanabe Y, Kikuchi M & Kamijo K. Experimental Study on High-Pressure Gas Seals for a Liquid Oxygen Turbopump, *STLE Trans.*, 31, 1, (1988),

[23] Nosaka M, Oike M & Kikuchi M. Cryogenic Tribology of Turbopumps for Rockets,

[20] Edmond E B & William J A. Advanced Bearing Technology, NASA SP-38, 1965.

bearings in Snecma, AIAA-2005-3950, (2005).

pump, *Lubr. Eng.*, 44, 1, (1988), 30-44 .

AIAA-2006-4904, (2006).

*Trans.*, 36, 3, (1993), 432-442.

*Lubr. Eng.*, 49, 9, (1993), 677-688.

*tion Engineers*, 32, 12, (1987), 833-838, in Japanese.

Cryogenic Turbo-Pumps, NASA CR-134615, 1974.

*ety of Tribologists*, 35, 4, (1990), 233-238, in Japanese.

*Cryogenic Engineering*, 31, 10, (1996), 500-511, in Japanese.

(1996), 221-233.

90-96, in Japanese.

689-695, in Japanese.

91-97.

#### **Author details**

Masataka Nosaka and Takahisa Kato

Department of Mechanical Engineering, University of Tokyo, Tokyo, Japan

#### **References**


[10] Collongeat L, Edeline E, Frocot M & Dehouve J. Development status of high DN LH2 bearings in Snecma, AIAA-2005-3950, (2005).

**Acknowledgements**

150 Tribology - Fundamentals and Advancements

**Author details**

**References**

Masataka Nosaka and Takahisa Kato

This paper is based on previous cryogenic tribology studies carried out by Japan Aerospace Exploration Agency (JAXA) at Kakuda Space Center. These studies were also supported by IHI Corporation for turbopumps, by NTN corporation for bearings and by Eagle Industry Co., LTD. for shaft seals, respectively. The author is indebted to researchers engaged for their valuable support, to organizations for their enthusiastic cooperation. At last, the author has to thank late Prof. Miyakawa, Y. of Hhosei University, as a pioneer in space tribology in Japan,

[1] Dieter K H & David H H. Modern Engineering for Design of Liquid-Propellant Rock‐ et Engines, *Progress in Astronautics and Aeronautics, Vol. 147*, AIAA, (1992), 155-218.

[4] Nosaka, M. Cryogenic Tribology of High-Speed Bearings and Shaft Seals in Liquid

[5] Nosaka M, Takada S & Yoshida M. Research and Development of Cryogenic Tribolo‐ gy of Turbopumps for Rocket Engines, *J. of Aeronautical and Space Science Japan*, 58,

[6] Hale J R & Klatt F T. SSME Improvement for Routine Shuttle Operations,

[7] Gibson H. Lubriction of Space Shuttle Main Engine Turbopump Bearings, *Lubr. Eng.*

[8] Nosaka M, Takada S, Kikuchi M, Sudo T & Yoshida M. Ultra-High-Speed Perform‐ ance of Ball Bearings and Annular Seals in Liquid Hydrogen at Up to 3 Million DN

[9] Ohta T, Kimoto K, Kawai T, Motomura T, Russ M & Paulus T. Design, Fabrication

and Test of the RL60 Fuel Turbopump, AIAA-2003-5073, (2003).

[3] Liquid Rocket Engine Turbopump Rotating- Shaft Seals, NASA SP-8121, 1978.

for his guidance to cryogenic tribology with profound appreciation.

Department of Mechanical Engineering, University of Tokyo, Tokyo, Japan

[2] Liquid Rocket Engine Turbopump Bearing, NASA SP-8048, 1971.

Hydrogen, *Tribology Online*, 6, 2, (2011), 133-141.

(120,000 rpm), *Trib. Trans.*, 47, (2004), 43-53.

681, (2010), 303-313, in Japanese.

AIAA-85-1163, (1985).

57, 8, (2001), 10-12.


[24] Nosaka M. Tribological Burn-Out of Wear, *J. of Japanese Society of Tribologists*, 36, 9, (1991), 689-691, in Japanese.

[36] Oike M, Nosaka M, Kikuchi M & Hasegawa S. Two-Phase Flow in Floating-Ring

Cryogenic Tribology in High–Speed Bearings and Shaft Seals of Rocket Turbopumps

http://dx.doi.org/10.5772/55733

153

[37] Oike M, Nosaka M, Kikuchi M & Watanabe Y. Performance of a Segmented Circum‐ ferential Seal for a Liquid Oxygen Turbopump (Part 1), Sealing Performance, *J. of Jap‐*

[38] Oike M, Nosaka M, Kikuchi M & Watanabe Y. Performance of a Segmented Circum‐ ferential Seal for a Liquid Oxygen Turbopump (Part 2), Durability, *J. of Japanese Soci‐*

[39] Nosaka M, Kikuchi M, Oike M & Kawai N. Tribo-Characteristics of Cryogenic Hy‐ brid Ceramic Ball Bearings for Rocket Turbopumps: Self-Lubricating Performance,

[40] Nosaka M, Takada S, Yoshida M, Kikuchi M, Sudo T & Nakamura S. Improvement of Durability of Hybrid Ceramic Ball Bearings in Liquid Hydrogen at 3 Million DN

Seals for Cryogenic Turbopumps, *Tribo. Trans.*, 42, 2, (1999), 273-281.

*anese Society of Tribologists*, 37, 4, (1992), 339-346, in Japanese.

*ety of Tribologists*, 37, 5, (1992), 389-396, in Japanese.

(120,000 rpm), *Tribology Online*, 5, 1, (2010), 60-70.

*Trib. Trans.*, 40, 1, (1997), 21-30.


[36] Oike M, Nosaka M, Kikuchi M & Hasegawa S. Two-Phase Flow in Floating-Ring Seals for Cryogenic Turbopumps, *Tribo. Trans.*, 42, 2, (1999), 273-281.

[24] Nosaka M. Tribological Burn-Out of Wear, *J. of Japanese Society of Tribologists*, 36, 9,

[25] Nosaka M. Tribology in Low Temperature Environment, *J. of Japanese Society of Tri‐*

[26] Nosaka M, Takada S, Yoshida M, Kikuchi M, Sudo T & Nakamura S. Effect of Tilted Misalignment of Tribo-Characteristics of High-Speed Ball Bearings in Liquid Hydro‐

[27] Nosaka M, Kikuchi M, Oike M & Kawai N. Tribo-Characteristics of Cryogenic Hy‐ brid Ceramic Ball Bearings for Rocket Turbopumps: Bearing Wear and Transfer Film,

[28] Nosaka M, Kikuchi M, Kawai N & Kikuyama H. Effect of Iron Fluoride Layer on Du‐ rability of Cryogenic High-Speed Ball Bearings for Rocket Turbopumps, *Trib. Trans.*,

[29] Suzuki M, Nosaka M, Kamijo K & Kikuchi M. Research and Development of a Rotat‐ ing- Shaft Seals for a Liquid Hydrogen Turbopump, *Lubr. Eng.*, 42, 3, (1986), 162-169.

[30] Nosaka M, Miyakawa Y, Kamijo K, Suzuki M & Kikuchi M. Study on Sealing Charac‐ teristics of High Speed, Contacting Mechanical Seals for Liquid Hydrogen (Part 1), Development of Mechanical Seal for Liquid Hydrogen Turbopump, *J. of Japan Society*

[31] Nosaka M, Kamijo K, Suzuki M, Kikuchi M & Miyakawa Y. Study on Sealing Charac‐ teristics of High Speed, Contacting Mechanical Seals for Liquid Hydrogen (Part 2), Starting Torque and Static Sealing Performance, *J. of Japan Society of Lubrication Engi‐*

[32] Nosaka M, Kamijo K, Suzuki M, Kikuchi M & Miyakawa Y. Study on Sealing Charac‐ teristics of High Speed, Contacting Mechanical Seals for Liquid Hydrogen (Part 3), Friction Power Loss and Dynamic Sealing Performance, *J. of Japan Society of Lubrica‐*

[33] Nosaka M, Kamijo K, Suzuki M, Kikuchi M & Miyakawa Y. Study on Sealing Charac‐ teristics of High Speed, Contacting Mechanical Seals for Liquid Hydrogen (Part 4), Characteristics of Running Process and Wear of Rubbing Seal Faces, *J. of Japan Society*

[34] Nosaka M, Kamijo K, Suzuki M, Kikuchi M & Miyakawa Y. Study on Sealing Charac‐ teristics of High Speed, Contacting Mechanical Seals for Liquid Hydrogen (Part 5), The Formation of Thermal Crack and Wear in Chromium Plate on Rotating Ring, *J. of*

[35] Oike M, Nosaka M, Kikuchi M & Watanabe Y. Performance of A Shaft Seal System for The LE-7 Rocket Engine Oxidizer Turbopump, *Proc. of The 18th Inter. Symposium on*

*Japan Society of Lubrication Engineers*, 29, 3, (1984), 187-194, in Japanese.

*of Lubrication Engineers*, 29, 1, (1984), 35-42, in Japanese.

(1991), 689-691, in Japanese.

152 Tribology - Fundamentals and Advancements

*bologists*, 52, 11, (2007), 759-764, in Japanese.

gen, *Tribology Online*, 5, 2, (2010), 71-79.

*Trib. Trans.*, 42, 1, (1999), 106-115.

*neers*, 29, 1, (1984), 43-49, in Japanese.

*tion Engineers*, 29, 2, (1984), 113-120, in Japanese.

*Space Tech. and Sci., Kagoshima*, (1992), 143-154.

*of Lubrication Engineers*, 29, 2, (1984), 121-128, in Japanese.

43, 2, (2000), 163-174.


**Chapter 5**

**Titanium and Titanium Alloys as Biomaterials**

Bone and its several associated elements – cartilage, connective tissue, vascular elements and nervous components – act as a functional organ. They provide support and protection for soft tissues and act together with skeletal muscles to make body movements possible. Bones are relatively rigid structures and their shapes are closely related to their functions. Bone metab‐ olism is mainly controlled by the endocrine, immune and neurovascular systems, and its metabolism and response to internal and external stimulations are still under assessment [1].

Long bones of the skeletal system are prone to injury, and internal or external fixation is a part of their treatment. Joint replacement is another major intervention where the bone is expected to host biomaterials. Response of the bone to biomaterial intervenes with the regeneration process. Materials implanted into the bone will, nevertheless, cause local and systemic biological responses even if they are known to be inert. Host responses with joint replacement

The field of biomaterials is on a continuous increase due to the high demand of an aging population as well as the increasing average weight of people. Biomaterials are artificial or natural materials that are used to restore or replace the loss or failure of a biological structure to recover its form and function in order to improve the quality and longevity of human life. Biomaterials are used in different parts of the human body as artificial valves in the heart, stents in blood vessels, replacement implants in shoulders, knees, hips, elbows, ears and dental structures [3] [4] [5]. They are also employed as cardiac simulators and for urinary and digestive tract reconstructions. Among all of them, the highest number of implants is for spinal, hip and knee replacements. It is estimated that by the end of 2030, the number of total hip replacements will rise by 174% (572,000 procedures) and total knee arthroplasties are projected to grow by 673% from the present rate (3.48 million procedures) [6]. This is due to the fact that human joints suffer from degenerative diseases such as osteoarthritis (inflammation in the

> © 2013 Sáenz de Viteri and Fuentes; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2013 Sáenz de Viteri and Fuentes; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,

distribution, and reproduction in any medium, provided the original work is properly cited.

and fixation materials will initiate an adaptive and reactive process [2].

Virginia Sáenz de Viteri and Elena Fuentes

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/55860

**1. Introduction**

### **Titanium and Titanium Alloys as Biomaterials**

Virginia Sáenz de Viteri and Elena Fuentes

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/55860

#### **1. Introduction**

Bone and its several associated elements – cartilage, connective tissue, vascular elements and nervous components – act as a functional organ. They provide support and protection for soft tissues and act together with skeletal muscles to make body movements possible. Bones are relatively rigid structures and their shapes are closely related to their functions. Bone metab‐ olism is mainly controlled by the endocrine, immune and neurovascular systems, and its metabolism and response to internal and external stimulations are still under assessment [1].

Long bones of the skeletal system are prone to injury, and internal or external fixation is a part of their treatment. Joint replacement is another major intervention where the bone is expected to host biomaterials. Response of the bone to biomaterial intervenes with the regeneration process. Materials implanted into the bone will, nevertheless, cause local and systemic biological responses even if they are known to be inert. Host responses with joint replacement and fixation materials will initiate an adaptive and reactive process [2].

The field of biomaterials is on a continuous increase due to the high demand of an aging population as well as the increasing average weight of people. Biomaterials are artificial or natural materials that are used to restore or replace the loss or failure of a biological structure to recover its form and function in order to improve the quality and longevity of human life. Biomaterials are used in different parts of the human body as artificial valves in the heart, stents in blood vessels, replacement implants in shoulders, knees, hips, elbows, ears and dental structures [3] [4] [5]. They are also employed as cardiac simulators and for urinary and digestive tract reconstructions. Among all of them, the highest number of implants is for spinal, hip and knee replacements. It is estimated that by the end of 2030, the number of total hip replacements will rise by 174% (572,000 procedures) and total knee arthroplasties are projected to grow by 673% from the present rate (3.48 million procedures) [6]. This is due to the fact that human joints suffer from degenerative diseases such as osteoarthritis (inflammation in the

© 2013 Sáenz de Viteri and Fuentes; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Sáenz de Viteri and Fuentes; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

bone joints), osteoporosis (weakening of the bones) and trauma leading to pain or loss in function. The degenerative diseases lead to degradation of the mechanical properties of the bone due to excessive loading or absence of normal biological self-healing process. Artificial biomaterials are the solutions to these problems and the surgical implantation of these artificial biomaterials of suitable shapes help restore the function of the otherwise functionally com‐ promised structures. However, not only the replacement surgeries have increased, simulta‐ neously the revision surgery of hip and knee implants have also increased. These revision surgeries which cause pain for the patient are very expensive and also their success rate is rather small. The target of present researches is developing implants that can serve for much longer period or until lifetime without failure or revision surgery [7]. Thus, development of appropriate material with high longevity, superior corrosion resistance in body environment, excellent combination of high strength and low Young´s modulus, high fatigue and wear resistance, high ductibility, excellent biocompatibility and be without citotoxicity is highly essential [8] [9].

patibility of a material [11]. The two main factors that influence the biocompatibility of a material are the host response induced by the material and the materials degradation in the body environment (Figure 1). According to the tissue reaction phenomena, the biocompati‐ bility of orthopedic implant materials was classified into three categories by Heimke [12], such as "biotolerant", showing distant osteogenesis (bone formation with indirect contact to the material); "bioinert", showing contact osteogenesis (bone formation with direct contact to the material), and "bioactive", showing bonding osteogenesis (bone formation with chemical or

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

157

When implants are exposed to human tissues and fluids, several reactions take place between the host and the implant material and these reactions dictate the acceptability of these materials by our system. The issues with regard to biocompatibility are (1) thrombosis, which involves blood coagulation and adhesion of blood platelets to biomaterial surface, and (2) the fibrous

The most important mechanical properties that help to decide the type of material are hardness, tensile strength, Young´s modulus and elongation. An implant fracture due to a mechanical failure is related to a biomechanical incompatibility. For this reason, it is expected that the

tissue encapsulation of biomaterials that are implanted in soft tissues.

biological bonding to the material).

**Figure 1.** Biological effects of a biomaterial

**2.2. Mechanical properties**

In general, metallic biomaterials are used for load bearing applications and must have sufficient fatigue strength to endure the rigors of daily activity. Ceramic biomaterials are generally used for their hardness and wear resistance for applications such as articulating surfaces in joints and in teeth as well as bone bonding surfaces in implants. Polymeric materials are usually used for their flexibility and stability, but have also been used for low friction articulating surfaces. Titanium is becoming one of the most promising engineering materials and the interest in the application of titanium alloys to mechanical and tribological components is growing rapidly in the biomedical field [10], due to their excellent properties.

This chapter is focused on the use of titanium and its alloys as biomaterials from a tribological point of view. The main limitation of these materials is their poor tribological behavior characterized by high friction coefficient and severe adhesive wear. A number of different surface modification techniques have been recently applied to titanium alloys in order to improve their tribological performance as well as osseointegration. This chapter includes the most recent developments carried out in the field of surface treatments on titanium with very promising results.

#### **2. Biomaterial properties**

The main property required of a biomaterial is that it does not illicit an adverse reaction when placed into services, that means to be a biocompatible material. As well, good mechanical properties, osseointegration, high corrosion resistance and excellent wear resistance are required.

#### **2.1. Biocompatibility**

The materials used as implants are expected to be highly non toxic and should not cause any inflammatory or allergic reactions in the human body. The success of the biomaterials is mainly dependent on the reaction of the human body to the implant, and this measures the biocom‐ patibility of a material [11]. The two main factors that influence the biocompatibility of a material are the host response induced by the material and the materials degradation in the body environment (Figure 1). According to the tissue reaction phenomena, the biocompati‐ bility of orthopedic implant materials was classified into three categories by Heimke [12], such as "biotolerant", showing distant osteogenesis (bone formation with indirect contact to the material); "bioinert", showing contact osteogenesis (bone formation with direct contact to the material), and "bioactive", showing bonding osteogenesis (bone formation with chemical or biological bonding to the material).

**Figure 1.** Biological effects of a biomaterial

bone joints), osteoporosis (weakening of the bones) and trauma leading to pain or loss in function. The degenerative diseases lead to degradation of the mechanical properties of the bone due to excessive loading or absence of normal biological self-healing process. Artificial biomaterials are the solutions to these problems and the surgical implantation of these artificial biomaterials of suitable shapes help restore the function of the otherwise functionally com‐ promised structures. However, not only the replacement surgeries have increased, simulta‐ neously the revision surgery of hip and knee implants have also increased. These revision surgeries which cause pain for the patient are very expensive and also their success rate is rather small. The target of present researches is developing implants that can serve for much longer period or until lifetime without failure or revision surgery [7]. Thus, development of appropriate material with high longevity, superior corrosion resistance in body environment, excellent combination of high strength and low Young´s modulus, high fatigue and wear resistance, high ductibility, excellent biocompatibility and be without citotoxicity is highly

In general, metallic biomaterials are used for load bearing applications and must have sufficient fatigue strength to endure the rigors of daily activity. Ceramic biomaterials are generally used for their hardness and wear resistance for applications such as articulating surfaces in joints and in teeth as well as bone bonding surfaces in implants. Polymeric materials are usually used for their flexibility and stability, but have also been used for low friction articulating surfaces. Titanium is becoming one of the most promising engineering materials and the interest in the application of titanium alloys to mechanical and tribological components

This chapter is focused on the use of titanium and its alloys as biomaterials from a tribological point of view. The main limitation of these materials is their poor tribological behavior characterized by high friction coefficient and severe adhesive wear. A number of different surface modification techniques have been recently applied to titanium alloys in order to improve their tribological performance as well as osseointegration. This chapter includes the most recent developments carried out in the field of surface treatments on titanium with very

The main property required of a biomaterial is that it does not illicit an adverse reaction when placed into services, that means to be a biocompatible material. As well, good mechanical properties, osseointegration, high corrosion resistance and excellent wear resistance are

The materials used as implants are expected to be highly non toxic and should not cause any inflammatory or allergic reactions in the human body. The success of the biomaterials is mainly dependent on the reaction of the human body to the implant, and this measures the biocom‐

is growing rapidly in the biomedical field [10], due to their excellent properties.

essential [8] [9].

156 Tribology - Fundamentals and Advancements

promising results.

required.

**2.1. Biocompatibility**

**2. Biomaterial properties**

When implants are exposed to human tissues and fluids, several reactions take place between the host and the implant material and these reactions dictate the acceptability of these materials by our system. The issues with regard to biocompatibility are (1) thrombosis, which involves blood coagulation and adhesion of blood platelets to biomaterial surface, and (2) the fibrous tissue encapsulation of biomaterials that are implanted in soft tissues.

#### **2.2. Mechanical properties**

The most important mechanical properties that help to decide the type of material are hardness, tensile strength, Young´s modulus and elongation. An implant fracture due to a mechanical failure is related to a biomechanical incompatibility. For this reason, it is expected that the material employed to replace the bone has similar mechanical properties to that of bone. The bone Young´s modulus varies in a range of 4 to 30 GPa depending on the type of the bone and the direction of measurement [13] [14].

**1.** Reaction of the implant with a foreign body as debris from implant component degrada‐

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

159

**3.** Imposition of abnormal or unphysiological conditions on the bone, such as fluid pressures

**4.** Alteration to the mechanical signals encouraging bone densification; strain reductions or

All metallic implants electrochemically corrode to some extent. This is disadvantageous for two main reasons: (1) the process of degradation reduces the structural integrity and (2) degradation products may react unfavorably with the host. Metallic implant degradation results from both electrochemical dissolution and wear, but most frequently occurs through a synergistic combination of the two [23] [24]. Electrochemical corrosion process includes both generalized dissolution uniformly affecting the entire surface and localized areas of

Metal implant corrosion is controlled by (1) the extent of the thermodynamic driving forces which cause corrosion (oxidation/reduction reactions) and (2) physical barriers which limit the kinetics of corrosion. In practice these two parameters that mediate the corrosion of orthopedic biomaterials can be broken down into a number of variables: geometric variables (e.g., taper geometry in modular component hip prostheses), metallurgical variables (e.g., surface microstructure, oxide structure and composition), mechanical variables (e.g., stress and/or relative

The corrosion resistance of a surgically implanted alloy is an essential characteristic since the metal alloys are in contact with a very aggressive media such as the body fluid due to the presence of chloride ions and proteins. In the corrosion process, the metallic components of the alloy are oxidized to their ionic forms and dissolved oxygen is reduced to hydroxide ions.

The corrosion characteristics of an alloy are greatly influenced by the passive film formed on

Wear always occurs in the articulation of artificial joints as a result of the mixed lubrication regime. The movement of an artificial hip joint produces billions of microscopic particles that are rubbed off cutting motions. These particles are trapped inside the tissues of the joint capsule and may lead to unwanted foreign body reactions. Histocytes and giant cells phagocytose and "digest" the released particles and form granulomas or granuloma-like tissues. At the boundary layer between the implant and bone, these interfere with the transformation process of the bone leading to osteolysis. Hence, the materials used to make the femoral head and cup play a significant role in the device performance. Since the advent of endoprosthetics, attempts

motion) and solution variables (e.g., pH, solution proteins and enzymes) [25].

the surface of the alloy and the presence of the alloying elements.

tion or wear, or to toxic emissions from the implant [22]

or motion against implant components

**2.4. High corrosion resistance**

a component.

**2.5. Wear resistance**

stress-shielding of replaced or adjacent bone.

**2.** Damage or lesion to the bone through mechanical trauma surgery

#### **2.3. Osseointegration**

The inability of an implant surface to integrate with the adjacent bone and other tissues due to micromotions, results in implant loosening [15]. Osseointegration (capacity for joining with bone and other tissue) is another important aspect of the use of metallic alloys in bone applications (Figure 2). A good integration of implant with the bone is essential to ensure the safety and efficacy of the implant over its useful life. It has been shown in previous studies [16], that enhancement of the bone response to implant surfaces can be achieved by increasing the roughness or by other surface treatments [17]. Although the precise molecular mechanisms are not well understood, it is clear that the chemical and physical properties of the surface play a major role in the implant – surface interactions through modulation of cell behavior, growth factor production and osteogenic gene expression [18] [19] [20].

**Figure 2.** Schematic drawing of the principles of osseointegration [21]

Furthermore, it is known that even if initial implant stability is achieved, the bone may retreat from or be isolated from the implant because of different reasons or situations listed below:


#### **2.4. High corrosion resistance**

material employed to replace the bone has similar mechanical properties to that of bone. The bone Young´s modulus varies in a range of 4 to 30 GPa depending on the type of the bone and

The inability of an implant surface to integrate with the adjacent bone and other tissues due to micromotions, results in implant loosening [15]. Osseointegration (capacity for joining with bone and other tissue) is another important aspect of the use of metallic alloys in bone applications (Figure 2). A good integration of implant with the bone is essential to ensure the safety and efficacy of the implant over its useful life. It has been shown in previous studies [16], that enhancement of the bone response to implant surfaces can be achieved by increasing the roughness or by other surface treatments [17]. Although the precise molecular mechanisms are not well understood, it is clear that the chemical and physical properties of the surface play a major role in the implant – surface interactions through modulation of cell behavior, growth

the direction of measurement [13] [14].

158 Tribology - Fundamentals and Advancements

factor production and osteogenic gene expression [18] [19] [20].

**Figure 2.** Schematic drawing of the principles of osseointegration [21]

Furthermore, it is known that even if initial implant stability is achieved, the bone may retreat from or be isolated from the implant because of different reasons or situations listed below:

**2.3. Osseointegration**

All metallic implants electrochemically corrode to some extent. This is disadvantageous for two main reasons: (1) the process of degradation reduces the structural integrity and (2) degradation products may react unfavorably with the host. Metallic implant degradation results from both electrochemical dissolution and wear, but most frequently occurs through a synergistic combination of the two [23] [24]. Electrochemical corrosion process includes both generalized dissolution uniformly affecting the entire surface and localized areas of a component.

Metal implant corrosion is controlled by (1) the extent of the thermodynamic driving forces which cause corrosion (oxidation/reduction reactions) and (2) physical barriers which limit the kinetics of corrosion. In practice these two parameters that mediate the corrosion of orthopedic biomaterials can be broken down into a number of variables: geometric variables (e.g., taper geometry in modular component hip prostheses), metallurgical variables (e.g., surface microstructure, oxide structure and composition), mechanical variables (e.g., stress and/or relative motion) and solution variables (e.g., pH, solution proteins and enzymes) [25].

The corrosion resistance of a surgically implanted alloy is an essential characteristic since the metal alloys are in contact with a very aggressive media such as the body fluid due to the presence of chloride ions and proteins. In the corrosion process, the metallic components of the alloy are oxidized to their ionic forms and dissolved oxygen is reduced to hydroxide ions.

The corrosion characteristics of an alloy are greatly influenced by the passive film formed on the surface of the alloy and the presence of the alloying elements.

#### **2.5. Wear resistance**

Wear always occurs in the articulation of artificial joints as a result of the mixed lubrication regime. The movement of an artificial hip joint produces billions of microscopic particles that are rubbed off cutting motions. These particles are trapped inside the tissues of the joint capsule and may lead to unwanted foreign body reactions. Histocytes and giant cells phagocytose and "digest" the released particles and form granulomas or granuloma-like tissues. At the boundary layer between the implant and bone, these interfere with the transformation process of the bone leading to osteolysis. Hence, the materials used to make the femoral head and cup play a significant role in the device performance. Since the advent of endoprosthetics, attempts have been made to reduce wear by using a variety of different combinations of materials and surface treatments.

In cases where good mechanical characteristics are required as in hip implants, knee implants, bone screws, and plates, Ti-6Al-4V alloy is being used [27] [28]. One of the most common applications of titanium alloys is artificial hip joints that consist of an articulating bearing (femoral head and cup) and stem [24], where metallic cup and hip stem components are made of titanium. As well, they are also often used in knee joint replacements, which consist of a femoral and tibial component made of titanium and a polyethylene articulating surface.

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

161

**Figure 3.** Schematic diagram of artificial hip joint (left) and knee implant [29] (right)

the implant and hence the implant has to be replaced by a new one.

The fundamental drawback of titanium and its alloys which limits wider use of these materials include their poor fretting fatigue resistance and poor tribological properties [30] [31], because of its low hardness [32]. Their poor tribological behavior is characterized by high coefficient of friction, severe adhesive wear with a strong tendency to seizing and low abrasion resistance [33]. Titanium tends to undergo severe wear when it is rubbed between itself or between other materials. Titanium has tendency for moving or sliding parts to gall and eventually seize. This causes a more intensive wear as a result of creation of adhesion couplings and mechanical instability of passive layer of oxides, particularly in presence of third bodies (Figure 4). Owing to this effect, in cases of total joint replacements made of titanium head and polymer cup, the 10%-20% of joints needs to be replaced within 15-20 years and the aseptic loosening accounts for approximately 80% of the revisions [34]. The reason for the failure of the implants is due to the high friction coefficient of these materials that can lead to the release of wear debris from the implant into the bloodstream that results in an inflammation of the surrounding tissue and gives rise to the bone resorption (osteolysis) [35] [36], which ultimately leads to loosening of

**3.1. Wear problems in titanium and titanium alloys**

Nowadays, the materials used for biomedical applications are mainly metallic materials such as 316L stainless steel, cobalt chromium alloys (CoCrMo), titanium-based alloys (Ti-6Al-4V) and miscellaneous others (including tantalum, gold, dental amalgams and other "specialty" metals). Titanium alloys are fast emerging as the first choice for majority of applications due to the combination of their outstanding characteristics such as high strength, low density, high immunity to corrosion, complete inertness to body environment, enhanced compatibility, low Young´s modulus and high capacity to join with bone or other tissues. Their lower Young´s modulus, superior biocompatibility and better corrosion resistance in comparison with conventional stainless steels and cobalt-based alloys, make them an ideal choice for bioapplications [26]. Because of the mentioned desirable properties, titanium and titanium alloys are widely used as hard tissue replacements in artificial bones, joints and dental implants.

#### **3. Titanium and titanium alloys**

The elemental metal titanium was first discovered in England by William Gregor in 1790, but in 1795 Klaproth gave it the name of titanium. Combination of low density, high strength to weight ratio, good biocompatibility and improved corrosion resistance with good plasticity and mechanical properties determines the application of titanium and its alloys in such industries as aviation, automotive, power and shipbuilding industries or architecture as well as medicine and sports equipment.

Increased use of titanium and its alloys as biomaterials comes from their superior biocompat‐ ibility and excellent corrosion resistance because of the thin surface oxide layer, and good mechanical properties, as a certain elastic modulus and low density that make that these metals present a mechanical behaviour close to those of bones. Light, strong and totally biocompatible, titanium is one of the few materials that naturally match the requirements for implantation in the human body. Among all titanium and its alloys, the mainly used materials in biomedical field are the commercially pure titanium (cp Ti, grade 2) and Ti-6Al-4V (grade 5) alloy. They are widely used as hard tissue replacements in artificial bones, joints and dental implants. As a hard tissue replacement, the low elastic modulus of titanium and its alloys is generally viewed as a biomechanical advantage because the smaller elastic modulus can result in smaller stress shielding.

Other property that makes titanium and its alloys the most promising biomaterials for implants is that titanium-based materials in general rely on the formation of an extremely thin, adherent, protective titanium oxide film. The presence of this oxide film that forms spontaneously in the passivation or repassivation process is a major criterion for the excellent biocompatibility and corrosion resistance of titanium and its alloys.

Concerning the medical applications of these materials, the use of cp (commercially pure) Titanium is more limited to the dental implants because of its limited mechanical properties. In cases where good mechanical characteristics are required as in hip implants, knee implants, bone screws, and plates, Ti-6Al-4V alloy is being used [27] [28]. One of the most common applications of titanium alloys is artificial hip joints that consist of an articulating bearing (femoral head and cup) and stem [24], where metallic cup and hip stem components are made of titanium. As well, they are also often used in knee joint replacements, which consist of a femoral and tibial component made of titanium and a polyethylene articulating surface.

**Figure 3.** Schematic diagram of artificial hip joint (left) and knee implant [29] (right)

#### **3.1. Wear problems in titanium and titanium alloys**

have been made to reduce wear by using a variety of different combinations of materials and

Nowadays, the materials used for biomedical applications are mainly metallic materials such as 316L stainless steel, cobalt chromium alloys (CoCrMo), titanium-based alloys (Ti-6Al-4V) and miscellaneous others (including tantalum, gold, dental amalgams and other "specialty" metals). Titanium alloys are fast emerging as the first choice for majority of applications due to the combination of their outstanding characteristics such as high strength, low density, high immunity to corrosion, complete inertness to body environment, enhanced compatibility, low Young´s modulus and high capacity to join with bone or other tissues. Their lower Young´s modulus, superior biocompatibility and better corrosion resistance in comparison with conventional stainless steels and cobalt-based alloys, make them an ideal choice for bioapplications [26]. Because of the mentioned desirable properties, titanium and titanium alloys are widely used as hard tissue replacements in artificial bones, joints and dental implants.

The elemental metal titanium was first discovered in England by William Gregor in 1790, but in 1795 Klaproth gave it the name of titanium. Combination of low density, high strength to weight ratio, good biocompatibility and improved corrosion resistance with good plasticity and mechanical properties determines the application of titanium and its alloys in such industries as aviation, automotive, power and shipbuilding industries or architecture as well

Increased use of titanium and its alloys as biomaterials comes from their superior biocompat‐ ibility and excellent corrosion resistance because of the thin surface oxide layer, and good mechanical properties, as a certain elastic modulus and low density that make that these metals present a mechanical behaviour close to those of bones. Light, strong and totally biocompatible, titanium is one of the few materials that naturally match the requirements for implantation in the human body. Among all titanium and its alloys, the mainly used materials in biomedical field are the commercially pure titanium (cp Ti, grade 2) and Ti-6Al-4V (grade 5) alloy. They are widely used as hard tissue replacements in artificial bones, joints and dental implants. As a hard tissue replacement, the low elastic modulus of titanium and its alloys is generally viewed as a biomechanical advantage because the smaller elastic modulus can result in smaller

Other property that makes titanium and its alloys the most promising biomaterials for implants is that titanium-based materials in general rely on the formation of an extremely thin, adherent, protective titanium oxide film. The presence of this oxide film that forms spontaneously in the passivation or repassivation process is a major criterion for the excellent biocompatibility and

Concerning the medical applications of these materials, the use of cp (commercially pure) Titanium is more limited to the dental implants because of its limited mechanical properties.

surface treatments.

160 Tribology - Fundamentals and Advancements

**3. Titanium and titanium alloys**

as medicine and sports equipment.

corrosion resistance of titanium and its alloys.

stress shielding.

The fundamental drawback of titanium and its alloys which limits wider use of these materials include their poor fretting fatigue resistance and poor tribological properties [30] [31], because of its low hardness [32]. Their poor tribological behavior is characterized by high coefficient of friction, severe adhesive wear with a strong tendency to seizing and low abrasion resistance [33]. Titanium tends to undergo severe wear when it is rubbed between itself or between other materials. Titanium has tendency for moving or sliding parts to gall and eventually seize. This causes a more intensive wear as a result of creation of adhesion couplings and mechanical instability of passive layer of oxides, particularly in presence of third bodies (Figure 4). Owing to this effect, in cases of total joint replacements made of titanium head and polymer cup, the 10%-20% of joints needs to be replaced within 15-20 years and the aseptic loosening accounts for approximately 80% of the revisions [34]. The reason for the failure of the implants is due to the high friction coefficient of these materials that can lead to the release of wear debris from the implant into the bloodstream that results in an inflammation of the surrounding tissue and gives rise to the bone resorption (osteolysis) [35] [36], which ultimately leads to loosening of the implant and hence the implant has to be replaced by a new one.

The relatively poor tribological properties and possible corrosion problems have led to the development of surface treatments to effectively increase near-surface strength, improving the hardness and abrasive wear resistance thereby reducing the coefficient of friction as well as avoiding or reducing the transference of ions from the surface or bulk material to the sur‐

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

163

When an implant is surgically placed within bone there are numerous biological, physical, chemical, thermal and other factors functioning that determine whether or not osseointegra‐

Titanium and its alloys have been widely used for dental and orthopedic implants under loadbearing conditions because of their good biocompatibility coupled with high strength and fracture toughness. Despite reports of direct bonding to bone, they do not form a chemical bond with bone tissue. For the last decade, various coatings have been attempted to provide titanium and its alloys with bond-bonding ability, which spontaneously bond to living bone. Hydroxyapatite plasma spray coatings are widely used in cementless hip replacement surgery, but the hydroxyapatite coating, although exhibiting a very good biocompatibility, presents some disadvantages including delamination of the coating layer from the substrate, difficulties in controlling the composition of the coating layer and degradation of the coating layer itself,

A strong and durable bone to implant connection can be achieved by the formation of a stable bone tissue at the bone-implant interface by proper implant surface treatments, as can be

Surface engineering can play a significant role in extending the performance of orthopedic

The main objectives of surface treatments mainly consist of the improvement of the tribological behaviour, corrosion resistance and osseointegration of the implant. There are coatings for enhanced wear and corrosion resistance by improving the surface hardness of the material that can be applied by different surface modifications techniques such as surface oxidation, physical deposition methods like ion implantation and plasma spray coatings, as well as thermo-chemical surface treatments such as nitriding, carburizing and boriding [43] [44].

Great efforts have been devoted to thickening and stabilizing surface oxides on titanium to achieve desired biological responses. The biological response to titanium depends on the surface chemical composition, and the ability of titanium oxides to absorb molecules and incorporate elements. Surface topography plays a fundamental role in regulating cell behavior,

One possible alternative to solve tribological problems and which is going to explain more detail consists of protecting the alloy surface by means of biocompatible Diamond-Like Carbon

electrochemical deposition, dipping and physical vapor deposition techniques [42].

rounding tissue.

tion will occur.

**3.3. Osseointegration of titanium and titanium alloys**

which can release debris becoming a source of third body wear [41].

devices made of titanium several times beyond its natural capability.

**3.4. Surface treatments of titanium and titanium alloys**

e.g. the shape, orientation and adhesion of cells.

**Figure 4.** Schematic representation of a sliding tribological coating with the presence of third bodies [37]

#### **3.2. Corrosion behaviour of titanium and titanium alloys**

All metals and alloys are subjected to corrosion when in contact with body fluid as the body environment is very aggressive owing to the presence of chloride ions and proteins. A variety of chemical reactions occur on the surface of a surgically implanted alloy. The metallic components of the alloy are oxidized to their ionic forms and dissolved oxygen is reduced to hydroxide ions.

Most metals and alloys that resist well against corrosion are in the passive state. Metals in the passive state (passive metals) have a thin oxide layer (TiO2 in case of titanium) on their surface, the passive film, which separates the metal from its environment [38]. Typically, the thickness of passive films formed on these metals is about 3-10 nm [39] and they consist of metal oxides (ceramic films). The natural oxide is amorphous and stoichiometrically defective. It is known that the protective and stable oxides on titanium surfaces (TiO2) are able to provide favorable osseointegration. The stability of the oxide depends strongly on the composition structure and thickness of the film [40].

Because of the presence of an oxide film, the dissolution rate of a passive metal at a given potential is much lower than that of an active metal. It depends mostly on the properties of the passive film and its solubility in the environment. These films which form spontaneously on the surface of the metal prevent further transport of metallic ions and/or electrons across the film. To be effective barriers, the films must be compact and fully cover the metal surface; they must have an atomic structure that limits the migration of ions and/or electrons across the metal oxide–solution interface; and they must be able to remain on the surface of these alloys even with mechanical stressing or abrasion, expected with orthopedic devices [25].

The relatively poor tribological properties and possible corrosion problems have led to the development of surface treatments to effectively increase near-surface strength, improving the hardness and abrasive wear resistance thereby reducing the coefficient of friction as well as avoiding or reducing the transference of ions from the surface or bulk material to the sur‐ rounding tissue.

#### **3.3. Osseointegration of titanium and titanium alloys**

Contact

Tribo-modified wear track and possible subsurface

damage

Wear debris

Direction load force

**Figure 4.** Schematic representation of a sliding tribological coating with the presence of third bodies [37]

All metals and alloys are subjected to corrosion when in contact with body fluid as the body environment is very aggressive owing to the presence of chloride ions and proteins. A variety of chemical reactions occur on the surface of a surgically implanted alloy. The metallic components of the alloy are oxidized to their ionic forms and dissolved oxygen is reduced to

Most metals and alloys that resist well against corrosion are in the passive state. Metals in the passive state (passive metals) have a thin oxide layer (TiO2 in case of titanium) on their surface, the passive film, which separates the metal from its environment [38]. Typically, the thickness of passive films formed on these metals is about 3-10 nm [39] and they consist of metal oxides (ceramic films). The natural oxide is amorphous and stoichiometrically defective. It is known that the protective and stable oxides on titanium surfaces (TiO2) are able to provide favorable osseointegration. The stability of the oxide depends strongly on the composition structure and

Because of the presence of an oxide film, the dissolution rate of a passive metal at a given potential is much lower than that of an active metal. It depends mostly on the properties of the passive film and its solubility in the environment. These films which form spontaneously on the surface of the metal prevent further transport of metallic ions and/or electrons across the film. To be effective barriers, the films must be compact and fully cover the metal surface; they must have an atomic structure that limits the migration of ions and/or electrons across the metal oxide–solution interface; and they must be able to remain on the surface of these alloys even with mechanical stressing or abrasion, expected with orthopedic devices [25].

of sliding

162 Tribology - Fundamentals and Advancements

Transfer film Depends on environment (chemistry, humidity, temperature, etc.) Substrate

**3.2. Corrosion behaviour of titanium and titanium alloys**

hydroxide ions.

thickness of the film [40].

When an implant is surgically placed within bone there are numerous biological, physical, chemical, thermal and other factors functioning that determine whether or not osseointegra‐ tion will occur.

Titanium and its alloys have been widely used for dental and orthopedic implants under loadbearing conditions because of their good biocompatibility coupled with high strength and fracture toughness. Despite reports of direct bonding to bone, they do not form a chemical bond with bone tissue. For the last decade, various coatings have been attempted to provide titanium and its alloys with bond-bonding ability, which spontaneously bond to living bone. Hydroxyapatite plasma spray coatings are widely used in cementless hip replacement surgery, but the hydroxyapatite coating, although exhibiting a very good biocompatibility, presents some disadvantages including delamination of the coating layer from the substrate, difficulties in controlling the composition of the coating layer and degradation of the coating layer itself, which can release debris becoming a source of third body wear [41].

A strong and durable bone to implant connection can be achieved by the formation of a stable bone tissue at the bone-implant interface by proper implant surface treatments, as can be electrochemical deposition, dipping and physical vapor deposition techniques [42].

#### **3.4. Surface treatments of titanium and titanium alloys**

Surface engineering can play a significant role in extending the performance of orthopedic devices made of titanium several times beyond its natural capability.

The main objectives of surface treatments mainly consist of the improvement of the tribological behaviour, corrosion resistance and osseointegration of the implant. There are coatings for enhanced wear and corrosion resistance by improving the surface hardness of the material that can be applied by different surface modifications techniques such as surface oxidation, physical deposition methods like ion implantation and plasma spray coatings, as well as thermo-chemical surface treatments such as nitriding, carburizing and boriding [43] [44].

Great efforts have been devoted to thickening and stabilizing surface oxides on titanium to achieve desired biological responses. The biological response to titanium depends on the surface chemical composition, and the ability of titanium oxides to absorb molecules and incorporate elements. Surface topography plays a fundamental role in regulating cell behavior, e.g. the shape, orientation and adhesion of cells.

One possible alternative to solve tribological problems and which is going to explain more detail consists of protecting the alloy surface by means of biocompatible Diamond-Like Carbon (DLC) coatings. "Diamond-Like Carbon" is a generic term referring to amorphous carbon films, deposited by either Physical Vapor Deposition (PVD) or Plasma-Enhanced Chemical Vapor Deposition (PECVD). DLC coatings basically consist of a mixture of diamond (sp3 ) and graphite (sp2 ). The relative amounts of these two phases will determine much of the coating properties. They are thus metastable and mostly amorphous, "crystalline" clusters being too small or too defective to reach graphite or diamond structures. Both the mechanical and the tribological properties of DLC coatings have been studied for about 30 years, and several different types of DLC coatings can currently be found. DLC films are attractive biomedical materials due to their relatively high hardness, low friction coefficient, owing to the solid lubricant because of its graphite and amorphous carbon contents [31], good chemical stability and excellent bio and hemocompatibility [45] [44] [46] [47]. Cells are seen to grow well on these films coated on titanium and other materials without any cytotoxicity and inflammation.

try, surface topography, surface roughness and mainly the surface energy. TiO2, calcium phosphate, titania/hydroxiapatite composite and silica coating by the sol-gel method can be applied on the surface of the titanium and titanium alloys. Plasma Electrolytic Oxidation (PEO) or Micro-Arc Oxidation (MAO) technique is used for the synthesize TiO2 layer. This technique is based on the modification of the growing anodic film by arc micro-discharges, which are initiated at potentials above the breakdown voltage of the growing oxide film and move rapidly across the anode surface. This technology provides a solution by transforming the surface into a dense layer of ceramic which not only prevents galling but also provides excellent dielectric insulation for contact metals, helping to protect them against aggressive galvanic corrosion. PEO process transforms the surface of titanium alloys into a complex ceramic matrix by passing a pulsed, bi-polar electrical current in a specific wave formation through a bath of low concentration aqueous solution. A plasma discharge is formed on the surface of the substrate, transforming it into a thin, protective layer of titanium oxide, without subjecting the

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

165

Among all the above mentioned surface treatments, Diamond-Like Carbon coating and Plasma Electrolytic Oxidation are the most promising ones applied on titanium surfaces. These two

In some biomedical applications continuously sliding contact is required, subjecting the implant to aggressive situations. To achieve and maintain higher efficiency and durability under such increasingly more severe sliding conditions, protective and/or solid coatings are

These coatings can generally be divided in two broad categories [59] : "soft coatings", which are usually good for solid lubrication and exhibit low friction coefficients, and "hard coatings", which are usually good for protection against wear, and exhibit low wear rates and hence

It would thus seem to be difficult to associate low friction and high wear resistance with all types of coating in most tribological contacts. Some trade-offs can be found in combining both hard and soft materials in composite or multilayer coatings, which require complex procedures and further optimization of the deposition process. Nevertheless, a diverse family of carbonbased materials seems to "naturally" combine the desired set of tribological properties, providing not only low friction but also high wear resistance. These materials are widely known as the diamond and Diamond-Like Carbon (DLC) coatings. They are usually harder than most metals and/or alloys, thus affording very high wear resistance and, at the same time, impressive friction coefficients generally in the range of 0.05-0.2 [60] [61] [62]. In some cases, friction values lower than 0.01 have been reported [63] [64], offering a sliding regime often referred to as "superlubricity". These exceptional tribological abilities explain the increasing success of Diamond-Like Carbon coatings over the years, both in industrial applications and in the laboratory. The exceptional tribological behavior of Diamond-Like Carbon films appears to be due to a unique combination of surface chemical, physical, and mechanical interactions

substrate itself to damaging thermal exposure.

*3.4.1. Diamond-like carbon coatings*

becoming prevalent.

longer durability (Figure 5).

at their sliding interfaces [65].

treatments are explained in more detail in the following sections.

Oxidation remains the most popular technique for the surface modification of Ti alloys; these oxide layers on titanium are commonly produced by either heat treatment [48] [49] [50] or electrolytic anodizing [51]. Thermal oxidation results in the formation of a 15-30 µm thick titanium dioxide layer of the rutile phase. However, due to their long-term high temperature action, thermal diffusion processes can also lead to the formation of a diffusion sub-layer consisting of an oxygen solid solution in α-Ti, and development of phase segregation and coalescence which may cause substrate embrittlement and worsened mechanical and/or corrosion performance.

Conventional anodic oxidation, which is carried out in various solutions providing passivation of the titanium surface, generates thin films of amorphous hydrated oxide or crystalline TiO2 in the anatase form [52]. These films exhibit poor corrosion resistance in some reducing acids and halide solutions, while rutile generally possesses much better protective properties. However, recent developments in high voltage anodizing allow the production of crystalline rutile/anatase films at near to ambient temperature [53]. By anodic oxidation, elements such as Ca and P can be imported into the surface oxide on titanium and the micro-topography can be varied through regulating electrolyte and electrochemical conditions. The presence of Caions has been reported to be advantageous to cell growth, and in vivo data show implant surfaces containing both Ca and P enhance bone apposition on the implant surface.

Furthermore, there are alternative methods to improve the biocompatibility such as biocom‐ patible chemicals [54] and materials such as ceramics for coating. In some studies, titanium surfaces were modified using phosphoric acid in an "in vitro" study to improve the biocom‐ patibility of dental implants. Results indicated that pretreatment of the implant with phos‐ phoric acid caused no citotoxicity to the osteoblasts [55]. Micro arc oxidation method in phosphoric acid on titanium implants provided chemical bonding sites for calcium ions during mineralization [56].

Moreover, there have been developed coatings for high osseointegration. Hydroxyapatite (HA) coating is a proven method to improve the implants´ mechanical bonding [57] [58], biocompatibility and improve the osseointegration. The higher the degree of osseointegration, the higher is the mechanical stability and the probability of implant loosening becomes smaller. The process of osseointegration depends upon the surface properties such as surface chemis‐ try, surface topography, surface roughness and mainly the surface energy. TiO2, calcium phosphate, titania/hydroxiapatite composite and silica coating by the sol-gel method can be applied on the surface of the titanium and titanium alloys. Plasma Electrolytic Oxidation (PEO) or Micro-Arc Oxidation (MAO) technique is used for the synthesize TiO2 layer. This technique is based on the modification of the growing anodic film by arc micro-discharges, which are initiated at potentials above the breakdown voltage of the growing oxide film and move rapidly across the anode surface. This technology provides a solution by transforming the surface into a dense layer of ceramic which not only prevents galling but also provides excellent dielectric insulation for contact metals, helping to protect them against aggressive galvanic corrosion. PEO process transforms the surface of titanium alloys into a complex ceramic matrix by passing a pulsed, bi-polar electrical current in a specific wave formation through a bath of low concentration aqueous solution. A plasma discharge is formed on the surface of the substrate, transforming it into a thin, protective layer of titanium oxide, without subjecting the substrate itself to damaging thermal exposure.

Among all the above mentioned surface treatments, Diamond-Like Carbon coating and Plasma Electrolytic Oxidation are the most promising ones applied on titanium surfaces. These two treatments are explained in more detail in the following sections.

#### *3.4.1. Diamond-like carbon coatings*

(DLC) coatings. "Diamond-Like Carbon" is a generic term referring to amorphous carbon films, deposited by either Physical Vapor Deposition (PVD) or Plasma-Enhanced Chemical Vapor Deposition (PECVD). DLC coatings basically consist of a mixture of diamond (sp3

properties. They are thus metastable and mostly amorphous, "crystalline" clusters being too small or too defective to reach graphite or diamond structures. Both the mechanical and the tribological properties of DLC coatings have been studied for about 30 years, and several different types of DLC coatings can currently be found. DLC films are attractive biomedical materials due to their relatively high hardness, low friction coefficient, owing to the solid lubricant because of its graphite and amorphous carbon contents [31], good chemical stability and excellent bio and hemocompatibility [45] [44] [46] [47]. Cells are seen to grow well on these films coated on titanium and other materials without any cytotoxicity and inflammation.

Oxidation remains the most popular technique for the surface modification of Ti alloys; these oxide layers on titanium are commonly produced by either heat treatment [48] [49] [50] or electrolytic anodizing [51]. Thermal oxidation results in the formation of a 15-30 µm thick titanium dioxide layer of the rutile phase. However, due to their long-term high temperature action, thermal diffusion processes can also lead to the formation of a diffusion sub-layer consisting of an oxygen solid solution in α-Ti, and development of phase segregation and coalescence which may cause substrate embrittlement and worsened mechanical and/or

Conventional anodic oxidation, which is carried out in various solutions providing passivation of the titanium surface, generates thin films of amorphous hydrated oxide or crystalline TiO2 in the anatase form [52]. These films exhibit poor corrosion resistance in some reducing acids and halide solutions, while rutile generally possesses much better protective properties. However, recent developments in high voltage anodizing allow the production of crystalline rutile/anatase films at near to ambient temperature [53]. By anodic oxidation, elements such as Ca and P can be imported into the surface oxide on titanium and the micro-topography can be varied through regulating electrolyte and electrochemical conditions. The presence of Caions has been reported to be advantageous to cell growth, and in vivo data show implant

surfaces containing both Ca and P enhance bone apposition on the implant surface.

Furthermore, there are alternative methods to improve the biocompatibility such as biocom‐ patible chemicals [54] and materials such as ceramics for coating. In some studies, titanium surfaces were modified using phosphoric acid in an "in vitro" study to improve the biocom‐ patibility of dental implants. Results indicated that pretreatment of the implant with phos‐ phoric acid caused no citotoxicity to the osteoblasts [55]. Micro arc oxidation method in phosphoric acid on titanium implants provided chemical bonding sites for calcium ions during

Moreover, there have been developed coatings for high osseointegration. Hydroxyapatite (HA) coating is a proven method to improve the implants´ mechanical bonding [57] [58], biocompatibility and improve the osseointegration. The higher the degree of osseointegration, the higher is the mechanical stability and the probability of implant loosening becomes smaller. The process of osseointegration depends upon the surface properties such as surface chemis‐

). The relative amounts of these two phases will determine much of the coating

graphite (sp2

164 Tribology - Fundamentals and Advancements

corrosion performance.

mineralization [56].

) and

In some biomedical applications continuously sliding contact is required, subjecting the implant to aggressive situations. To achieve and maintain higher efficiency and durability under such increasingly more severe sliding conditions, protective and/or solid coatings are becoming prevalent.

These coatings can generally be divided in two broad categories [59] : "soft coatings", which are usually good for solid lubrication and exhibit low friction coefficients, and "hard coatings", which are usually good for protection against wear, and exhibit low wear rates and hence longer durability (Figure 5).

It would thus seem to be difficult to associate low friction and high wear resistance with all types of coating in most tribological contacts. Some trade-offs can be found in combining both hard and soft materials in composite or multilayer coatings, which require complex procedures and further optimization of the deposition process. Nevertheless, a diverse family of carbonbased materials seems to "naturally" combine the desired set of tribological properties, providing not only low friction but also high wear resistance. These materials are widely known as the diamond and Diamond-Like Carbon (DLC) coatings. They are usually harder than most metals and/or alloys, thus affording very high wear resistance and, at the same time, impressive friction coefficients generally in the range of 0.05-0.2 [60] [61] [62]. In some cases, friction values lower than 0.01 have been reported [63] [64], offering a sliding regime often referred to as "superlubricity". These exceptional tribological abilities explain the increasing success of Diamond-Like Carbon coatings over the years, both in industrial applications and in the laboratory. The exceptional tribological behavior of Diamond-Like Carbon films appears to be due to a unique combination of surface chemical, physical, and mechanical interactions at their sliding interfaces [65].

**Figure 5.** Classification of coatings with respect to hardness and coefficient of friction, highlighting the special case of carbon-based coatings

Since their initial discovery in the early 1950s, Diamond-Like Carbon coatings have attracted the most attention in recent years, mainly because they are cheap and easy to produce and offer exceptional properties for demanding engineering and medical applications. They can be used in invasive and implantable medical devices. These films are currently being evaluated for their durability and performance characteristics in certain biomedical implants including hip and knee joints and coronary stents.

Diamond-Like Carbon is the only coating that can provide both high hardness and low friction under dry sliding conditions. These films are metastable forms of carbon combining both sp2 and sp3 hybridizations, including hydrogen when a hydrocarbon precursor is used during deposition. The tribological behavior of Diamond-Like Carbon films requires a solid back‐ ground on the chemical and structural nature of these films, which, in turn, depends on the deposition process and/or parameters. The chemical composition, such as the hydrogen and/ or nitrogen content or the presence of other alloying elements, controls the mechanical and tribological properties of a sliding pair consisting of DLC on one or both sliding surfaces [66]. For example, DLC samples containing different concentrations of titanium (Figure 6) have also been examined "in vitro" to obtain a biocompatible surface that is hard, preventing abrasion and scratching [67].

**Figure 7.** SEM (Scanning electron microscopy) micrograph of Ti-DLC coating deposited by physical vapour deposition

**Figure 6.** Scheme of titanium doped DLC coating. In this case, the first titanium layer was deposited in order to im‐

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

167

prove adhesion of DLC coating to the substrate and relax stress of the coating

A frequently observed feature in tribological testing of Diamond-Like Carbon films is the formation of transfer layer. The formation of carbonous transfer layer on the sliding surface

DLC coatings are usually applied by means of Cathodic Arc Evaporation Physical Vapor Deposition technology. An arc can be defined as a discharge of electricity between two electrodes. The arc evaporation process begins with the striking of a high current, low voltage arc on the surface of a cathode that gives rise to a small (usually a few microns wide) highly energetic emitting area known as a cathode spot. The localised temperature at the cathode spot is extremely high (around 15000 °C), which results in a high velocity (10 km/s) jet of vaporised

technique using cathodic arc evaporation method

was observed to reduce the friction coefficient [68].

cathode material, leaving a crater behind on the cathode surface.

It is well known that Diamond-Like Carbon films usually present smooth surfaces, except maybe in the case of films formed by unfiltered cathodic vacuum arc deposition (Figure 7). Roughness of the films on industrial surfaces will then be mainly controlled by the substrate roughness and can therefore be minimized.

**Figure 6.** Scheme of titanium doped DLC coating. In this case, the first titanium layer was deposited in order to im‐ prove adhesion of DLC coating to the substrate and relax stress of the coating

Since their initial discovery in the early 1950s, Diamond-Like Carbon coatings have attracted the most attention in recent years, mainly because they are cheap and easy to produce and offer exceptional properties for demanding engineering and medical applications. They can be used in invasive and implantable medical devices. These films are currently being evaluated for their durability and performance characteristics in certain biomedical implants including

**Figure 5.** Classification of coatings with respect to hardness and coefficient of friction, highlighting the special case of

Diamond-Like Carbon is the only coating that can provide both high hardness and low friction under dry sliding conditions. These films are metastable forms of carbon combining both sp2 and sp3 hybridizations, including hydrogen when a hydrocarbon precursor is used during deposition. The tribological behavior of Diamond-Like Carbon films requires a solid back‐ ground on the chemical and structural nature of these films, which, in turn, depends on the deposition process and/or parameters. The chemical composition, such as the hydrogen and/ or nitrogen content or the presence of other alloying elements, controls the mechanical and tribological properties of a sliding pair consisting of DLC on one or both sliding surfaces [66]. For example, DLC samples containing different concentrations of titanium (Figure 6) have also been examined "in vitro" to obtain a biocompatible surface that is hard, preventing abrasion

It is well known that Diamond-Like Carbon films usually present smooth surfaces, except maybe in the case of films formed by unfiltered cathodic vacuum arc deposition (Figure 7). Roughness of the films on industrial surfaces will then be mainly controlled by the substrate

hip and knee joints and coronary stents.

roughness and can therefore be minimized.

and scratching [67].

carbon-based coatings

166 Tribology - Fundamentals and Advancements

**Figure 7.** SEM (Scanning electron microscopy) micrograph of Ti-DLC coating deposited by physical vapour deposition technique using cathodic arc evaporation method

A frequently observed feature in tribological testing of Diamond-Like Carbon films is the formation of transfer layer. The formation of carbonous transfer layer on the sliding surface was observed to reduce the friction coefficient [68].

DLC coatings are usually applied by means of Cathodic Arc Evaporation Physical Vapor Deposition technology. An arc can be defined as a discharge of electricity between two electrodes. The arc evaporation process begins with the striking of a high current, low voltage arc on the surface of a cathode that gives rise to a small (usually a few microns wide) highly energetic emitting area known as a cathode spot. The localised temperature at the cathode spot is extremely high (around 15000 °C), which results in a high velocity (10 km/s) jet of vaporised cathode material, leaving a crater behind on the cathode surface.

The plasma jet intensity is greatest normal to the surface of the cathode and contains a high level of ionization (30%-100%) multiply charged ions, neutral particles, clusters and macroparticles (droplets). The metal is evaporated by the arc in a single step, and ionized and accelerated within an electric field. Theoretically the arc is a self-sustaining discharge capable of sustaining large currents through electron emission from the cathode surface and the rebombardment of the surface by positive ions under high vacuum conditions.

*3.4.2. Plasma electrolytic oxidation treatment*

composition structure and thickness of the film [75].

temperature.

coating.

surface roughness.

In biomedical application titanium is the most employed alloy due to its biocompatibility as an implant material, attributed to surface oxides spontaneously formed in air and/or physio‐ logical fluids [70]. Cellular behaviors, e.g. adhesion, morphologic change, functional alteration, proliferation and differentiation are greatly affected by surface properties, including compo‐ sition, roughness, hydrophilicity, texture and morphology of the oxide on titanium [71] [72]. The natural oxide is thin (about 3–10nm in thickness [39] ) amorphous and stoichiometrically defective. It is known that the protective and stable oxides on titanium surfaces are able to provide favorable osseointegration [73] [74]. The stability of the oxide depends strongly on the

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

169

On titanium and its alloys a thin oxide layer is formed naturally on the surface of titanium metal in exposure to air at room temperature [76] [77] [78]. Titania (TiO2) exists in three polymorphic forms: rutile, anatase and brookite. Rutile, stable form of titania at ambient condition, possesses unique properties [79]. The metastable anatase and brookite phases convert to rutile upon heating. However, contact loads damage this thin native oxide film and cause galvanic and crevice corrosion as well as corrosion embrittlement. Moreover, the low wear resistance and high friction coefficient without applied protective coatings on the surface gravely limit its extensive applications. The most accepted technique for the surface modification of Ti alloys is oxidation. Anodizing produces anatase phase of titania that shows poor corrosion resistance in comparison with rutile phase. Recent develop‐ ments in high voltage anodizing cause a crystalline rutile/anatase film at near to room

Attempts to improve surface properties of titanium and its alloys over the last few decades have led to development of Plasma Electrolytic Oxidation (PEO) technique by Kurze et al. [80] [81], which is a process to synthesize the ceramic-like oxide films at high voltages. This technique is based on the modification of the growing anodic film by spark/arc microdischarges in aqueous solutions (Figure 9), which are initiated at potentials above the breakdown voltage of the growing oxide film and move rapidly across the anode surface [53]. Since they rapidly develop and extinguish (within 10-4-10-5 s), the discharges heat the metal substrate to less than 100-150 ºC. At the same time the local temperature and pressure inside the discharge channel can reach 10-3-10-4 K and 10-2-10-3 MPa, respectively, which is high enough to give rise to plasma thermo-chemical interactions between the substrate and the electrolyte. These interactions result in the formation of melt-quenched high-tempera‐ ture oxides and complex compounds on the surface, composed of oxides of both the substrate material and electrolyte-borne modifying elements. The result is a porous oxide

The PEO coating shows a significantly higher thickness (18 µm ± 4 µm) than PVD coatings and also a different morphology. The external part of the layer is porous (with pore diameter ranging from 3 to 8 µm) (Figure 10). The coating becomes increasingly compact on going towards the interface with the substrate. This kind of morphology leads to a relatively high

If a reactive gas is introduced during the evaporation process dissociation, ionization and excitation can occur during interaction with the ion flux and a compound film will be depos‐ ited. Without the influence of an applied magnetic field the cathode spot moves around randomly evaporating microscopic asperities and creating craters. However if the cathode spot stays at one of these evaporative points for too long it can eject a large amount of macroparticles or droplets as seen above. These droplets are detrimental to the performance of the coating as they are poorly adhered and can extend through the coating.

A recent tribological study carried out about the effect of deposition of Diamond-Like Carbon coatings on a substrate of Ti-6Al-4V for knee implants has confirmed that these types of coating improve the tribological response of substrate decreasing the coefficient of friction (µ) (Table 1) and reducing the wear of the surface (Figure 8) [69]. For this study fretting tests were performed using alumina balls as counter body, bovine serum as lubricant and a continuous temperature of 37 ºC, trying to simulate real environment.


**Table 1.** Friction coefficients values and ball and disc wear scars measurements

**Figure 8.** SEM micrographs of the fretting tests wear scars. Ti-6Al-4V (left), Ti-DLC (right)

#### *3.4.2. Plasma electrolytic oxidation treatment*

The plasma jet intensity is greatest normal to the surface of the cathode and contains a high level of ionization (30%-100%) multiply charged ions, neutral particles, clusters and macroparticles (droplets). The metal is evaporated by the arc in a single step, and ionized and accelerated within an electric field. Theoretically the arc is a self-sustaining discharge capable of sustaining large currents through electron emission from the cathode surface and the re-

If a reactive gas is introduced during the evaporation process dissociation, ionization and excitation can occur during interaction with the ion flux and a compound film will be depos‐ ited. Without the influence of an applied magnetic field the cathode spot moves around randomly evaporating microscopic asperities and creating craters. However if the cathode spot stays at one of these evaporative points for too long it can eject a large amount of macroparticles or droplets as seen above. These droplets are detrimental to the performance of the

A recent tribological study carried out about the effect of deposition of Diamond-Like Carbon coatings on a substrate of Ti-6Al-4V for knee implants has confirmed that these types of coating improve the tribological response of substrate decreasing the coefficient of friction (µ) (Table 1) and reducing the wear of the surface (Figure 8) [69]. For this study fretting tests were performed using alumina balls as counter body, bovine serum as lubricant and a continuous

(a) (b)

bombardment of the surface by positive ions under high vacuum conditions.

coating as they are poorly adhered and can extend through the coating.

**Sample µ ± SD (standard deviation) Disc Wear Scar, Maximum Depth (µm)**

temperature of 37 ºC, trying to simulate real environment.

Ti-DLC 0.24 ± 0.01 Polishing Effect

**Table 1.** Friction coefficients values and ball and disc wear scars measurements

**Figure 8.** SEM micrographs of the fretting tests wear scars. Ti-6Al-4V (left), Ti-DLC (right)

Ti-6Al-4V 0.86 ± 0.08 10 ± 3

168 Tribology - Fundamentals and Advancements

In biomedical application titanium is the most employed alloy due to its biocompatibility as an implant material, attributed to surface oxides spontaneously formed in air and/or physio‐ logical fluids [70]. Cellular behaviors, e.g. adhesion, morphologic change, functional alteration, proliferation and differentiation are greatly affected by surface properties, including compo‐ sition, roughness, hydrophilicity, texture and morphology of the oxide on titanium [71] [72]. The natural oxide is thin (about 3–10nm in thickness [39] ) amorphous and stoichiometrically defective. It is known that the protective and stable oxides on titanium surfaces are able to provide favorable osseointegration [73] [74]. The stability of the oxide depends strongly on the composition structure and thickness of the film [75].

On titanium and its alloys a thin oxide layer is formed naturally on the surface of titanium metal in exposure to air at room temperature [76] [77] [78]. Titania (TiO2) exists in three polymorphic forms: rutile, anatase and brookite. Rutile, stable form of titania at ambient condition, possesses unique properties [79]. The metastable anatase and brookite phases convert to rutile upon heating. However, contact loads damage this thin native oxide film and cause galvanic and crevice corrosion as well as corrosion embrittlement. Moreover, the low wear resistance and high friction coefficient without applied protective coatings on the surface gravely limit its extensive applications. The most accepted technique for the surface modification of Ti alloys is oxidation. Anodizing produces anatase phase of titania that shows poor corrosion resistance in comparison with rutile phase. Recent develop‐ ments in high voltage anodizing cause a crystalline rutile/anatase film at near to room temperature.

Attempts to improve surface properties of titanium and its alloys over the last few decades have led to development of Plasma Electrolytic Oxidation (PEO) technique by Kurze et al. [80] [81], which is a process to synthesize the ceramic-like oxide films at high voltages. This technique is based on the modification of the growing anodic film by spark/arc microdischarges in aqueous solutions (Figure 9), which are initiated at potentials above the breakdown voltage of the growing oxide film and move rapidly across the anode surface [53]. Since they rapidly develop and extinguish (within 10-4-10-5 s), the discharges heat the metal substrate to less than 100-150 ºC. At the same time the local temperature and pressure inside the discharge channel can reach 10-3-10-4 K and 10-2-10-3 MPa, respectively, which is high enough to give rise to plasma thermo-chemical interactions between the substrate and the electrolyte. These interactions result in the formation of melt-quenched high-tempera‐ ture oxides and complex compounds on the surface, composed of oxides of both the substrate material and electrolyte-borne modifying elements. The result is a porous oxide coating.

The PEO coating shows a significantly higher thickness (18 µm ± 4 µm) than PVD coatings and also a different morphology. The external part of the layer is porous (with pore diameter ranging from 3 to 8 µm) (Figure 10). The coating becomes increasingly compact on going towards the interface with the substrate. This kind of morphology leads to a relatively high surface roughness.

density regime have an influence on the phase composition and morphology of the anodic oxide layer [87]. A higher spark voltage causes a higher level of discharge energy, which

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

171

The influence of electrolyte characteristics on the phase composition of PEO films on titanium has previously been studied [89] [90]. It has been shown that surface layers composed of rutile, anatase, rutile/anatase, as well as oxides of electrolyte elements (e.g. Al2O3, MgO, WO3), their hydroxides and complex oxides (e.g. Al2TiO5, AlPO4, CaWO4, BaTiO3, MnTiO3, etc.) can be

Surfaces containing Ca and/or P induce osteoinduction of new bones and become bioactive. Ca and P ions can be incorporated into the layer, controlling the electrolyte employed during the electro oxidation process, and they further transform it into hydroxyapatite by a hydro‐

One technique that could show the effect of the electrolyte in the chemical composition of the coating could be the EDS (Energy Dispersive Spectroscopy) technique. In the following graphs a comparative study can be observed. The results of different samples, uncoated cp Ti, a coating obtained with a commercial electrolyte and a coating prepared in an aqueous electrolyte containing calcium phosphate and β-glycerophosphate, are showed in the following spec‐ trums. The Ca- and P-containing titania coatings produced by PEO improve the bioactivity of the titanium-constructed orthopedic implant [91]. In Figure 11, in spectrum b) and c) can be

The biological response to titanium depends on the surface chemical composition and the ability of titanium oxides to absorb molecules and incorporate elements [92]. Surface topog‐ raphy plays a fundamental role in regulating cell behaviour, e.g. the shape, orientation and adhesion of cells [93] [94]. As a surface begins to contact with biological tissues, water molecules first reach the surface. Hence, surface wettability, initially, may play a major role in adsorption of proteins onto the surface, as well as cell adhesion. Cell adhesion is generally better on hydrophilic surfaces. It is known that changes in the physicochemical properties, which influence the hydrophilicity of Ti dioxide, will modulate the protein adsorption and further cell attachment [39]. By anodic oxidation, elements such as Ca and P can be imported into the surface oxide on titanium and the micro-topography can be varied through regulating electrolyte and electrochemical conditions. The presence of Ca-ions has been reported to be advantageous to cell growth, and "in vivo" data show implant surfaces containing both Ca

Some experiments carried out to study the tribological behaviour of the PEO-treated Ti-6Al-4V by means of dry sliding tests against PS (plasma sprayed) Al2O3–TiO2 and compared with that of thin PVD coatings showed that the best tribological behavior, both in terms of low coefficient of friction and high wear resistance (i.e. low wear damage) was displayed by the PEO treated samples. The highest wear resistance was displayed by the PEO-treated samples, with negligible wear loss even under the highest applied load of 35 N. This good tribological behavior should be mainly related to the superior thickness of this coating that can better

observed the difference in the calcium quantity presented into the coating.

and P enhance bone apposition on the implant surface.

support the applied load.

provides a larger pore [88].

thermal treatment [41].

produced.

**Figure 9.** Photography of the arc micro-discharges in PEO process

**Figure 10.** SEM micrographs of porosity of the external layer in PEO treatment. a) overview and b) detail

This method is characterized by the titanium surface, at near-to-ambient bulk temperature, into the high temperature titanium oxide (rutile) modified by other oxide constituents. Economic efficiency, ecological friendliness, corrosion resistance, high hardness, good wear resistance, and excellent bonding strength with the substrate are the other characteristics of this treatment [82] [83] [84].

The main conversion products formed by the PEO treatment are titanium oxides: rutile and anatase, typical anodic oxidation products of titanium. The structure and composition of anodic oxide films are known to be strongly dependent on film formation temperature and potential [85] [86]. In the case of PEO coatings, both the electrolyte composition and the current density regime have an influence on the phase composition and morphology of the anodic oxide layer [87]. A higher spark voltage causes a higher level of discharge energy, which provides a larger pore [88].

The influence of electrolyte characteristics on the phase composition of PEO films on titanium has previously been studied [89] [90]. It has been shown that surface layers composed of rutile, anatase, rutile/anatase, as well as oxides of electrolyte elements (e.g. Al2O3, MgO, WO3), their hydroxides and complex oxides (e.g. Al2TiO5, AlPO4, CaWO4, BaTiO3, MnTiO3, etc.) can be produced.

Surfaces containing Ca and/or P induce osteoinduction of new bones and become bioactive. Ca and P ions can be incorporated into the layer, controlling the electrolyte employed during the electro oxidation process, and they further transform it into hydroxyapatite by a hydro‐ thermal treatment [41].

One technique that could show the effect of the electrolyte in the chemical composition of the coating could be the EDS (Energy Dispersive Spectroscopy) technique. In the following graphs a comparative study can be observed. The results of different samples, uncoated cp Ti, a coating obtained with a commercial electrolyte and a coating prepared in an aqueous electrolyte containing calcium phosphate and β-glycerophosphate, are showed in the following spec‐ trums. The Ca- and P-containing titania coatings produced by PEO improve the bioactivity of the titanium-constructed orthopedic implant [91]. In Figure 11, in spectrum b) and c) can be observed the difference in the calcium quantity presented into the coating.

The biological response to titanium depends on the surface chemical composition and the ability of titanium oxides to absorb molecules and incorporate elements [92]. Surface topog‐ raphy plays a fundamental role in regulating cell behaviour, e.g. the shape, orientation and adhesion of cells [93] [94]. As a surface begins to contact with biological tissues, water molecules first reach the surface. Hence, surface wettability, initially, may play a major role in adsorption of proteins onto the surface, as well as cell adhesion. Cell adhesion is generally better on hydrophilic surfaces. It is known that changes in the physicochemical properties, which influence the hydrophilicity of Ti dioxide, will modulate the protein adsorption and further cell attachment [39]. By anodic oxidation, elements such as Ca and P can be imported into the surface oxide on titanium and the micro-topography can be varied through regulating electrolyte and electrochemical conditions. The presence of Ca-ions has been reported to be advantageous to cell growth, and "in vivo" data show implant surfaces containing both Ca and P enhance bone apposition on the implant surface.

(a) (b)

This method is characterized by the titanium surface, at near-to-ambient bulk temperature, into the high temperature titanium oxide (rutile) modified by other oxide constituents. Economic efficiency, ecological friendliness, corrosion resistance, high hardness, good wear resistance, and excellent bonding strength with the substrate are the other characteristics of

The main conversion products formed by the PEO treatment are titanium oxides: rutile and anatase, typical anodic oxidation products of titanium. The structure and composition of anodic oxide films are known to be strongly dependent on film formation temperature and potential [85] [86]. In the case of PEO coatings, both the electrolyte composition and the current

**Figure 10.** SEM micrographs of porosity of the external layer in PEO treatment. a) overview and b) detail

**Figure 9.** Photography of the arc micro-discharges in PEO process

170 Tribology - Fundamentals and Advancements

this treatment [82] [83] [84].

Some experiments carried out to study the tribological behaviour of the PEO-treated Ti-6Al-4V by means of dry sliding tests against PS (plasma sprayed) Al2O3–TiO2 and compared with that of thin PVD coatings showed that the best tribological behavior, both in terms of low coefficient of friction and high wear resistance (i.e. low wear damage) was displayed by the PEO treated samples. The highest wear resistance was displayed by the PEO-treated samples, with negligible wear loss even under the highest applied load of 35 N. This good tribological behavior should be mainly related to the superior thickness of this coating that can better support the applied load.

Figure 11.a) Microchemical analysis of cp Ti, b) microchemical analysis of coating prepared with commercial electrolyte, c) microchemical analysis of coating prepared with calcium phosphate and β-glycerophosphate electrolyte. **Figure 11.** a) Microchemical analysis of cp Ti, b) microchemical analysis of coating prepared with commercial electro‐ lyte, c) microchemical analysis of coating prepared with calcium phosphate and β-glycerophosphate electrolyte.

The PEO treatment leads to a very good tribological behavior, significantly reducing both wear and friction of the Ti-6Al-4V alloy, even under high applied loads (up to 35 N). This good tribological behaviour should be mainly related to the superior thickness of this coating, which The biological response to titanium depends on the surface chemical composition and the ability of titanium oxides to absorb molecules and incorporate elements [92] . Surface topography plays a fundamental role in regulating cell behaviour, e.g. the shape, orientation and adhesion of cells [93] [94] . As a surface begins to contact with biological tissues, water molecules first reach the surface. Hence, surface wettability, initially, may play a major role in adsorption of proteins onto the surface, as well as cell

better support the applied load.

adhesion. Cell adhesion is generally better on hydrophilic surfaces. It is known that changes in the physicochemical properties, which influence the hydrophilicity of Ti dioxide, will modulate the protein adsorption and further cell attachment [39] . By anodic oxidation, elements such as Ca and P can be imported into the surface oxide on titanium and the micro-topography can be varied through regulating electrolyte and electrochemical conditions. The presence of Ca-ions has been reported to be advantageous to cell growth, and "in vivo" data show implant surfaces containing both Ca and P enhance bone apposition on the implant surface.

can better support the applied load. The main wear mechanism is micro-polishing and the

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

173

Last studies carried out have concluded that the PEO surface treatments enhance the biological response "in vitro", promoting early osteoblast adhesion, and the osseointegrative properties "in vivo", accelerating the primary osteogenic response, as they confirmed by the more

Titanium and its alloys are considered to be among the most promising engineering materials across a range of application sectors. Due to a unique combination of high strength-to-weight ratio, melting temperature and corrosion resistance, interest in the application of titanium alloys to mechanical and tribological components is growing rapidly in a wide range of industries, especially in biomedical field, also due to their excellent biocompatibility and good osseointegration. In such application, components made from Ti-alloys are often in tribological contact with different materials (metals, polymers or ceramics) and media, under stationary or dynamic loading and at various temperatures. These contact loads can cause damage of the thin native oxide film which passivates the titanium surface; and the metal can undergo intensive interactions with the counterface material and/or the surrounding environment. These interactions can generate various adverse effects on titanium components, such as high friction or even seizure (galvanic and crevice corrosion) as well as corrosion embrittlement, which lead to the premature failure of the implanted systems. The development of new specialized surface modification techniques for titanium and its alloys is therefore an increas‐ ingly critical requirement in order to control or prevent these effects and improve osseointe‐

Physical Vapour Deposition (PVD) technique allows develop Diamond-Like Carbon coatings that can be doped with different elements as titanium, tantalum, silver… which are biocom‐ patible and increase the corrosion and wear resistance of the substrate, diminishing friction

Plasma Electrolytic Oxidation (PEO) technique provides a possibility for the variation of composition and structure of the surface oxide film and attracts special interest for the corrosion protection and the optimization of friction and wear of titanium alloys as well as

The authors acknowledge financial support from the Spanish Ministry of Science and Inno‐ vation obtained in the project: CSD2008-00023 FUNCOAT (in the frame of the CONSOLIDER

coating thickness dictates its tribological life [95].

gration, hence extending the lifetime of the implant.

INGENIO-2010 program) and from the Basque Government.

**4. Conclusions**

coefficient.

enhance the osseointegration.

**Acknowledgements**

extensive bone-implant contact reached after 2 weeks of study [94].

Some experiments carried out to study the tribological behaviour of the PEO-treated Ti-6Al-4V by means of dry sliding tests against PS (plasma sprayed) Al2O3–TiO2 and compared with that of thin PVD coatings showed that the best tribological behavior, both in terms of low coefficient of friction and high wear resistance (i.e. low wear damage) was displayed by the PEO treated samples. The highest wear resistance was displayed by the PEO-treated samples, with negligible wear loss even under the highest applied load of 35 N. This good tribological behavior should be mainly related to the superior thickness of this coating that can can better support the applied load. The main wear mechanism is micro-polishing and the coating thickness dictates its tribological life [95].

Last studies carried out have concluded that the PEO surface treatments enhance the biological response "in vitro", promoting early osteoblast adhesion, and the osseointegrative properties "in vivo", accelerating the primary osteogenic response, as they confirmed by the more extensive bone-implant contact reached after 2 weeks of study [94].

#### **4. Conclusions**

Titanium and its alloys are considered to be among the most promising engineering materials across a range of application sectors. Due to a unique combination of high strength-to-weight ratio, melting temperature and corrosion resistance, interest in the application of titanium alloys to mechanical and tribological components is growing rapidly in a wide range of industries, especially in biomedical field, also due to their excellent biocompatibility and good osseointegration. In such application, components made from Ti-alloys are often in tribological contact with different materials (metals, polymers or ceramics) and media, under stationary or dynamic loading and at various temperatures. These contact loads can cause damage of the thin native oxide film which passivates the titanium surface; and the metal can undergo intensive interactions with the counterface material and/or the surrounding environment. These interactions can generate various adverse effects on titanium components, such as high friction or even seizure (galvanic and crevice corrosion) as well as corrosion embrittlement, which lead to the premature failure of the implanted systems. The development of new specialized surface modification techniques for titanium and its alloys is therefore an increas‐ ingly critical requirement in order to control or prevent these effects and improve osseointe‐ gration, hence extending the lifetime of the implant.

Physical Vapour Deposition (PVD) technique allows develop Diamond-Like Carbon coatings that can be doped with different elements as titanium, tantalum, silver… which are biocom‐ patible and increase the corrosion and wear resistance of the substrate, diminishing friction coefficient.

Plasma Electrolytic Oxidation (PEO) technique provides a possibility for the variation of composition and structure of the surface oxide film and attracts special interest for the corrosion protection and the optimization of friction and wear of titanium alloys as well as enhance the osseointegration.

#### Figure 11.a) Microchemical analysis of cp Ti, b) microchemical analysis of coating prepared with commercial electrolyte, c) microchemical analysis **Acknowledgements**

The PEO treatment leads to a very good tribological behavior, significantly reducing both wear and friction of the Ti-6Al-4V alloy, even under high applied loads (up to 35 N). This good tribological behaviour should be mainly related to the superior thickness of this coating, which

**Figure 11.** a) Microchemical analysis of cp Ti, b) microchemical analysis of coating prepared with commercial electro‐ lyte, c) microchemical analysis of coating prepared with calcium phosphate and β-glycerophosphate electrolyte.

> adhesion. Cell adhesion is generally better on hydrophilic surfaces. It is known that changes in the physicochemical properties, which influence the hydrophilicity of Ti dioxide, will modulate the protein adsorption and further cell attachment [39] . By anodic oxidation, elements such as Ca and P can be imported into the surface oxide on titanium and the micro-topography can be varied through regulating electrolyte and electrochemical conditions. The presence of Ca-ions has been reported to be advantageous to cell growth, and "in vivo" data show implant surfaces containing both Ca and P enhance bone apposition on the implant surface.

> Some experiments carried out to study the tribological behaviour of the PEO-treated Ti-6Al-4V by means of dry sliding tests against PS (plasma sprayed) Al2O3–TiO2 and compared with that of thin PVD coatings showed that the best tribological behavior, both in terms of low coefficient of friction and high wear resistance (i.e. low wear damage) was displayed by the PEO treated samples. The highest wear resistance was displayed by the PEO-treated samples, with negligible wear loss even under the highest applied load of 35 N. This good tribological behavior should be mainly related to the superior thickness of this coating that can

of coating prepared with calcium phosphate and β-glycerophosphate electrolyte.

(c)

(a)

172 Tribology - Fundamentals and Advancements

(b)

better support the applied load.

The biological response to titanium depends on the surface chemical composition and the ability of titanium oxides to absorb molecules and incorporate elements [92] . Surface topography plays a fundamental role in regulating cell behaviour, e.g. the shape, orientation and adhesion of cells [93] [94] . As a surface begins to contact with biological tissues, water molecules first reach the surface. Hence, surface wettability, initially, may play a major role in adsorption of proteins onto the surface, as well as cell The authors acknowledge financial support from the Spanish Ministry of Science and Inno‐ vation obtained in the project: CSD2008-00023 FUNCOAT (in the frame of the CONSOLIDER INGENIO-2010 program) and from the Basque Government.

#### **Author details**

Virginia Sáenz de Viteri\* and Elena Fuentes

\*Address all correspondence to: virginia.saenzdeviteri@tekniker.es

IK4-Tekniker, Eibar, Spain

#### **References**

[1] Korkusuz, P. & Korkusuz, F. Hard Tissue – Biomaterial Interactions. In: Michael J. Yaszemski; Debra J. Trantolo; Kai-Uwe Lewandrowski; Vasif Hasirci, David E. Alto‐ belli & Donald L. Wise. (ed.) Biomaterials in Orthopedics. United States of America: Marcel Dekker, Inc.; 2004. p1-40.

[12] Heimke, G. & Stock, D. Clinical application of ceramic osseo – or soft tissue - inte‐

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

175

[13] Black, J. & Hastings G.W. Handbook of biomaterials properties. London UK: Chap‐

[14] Lawrence Katz, J. Anisotropy of Young´s modulus of bone. Nature 1980;283 106-107.

[15] Viceconti, M., Muccini, R., Bernakiewicz, M., Baleani M. & Cristofolini, L. Large-slid‐ ing contact elements accurately predict levels of bone-implant micromotion relevant

[16] Wennerberg, A. On surface roughness and implant incorporation. Göteborg. Swe‐ den: Biomaterials/Handicap Research; Institute of Surgical Sciences, Göteborgs Uni‐

[17] Calrsson, L.V., Macdonald, W., Magnus Jacobsson, C. &. Albrektsson T. Osseointe‐ gration Principles in Orthopedics: Basic Research and Clinical Applications. In: Mi‐ chael J. Yaszemski; Debra J. Trantolo; Kai-Uwe Lewandrowski; Vasif Hasirci, David E. Altobelli & Donald L. Wise. (ed.) Biomaterials in Orthopedics. United States of

[18] Junker, R., Dimakis, A., Thoneick, M. &. Jansen, J.A. Effects of implant surface coat‐ ings and composition on bone integration: a systematic review. Clinical Oral Im‐

[19] Kim, H.J., Kim, S.H., Kim, M.S., LeeJ, E.J., Oh, H.G., Oh, W.M., et al. Varying Ti-6Al-4V surface roughness induces different early morphologic and molecular re‐ sponses in MG63 osteoblast-like cells. Journal of Biomedical Materials Research

[20] Vlacic-Zischke, J., Hamlet, S.M., Friis, T., Tonetti, M.S. & Ivanovski, S. The influence of surface microroughness and hydrophilicity of titanium on the up-regulation of

[22] Montanaro, L., Campoccia, D. & Arciola, C.R. Nanostructured materials for inhibi‐ tion of bacterial adhesion in orthopedic implants: a minireview. International Journal

[23] Black. J. Othopaedic Biomaterials in Research and Practice. New York: Churchill Liv‐

[24] Jacobs, J.J., Gilbert, J.L. & Urban, R.M. Corrosion of metal orthopaedic implants. Jour‐

[25] Hallab, N.J., Urban, R. M. & Jacobs, J.J. (2004). Corrosion and Biocompatibility of Or‐ thopedic Implants, In: Michael J. Yaszemski; Debra J. Trantolo; Kai-Uwe Lewan‐

TGFb/BMP signalling in osteoblasts. Biomaterials 2011;32 665-671.

nal of Bone and Joint Surgery – American Volume 1998;80 268-282.

[21] http://360oandp.com/Technology-Osseointegration.aspx

of Artificial Organs 2008;31 771-776.

grated implant. Orthopedic Ceramic Implants 1984;4 1-19.

to osseointegration. Journal of Biomechanics 2000;33 1611-1618.

America: Marcel Dekker, Inc.; 2004 p223-240.

plants Research 2009;20 185-206.

2005,74A 366-373.

ingstone; 1988.

man and Hall; 1998.

versitet, 1996.


[12] Heimke, G. & Stock, D. Clinical application of ceramic osseo – or soft tissue - inte‐ grated implant. Orthopedic Ceramic Implants 1984;4 1-19.

**Author details**

**References**

Virginia Sáenz de Viteri\*

174 Tribology - Fundamentals and Advancements

IK4-Tekniker, Eibar, Spain

1992;63 225-232.

2001;61(9) 1189-1224.

397-425.

1996;213 134-137.

1941-2953.

Marcel Dekker, Inc.; 2004. p1-40.

ton, Florida: CC Press; 2003 p. 1-241.

 and Elena Fuentes \*Address all correspondence to: virginia.saenzdeviteri@tekniker.es

[1] Korkusuz, P. & Korkusuz, F. Hard Tissue – Biomaterial Interactions. In: Michael J. Yaszemski; Debra J. Trantolo; Kai-Uwe Lewandrowski; Vasif Hasirci, David E. Alto‐ belli & Donald L. Wise. (ed.) Biomaterials in Orthopedics. United States of America:

[2] Santavirta, S., Gristina, A. & Konttinen, YT. Cemented versus cementless hip arthro‐ plasty: a review of prosthetic biocompatibility. Acta Orthopaedica Scandinavica

[3] Park, J.B. & Bronzino, L.D., (ed.) Biomaterials: principles and applications. Boca Ra‐

[4] Ramakrishna, S., Mayer, J., Wintermantel, E. & Leong K. W. Biomedical applications of polymer-composite materials: A review. Composites Science and Technology

[6] Kurtz, S., Ong, K., Jau, E., Mowat, F. & Halpern, M. Projections of primary and revi‐ sion hip and knee arthroplasty in the United States from 2005 to 2030. Journal of

[7] Geetha, M., Singh, A.K., Asokamani, R. & Gogia, A.K. Ti based biomaterials, the ulti‐ mate choice for orthopaedic implants- A review. Progress in Material Science 2009;54

[8] Long, M. & Rack. H.J. Titanium alloys in total joint replacement – A materials science

[9] Wang, K. The use of titanium for medical applications in the USA. Materials Science and Engineering A - Structural Materials Properties Microstructure and Processing

[11] Williams, D.F. On the mechanisms of biocompatibility. Biomaterial 2008;29(20)

[5] Wise, D.L. Biomaterials engineering and devices. Berlin: Human Press; 2000.

Bone and Joint Surgery – American Volume 2007;89 780-785.

perspective. Biomaterials 1998;19 1621-1639.

[10] http://azom.com/article.aspx?ArticleID=108


drowski; Vasif Hasirci, David E. Altobelli & Donald L. Wise. (ed.) Biomaterials in Orthopedics. United States of America: Marcel Dekker, Inc.; 2004 p63-92.

[39] Neoh, K.G., Hu, X., Zheng, D. & Tang Kang, E. Balancing osteoblast functions and bacterial adhesion on functionalized titanium surfaces. Biomaterials 2012;33

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

177

[40] Zhu, X., Chen, J., Scheideler, L., R. Reichl, R. & Geis-Gerstorfer, J. Effects of topogra‐ phy and composition of titanium surface oxides on osteoblast responses. Biomateri‐

[41] Liu, F., Wang, F., Shimizu, T., Igarashi, K. & Zhao, L. (2005). Formation of hydroxya‐ patite on Ti-6Al-4V alloy by microarc oxidation and hydrothermal treatment. Surface

[42] Kokubo, T., Kim, H-M., Miyaji, F. & Nakamura, T. Preparation of bioactive Ti and its alloys via simple chemical surface treatment. Journal of Biomedical Materials Re‐

[43] Carapeto, A.P., Serro, A.P., Nunes, B.M.F., Martins, M.C.L., Todorovic, S., Duarte, M.T., André, V., Colaço, R. & Saramago, B. Characterization of two DLC coatings for joint prosthesis: The role of albumin on the tribological behavior. Surface & Coatings

[44] Ma, G., Gong, S., Lin, G., Zhang, L. & Sun, G. A study of structure and properties of Ti-doped DLC film by reactive magnetron sputtering with ion implantation. Applied

[45] Dowling, D.P. Evaluation of diamond-like carbon coated orthopedic implants. Dia‐

[46] Zhang, L., Lv, P., Huang, Z.Y., Lin, S.P., Chen, D.H., Pan, S.R. & Chen, M. Blood compatibility of La2O3 doped diamond-like carbon films. Diamond and Related Ma‐

[47] Zheng, Y., Liu, D., Liu, X. & Li, L. Ti-TiC-TiC/DLC gradient nano-composite film on a

[48] Han, Y., Hong, S.H. & Xu, K.W. Porous nanocrystalline titania films by plasma elec‐

[49] Huang, P., Wang, F., Xu, K. & Han, Y. Mechanical properties of titania prepared by plasma electrolytic oxidation at different voltages. Surface & Coatings Technology

[50] Lange, R., Lüthen, F., Beck, U., Rychly, J., Baumann, A. & Nebe, B. Cell-extracellular matrix interaction and physic-chemical characteristics of titanium surfaces depend

[51] Huang, P., Xu, K-W. & Han, Y. Preparation and apatite layer formation of plasma electrolytic oxidation film on titanium for biomedical application. Materials Letters

on the roughness of the material. Biomolecular Engineering 2002;19 255-261.

biomedical NiTi alloy. Biomedical Materials 2008;3 044103-044109.

trolytic oxidation. Surface & Coatings Technology 2002;154 314–318.

2813-2822.

als 2004;25 4087-4103.

search 1996;32 409-417.

Technology 2010;204 3451-3458.

Surface Science 2012;258 3045-3050.

terials 2008;17 1922-1926.

2007; 201 5168–5171.

2005;59 185-189.

mond and Related Materials 1997;6 390-393.

& Coatings Technology 2005;199 220-224.


[39] Neoh, K.G., Hu, X., Zheng, D. & Tang Kang, E. Balancing osteoblast functions and bacterial adhesion on functionalized titanium surfaces. Biomaterials 2012;33 2813-2822.

drowski; Vasif Hasirci, David E. Altobelli & Donald L. Wise. (ed.) Biomaterials in

[26] Liu, X., Chu, P.K. & Ding C. (2004). Surface modification of titanium, titanium alloys, and related materials for biomedical applications. Mater Sci Eng, Vo. R 47, (2004), pp.

[27] Stadlinger, B., Ferguson, S.J., Eckelt, U., Mai, R., Lode, A.T., Loukota, R. & Sclotting F. Biomechanical evaluation of a titanium implant surface conditioned by a hydrox‐

[28] Subramani, K. & Mathew, R.T. Titanium Surface Modification. Techniques for Dental Implants – From Microscale to Nanoscale. Emerging Nanotechnologies in Dentistry.

[30] Fraczek, T., Olejnik, M. & Tokarz, A. Evaluations of plasma nitriding efficiency of ti‐

[31] Kustas, F.M. & Misra, M.S. Friction and Wear of Titanium Alloys, In: Scott D. Henry (ed.) Volume 18, Friction, Lubrication and Wear Technology. United States of Ameri‐

[32] Freese, H., Volas, M.G. & Wood, J.R. (2001). In: Brunette D.M., Tengvall P., Textor M.,

[33] Yerokhin, A.L., Niea, X., Leyland, A. & Matthews, A. Characterization of oxide films produced by plasma electrolytic oxidation of a Ti–6Al–4V alloy. Surface & Coating

[34] Malchau, H. & Herberts, P. Revision and re-revision rate in THR: a revision-risk study of 148,359 primary operations. Scientific exhibition, 65th annual meeting of the

[35] Chandra, A., Ryu, J.J., Karra, P., Shrotriya, P., Tvergaard,V., Gaisser, M. & Weik, T. Life expectancy of modular Ti6Al4V hip implants: Influence of stress and environ‐ ment. Journal of the Mechanical Behavior of Biomedical Materials 2011;4 1990-2001.

[36] Wolford, L.M. Factors to consider in joint prosthesis systems, Proceedings (Baylor

[37] Zabinski, J.S. & Voevodin, A.A. Ceramic and other hard coatings, In: Joze Vizintin, Mitjan Kalin, Kuniaki Dohda & Said Jahanmir (eds) Trybology of Mechanical Sys‐ tems: A Guide to Present and Future Technologies. United States of America: ASME

[38] Landolt, D. Corrosion and Surface Chemistry of Metals. Lausanne Switzerland: EPFL

Thomsen P. (eds.) Titanium in Medicine. Springer: Berlin; 2001. p25-51.

ide ion solution. British Journal of Oral & Maxillofacial Surgery 2012;50 74-79.

[29] http://hss.edu/conditions\_arthritis-of-the-knee-total-knee-replacement.asp

tanium alloys for medical applications. Metalurgija 2009;48(2) 83-86.

DOI: 10.1016/B978-1-4557-7862-1.00006-7.

ca: ASM International; 1992. p. 1585-1598.

University Medical Center) 2006;19 232-238.

Technology 2000;130 195–206.

AAOS, New-Orleans, 1998.

Press; 2004 p157-182.

Press; 2007.

Orthopedics. United States of America: Marcel Dekker, Inc.; 2004 p63-92.

49-121.

176 Tribology - Fundamentals and Advancements


[52] Cigada, A., Cabrini, M. & Pedferri, P. Increase of the corrosion resistance of Ti6Al4V alloy by high thickness anodic oxidation. Journal of Materials Science – Materials in Medicine 1992;3 408-412.

[66] Donnet, C. & Erdemir, A. Diamond-like Carbon Films: A Historical Overview. In: Christophe Donnet & Ali Erdemir (ed.) Tribology of Diamond-Like Carbon Films.

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

179

[67] Hauert, R., Knoblauch-Meyer, L., Francz, G., Schroeder, A. & Wintermantel, E. Sur‐

[68] Ronkainen, H. & Holmberg, K. Environmental and Thermal Effects on the Tribologi‐ cal Performance of DLC Coatings. In: Christophe Donnet & Ali Erdemir (ed.) Tribol‐

[69] Sáenz de Viteri, V., Barandika, M.G., Ruiz de Gopegui, U., Bayón, R., Zubizarreta, C., Fernández, X., Igartua, A. & Agullo-Rueda, F. Characterization of Ti-C-N coatings deposited on Ti6Al4V for biomedical applications. Journal of Inorganic Biochemistry

[70] Williams, D.F. Titanium and titanium alloys. In: Williams DF (ed.) Biocompatibility of clinical implant materials, Vol. I. Boca Raton, Florida: CRC Press, Inc; 1981. p. 9-44

[71] Lampin, M., Warocquier-Clerout, R., Legris, C., Degrange, M. & Sigot-Luizard, M.F. Correlation between substratum roughness and wettability, cell adhesion, and cell

[72] Lim, Y.J., Oshida, Y., Andres, C.J. & Barco, M.T. Surface characterization of variously treated titanium materials. International Journal of Oral & Maxillofacial Implants

[73] Keller, J.C., Stanford, C.M., Wightman, J.P., Draughn, R.A. & Zaharias, R. Characteri‐ zation of titanium implant surfaces. III. Journal of Biomedical Materials Research

[74] Kieswetter, K., Schwartz, Z., Dean, D.D. & Boyan, B.D. The role of implant surface characteristic in the healing of bone. Critical Reviews in Oral Biology & Medicine

[75] Pouilleau, J., Devilliers, D., Garrido, F., Durand-Vidal, S. & Mahe, E. Structure and composition of passive titanium oxide films. Materials Science and Engineering

[76] Fei, C., Hai, Z., Chen, C. & Yangjian, X. Study on the tribological performance of ce‐ ramic coatings on titanium alloy surfaces obtained through microarc oxidation. Prog‐

[77] Kuromoto, N.K., Simão, R.A. & Soares, G.A. Titanium oxide films produced on com‐ mercially pure titanium by anodic oxidation with different voltages. Materials Char‐

ogy of Diamond-Like Carbon Films. USA: Springer; 2008. p155-200.

migration. Journal of Biomedical Materials Research 1997;36 99-108.

USA: Springer; 2008. p1-12.

2012;117 359-366.

2001;16 333–342.

1994;28 939–946.

1996;7 329–345.

1997;B47 235–243.

ress in Organic Coatings 2009;64 264–267.

acterization 2007;58 114–121.

face & Coatings Technology 1999;120-121 291-296.


[66] Donnet, C. & Erdemir, A. Diamond-like Carbon Films: A Historical Overview. In: Christophe Donnet & Ali Erdemir (ed.) Tribology of Diamond-Like Carbon Films. USA: Springer; 2008. p1-12.

[52] Cigada, A., Cabrini, M. & Pedferri, P. Increase of the corrosion resistance of Ti6Al4V alloy by high thickness anodic oxidation. Journal of Materials Science – Materials in

[53] Yerokhin, A.L., Nie, X., Leyland, A., Matthews, A. & Dowey, S.J. Surface & Coatings

[54] Nanci, A., Wuest, J.D., Peru, L., Brunet, P., Sharma, V., Zalzal, S. & McKee, M.D. Chemical modification of titanium surfaces for covalent attachment of biological

[55] Viorney, C., Guenther, H.L., Aronsson, B.O., Pechy, P., Descouts, P. & Gratzel, M. (2002). Osteoblast culture on polished titanium disks modified with phosphoric

[56] Sul, Y.T., Johansson, C.B., Kang, Y., Jeon, D.G., Kang, Y., Jeong, D.G. & Albrektsson, T. Bone reaction to oxidized titanium implants with electrochemical anion sulphuric acid and phosphoric acid incorporation. Clinical Implant Dentistry and Related Re‐

[57] Cook, S.D., Thomas, K.A., Kay, J.F. & Jarcho, M. Hydroxyapatite-coated porous tita‐ nium for use as an orthopedic biologic attachment system. Clinical Orthopaedic and

[58] Rashmir- Raven, M.A., Richardson, D.C., Aberman, H.M. & DeYoung, D.J. The re‐ sponse of cancellous and cortical canine bone to hydroxyapatite-coated and uncoated

[59] Holmberg, K. & Matthews, A. Coatings Tribology – Properties, Techniques and Ap‐ plications in Surface Engineering. Amsterdam, The Netherlands: Elsevier; 1994.

[62] Erdemir, A. & Donnet, C. Trybology of diamond, diamond-like carbon, and related films. In: B. Bhushan (ed.) Handbook of Modern Tribology, Vol. 2. Materials Coat‐

[63] Donnet, C., Belin, M., Augé, J.C., Martin, J.M., Grill, A. & Patel, V. Surface & Coatings

[64] Erdemir, A., Erylmaz, O.L. & Fenske, G. (2000). Journal of Vacuum Science of Tech‐

[65] Fontaine, J., Donnet, C. & Erdemir, A. Fundamentals of the Tribology of DLC Coat‐ ings. In: Christophe Donnet & Ali Erdemir (ed.) Tribology of Diamond-Like Carbon

molecules. J Biomed Mater, Vol. 40, (1998), pp. 237-242.

acids. Journal of Biomedical Materials Research 2002;62 149-155.

titanium rods. Journal of Applied Biomaterials 1995;6 237-242.

[61] Grill, A. Surface & Coatings Technology 1997;94-95(1-3) 507.

nology A – Vacuum Surfaces and Films 2000;18(4) 1987.

ings. Boca Raton, Florida: CRC Press; 2001.

Films. USA: Springer; 2008. p139-154.

Medicine 1992;3 408-412.

Technology 1999;116.

178 Tribology - Fundamentals and Advancements

search 2002;4 78-87.

Related Research 1988;230 303-312.

[60] Grill, A. Wear 1993;168(1-2) 143.

Technology 1994;68-69 626.


[78] Wanga, Y., Jiang, B., Lei, T. & Guo, L. Dependence of growth features of microarc ox‐ idation coatings of titanium alloy on control modes of alternate pulse. Materials Let‐ ters 2004;58 1907–1911.

[93] Eriksson, C., Lausmaa, J. & Nygren, H. Interactions between human whole blood and modified TiO2-surfaces: influence of surface topography and oxide thickness on leu‐

Titanium and Titanium Alloys as Biomaterials

http://dx.doi.org/10.5772/55860

181

[94] Ravanetti, F., Borghetti, P., De Angelis, E., Chiesa, R., Martini, F.M., Gabbi, C. & Cac‐ chioli, A. (2010). In vitro cellular response and in vivo primary osteointegration of

[95] Ceschini, L., Lanzoni, E., Martini, C., Prandstraller, D. & Sambogna, G. Comparison of dry sliding friction and wear of Ti6Al4V alloys treated by plasma electrolytic oxi‐

electrochemically modified titanium. Acta Biomaterialia 2010;6 1014-1024.

kocyte adhesion and activation. Biomaterials 2001;22 1987–1996.

dation and PVD coating. Wear 2008;26 86-95.


[93] Eriksson, C., Lausmaa, J. & Nygren, H. Interactions between human whole blood and modified TiO2-surfaces: influence of surface topography and oxide thickness on leu‐ kocyte adhesion and activation. Biomaterials 2001;22 1987–1996.

[78] Wanga, Y., Jiang, B., Lei, T. & Guo, L. Dependence of growth features of microarc ox‐ idation coatings of titanium alloy on control modes of alternate pulse. Materials Let‐

[79] Han, Y., Hong, S. & Xu, K. Synthesis of nanocrystalline titania films by micro-arc oxi‐

[80] Kurze, P., Krysman, W., Dittrich, K.H. & Schneider, H.G. Process characteristics and parameters of anodic oxidation by spark deposition (ANOF). Crystal Research and

[81] Kurze, P., Dittrich, K.H., Krysman, W. & Schneider, H.G. Structure and properties of

[82] Han, I., Choi, J.H., Zhao, B.H., Baik, H.K. & Lee, I. Micro-arc oxidation in various concentration of KOH and structural change by different cut off potential. Current

[83] Matykina, E., Berkani, A., Skeldon, P. & Thompson, G.E. Real-time imaging of coat‐ ing growth during plasma electrolytic oxidation of titanium. Electrochimica Acta

[84] Wang, Y., Lei, T., Jiang, B. & Guo, L. Growth, microstructure and mechanical proper‐ ties of microarc oxidation coatings on titanium alloy in phosphate-containing solu‐

[87] Yerokhin, A.L., Nie, X., Leyland, A., Matthews, A. & Dowey, S.J. (1999). Surface &

[88] Shokouhfar, M., Dehghanian, C., Montazeri, M. & Baradaran, A. Preparation of ce‐ ramic coating on Ti substrate by plasma electrolytic oxidation in different electrolytes and evaluation of its corrosion resistance: Part II. Applied Surface Science 2012;258

[89] Amin, M.S., Randeniya, L.K., Bendavid, A., Martin, P.J. & Preston, E.W. Amorphous carbonated apatite formation on diamond-like carbon containing titanium oxide. Di‐

[90] Yang, B., Uchida, M., Kim, H-M., Zhang, X. & Kokubo, T. Preparation of bioactive titanium metal via anodic oxidation treatment. Biomaterials 2004;25 1003-1010.

[91] Han, Y., Sun, J. & Huang, X. Formation mechanism of HA-based coatings by micro-

[92] Letic-Gavrilovic, A., Scandurra, R. & Abe, K. Genetic potential of interfacial guided osteogenesis in implant devices. Dental Materials Journal 2000;19 99–132.

arc oxidation. Electrochemistry Communications 2008;10 510-513.

ANOF layers. Crystal Research and Technology 1984;19 93–99.

ters 2004;58 1907–1911.

180 Tribology - Fundamentals and Advancements

Technology 1984;19 973–979.

Applied Physics 2007;7S1 23–27.

tion. Applied Surface Science 2004;233 258–267.

amond and Related Materials 2009;18 1139-1144.

Coatings Technolog 1999;122 73–93.

[85] Shibata, T. & Zhu, Y. C., Corrosion Science 1995;37(1) 133–144.

[86] Shibata, T. & Zhu, Y. C., Corrosion Science 1995;37(2) 253–270.

2007;53 1987–1994.

2416-2423.

dation. Materials Letters 2002;56 744–747.


**Section 3**

**Testing and Modeling**

### **Testing and Modeling**

**Chapter 6**

**New Scuffing Test Methods**

**for the Determination of the**

**Scuffing Resistance of Coated Gears**

Remigiusz Michalczewski, Marek Kalbarczyk,

Marian Szczerek, Waldemar Tuszynski and

Additional information is available at the end of the chapter

In modern machines the problems of the prevention of scuffing of the gear teeth is still very important. One of the reasons is that for many years the technique development is related to increasing the loading of the friction surfaces accompanied by decreasing their size [1]. In the case of gears, the risk of scuffing occurrence rises because of potential design and assembly mistakes, unexpected overloads, as well as extremely different speeds of the rotation of gears, because both very high speeds and very low speeds may cause scuffing [2]. The occurrence of

Apart from the above mentioned factors, the problems of using proper lubricating oils, with

In gears, the surface destroyed by scuffing appears at the addendum and dedendum of the tooth. This results from the sliding speed of the meshing teeth that reaches the highest values

> © 2013 Michalczewski et al.; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2013 Michalczewski et al.; licensee InTech. This is a paper distributed under the terms of the Creative Commons

one of the mentioned factors may lead to very serious gear failures.

Failures of the gear teeth flanks due to scuffing are shown in Figure 1.

high extreme-pressure (EP) properties cannot be neglected.

Michal Michalak, Witold Piekoszewski,

Jan Wulczynski

**1. Introduction**

**1.1. Scuffing of gear teeth**

at these places of the gear tooth.

http://dx.doi.org/10.5772/54569

#### **Chapter 6**

### **New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears**

Remigiusz Michalczewski, Marek Kalbarczyk, Michal Michalak, Witold Piekoszewski, Marian Szczerek, Waldemar Tuszynski and Jan Wulczynski

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/54569

**1. Introduction**

#### **1.1. Scuffing of gear teeth**

In modern machines the problems of the prevention of scuffing of the gear teeth is still very important. One of the reasons is that for many years the technique development is related to increasing the loading of the friction surfaces accompanied by decreasing their size [1]. In the case of gears, the risk of scuffing occurrence rises because of potential design and assembly mistakes, unexpected overloads, as well as extremely different speeds of the rotation of gears, because both very high speeds and very low speeds may cause scuffing [2]. The occurrence of one of the mentioned factors may lead to very serious gear failures.

Apart from the above mentioned factors, the problems of using proper lubricating oils, with high extreme-pressure (EP) properties cannot be neglected.

In gears, the surface destroyed by scuffing appears at the addendum and dedendum of the tooth. This results from the sliding speed of the meshing teeth that reaches the highest values at these places of the gear tooth.

Failures of the gear teeth flanks due to scuffing are shown in Figure 1.

© 2013 Michalczewski et al.; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Michalczewski et al.; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

concern the contact between two balls of the four-ball tribosystem (the rotating upper ball with one of the three stationary lower balls) during the testing of the automotive gear oils of API GL-4 and GL-5 performance levels. Such oils contain chemically active extreme-pressure (EP) lubricating additives to prevent scuffing. API GL-4 oils are used to lubricate synchronised manual transmissions of European cars and contain up to 4% of EP additives. API GL-5 oils containing up to 6.5% of EP additives are employed to lubricate automotive gears especially susceptible to scuffing, i.e. hypoid gears, in axles operating under various combinations of

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

It should be emphasised here that a four-ball tribosystem is very often used for tribological

increasing load (P). The brackets over the graph indicate particular phases of the scuffing process. In these phases, the friction coefficient values (µ) were determined, and they are given in the red rectangles in the graph area. The thick red line below the graph denotes the time from the beginning of the run until the occurrence of the scuffing initiation reflected by a sharp

The interpretative models of phenomena related to scuffing are presented over the graph in Figure 3. Because the models concern the contact zone between two balls of the four-ball tribosystem where the upper ball rotates and the lower ball is stationary, the direction of the movement was indicated in the upper part of the models by an arrow. If there is no arrow, the given model illustrates no movement of the balls, i.e. at the beginning of the run (before the

For the phase "scuffing initiation," the upper model in Figure 3 illustrates the surface that did not exhibit very rough topography typical of scuffing (shown in the surface topography

In the models, three characteristic zones in the wear scar surface layer were identified: a chemically modified zone through the action of the lubricating additives and the steel surface, a zone of plastic deformation, and a zone of elastic deformation. All of these zones are described

After immersing the test balls in the tested gear oil and applying the initial load close to 0, a phenomenon known as physical adsorption or physisorption appears. In this phase, adsorbed molecules constitute the boundary layer on the friction surface, which protects the surface asperities against direct contact. The model with the heading "Beginning of run" in Figure 3 illustrates this, reflecting the situation before the start of the relative

image), while the lower one concerns the surface already destroyed by scuffing.

) obtained at continuously

http://dx.doi.org/10.5772/54569

187

high-speed/shock-load and low-speed/high-torque conditions.

The lower graph in Figure 3 presents the friction torque curve (Mt

testing of the performance of automotive gear oils.

rise in the friction torque.

motor of the tribotester starts).

*1.2.1. Phase: "Beginning of run"*

movement of the test balls.

in the legend above the models in Figure 3.

**Figure 1.** Photographs of failures of the gear teeth flanks due to scuffing: a) "non-symmetric" scuffing observed in gear service, resulting from the incorrect distribution of load along the tooth [3], b) scuffing on the flank of the test gear due to poor extreme-pressure (EP) properties of the tested gear oil during the gear scuffing experiments per‐ formed by the authors

Another example of scuffing of gears concerns the rudder speed brake power drive unit of a space shuttle, observed during its inspection after grounding [4]. Figure 2 a) shows the pinion and ring gear of the power drive unit of the space shuttle. Figure 2 b) presents the pinion tooth with wear at the tip and scuffing on dedendum. It was postulated that early shutdown of one of three hydraulic motors driving the gearbox could cause scuffing - in a differential gearbox, early shutdown of one motor could cause the overloading with potential for scuffing.

**Figure 2.** Photographs of the components of the rudder speed brake power drive unit of a space shuttle: a) pinion and ring gear of the power drive unit, b) damaged pinion tooth [4]

From the above example, it is absolutely apparent that the prevention of scuffing is still an important challenge, even in the high-tech sector.

#### **1.2. Scuffing — How is it brought about?**

To better understand scuffing, Figure 3 presents the interpretative models of the phenomena in different phases of this process, caused by the continuously increasing load. The models concern the contact between two balls of the four-ball tribosystem (the rotating upper ball with one of the three stationary lower balls) during the testing of the automotive gear oils of API GL-4 and GL-5 performance levels. Such oils contain chemically active extreme-pressure (EP) lubricating additives to prevent scuffing. API GL-4 oils are used to lubricate synchronised manual transmissions of European cars and contain up to 4% of EP additives. API GL-5 oils containing up to 6.5% of EP additives are employed to lubricate automotive gears especially susceptible to scuffing, i.e. hypoid gears, in axles operating under various combinations of high-speed/shock-load and low-speed/high-torque conditions.

It should be emphasised here that a four-ball tribosystem is very often used for tribological testing of the performance of automotive gear oils.

The lower graph in Figure 3 presents the friction torque curve (Mt ) obtained at continuously increasing load (P). The brackets over the graph indicate particular phases of the scuffing process. In these phases, the friction coefficient values (µ) were determined, and they are given in the red rectangles in the graph area. The thick red line below the graph denotes the time from the beginning of the run until the occurrence of the scuffing initiation reflected by a sharp rise in the friction torque.

The interpretative models of phenomena related to scuffing are presented over the graph in Figure 3. Because the models concern the contact zone between two balls of the four-ball tribosystem where the upper ball rotates and the lower ball is stationary, the direction of the movement was indicated in the upper part of the models by an arrow. If there is no arrow, the given model illustrates no movement of the balls, i.e. at the beginning of the run (before the motor of the tribotester starts).

For the phase "scuffing initiation," the upper model in Figure 3 illustrates the surface that did not exhibit very rough topography typical of scuffing (shown in the surface topography image), while the lower one concerns the surface already destroyed by scuffing.

In the models, three characteristic zones in the wear scar surface layer were identified: a chemically modified zone through the action of the lubricating additives and the steel surface, a zone of plastic deformation, and a zone of elastic deformation. All of these zones are described in the legend above the models in Figure 3.

#### *1.2.1. Phase: "Beginning of run"*

(a) (b)

**Figure 1.** Photographs of failures of the gear teeth flanks due to scuffing: a) "non-symmetric" scuffing observed in gear service, resulting from the incorrect distribution of load along the tooth [3], b) scuffing on the flank of the test gear due to poor extreme-pressure (EP) properties of the tested gear oil during the gear scuffing experiments per‐

Another example of scuffing of gears concerns the rudder speed brake power drive unit of a space shuttle, observed during its inspection after grounding [4]. Figure 2 a) shows the pinion and ring gear of the power drive unit of the space shuttle. Figure 2 b) presents the pinion tooth with wear at the tip and scuffing on dedendum. It was postulated that early shutdown of one of three hydraulic motors driving the gearbox could cause scuffing - in a differential gearbox,

early shutdown of one motor could cause the overloading with potential for scuffing.

(a) (b)

**Figure 2.** Photographs of the components of the rudder speed brake power drive unit of a space shuttle: a) pinion

From the above example, it is absolutely apparent that the prevention of scuffing is still an

To better understand scuffing, Figure 3 presents the interpretative models of the phenomena in different phases of this process, caused by the continuously increasing load. The models

and ring gear of the power drive unit, b) damaged pinion tooth [4]

important challenge, even in the high-tech sector.

**1.2. Scuffing — How is it brought about?**

formed by the authors

186 Tribology - Fundamentals and Advancements

After immersing the test balls in the tested gear oil and applying the initial load close to 0, a phenomenon known as physical adsorption or physisorption appears. In this phase, adsorbed molecules constitute the boundary layer on the friction surface, which protects the surface asperities against direct contact. The model with the heading "Beginning of run" in Figure 3 illustrates this, reflecting the situation before the start of the relative movement of the test balls.

*1.2.2. Phase: "Mixed friction"*

specific friction coefficient.

friction torque (Mt

low load and high speed.

surface layer.

the heading "Mixed friction" in Figure 3.

friction.

In publications, the terms "mixed friction" and "mixed lubrication" are often used equivalently and concern the same phenomena. For the purpose of this chapter, one can assume that occurrences during the regime of the mixed lubrication result in the mixed friction with its

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

189

The phase "Mixed friction" concerns the first stage of the run from the moment of the start of the relative movement between the test balls to the scuffing initiation reflected by a sharp rise in the friction torque. Its duration is denoted by the thick red line below the graph with the

In this phase the mixed friction occurs. This can be stated on the basis of the fundamental criterion that is the friction coefficient value. The friction coefficients typical of particular types of friction were adopted from the work [5], where the four-ball tribosystem was also employed. From that work, it implies that the mixed friction occurs in the four-ball tribosystem when the friction coefficient is in the range between 0.07 and 0.1. Thus, the authors determined the friction coefficient at the 2nd second of the four-ball experiment, being 0.1, denote the mixed

It is worth noting here that the idea of the occurrence of the mixed friction regime (instead of EHL, i.e. elastohydrodynamic lubrication) at the very start of the relative movement between the test balls (load is close to 0) is also supported in the mentioned work [5]. From that work it is apparent that "pure" EHL occurs in the four-ball tribosystem only under conditions of a

At mixed friction, the micro-EHL films mainly carry the load and the mating surfaces are protected from direct contact by the boundary layer. But at some micro-zones, due to the failure of the micro-EHL film surface, asperities locally collide, which is illustrated in the model with

Due to collisions of surface asperities, the temperature in the micro-contact rises. At a higher temperature, physically adsorbed molecules may be attracted to the surface with greater forces, and chemical adsorption or chemisorption appears. The decomposition of the active compounds in the lubricating additives catalyses the transformation of some chemically

The collision of the surface asperities and the local high pressure of the oil induced by the approaching asperities bring about elastic (reversible) and plastic (irreversible) deformations of the contacting surface. Due to the thermal (temperature rise) and mechanical activation (plastic deformation causing surface defects), the conditions exist for the initiation of the diffusion of "active" atoms from the lubricating compounds (e.g. sulphur atoms) into the

The described phenomena lead to the formation of inorganic chemical compounds of iron with sulphur, phosphorus, and oxygen, coming from EP lubricating additives in the tested gear oil. Such additives (based on organic S-P compounds) form e.g. iron sulphide FeS [6]. FeS compounds, apart from hampering the creation of adhesive bonds with their shear strength

adsorbed molecules into chemical compounds at higher temperatures.

) and applied load (P) - Figure 3.

Figure 3. Models of scuffing in different phases for the automotive gear oils of the API GL-4 and GL-5 performance levels **Figure 3.** Models of scuffing in different phases for the automotive gear oils of the API GL-4 and GL-5 performance levels

In publications, the terms "mixed friction" and "mixed lubrication" are often used equivalently and concern the same phenomena. For the purpose of this chapter, one can assume that occurrences during the regime of the mixed lubrication result in the mixed

**1.2.2. Phase: "Mixed friction"** 

friction with its specific friction coefficient.

#### *1.2.2. Phase: "Mixed friction"*

Figure 3. Models of scuffing in different phases for the automotive gear oils of the API GL-4 and GL-5 performance levels

**Figure 3.** Models of scuffing in different phases for the automotive gear oils of the API GL-4 and GL-5 performance

**GL-4/GL-5**

**Mt**

= 0.12

0 2 4 6 8 10 12 14 16 18 Time in s

**. . . . . . .**

**.**

**Scuffing**

**P**

= 0.11

= 0.15

Applied load, P in N

**Scuffing initiation** *microscale* **Mixed friction**

Boundary layer Elastically deformed zone

Plastically deformed zone

*(microscale)*

**. . . . . .**

**. . . .**

**. . . .**

**. . . . . .**

**. . . .** **. . . .**

**. . . .**

**. . . .**

**. . .**

**. . .**

**. . . .**

**. . .**

**.**

**. . .**

**.**

... ... ... . .

> **. .**

**. . .** **. . .**

**. . .**

**. .**

**. .** **.. . .**

**.**

*Upper ball - rotating*

**. . . . . . . . .**

**. . . .**

**. . . .** **. . . .**

**. . . .**

**. . .**

**. . .**

> **. . . .**

**. .**

**.**

**. . .**

**. . . . . . .**

**. . . .**

**. . .**

**. . .**

**. . .**

**.**

**.**

**.. . . . . . . . .**

**. . .**

**. . .**

**. .**

**. . .**

**. . . . . . . .**

**. . . . . . .**

**. . . .**

**. . . .**

**.**

**.**

**. .**

**. .**

**. . . .**

**. . . .**

**Scuffing propagation**

**. . . . . . . .**

**. . . .**

**. .**

**. .**

**.**

**. .**

**. . . . . . .**

**. . . .**

**. . . .**

**. . . .**

**. .**

**. . . .**

**. . .**

**. . . .**

**. . . .**

**. . . . . . . . . .**

> **. . . .**

**. . . .**

**. . . .**

**. . . .**

**.**

**. .**

**.**

**.**

**.**

**.**

**.**

**. . .**

**. . .**

**. . .**

**.**

Core materials

*Lower balls (3) - stationary*

Zone of plastic deformation and chemical modification

**. . .. .**

> **. . . .**

**. . . .** **. . . .**

**. . . .**

**. . .**

**. . . . . .**

**. . .**

**. .**

*Contact zone*

Lubricating oil Chemically modified zone

Desorbed molecules

188 Tribology - Fundamentals and Advancements

**Beginning of run** *(microscale)*

**Mixed friction** *(microscale)*

In publications, the terms "mixed friction" and "mixed lubrication" are often used equivalently and concern the same phenomena. For the purpose of this chapter, one can assume that occurrences during the regime of the mixed lubrication result in the mixed

**1.2.2. Phase: "Mixed friction"** 

levels

= 0.10

Friction torque, Mt in Nm

friction with its specific friction coefficient.

In publications, the terms "mixed friction" and "mixed lubrication" are often used equivalently and concern the same phenomena. For the purpose of this chapter, one can assume that occurrences during the regime of the mixed lubrication result in the mixed friction with its specific friction coefficient.

The phase "Mixed friction" concerns the first stage of the run from the moment of the start of the relative movement between the test balls to the scuffing initiation reflected by a sharp rise in the friction torque. Its duration is denoted by the thick red line below the graph with the friction torque (Mt ) and applied load (P) - Figure 3.

In this phase the mixed friction occurs. This can be stated on the basis of the fundamental criterion that is the friction coefficient value. The friction coefficients typical of particular types of friction were adopted from the work [5], where the four-ball tribosystem was also employed. From that work, it implies that the mixed friction occurs in the four-ball tribosystem when the friction coefficient is in the range between 0.07 and 0.1. Thus, the authors determined the friction coefficient at the 2nd second of the four-ball experiment, being 0.1, denote the mixed friction.

It is worth noting here that the idea of the occurrence of the mixed friction regime (instead of EHL, i.e. elastohydrodynamic lubrication) at the very start of the relative movement between the test balls (load is close to 0) is also supported in the mentioned work [5]. From that work it is apparent that "pure" EHL occurs in the four-ball tribosystem only under conditions of a low load and high speed.

At mixed friction, the micro-EHL films mainly carry the load and the mating surfaces are protected from direct contact by the boundary layer. But at some micro-zones, due to the failure of the micro-EHL film surface, asperities locally collide, which is illustrated in the model with the heading "Mixed friction" in Figure 3.

Due to collisions of surface asperities, the temperature in the micro-contact rises. At a higher temperature, physically adsorbed molecules may be attracted to the surface with greater forces, and chemical adsorption or chemisorption appears. The decomposition of the active compounds in the lubricating additives catalyses the transformation of some chemically adsorbed molecules into chemical compounds at higher temperatures.

The collision of the surface asperities and the local high pressure of the oil induced by the approaching asperities bring about elastic (reversible) and plastic (irreversible) deformations of the contacting surface. Due to the thermal (temperature rise) and mechanical activation (plastic deformation causing surface defects), the conditions exist for the initiation of the diffusion of "active" atoms from the lubricating compounds (e.g. sulphur atoms) into the surface layer.

The described phenomena lead to the formation of inorganic chemical compounds of iron with sulphur, phosphorus, and oxygen, coming from EP lubricating additives in the tested gear oil. Such additives (based on organic S-P compounds) form e.g. iron sulphide FeS [6]. FeS compounds, apart from hampering the creation of adhesive bonds with their shear strength being 1/5th that of steel and their hardness being 1/4th that of steel, facilitate shearing of the chemically modified surface asperities, and the shear plane is transferred to the thin FeS layer, which protects the surface from tearing out the material from deeper layers, reducing the wear intensity.

*1.2.4. Scuffing phase: "Scuffing propagation"*

in the friction torque (Mt

Figure 3.

This phase refers to the scuffing process, after its initiation. It is reflected by a sharp increase

4 a, b). This situation is illustrated in the models with the heading "Scuffing propagation" in

(a) (b) (c)

**Figure 4.** Development of the wear of the lower test balls due to scuffing: a) at scuffing initiation, b) at 12th seconds of

For the tested gear oils, after the scuffing initiation due to rapid chemical reactions of their EP additives with the surface, a rise in the friction torque is mitigated to quickly stabilise at relatively low value - Figure 3. It is accompanied by continuously evolving wear of the lower test balls that is not intensive - Figure 4 b, c). A drop in the pressure in the contact zone due to wear, brings about the possibility of oil introduction into the contact zone and the regeneration of the boundary layer on much of the friction surface. Such an action is indicated by the friction coefficient within the range 0.11 to 0.15, typical of boundary friction. On the basis of the work [5], which also concerns four-ball experiments, it was assumed that the boundary friction occurs in the four-ball tribosystem when the friction coefficient is in the range between 0.09 and 0.15. The determined values of the friction coefficient being in the middle and upper limit typical of boundary friction denote that some part of the friction surface must have undergone scuffing; It can be assumed from [5] that "full scuffing" occurs when the friction coefficient exceeds 0.3. The specific state of the surface layer in this phase is called the "Secondary Boundary Layer" (SBL) in the work [8]. The round model in the micro-scale concerning the scuffing propagation (Figure 3) illustrates the places of oil appearance in the contact zone. Let us call them "the micro-pockets." One can presume that inside the oil micro-pockets the following phenomena take place: the intensive adsorption and desorption of the base oil and lubricating additives molecules on/from the steel surface, chemical reactions of the lubricating additives with the surface, and - in view of plastic deformation that causes surface defects the diffusion of "active" atoms from the lubricating compounds (e.g. sulphur atoms) into the surface layer. In view of the transfer and mingling of the material of the rubbing test balls, the chemically modified zones appear across the entire zone of plastic deformations - orange spots.

the run (scuffing propagation), c) at the end of the run; images obtained at the same magnification

), accompanied by a high intensity of the lower test balls wear - Figure

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

191

For the tested oil, containing EP lubricating additives, the surface asperities are covered by the protective layer of the above mentioned chemical compounds. This is illustrated in the respective model in Figure 3. Due to this, for the gear oils with EP lubricating additives, the scuffing initiation is delayed to appear at much higher loads than in the case of oils without lubricating additives (e.g. API GL-1 ones, not presented here).

#### *1.2.3. Scuffing phase: "Scuffing initiation"*

In this phase, scuffing initiates - the friction torque (Mt ) sharply increases and measured friction coefficient values exceed the maximum value assumed for the mixed friction, i.e. 0.1 [5].

The scuffing initiation occurs at a load called the scuffing load, which is characteristic for each tested lubricating oil. At this load, the lubricating film collapses, the number of colliding surface asperities drastically increases, and the destruction changes its occurrence from the micro- to macro-scale and scuffing appears. Initially only part of the friction surface undergoes scuffing. It can be observed in the surface topography image of the border between the surface that did not exhibit very rough topography typical of scuffing (left side) and the surface destroyed by scuffing (right side) - Figure 3.

The described phenomena leading to scuffing are illustrated in the models with the heading "Scuffing initiation" in Figure 3. The upper model concerns the surface that did not exhibit very rough topography typical of scuffing, where still the mixed friction exists, while the lower one refers to the surface already destroyed by scuffing.

The upper model shows that the micro-scale phenomena in the zone intact by scuffing are similar to those described in the phase "Mixed friction" apart from the thickness of elastic and plastic deformations which increased due to rising load. Probably, in view of plastic defor‐ mation that causes surface defects, the reactive diffusion of "active" atoms from the EP lubricating additives (e.g. sulphur atoms) into the surface layer takes place and iron sulphides form, which is confirmed by other researchers, e.g. in the work [7]. The diffusively modified micro-zones inside the highest asperities are plastically deformed and are indicated in the respective model as orange spots - Figure 3.

By observing phenomena in the part of the friction surface that undergoes scuffing, one can indicate that the situation changes radically. The lower model illustrates that, in the first phase of scuffing, the lubricating film no longer exists, nor is there any boundary layer. This leads to a rapid intensification of the material destruction. Much plastic deformation appears, turning into the transfer, flowing and mingling of the material of the rubbing test balls. For the tested oils with EP lubricating additives, much of the surface layer starts to be chemically modified. This will be decisive for the scuffing propagation character.

#### *1.2.4. Scuffing phase: "Scuffing propagation"*

being 1/5th that of steel and their hardness being 1/4th that of steel, facilitate shearing of the chemically modified surface asperities, and the shear plane is transferred to the thin FeS layer, which protects the surface from tearing out the material from deeper layers, reducing the wear

For the tested oil, containing EP lubricating additives, the surface asperities are covered by the protective layer of the above mentioned chemical compounds. This is illustrated in the respective model in Figure 3. Due to this, for the gear oils with EP lubricating additives, the scuffing initiation is delayed to appear at much higher loads than in the case of oils without

coefficient values exceed the maximum value assumed for the mixed friction, i.e. 0.1 [5].

The scuffing initiation occurs at a load called the scuffing load, which is characteristic for each tested lubricating oil. At this load, the lubricating film collapses, the number of colliding surface asperities drastically increases, and the destruction changes its occurrence from the micro- to macro-scale and scuffing appears. Initially only part of the friction surface undergoes scuffing. It can be observed in the surface topography image of the border between the surface that did not exhibit very rough topography typical of scuffing (left side) and the surface

The described phenomena leading to scuffing are illustrated in the models with the heading "Scuffing initiation" in Figure 3. The upper model concerns the surface that did not exhibit very rough topography typical of scuffing, where still the mixed friction exists, while the lower

The upper model shows that the micro-scale phenomena in the zone intact by scuffing are similar to those described in the phase "Mixed friction" apart from the thickness of elastic and plastic deformations which increased due to rising load. Probably, in view of plastic defor‐ mation that causes surface defects, the reactive diffusion of "active" atoms from the EP lubricating additives (e.g. sulphur atoms) into the surface layer takes place and iron sulphides form, which is confirmed by other researchers, e.g. in the work [7]. The diffusively modified micro-zones inside the highest asperities are plastically deformed and are indicated in the

By observing phenomena in the part of the friction surface that undergoes scuffing, one can indicate that the situation changes radically. The lower model illustrates that, in the first phase of scuffing, the lubricating film no longer exists, nor is there any boundary layer. This leads to a rapid intensification of the material destruction. Much plastic deformation appears, turning into the transfer, flowing and mingling of the material of the rubbing test balls. For the tested oils with EP lubricating additives, much of the surface layer starts to be chemically modified.

) sharply increases and measured friction

lubricating additives (e.g. API GL-1 ones, not presented here).

In this phase, scuffing initiates - the friction torque (Mt

*1.2.3. Scuffing phase: "Scuffing initiation"*

190 Tribology - Fundamentals and Advancements

destroyed by scuffing (right side) - Figure 3.

respective model as orange spots - Figure 3.

one refers to the surface already destroyed by scuffing.

This will be decisive for the scuffing propagation character.

intensity.

This phase refers to the scuffing process, after its initiation. It is reflected by a sharp increase in the friction torque (Mt ), accompanied by a high intensity of the lower test balls wear - Figure 4 a, b). This situation is illustrated in the models with the heading "Scuffing propagation" in Figure 3.

**Figure 4.** Development of the wear of the lower test balls due to scuffing: a) at scuffing initiation, b) at 12th seconds of the run (scuffing propagation), c) at the end of the run; images obtained at the same magnification

For the tested gear oils, after the scuffing initiation due to rapid chemical reactions of their EP additives with the surface, a rise in the friction torque is mitigated to quickly stabilise at relatively low value - Figure 3. It is accompanied by continuously evolving wear of the lower test balls that is not intensive - Figure 4 b, c). A drop in the pressure in the contact zone due to wear, brings about the possibility of oil introduction into the contact zone and the regeneration of the boundary layer on much of the friction surface. Such an action is indicated by the friction coefficient within the range 0.11 to 0.15, typical of boundary friction. On the basis of the work [5], which also concerns four-ball experiments, it was assumed that the boundary friction occurs in the four-ball tribosystem when the friction coefficient is in the range between 0.09 and 0.15. The determined values of the friction coefficient being in the middle and upper limit typical of boundary friction denote that some part of the friction surface must have undergone scuffing; It can be assumed from [5] that "full scuffing" occurs when the friction coefficient exceeds 0.3. The specific state of the surface layer in this phase is called the "Secondary Boundary Layer" (SBL) in the work [8]. The round model in the micro-scale concerning the scuffing propagation (Figure 3) illustrates the places of oil appearance in the contact zone. Let us call them "the micro-pockets." One can presume that inside the oil micro-pockets the following phenomena take place: the intensive adsorption and desorption of the base oil and lubricating additives molecules on/from the steel surface, chemical reactions of the lubricating additives with the surface, and - in view of plastic deformation that causes surface defects the diffusion of "active" atoms from the lubricating compounds (e.g. sulphur atoms) into the surface layer. In view of the transfer and mingling of the material of the rubbing test balls, the chemically modified zones appear across the entire zone of plastic deformations - orange spots. For the API GL-4 and GL-5 gear oils, the effective chemical modification of the surface mitigates the increase of the wear scar diameter - Figure 4 b, c) - in the phase of the SBL formation, accompanied by a mitigated rise in the friction torque and a decreasing friction coefficient (Figure 3).

#### **1.3. Gear tests of scuffing**

Nowadays, two manners of the improvement of the resistance to scuffing of gears are in use in the world. One is focused on the improvement of extreme-pressure (EP) properties of gear oils. The other one is related to the improvement of the properties of gear materials, e.g. by the deposition of thin hard coatings onto the tooth flank surface.

The verification of the quality of gear oils and new techniques of surface engineering of the tooth surface of gears requires that gear testing should be used. The most known is a unique complex of gear test methods developed in the Gear Research Center (FZG) at the Technical University of Munich. Approximately, 500 FZG gear test rigs are used around the world [9].

data compiled from [23-25]

**2. New test methods**

**2.1. Idea of the methods**

the coated gears using the FZG A/8.3/90 test method.

a-C:H:W - a-C:H:W MoS2/Ti - MoS2/Ti a-C:H - a-C:H

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

193

**Figure 5.** Failure load stages (FLS) obtained for the tested coatings (both gears coated) - FZG A/8.3/90 test method;

It is apparent from Figure 5 that the failure load stages (FLS), indicating the gear resistance to scuffing exceed the maximum number 12, so that the it is impossible to differentiate between

To solve this problem, in the Tribology Department of ITeE-PIB, research was undertaken to apply the new FZG scuffing tests for coated gears to differentiate between their resistance to scuffing. Because the FZG test methods are dedicated exclusively to lubricating oils, their application for testing coated gears required introducing significant modifications - unique test methods have been developed, being the subject of this chapter. They are called the "Gear

The main difference between the test methods designed by the authors and the gear scuffing tests A10/16.6R/90 and S-A10/16.6R/90, developed by FZG, is a rise in the initial oil temperature to 120 °C, adoption of a failure criterion related to wear of the wheel (big gear), and resigning from the criterion of invalidation of the test results when wear of the wheel exceeds 20 mg. The tests are performed on a pair of lubricated test gears with a coating (it can be applied on one or both the gears) at a constant rotational speed, and at the initial temperature of the lubricating oil identical for all the runs - until a failure load stage (FLS) is determined, i.e., such a load at which at least one of the failure criteria is met. In the Gear Scuffing EP Test for Coatings, based on the FZG S-A10/16.6R/90 test, the load is increased stepwise, from the lowest

Scuffing EP Test for Coatings" and "Gear Scuffing Shock Test for Coatings."

FLS

The most often used and popular gear tests for lubricating oils are performed using the FZG A/8.3/90 scuffing test method. Unfortunately, this method makes it impossible to differentiate between gear oils having very good extreme-pressure (EP) properties, from the point of view of the resistance to gear scuffing [10]. This is why various scientific centres have developed their own test methods [10-13].

Recognising the problem of the low resolution of A/8.3/90 scuffing test, the FZG has developed two new scuffing methods denoted as A10/16.6R/90 and S-A10/16.6R/90 (S - *shock*). The new test methods are described in detail in the literature, e.g. [14-19]. They are carried out under much severer conditions compared to the A/8.3/90 test. This is a result of the reduced face width of the small gear (pinion), doubling rotational speed, and reversing the direction of rotation. Additionally, according to the S-A10/16.6R/90 method, the test is started at once with a load at which the failure is expected, hence the name "scuffing shock test." Shock loading prevents the test gears from running-in and in turn increases their susceptibility to scuffing, which further increases the method resolution.

Nowadays, one of the research directions in numerous scientific centres in the world is an improvement in the scuffing resistance of toothed gears, achievable by the deposition of thin, hard, low-friction coatings onto the gear teeth, e.g. the a-C:H:W or MoS2/Ti coatings [20-22]. For the last several years, intensive research work has also been performed on this subject in the Tribology Department of ITeE-PIB. Until now, the FZG A/8.3/90 gear scuffing test method has been used most often in various scientific centres, which, like in the case of testing gear oils, exhibits a resolution that is too low to differentiate between the coated gears from the point of view of their resistance to scuffing [23-25] - Figure 5. It should be explained here that a-C:H:W and a-C:H coatings are DLC (diamond-like carbon) coatings, and the a-C:H:W coating has an outermost DLC layer doped with W (tungsten).

**Figure 5.** Failure load stages (FLS) obtained for the tested coatings (both gears coated) - FZG A/8.3/90 test method; data compiled from [23-25]

It is apparent from Figure 5 that the failure load stages (FLS), indicating the gear resistance to scuffing exceed the maximum number 12, so that the it is impossible to differentiate between the coated gears using the FZG A/8.3/90 test method.

To solve this problem, in the Tribology Department of ITeE-PIB, research was undertaken to apply the new FZG scuffing tests for coated gears to differentiate between their resistance to scuffing. Because the FZG test methods are dedicated exclusively to lubricating oils, their application for testing coated gears required introducing significant modifications - unique test methods have been developed, being the subject of this chapter. They are called the "Gear Scuffing EP Test for Coatings" and "Gear Scuffing Shock Test for Coatings."

#### **2. New test methods**

For the API GL-4 and GL-5 gear oils, the effective chemical modification of the surface mitigates the increase of the wear scar diameter - Figure 4 b, c) - in the phase of the SBL formation, accompanied by a mitigated rise in the friction torque and a decreasing friction coefficient

Nowadays, two manners of the improvement of the resistance to scuffing of gears are in use in the world. One is focused on the improvement of extreme-pressure (EP) properties of gear oils. The other one is related to the improvement of the properties of gear materials, e.g. by the

The verification of the quality of gear oils and new techniques of surface engineering of the tooth surface of gears requires that gear testing should be used. The most known is a unique complex of gear test methods developed in the Gear Research Center (FZG) at the Technical University of Munich. Approximately, 500 FZG gear test rigs are used around the world [9].

The most often used and popular gear tests for lubricating oils are performed using the FZG A/8.3/90 scuffing test method. Unfortunately, this method makes it impossible to differentiate between gear oils having very good extreme-pressure (EP) properties, from the point of view of the resistance to gear scuffing [10]. This is why various scientific centres have developed

Recognising the problem of the low resolution of A/8.3/90 scuffing test, the FZG has developed two new scuffing methods denoted as A10/16.6R/90 and S-A10/16.6R/90 (S - *shock*). The new test methods are described in detail in the literature, e.g. [14-19]. They are carried out under much severer conditions compared to the A/8.3/90 test. This is a result of the reduced face width of the small gear (pinion), doubling rotational speed, and reversing the direction of rotation. Additionally, according to the S-A10/16.6R/90 method, the test is started at once with a load at which the failure is expected, hence the name "scuffing shock test." Shock loading prevents the test gears from running-in and in turn increases their susceptibility to scuffing,

Nowadays, one of the research directions in numerous scientific centres in the world is an improvement in the scuffing resistance of toothed gears, achievable by the deposition of thin, hard, low-friction coatings onto the gear teeth, e.g. the a-C:H:W or MoS2/Ti coatings [20-22]. For the last several years, intensive research work has also been performed on this subject in the Tribology Department of ITeE-PIB. Until now, the FZG A/8.3/90 gear scuffing test method has been used most often in various scientific centres, which, like in the case of testing gear oils, exhibits a resolution that is too low to differentiate between the coated gears from the point of view of their resistance to scuffing [23-25] - Figure 5. It should be explained here that a-C:H:W and a-C:H coatings are DLC (diamond-like carbon) coatings, and the a-C:H:W coating

deposition of thin hard coatings onto the tooth flank surface.

(Figure 3).

**1.3. Gear tests of scuffing**

192 Tribology - Fundamentals and Advancements

their own test methods [10-13].

which further increases the method resolution.

has an outermost DLC layer doped with W (tungsten).

#### **2.1. Idea of the methods**

The main difference between the test methods designed by the authors and the gear scuffing tests A10/16.6R/90 and S-A10/16.6R/90, developed by FZG, is a rise in the initial oil temperature to 120 °C, adoption of a failure criterion related to wear of the wheel (big gear), and resigning from the criterion of invalidation of the test results when wear of the wheel exceeds 20 mg.

The tests are performed on a pair of lubricated test gears with a coating (it can be applied on one or both the gears) at a constant rotational speed, and at the initial temperature of the lubricating oil identical for all the runs - until a failure load stage (FLS) is determined, i.e., such a load at which at least one of the failure criteria is met. In the Gear Scuffing EP Test for Coatings, based on the FZG S-A10/16.6R/90 test, the load is increased stepwise, from the lowest to the highest value. According to the Gear Scuffing Shock Test for Coatings, based on the FZG S-A10/16.6R/90 test the load is not increased in stages from the lowest value, but the expected failure load is applied to an unused gear flank (hence, the name "shock test"). In the shock test, each change of the load requires an unused gear flank; therefore, before subsequent runs, the test gears should be disassembled and reversed or replaced with new ones.

**Gear scuffing test FZG A/8.3/90**

(pinion and wheel width 20 mm)

Test materials 20MnCr5 20MnCr5, but at least one gear

Motor rotational speed 1500 rpm 3000 rpm 3000 rpm Circumferential speed 8.3 m/s 16.6 m/s 16.6 m/s

Direction of motor rotation "Normal" "Reversed" (R) "Reversed" (R) Run duration 15 min. 7 min. 30 s 7 min. 30 s

Maximum load stage 12 10 12

Maximum loading torque 535 N·m 373 N·m 535 N·m Maximum Hertzian pressure1.8 GPa 2.2 GPa 2.6 GPa

**Purpose of test Testing lubricating**

Loading type Stepwise, from load

Initial lubricating oil temperature

Temperature stabilisation during the run by cooling

Main failure criterion for FLS determination

Additional criteria of failure

Criterion of invalidation of

Ap - total area of failures on the pinion

b Ww - wear (mass loss) of the wheel

assessment

the run

a

stage 1

Ap ≥ area of one pinion tooth (≈200

mm2) a

Test gear type FZG A-type

**oils**

**Gear Scuffing EP Test for**

**based on FZG A10/16.6R/90**

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

**Testing coatings deposited on**

(pinion width 10 mm, wheel

Stepwise, from load stage 1 Shock

None Failures on the pinion teeth Failures on the pinion teeth

90 ºC 120 ºC 120 ºC

No No No

Ap > area of one pinion tooth (≈100 mm2)

None Significant decohesion of the coating

**Table 1.** Comparison of the FZG gear scuffing test and the methods designed by the authors

Type of lubrication Dip lubrication Dip lubrication Dip lubrication

**Gear Scuffing Shock Test for**

http://dx.doi.org/10.5772/54569

195

**based on FZG S-A10/16.6R/90**

**Testing coatings deposited on**

(pinion width 10 mm, wheel width

20MnCr5, but at least one gear

(i.e. starting with a load at which the failure is expected)

pinion tooth (≈100 mm2), or Ww >

Significant decohesion of the

Ap > area of one

200 mgb

coating

**Coatings**

**gears**

20 mm)

coated

FZG A10-type

**Coatings**

**gears**

coated

FZG A10-type

width 20 mm)

Although the authors have introduced some significant changes to the FZG gear scuffing tests, the core procedures of performing the tests are the same as in the FZG tests, and they can be found in the relevant publications, e.g. in [14].

To better explain the differences between the "old" FZG gear scuffing test A/8.3/90 and the new test methods designed by the authors, the test conditions according to each method and the failure criteria are specified in Table 1.

After starting the run, the oil in the test chamber is heated by the heaters and friction. The oil temperature is allowed to rise freely. No cooling system is used in the tests.

Like in the FZG gear scuffing tests, if the failures are observed only within 1 mm from the tooth addendum, they are only scratches, or the failures are so small that the original criss-crossgrinding pattern (Figure 6) is still intact, they should be neglected when calculating the total area of the failures.

**Figure 6.** Original criss-cross-grinding pattern on the test gear teeth - stylus profilometry image

The failure load stage (FLS) is the main measure of the resistance of the test gears to scuffing. According to the Gear Scuffing EP Test for Coatings, the FLS is such a load at which the main failure criterion specified in Table 1 has been met. According to the Gear Scuffing Shock Test for Coatings, the FLS is such a load at which at least one of the failure criteria has been met and, when at the load stage lower by 1, neither of the failure criteria has been met.

When there is significant decohesion of the coating due to poor adhesion to the surface, the run should be invalidated.


a Ap - total area of failures on the pinion

b Ww - wear (mass loss) of the wheel

to the highest value. According to the Gear Scuffing Shock Test for Coatings, based on the FZG S-A10/16.6R/90 test the load is not increased in stages from the lowest value, but the expected failure load is applied to an unused gear flank (hence, the name "shock test"). In the shock test, each change of the load requires an unused gear flank; therefore, before subsequent runs, the

Although the authors have introduced some significant changes to the FZG gear scuffing tests, the core procedures of performing the tests are the same as in the FZG tests, and they can be

To better explain the differences between the "old" FZG gear scuffing test A/8.3/90 and the new test methods designed by the authors, the test conditions according to each method and

After starting the run, the oil in the test chamber is heated by the heaters and friction. The oil

Like in the FZG gear scuffing tests, if the failures are observed only within 1 mm from the tooth addendum, they are only scratches, or the failures are so small that the original criss-crossgrinding pattern (Figure 6) is still intact, they should be neglected when calculating the total

test gears should be disassembled and reversed or replaced with new ones.

temperature is allowed to rise freely. No cooling system is used in the tests.

**Figure 6.** Original criss-cross-grinding pattern on the test gear teeth - stylus profilometry image

and, when at the load stage lower by 1, neither of the failure criteria has been met.

The failure load stage (FLS) is the main measure of the resistance of the test gears to scuffing. According to the Gear Scuffing EP Test for Coatings, the FLS is such a load at which the main failure criterion specified in Table 1 has been met. According to the Gear Scuffing Shock Test for Coatings, the FLS is such a load at which at least one of the failure criteria has been met

When there is significant decohesion of the coating due to poor adhesion to the surface, the

found in the relevant publications, e.g. in [14].

194 Tribology - Fundamentals and Advancements

the failure criteria are specified in Table 1.

area of the failures.

run should be invalidated.

**Table 1.** Comparison of the FZG gear scuffing test and the methods designed by the authors

After run completion at a given load stage, the failures on the pinion teeth should be noted using the symbols from Table 2. These data are used for additional failure assessment, complementarily to FLS.

When both test gears are uncoated, a respective standardised test method A10/16.6R/90 or S-A10/16.6R/90, developed by FZG should be used. However, to compare the results with the new test methods, it is necessary to start a run at the initial oil temperature of 120 °C

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

197

A photograph of the FZG A10 scuffing test gears employed in the tests according to the

The A10 test gears are made of 20MnCr5 steel. They are carburized, case hardened, tempered and Maag criss-cross ground. The surface hardness is HRC = 60 + 2 and the case hardness depth (CHD) is 0.6 to 0.9 mm (Eht). The effective face width of the pinion is 10 mm, and the wheel is 20 mm. The number of pinion teeth is 16, and wheel 24. The gears are identical to the ones used

For the complex testing of gears, a back-to-back gear test rig, denoted as T-12U, was designed in the Tribology Department of ITeE-PIB in Radom. Its photograph is presented in Figure 8

The T-12U test rig is equipped with a control-measuring system, which consists of measuring

to perform tests according to the FZG A10/16.6R/90 and S-A10/16.6R/90 methods.

transducers (thermocouple, speed transducer) and the controller (Figure 8).

rather than 90 °C.

developed methods is shown in Figure 7.

**Figure 7.** Photograph of the FZG A10 scuffing test gears

and kinematic schemes are presented in Figure 9.

**2.3. Test equipment**

**2.2. Test gears**


**Table 2.** Modes of wear of the test pinion (small gear)

Polishing can be identified when the "mirror-like" surface on the tooth flank is observed with the disappearing criss-cross-grinding pattern shown in Figure 6.

Scratches appear as shorter or longer fine lines in the sliding direction of the tooth flanks.

Scoring marks run in the same direction as scratches. On the basis of CEC L-07-95 standard [26], it can be adopted that they occur singly or in zones as light, medium or deep grooves continuing towards the tip of the tooth and having a rougher appearance than the criss-crossgrinding pattern (Figure 6).

Scuffing marks occur as single, fine marks or strips, or areas covering a part or all of the flank width. According to CEC L-07-95 standard, they appear as dull areas with the roughness much greater than the original criss-cross-grinding pattern shown in Figure 6. In this case, the grinding pattern is no longer visible.

The difference between scuffing and scoring is that scuffing originates from the adhesive bond creation between the mating surfaces, which are then sheared, and scoring results from mechanical abrasion of the surface by the very hard wear particles under conditions of a very high load. Like scuffing, scoring is one of the most dangerous modes of gear wear.

When both test gears are uncoated, a respective standardised test method A10/16.6R/90 or S-A10/16.6R/90, developed by FZG should be used. However, to compare the results with the new test methods, it is necessary to start a run at the initial oil temperature of 120 °C rather than 90 °C.

#### **2.2. Test gears**

After run completion at a given load stage, the failures on the pinion teeth should be noted using the symbols from Table 2. These data are used for additional failure assessment,

**Mode of wear Symbol Appearance**

Polishing can be identified when the "mirror-like" surface on the tooth flank is observed with

Scoring marks run in the same direction as scratches. On the basis of CEC L-07-95 standard [26], it can be adopted that they occur singly or in zones as light, medium or deep grooves continuing towards the tip of the tooth and having a rougher appearance than the criss-cross-

Scuffing marks occur as single, fine marks or strips, or areas covering a part or all of the flank width. According to CEC L-07-95 standard, they appear as dull areas with the roughness much greater than the original criss-cross-grinding pattern shown in Figure 6. In this case, the

The difference between scuffing and scoring is that scuffing originates from the adhesive bond creation between the mating surfaces, which are then sheared, and scoring results from mechanical abrasion of the surface by the very hard wear particles under conditions of a very

high load. Like scuffing, scoring is one of the most dangerous modes of gear wear.

Scratches appear as shorter or longer fine lines in the sliding direction of the tooth flanks.

Polishing W

Scratches R

Scoring B

Scuffing Z

the disappearing criss-cross-grinding pattern shown in Figure 6.

**Table 2.** Modes of wear of the test pinion (small gear)

grinding pattern (Figure 6).

grinding pattern is no longer visible.

complementarily to FLS.

196 Tribology - Fundamentals and Advancements

A photograph of the FZG A10 scuffing test gears employed in the tests according to the developed methods is shown in Figure 7.

**Figure 7.** Photograph of the FZG A10 scuffing test gears

The A10 test gears are made of 20MnCr5 steel. They are carburized, case hardened, tempered and Maag criss-cross ground. The surface hardness is HRC = 60 + 2 and the case hardness depth (CHD) is 0.6 to 0.9 mm (Eht). The effective face width of the pinion is 10 mm, and the wheel is 20 mm. The number of pinion teeth is 16, and wheel 24. The gears are identical to the ones used to perform tests according to the FZG A10/16.6R/90 and S-A10/16.6R/90 methods.

#### **2.3. Test equipment**

For the complex testing of gears, a back-to-back gear test rig, denoted as T-12U, was designed in the Tribology Department of ITeE-PIB in Radom. Its photograph is presented in Figure 8 and kinematic schemes are presented in Figure 9.

The T-12U test rig is equipped with a control-measuring system, which consists of measuring transducers (thermocouple, speed transducer) and the controller (Figure 8).

During runs, the following quantities are measured: rotational speed, lubricating oil temper‐ ature, motor current load, time, and the number of motor revolutions. The measured values

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

199

The T-12U gear test rig is a back-to-back rig (Figure 9) where the test gears (2) and (3), located in the test chamber (5), are connected by two shafts to the slave gears, located in the chamber (9). The front shaft (8) has two parts. Between them there is the load clutch (7). To apply the loading torque between the meshing gears, before the run, one part of the shaft (the left part of the front shaft (8) is fixed to the base with the lock-pin via the clutch and its support. A round-shaped loading lever (12) is placed on the right part of the clutch (7), and then the weight hanger (13) is suspended and the appropriate number weights (14) put on it. They give a static loading torque by twisting the shafts, which is measured indirectly using the torsion angle indicator (6). When the load has been applied, the two halves of the clutch (7) are firmly fixed against each other with the bolts. Then, the lock-pin is removed to close the safety cover. During the run, this loading torque "circulates" between the gears. In the back-to-back solution the motor (11) must overcome only the friction between gears, rolling bearings, and some minor components of friction (friction against seals, internal friction in the oil). Thus, the whole design

An AC squirrel-cage motor (11) of the nominal rotational speed of 3000 rpm is used to drive the rig. It is controlled by the frequency converter, which enables to change the rotational speed

In the gear scuffing tests the test gears are dip lubricated. In the test chamber where the test gears are located, there are heaters (15) to heat up the lubricating oil. The thermocouple (1), with the measuring point inserted in the lubricating oil, is to measure the oil temperature. A

The motor (11) of the machine is automatically stopped when the preset time elapses. The required time is set on the controller panel. Additionally, the operator can read out the number of motor revolutions to confirm the correct duration of the run. The number of motor revolu‐ tions is displayed on the controller panel (Figure 8) connected to the speed transducer.

In the T-12U machine, the friction torque can be measured indirectly by measurement of the

The test rig has a special support on the side cover of the test chamber (5) for mounting vibration transducers (accelerometers) to enable the operator to monitor the level of vibrations along different axes. However, now there is no possibility to automatically stop the motor when the vibration level is very high. This feature (together with other features like direct measurement of the friction torque) will be included in a new test rig, denoted as T-12UF, being

Additional equipment includes a mass comparator for a very precise determination of the mass

motor current load, which can be assumed to be proportional to the friction torque.

PID regulator is used to protect against overheating of the lubricating oil.

The test rig is mounted on the concrete base equipped with vibration-dumping feet.

are displayed on the controller.

is very simple and compact.

within a wide range.

developed at present.

loss (wear) of the wheel.

**Figure 8.** Photograph of the T-12U gear test rig

**Figure 9.** Kinematic schemes of the T-12U gear test rig: a) front view, b) top view, c) loading equipment; 1 - thermo‐ couple, 2 - test wheel, 3 - test pinion, 4 - vent, 5 - test chamber, 6 - shafts torsion angle indicator, 7 - load clutch, 8 front shaft, 9 - slave chamber, 10 - drive clutch, 11 - electric motor, 12 - loading lever, 13 - weight hanger, 14 - weights, 15 - heaters, 16 - frame, 17 - concrete base

During runs, the following quantities are measured: rotational speed, lubricating oil temper‐ ature, motor current load, time, and the number of motor revolutions. The measured values are displayed on the controller.

The test rig is mounted on the concrete base equipped with vibration-dumping feet.

The T-12U gear test rig is a back-to-back rig (Figure 9) where the test gears (2) and (3), located in the test chamber (5), are connected by two shafts to the slave gears, located in the chamber (9). The front shaft (8) has two parts. Between them there is the load clutch (7). To apply the loading torque between the meshing gears, before the run, one part of the shaft (the left part of the front shaft (8) is fixed to the base with the lock-pin via the clutch and its support. A round-shaped loading lever (12) is placed on the right part of the clutch (7), and then the weight hanger (13) is suspended and the appropriate number weights (14) put on it. They give a static loading torque by twisting the shafts, which is measured indirectly using the torsion angle indicator (6). When the load has been applied, the two halves of the clutch (7) are firmly fixed against each other with the bolts. Then, the lock-pin is removed to close the safety cover. During the run, this loading torque "circulates" between the gears. In the back-to-back solution the motor (11) must overcome only the friction between gears, rolling bearings, and some minor components of friction (friction against seals, internal friction in the oil). Thus, the whole design is very simple and compact.

An AC squirrel-cage motor (11) of the nominal rotational speed of 3000 rpm is used to drive the rig. It is controlled by the frequency converter, which enables to change the rotational speed within a wide range.

**Figure 8.** Photograph of the T-12U gear test rig

198 Tribology - Fundamentals and Advancements

15 - heaters, 16 - frame, 17 - concrete base

(a) (b)

(c)

**Figure 9.** Kinematic schemes of the T-12U gear test rig: a) front view, b) top view, c) loading equipment; 1 - thermo‐ couple, 2 - test wheel, 3 - test pinion, 4 - vent, 5 - test chamber, 6 - shafts torsion angle indicator, 7 - load clutch, 8 front shaft, 9 - slave chamber, 10 - drive clutch, 11 - electric motor, 12 - loading lever, 13 - weight hanger, 14 - weights,

)

In the gear scuffing tests the test gears are dip lubricated. In the test chamber where the test gears are located, there are heaters (15) to heat up the lubricating oil. The thermocouple (1), with the measuring point inserted in the lubricating oil, is to measure the oil temperature. A PID regulator is used to protect against overheating of the lubricating oil.

The motor (11) of the machine is automatically stopped when the preset time elapses. The required time is set on the controller panel. Additionally, the operator can read out the number of motor revolutions to confirm the correct duration of the run. The number of motor revolu‐ tions is displayed on the controller panel (Figure 8) connected to the speed transducer.

In the T-12U machine, the friction torque can be measured indirectly by measurement of the motor current load, which can be assumed to be proportional to the friction torque.

The test rig has a special support on the side cover of the test chamber (5) for mounting vibration transducers (accelerometers) to enable the operator to monitor the level of vibrations along different axes. However, now there is no possibility to automatically stop the motor when the vibration level is very high. This feature (together with other features like direct measurement of the friction torque) will be included in a new test rig, denoted as T-12UF, being developed at present.

Additional equipment includes a mass comparator for a very precise determination of the mass loss (wear) of the wheel.

#### **2.4. Test materials**

The gears coated with the low-friction a-C:H:W coating (trade name: WC/C) of DLC type and composite low-friction MoS2/Ti coating (trade name: MoST) were tested. All material combi‐ nations were tested: coating-coating (both gears coated), coating-steel, steel-coating, and steelsteel for reference (both gears without the coating). In all cases, mineral, automotive gear oil of API GL-5 performance level and of SAE 80W-90-viscosity grade was used for lubrication.

#### **2.5. Statistical analysis**

To check statistical differences between the results obtained (FLS values), the uncertainty of measurement was assessed for the both developed test methods. This was done according to the procedures specified in the document EA-4/16 G:2003, which are binding in the accredited laboratories meeting the requirements of ISO/IEC 17025:2005.

Once the uncertainty of measurement has been calculated, the test result "y" and the uncer‐ tainty of measurement "U" should be reported as "y ± U."

As a normal practice, the uncertainty of measurement is given in relation to the average value of the measurement. For example, in the case of the gear scuffing shock tests, the respective formula derived by the authors is expressed as follows:

$$
\Delta U = 0.45 + 0.06 \ast FLS \tag{1}
$$

Load stage

4

binations with the a-C:H:W coating

symbols of wear were presented earlier in Table 2.

FLS

Combination of tested materials

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

201

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap ≈ 0

**Figure 10.** Failure load stages (FLS) obtained using the Gear Scuffing EP Test for Coatings for the tested material com‐

Figure 10 shows that the Gear Scuffing EP Test for Coatings is unable to differentiate between the tested material combinations from the point of view of the main criterion - FLS. All the FLS values exceed the maximum load stage, i.e. 10th. Thus, the additional criteria of failure assessment, related to the wear of the pinion after runs at particular load stages, were taken into account - Table 3. The table presents the symbolic modes of the wear of the test pinion at particular load stages for the tested material combinations with the a-C:H:W coating, and the mode of wear that appeared most often on the pinion teeth was considered. Below are the symbols of the wear modes, the total area of failures on the pinion (Ap) are given. The used

where:

U - uncertainty of measurement,

FLS - failure load stage.

According to ILAC-G8:03/2009, if the uncertainty intervals expressed by U do not overlap each other, one can say that the compared results are statistically different.

#### **3. Results and discussion**

#### **3.1. Gear Scuffing EP Test for Coatings**

#### *3.1.1. Material combinations with the a-C:H:W coating*

Failure load stages (FLS) obtained for the tested material combinations with the a-C:H:W coating are presented in Figure 10. The coated gear is dark grey coloured, and the uncoated one is light grey.

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears http://dx.doi.org/10.5772/54569 201

**2.4. Test materials**

200 Tribology - Fundamentals and Advancements

**2.5. Statistical analysis**

where:

U - uncertainty of measurement,

**3. Results and discussion**

**3.1. Gear Scuffing EP Test for Coatings**

*3.1.1. Material combinations with the a-C:H:W coating*

FLS - failure load stage.

one is light grey.

The gears coated with the low-friction a-C:H:W coating (trade name: WC/C) of DLC type and composite low-friction MoS2/Ti coating (trade name: MoST) were tested. All material combi‐ nations were tested: coating-coating (both gears coated), coating-steel, steel-coating, and steelsteel for reference (both gears without the coating). In all cases, mineral, automotive gear oil of API GL-5 performance level and of SAE 80W-90-viscosity grade was used for lubrication.

To check statistical differences between the results obtained (FLS values), the uncertainty of measurement was assessed for the both developed test methods. This was done according to the procedures specified in the document EA-4/16 G:2003, which are binding in the accredited

Once the uncertainty of measurement has been calculated, the test result "y" and the uncer‐

As a normal practice, the uncertainty of measurement is given in relation to the average value of the measurement. For example, in the case of the gear scuffing shock tests, the respective

According to ILAC-G8:03/2009, if the uncertainty intervals expressed by U do not overlap each

Failure load stages (FLS) obtained for the tested material combinations with the a-C:H:W coating are presented in Figure 10. The coated gear is dark grey coloured, and the uncoated

other, one can say that the compared results are statistically different.

*U* =+\* 0.45 0.06 *FLS* (1)

laboratories meeting the requirements of ISO/IEC 17025:2005.

tainty of measurement "U" should be reported as "y ± U."

formula derived by the authors is expressed as follows:

#### Combination of tested materials

**Figure 10.** Failure load stages (FLS) obtained using the Gear Scuffing EP Test for Coatings for the tested material com‐ binations with the a-C:H:W coating

Figure 10 shows that the Gear Scuffing EP Test for Coatings is unable to differentiate between the tested material combinations from the point of view of the main criterion - FLS. All the FLS values exceed the maximum load stage, i.e. 10th. Thus, the additional criteria of failure assessment, related to the wear of the pinion after runs at particular load stages, were taken into account - Table 3. The table presents the symbolic modes of the wear of the test pinion at particular load stages for the tested material combinations with the a-C:H:W coating, and the mode of wear that appeared most often on the pinion teeth was considered. Below are the symbols of the wear modes, the total area of failures on the pinion (Ap) are given. The used symbols of wear were presented earlier in Table 2.

As can be observed in Table 3 for the tested material combinations, the three modes of wear that appear most often on the pinion teeth are scratches, polishing, and scoring. The uncoated pinion undergoes the process of polishing through the rubbing by the hard a-C:H:W coating deposited on the meshing wheel. Similar action was observed on the uncoated wheel meshing the coated pinion (results not shown here). The role of such polishing is to be explained in

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

203

To sum up this part of the experiment, the Gear Scuffing EP Test for Coatings gives minor differences between the tested material combinations with the a-C:H:W coating, observed only when the pinion is uncoated and the wheel is coated. From the point of view of the practical applications of the a-C:H:W coating in gears, the situation when the both gears are coated seems to be better than in the case of one of the gears uncoated, because it is exposed to the

Failure load stages (FLS) obtained for the tested material combinations with the MoS2/Ti coating are presented in Figure 11. The coated gear is dark grey coloured, and the uncoated

Combination of tested materials

**Figure 11.** Failure load stages (FLS) obtained using the Gear Scuffing EP Test for Coatings for the tested material com‐

Figure 11 shows that the Gear Scuffing EP Test for Coatings is unable to differentiate between the tested material combinations from the point of view of the main criterion - FLS. As in the case of testing the a-C:H:W coating, all the FLS values exceed the maximum load

abrasive action of the meshing coated gear, which results in polishing and scoring.

further experiments planned by the authors.

*3.1.2. Material combinations with the MoS2/Ti coating*

one is light grey.

binations with the MoS2/Ti coating

FLS

**Table 3.** Modes of the wear of the test pinion at particular load stages for the tested material combinations with the a-C:H:W coating, together with the total area of failures on the pinion (Ap); Gear Scuffing EP Test for Coatings

As can be observed in Table 3 for the tested material combinations, the three modes of wear that appear most often on the pinion teeth are scratches, polishing, and scoring. The uncoated pinion undergoes the process of polishing through the rubbing by the hard a-C:H:W coating deposited on the meshing wheel. Similar action was observed on the uncoated wheel meshing the coated pinion (results not shown here). The role of such polishing is to be explained in further experiments planned by the authors.

To sum up this part of the experiment, the Gear Scuffing EP Test for Coatings gives minor differences between the tested material combinations with the a-C:H:W coating, observed only when the pinion is uncoated and the wheel is coated. From the point of view of the practical applications of the a-C:H:W coating in gears, the situation when the both gears are coated seems to be better than in the case of one of the gears uncoated, because it is exposed to the abrasive action of the meshing coated gear, which results in polishing and scoring.

#### *3.1.2. Material combinations with the MoS2/Ti coating*

Load stage

202 Tribology - Fundamentals and Advancements

5

6

7

8

9

10

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap ≈ 0

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap ≈ 0

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap ≈ 0

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap ≈ 0

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap ≈ 0

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap ≈ 0

**Table 3.** Modes of the wear of the test pinion at particular load stages for the tested material combinations with the a-C:H:W coating, together with the total area of failures on the pinion (Ap); Gear Scuffing EP Test for Coatings

Failure load stages (FLS) obtained for the tested material combinations with the MoS2/Ti coating are presented in Figure 11. The coated gear is dark grey coloured, and the uncoated one is light grey.

#### Combination of tested materials

**Figure 11.** Failure load stages (FLS) obtained using the Gear Scuffing EP Test for Coatings for the tested material com‐ binations with the MoS2/Ti coating

Figure 11 shows that the Gear Scuffing EP Test for Coatings is unable to differentiate between the tested material combinations from the point of view of the main criterion - FLS. As in the case of testing the a-C:H:W coating, all the FLS values exceed the maximum load stage, i.e. 10th. Thus, the additional criteria of failure assessment, related to the wear of the pinion after runs at particular load stages, were taken into account - Table 4. The table presents the symbolic modes of the wear of the test pinion at particular load stages for the tested material combinations with the MoS2/Ti coating, which is the mode of wear that appeared most often on the pinion teeth was considered. Below are the symbols of the wear modes, and the total area of failures on the pinion (Ap) are given. The used symbols of wear were presented earlier in Table 2.

Load stage

9

10

of the uncoated gear.

shown in the Figure.

**3.2. Gear scuffing shock test for coatings**

observed when both gears are coated.

*3.2.1. Material combinations with the a-C:H:W coating*

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap = 10 mm<sup>2</sup>

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

205

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap = 10 mm<sup>2</sup>

**Table 4.** Modes of the wear of the test pinion at particular load stages for the tested material combinations with the MoS2/Ti coating, together with the total area of failures on the pinion (Ap); Gear Scuffing EP Test for Coatings

As can be observed in Table 4 for the tested material combinations, the two modes of wear that appear most often on the pinion teeth are scratches and scoring. In the material combination of the uncoated pinion meshing the coated wheel, the pinion bears the mark of scoring caused

Thus, the Gear Scuffing EP Test for Coatings gives minor differences between the tested material combinations with the MoS2/Ti coating, observed only when the pinion is uncoated and the wheel is coated. From the point of view of the practical applications of the MoS2/Ti coating in gears, the situation when the both gears are coated seems to be better than in the case of one of the gears uncoated as it is exposed to the abrasive action of the meshing coated gear, which results in scoring. However, when one of the gears needs to remain uncoated, using the a-C:H:W coating is more preferable than MoS2/Ti, because a-C:H:W causes less wear

Failure load stages (FLS) obtained for the tested material combinations with the a-C:H:W coating are presented in Figure 12. The coated gear is dark grey coloured, and the uncoated one is light grey. The assessed uncertainties of measurement for each result obtained are also

Figure 12 shows that the Gear Scuffing Shock Test for Coatings makes it possible to differentiate between the tested material combinations. The best resistance to scuffing (highest FLS) is

by the rubbing by the hard coating deposited on the meshing wheel.


**Table 4.** Modes of the wear of the test pinion at particular load stages for the tested material combinations with the MoS2/Ti coating, together with the total area of failures on the pinion (Ap); Gear Scuffing EP Test for Coatings

As can be observed in Table 4 for the tested material combinations, the two modes of wear that appear most often on the pinion teeth are scratches and scoring. In the material combination of the uncoated pinion meshing the coated wheel, the pinion bears the mark of scoring caused by the rubbing by the hard coating deposited on the meshing wheel.

Thus, the Gear Scuffing EP Test for Coatings gives minor differences between the tested material combinations with the MoS2/Ti coating, observed only when the pinion is uncoated and the wheel is coated. From the point of view of the practical applications of the MoS2/Ti coating in gears, the situation when the both gears are coated seems to be better than in the case of one of the gears uncoated as it is exposed to the abrasive action of the meshing coated gear, which results in scoring. However, when one of the gears needs to remain uncoated, using the a-C:H:W coating is more preferable than MoS2/Ti, because a-C:H:W causes less wear of the uncoated gear.

#### **3.2. Gear scuffing shock test for coatings**

stage, i.e. 10th. Thus, the additional criteria of failure assessment, related to the wear of the pinion after runs at particular load stages, were taken into account - Table 4. The table presents the symbolic modes of the wear of the test pinion at particular load stages for the tested material combinations with the MoS2/Ti coating, which is the mode of wear that appeared most often on the pinion teeth was considered. Below are the symbols of the wear modes, and the total area of failures on the pinion (Ap) are given. The used symbols of wear

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap ≈ 0

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap ≈ 0

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap = 5 mm2

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap = 10 mm<sup>2</sup>

Ap ≈ 0 Ap ≈ 0 Ap ≈ 0 Ap = 10 mm<sup>2</sup>

were presented earlier in Table 2.

204 Tribology - Fundamentals and Advancements

Load stage

4

5

6

7

8

#### *3.2.1. Material combinations with the a-C:H:W coating*

Failure load stages (FLS) obtained for the tested material combinations with the a-C:H:W coating are presented in Figure 12. The coated gear is dark grey coloured, and the uncoated one is light grey. The assessed uncertainties of measurement for each result obtained are also shown in the Figure.

Figure 12 shows that the Gear Scuffing Shock Test for Coatings makes it possible to differentiate between the tested material combinations. The best resistance to scuffing (highest FLS) is observed when both gears are coated.

pinion teeth was taken into account. The photographs of the most often appearing mode of wear of the pinion at the highest load stage are shown also shown in the table. Red-shadowed cells in the table denote the failure load stage (FLS). Below are given the symbols of the wear modes, the total area of failures on the pinion (Ap), and wear of wheel (Ww). The used symbols

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

207

As can be observed in Table 5 for the tested material combinations, the three modes of wear that appear most often on the pinion teeth are scratches, scuffing, and scoring. When one or both gears are a-C:H:W-coated, only scratches and scoring predominate on the pinion teeth.

What was observed also during the Gear Scuffing EP Test for Coatings, and what seems to by typical of "the action" of the a-C:H:W coating, the uncoated gear undergoes the process of polishing or scoring through the rubbing by the hard coating deposited on the meshing gear.

**Figure 13.** Photograph of the tooth flank of the uncoated wheel, polished by the a-C:H:W-coated pinion; Gear Scuff‐

The polishing on the wheel teeth flanks can be seen in Figure 13.

of wear were presented earlier in Table 2.

ing Shock Test for Coatings

Load stage

7

Ap = 26 mm2 Ww = 1 mg

#### Combination of tested materials

**Figure 12.** Failure load stages (FLS) obtained using the Gear Scuffing Shock Test for Coatings for the tested material combinations with the a-C:H:W coating

Under "shock" conditions, when the pinion is uncoated and the wheel is coated with the a-C:H:W coating, the resistance to scuffing is slightly higher than in the case when the pinion is coated and the wheel is uncoated. Hypothetically, there is a transfer of graphite (solid lubricant) from the a-C:H:W coated gear to the teeth of the uncoated one, which is more effective for the wheel coated than in the opposite situation, because the area of the coated steel surface of the wheel (larger gear with 24 teeth) is greater than in the case the coating is deposited on the pinion (small gear having only 16 teeth). However, one must have in mind that the difference in the scuffing resistance of the two material combinations is not statistically significant, because the measurement uncertainties overlap each other.

In comparison to the case of the both gears uncoated, when the a-C:H:W coating is deposited on one or two gears, much higher resistance to scuffing is observed. This is a result of a high surface energy for metals (here for steel) promoting adhesive bonding in the steel-steel contact, and smaller affinity in the different materials than when both of them are identical (i.e. steelsteel), which protects the surface from adhesive bonding. Yet another phenomenon can be attributed to it. When one of the mating materials (coating) is much harder than the other one (steel), or when two very hard materials are in contact (coating-coating) there is a reduction in the tendency to adhesive bonding, hence scuffing.

The additional criteria of failure assessment, related to the wear of the pinion after runs at particular load stages, were also taken into account - Table 5. The table presents the symbolic modes of the wear of the test pinion at particular load stages for the tested material combina‐ tions with the a-C:H:W coating, which is the mode of wear that appeared most often on the pinion teeth was taken into account. The photographs of the most often appearing mode of wear of the pinion at the highest load stage are shown also shown in the table. Red-shadowed cells in the table denote the failure load stage (FLS). Below are given the symbols of the wear modes, the total area of failures on the pinion (Ap), and wear of wheel (Ww). The used symbols of wear were presented earlier in Table 2.

As can be observed in Table 5 for the tested material combinations, the three modes of wear that appear most often on the pinion teeth are scratches, scuffing, and scoring. When one or both gears are a-C:H:W-coated, only scratches and scoring predominate on the pinion teeth.

What was observed also during the Gear Scuffing EP Test for Coatings, and what seems to by typical of "the action" of the a-C:H:W coating, the uncoated gear undergoes the process of polishing or scoring through the rubbing by the hard coating deposited on the meshing gear. The polishing on the wheel teeth flanks can be seen in Figure 13.

Under "shock" conditions, when the pinion is uncoated and the wheel is coated with the a-C:H:W coating, the resistance to scuffing is slightly higher than in the case when the pinion is coated and the wheel is uncoated. Hypothetically, there is a transfer of graphite (solid lubricant) from the a-C:H:W coated gear to the teeth of the uncoated one, which is more effective for the wheel coated than in the opposite situation, because the area of the coated steel surface of the wheel (larger gear with 24 teeth) is greater than in the case the coating is deposited on the pinion (small gear having only 16 teeth). However, one must have in mind that the difference in the scuffing resistance of the two material combinations is not statistically

**Figure 12.** Failure load stages (FLS) obtained using the Gear Scuffing Shock Test for Coatings for the tested material

Combination of tested materials

In comparison to the case of the both gears uncoated, when the a-C:H:W coating is deposited on one or two gears, much higher resistance to scuffing is observed. This is a result of a high surface energy for metals (here for steel) promoting adhesive bonding in the steel-steel contact, and smaller affinity in the different materials than when both of them are identical (i.e. steelsteel), which protects the surface from adhesive bonding. Yet another phenomenon can be attributed to it. When one of the mating materials (coating) is much harder than the other one (steel), or when two very hard materials are in contact (coating-coating) there is a reduction in

The additional criteria of failure assessment, related to the wear of the pinion after runs at particular load stages, were also taken into account - Table 5. The table presents the symbolic modes of the wear of the test pinion at particular load stages for the tested material combina‐ tions with the a-C:H:W coating, which is the mode of wear that appeared most often on the

significant, because the measurement uncertainties overlap each other.

the tendency to adhesive bonding, hence scuffing.

combinations with the a-C:H:W coating

206 Tribology - Fundamentals and Advancements

FLS

**Figure 13.** Photograph of the tooth flank of the uncoated wheel, polished by the a-C:H:W-coated pinion; Gear Scuff‐ ing Shock Test for Coatings

Scuffing is observed only when both gears are uncoated - Table 5. This is one of the most dangerous modes of gear wear. As mentioned earlier, scuffing marks occur as single, fine marks or strips, or areas covering a part or all of the flank width. They appear as dull areas with the roughness much greater than the original criss-cross-grinding pattern shown in

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

209

To sum up this part of the experiment, the Gear Scuffing Shock Test for Coatings gives a much better resolution than the Gear Scuffing EP Test. However, one needs to have in mind that the cost of the former is about four-times higher than in the case of the latter, because the "shock tests" require more test gears to be used. From the point of view of the practical applications of the a-C:H:W coating in gears, the situation when the both gears are coated seems to be better than in the case of one of the gears uncoated, because it is exposed to the abrasive action of the meshing coated gear, which results in polishing or scoring. This positively verifies the

Failure load stages (FLS) obtained for the tested material combinations with the MoS2/Ti

Combination of tested materials

**Figure 14.** Failure load stages (FLS) obtained using the Gear Scuffing Shock Test for Coatings for the tested material

Figure 14 shows that the best resistance to scuffing (highest FLS) is observed when both gears are coated with the MoS2/Ti coating, or when the uncoated pinion meshes the coated wheel. As in the case of the a-C:H:W coating, when only the wheel is MoS2/Ti-coated, under "shock" conditions the resistance to scuffing is higher than in the situation when only the pinion coated.

observations taken during performing the Gear Scuffing EP Test for Coatings.

Figure 6. In this case, the grinding pattern is no longer visible.

*3.2.2. Material combinations with the MoS2/Ti coating*

coating are presented in Figure 14.

combinations with the MoS2/Ti coating

FLS

**Table 5.** Modes of the wear of the test pinion at particular load stages for the tested material combinations with the a-C:H:W coating, together with the total area of failures on the pinion (Ap), and wear of wheel (Ww), obtained in the Gear Scuffing Shock Test for Coatings; red-shadowed cells - the failure load stage (FLS)

Scuffing is observed only when both gears are uncoated - Table 5. This is one of the most dangerous modes of gear wear. As mentioned earlier, scuffing marks occur as single, fine marks or strips, or areas covering a part or all of the flank width. They appear as dull areas with the roughness much greater than the original criss-cross-grinding pattern shown in Figure 6. In this case, the grinding pattern is no longer visible.

To sum up this part of the experiment, the Gear Scuffing Shock Test for Coatings gives a much better resolution than the Gear Scuffing EP Test. However, one needs to have in mind that the cost of the former is about four-times higher than in the case of the latter, because the "shock tests" require more test gears to be used. From the point of view of the practical applications of the a-C:H:W coating in gears, the situation when the both gears are coated seems to be better than in the case of one of the gears uncoated, because it is exposed to the abrasive action of the meshing coated gear, which results in polishing or scoring. This positively verifies the observations taken during performing the Gear Scuffing EP Test for Coatings.

#### *3.2.2. Material combinations with the MoS2/Ti coating*

Load stage

208 Tribology - Fundamentals and Advancements

8

9

10

11

12

Ap = 703 mm2 Ww - not measured

> Ap ≈ 0 Ww = 180 mg

> Ap ≈ 0 Ww = 338 mg

**Table 5.** Modes of the wear of the test pinion at particular load stages for the tested material combinations with the a-C:H:W coating, together with the total area of failures on the pinion (Ap), and wear of wheel (Ww), obtained in the

Gear Scuffing Shock Test for Coatings; red-shadowed cells - the failure load stage (FLS)

Ap = 5 mm2 Ww = 76 mg

Ap = 6 mm2 Ww = 145 mg

Ap ≈ 0 Ww = 2 mg

Ap = 318 mm2 Ww = 3 mg

Failure load stages (FLS) obtained for the tested material combinations with the MoS2/Ti coating are presented in Figure 14.

Figure 14 shows that the best resistance to scuffing (highest FLS) is observed when both gears are coated with the MoS2/Ti coating, or when the uncoated pinion meshes the coated wheel.

As in the case of the a-C:H:W coating, when only the wheel is MoS2/Ti-coated, under "shock" conditions the resistance to scuffing is higher than in the situation when only the pinion coated. Hypothetically, there is a transfer of MoS2 (solid lubricant) from the teeth of the coated gear to the uncoated one. The transfer is more effective in the case of the MoS2/Ti-coated wheel meshing the uncoated pinion than in the opposite situation, because the area of the coated steel surface of the larger gear (wheel) is greater than in the case when the coating is deposited on the small gear (pinion).

Load stage

8

9

10

11

12

Ap = 703 mm2 Ww - not measured

> Ap = 32 mm2 Ww = 11 mg

Ap = 109 mm2 Ww = 25 mg

**Table 6.** Modes of wear of the test pinion at particular load stages for the tested material combinations with the MoS2/Ti coating, together with the total area of failures on the pinion (Ap), and wear of wheel (Ww), obtained in the

Gear Scuffing Shock Test for Coatings; red-shadowed cells - the failure load stage (FLS)

Ap ≈ 0 Ww = 16 mg

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

Ap ≈ 0 Ww = 16 mg

Ap ≈ 0 Ww = 9 mg

http://dx.doi.org/10.5772/54569

211

Ap ≈ 0 Ww = 9 mg

In comparison to the case when both gears are uncoated, a much higher resistance to scuffing is observed when the MoS2/Ti coating is deposited on one or two gears. The respective mechanisms of this behaviour were described earlier.

Table 6 presents the symbolic modes of wear of the test pinion at particular load stages for the tested material combinations with the MoS2/Ti coating. As in the case of the a-C:H:W coating, the mode of wear that appeared most often on the pinion teeth was taken into account. The photographs of the most often appearing mode of wear of the pinion at the highest load stage are also shown in the table. Red-shadowed cells in the table denote the failure load stage (FLS). Below are given the symbols of the wear modes, the total area of failures on the pinion (Ap) and wear of wheel (Ww). The symbols of wear were presented earlier in Table 2.

As can be observed in Table 6 for the tested material combinations, the two modes of wear that appear most often on the pinion teeth are scuffing and scoring. When one or both gears are coated, only scoring predominates on the pinion teeth.

Whenthepinioniscoatedandthewheelisuncoated,andwhenboththegearsarecoated,identical results were obtained for the two investigated coatings - FLS values are respectively 11 and higher than 12 (Figures 12 and 14). Therefore, the main criterion of assessment of the resist‐ ance to scuffing (FLS) makes it impossible to differentiate between these two situations. Under thesecircumstances,theanalysisofadditionalcriteriaoffailureassessment,relatedtothemodes of wear at particular load stages, like in the previous cases can give additional, valuable information.Inthe caseofthematerial combinationswiththe a-C:H:Wcoating,thepredominat‐ ing mode of wear of the pinion were only scratches. For the material combinations with the MoS2/Ti coating, the pinion wear was much more sever, and scoring instead of scratches could be met most often. Thus, the a-C:H:W coating provides better protection against severe wear than MoS2/Ti, especially when it is deposited on the both gears. This positively verifies the observations taken during performing the Gear Scuffing EP Test for Coatings.

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears http://dx.doi.org/10.5772/54569 211

Hypothetically, there is a transfer of MoS2 (solid lubricant) from the teeth of the coated gear to the uncoated one. The transfer is more effective in the case of the MoS2/Ti-coated wheel meshing the uncoated pinion than in the opposite situation, because the area of the coated steel surface of the larger gear (wheel) is greater than in the case when the coating is deposited on

In comparison to the case when both gears are uncoated, a much higher resistance to scuffing is observed when the MoS2/Ti coating is deposited on one or two gears. The respective

Table 6 presents the symbolic modes of wear of the test pinion at particular load stages for the tested material combinations with the MoS2/Ti coating. As in the case of the a-C:H:W coating, the mode of wear that appeared most often on the pinion teeth was taken into account. The photographs of the most often appearing mode of wear of the pinion at the highest load stage are also shown in the table. Red-shadowed cells in the table denote the failure load stage (FLS). Below are given the symbols of the wear modes, the total area of failures on the pinion (Ap)

As can be observed in Table 6 for the tested material combinations, the two modes of wear that appear most often on the pinion teeth are scuffing and scoring. When one or both gears are

Whenthepinioniscoatedandthewheelisuncoated,andwhenboththegearsarecoated,identical results were obtained for the two investigated coatings - FLS values are respectively 11 and higher than 12 (Figures 12 and 14). Therefore, the main criterion of assessment of the resist‐ ance to scuffing (FLS) makes it impossible to differentiate between these two situations. Under thesecircumstances,theanalysisofadditionalcriteriaoffailureassessment,relatedtothemodes of wear at particular load stages, like in the previous cases can give additional, valuable information.Inthe caseofthematerial combinationswiththe a-C:H:Wcoating,thepredominat‐ ing mode of wear of the pinion were only scratches. For the material combinations with the MoS2/Ti coating, the pinion wear was much more sever, and scoring instead of scratches could be met most often. Thus, the a-C:H:W coating provides better protection against severe wear than MoS2/Ti, especially when it is deposited on the both gears. This positively verifies the

and wear of wheel (Ww). The symbols of wear were presented earlier in Table 2.

observations taken during performing the Gear Scuffing EP Test for Coatings.

the small gear (pinion).

210 Tribology - Fundamentals and Advancements

Load stage

7

Ap = 26 mm2 Ww = 1 mg

mechanisms of this behaviour were described earlier.

coated, only scoring predominates on the pinion teeth.

**Table 6.** Modes of wear of the test pinion at particular load stages for the tested material combinations with the MoS2/Ti coating, together with the total area of failures on the pinion (Ap), and wear of wheel (Ww), obtained in the Gear Scuffing Shock Test for Coatings; red-shadowed cells - the failure load stage (FLS)

#### **4. Summary and conclusions**

The authors have developed unique test methods, being the subjects of this chapter. They are called the "Gear Scuffing EP Test for Coatings" and "Gear Scuffing Shock Test for Coatings." acquisition of the friction torque, which will make it possible to investigate, as postulated by gear transmissions manufacturers, the possibility of the reduction of friction between the meshing teeth by the application of a low-friction coating. At present, a new version of the T-12U test rig, denoted as T-12UF, is being developed within the framework of the Strategic Programme executed at ITeE-PIB in Radom, and the planned deadline of this work is in

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

213

**•** Both the differentiation between the tested objects (lubricating oils, material combinations) and the predictability of gear failures in real applications (transmissions) are important when assessing gear tests. This is why the authors plan to verify the results obtained by application of coated gears in transmissions (speed reducers) of different devices manufac‐ tured by one of the Polish producers. What is more, at present another test rig - a back-toback bevel gear test rig, denoted as T-30 - is being developed in the Tribology Department of ITeE-PIB in Radom with the deadline in 2012. The reason is that until now widely used test devices and methods have allowed researchers to perform runs on only spur gears having the tooth geometry significantly different than the geometry of bevel gears. The new

**•** From the means of the Minister of Science and Higher Education, executed within the Strategic Programme "Innovative Systems of Technical Support for Sustainable Develop‐ ment of the Country's Economy" within Innovative Economy Operational Programme.

**•** By the National Centre for Research and Development (NCBiR) within the scope of the R&D

The authors wish to express their thanks also to Dr. Maksim Antonov from Tallinn University (Estonia) for his support with the gear scuffing tests and helpful discussions, within the

Institute for Sustainable Technologies - National Research Institute (ITeE-PIB),Tribology De‐

, Marek Kalbarczyk, Michal Michalak, Witold Piekoszewski,

framework of Marie Curie RTN (6th EU FP); Contract No MRTN-CT-2006-035589.

Marian Szczerek, Waldemar Tuszynski and Jan Wulczynski

\*Address all correspondence to: remigiusz.michalczewski@itee.radom.pl

tribotester will allow researchers to better predict the failures of bevel gears.

2013.

**Acknowledgements**

Scientific work was financed:

project No. N R03 0019 06.

**Author details**

Remigiusz Michalczewski\*

partment, Radom, Poland

The analysis of the values of the failure load stage (FLS), reflecting the resistance to scuffing, shows that the developed Gear Scuffing EP Test for Coatings has too little resolution to differentiate between the tested material combinations - coating-coating (both gears coated), coating-steel, steel-coating, and also steel-steel (both gears without a coating). Additional criteria of failure assessment need to be employed to reveal minor differences between the tested material combinations observed only when the pinion is uncoated and the wheel is coated.

In comparison, Gear Scuffing Shock Test for Coatings makes it generally possible to differen‐ tiate between the tested material combinations from the point of view of the main criterion of assessment of the gear resistance to scuffing, i.e. FLS. Thus, this test method has a sufficient resolution. However, as in the case of the Gear Scuffing EP Test for Coatings, apart from the analysis of only FLS values, analysis of the additional criteria of failure assessment related to predominating modes of wear at particular load stages is recommended and may give additional, valuable information. For the two coatings tested (a-C:H:W and MoS2/Ti), the best resistance to scuffing/scoring (FLS > 12) is observed when both gears are coated; however, the a-C:H:W coating gives a slightly better protection against severe wear than MoS2/Ti - only scratches instead of scoring are observed for a-C:H:W.

Although the Gear Scuffing Shock Test for Coatings gives a much better resolution than the Gear Scuffing EP Test, one needs to have in mind that the cost of the former is about four-times higher than in the case of the latter, because the "shock tests" require more test gears to be used.

In the both tests, when one or both gears are coated, three modes of wear occur most often on the pinion teeth - polishing, scratches, or scoring. Scuffing is observed only when the two gears are uncoated.

The following conclusions can be drawn:


acquisition of the friction torque, which will make it possible to investigate, as postulated by gear transmissions manufacturers, the possibility of the reduction of friction between the meshing teeth by the application of a low-friction coating. At present, a new version of the T-12U test rig, denoted as T-12UF, is being developed within the framework of the Strategic Programme executed at ITeE-PIB in Radom, and the planned deadline of this work is in 2013.

**•** Both the differentiation between the tested objects (lubricating oils, material combinations) and the predictability of gear failures in real applications (transmissions) are important when assessing gear tests. This is why the authors plan to verify the results obtained by application of coated gears in transmissions (speed reducers) of different devices manufac‐ tured by one of the Polish producers. What is more, at present another test rig - a back-toback bevel gear test rig, denoted as T-30 - is being developed in the Tribology Department of ITeE-PIB in Radom with the deadline in 2012. The reason is that until now widely used test devices and methods have allowed researchers to perform runs on only spur gears having the tooth geometry significantly different than the geometry of bevel gears. The new tribotester will allow researchers to better predict the failures of bevel gears.

#### **Acknowledgements**

**4. Summary and conclusions**

212 Tribology - Fundamentals and Advancements

scratches instead of scoring are observed for a-C:H:W.

The following conclusions can be drawn:

coated.

used.

are uncoated.

rather expensive.

are a-C:H:W-coated.

The authors have developed unique test methods, being the subjects of this chapter. They are called the "Gear Scuffing EP Test for Coatings" and "Gear Scuffing Shock Test for Coatings."

The analysis of the values of the failure load stage (FLS), reflecting the resistance to scuffing, shows that the developed Gear Scuffing EP Test for Coatings has too little resolution to differentiate between the tested material combinations - coating-coating (both gears coated), coating-steel, steel-coating, and also steel-steel (both gears without a coating). Additional criteria of failure assessment need to be employed to reveal minor differences between the tested material combinations observed only when the pinion is uncoated and the wheel is

In comparison, Gear Scuffing Shock Test for Coatings makes it generally possible to differen‐ tiate between the tested material combinations from the point of view of the main criterion of assessment of the gear resistance to scuffing, i.e. FLS. Thus, this test method has a sufficient resolution. However, as in the case of the Gear Scuffing EP Test for Coatings, apart from the analysis of only FLS values, analysis of the additional criteria of failure assessment related to predominating modes of wear at particular load stages is recommended and may give additional, valuable information. For the two coatings tested (a-C:H:W and MoS2/Ti), the best resistance to scuffing/scoring (FLS > 12) is observed when both gears are coated; however, the a-C:H:W coating gives a slightly better protection against severe wear than MoS2/Ti - only

Although the Gear Scuffing Shock Test for Coatings gives a much better resolution than the Gear Scuffing EP Test, one needs to have in mind that the cost of the former is about four-times higher than in the case of the latter, because the "shock tests" require more test gears to be

In the both tests, when one or both gears are coated, three modes of wear occur most often on the pinion teeth - polishing, scratches, or scoring. Scuffing is observed only when the two gears

**•** The developed Gear Scuffing Shock Test for Coatings has been successfully verified by the testing of thin, hard coatings deposited on the gears; therefore, it can be implemented in the laboratories of the R&D centres devoted to surface engineering and the engineering of advanced materials intended for modern toothed gears, having in mind that this test is

**•** If the coating is intended for application on gears, from the point of view of the high‐ est achievable resistance to scuffing/scoring, it is recommended that both meshing gears

**•** Although the T-12U gear test rig has been effectively employed in the performed research, it is suggested that its research capacities should be extended by the measurement and data

Scientific work was financed:


The authors wish to express their thanks also to Dr. Maksim Antonov from Tallinn University (Estonia) for his support with the gear scuffing tests and helpful discussions, within the framework of Marie Curie RTN (6th EU FP); Contract No MRTN-CT-2006-035589.

#### **Author details**

Remigiusz Michalczewski\* , Marek Kalbarczyk, Michal Michalak, Witold Piekoszewski, Marian Szczerek, Waldemar Tuszynski and Jan Wulczynski

\*Address all correspondence to: remigiusz.michalczewski@itee.radom.pl

Institute for Sustainable Technologies - National Research Institute (ITeE-PIB),Tribology De‐ partment, Radom, Poland

#### **References**

[1] Pytko S., Sroda P. Classification and Evaluation of Machines for Investigation of Materials for Production of Gear Wheels. ZEM 1975; 1 39-58 (*in Polish*).

[16] Michaelis K., Höhn B.-R., Graswald C. Scuffing Tests for API GL-1 to GL-5 Gear Lubricants. In: proceedings of the 13th International Colloquium Tribology, 2002,

New Scuffing Test Methods for the Determination of the Scuffing Resistance of Coated Gears

http://dx.doi.org/10.5772/54569

215

[17] Michaelis K., Höhn B.-R., Oster P. Influence of Lubricant on Gear Failures - Test Methods and Application to Gearboxes in Practice. Tribotest journal 2004;11-1 43-56.

[18] Höhn B.-R., Oster P., Tobie T., Michaelis K. Test Methods for Gear Lubricants. Goriva

[19] Tuszynski W. Performance Classification of Automotive Gear Oils Using the Gear

[20] Kalin M., Vižintin J. The Tribological Performance of DLC-Coated Gears Lubricated with Biodegradable Oil in Various Pinion/Gear Material Combinations. Wear 2005;259

[21] Martins R.C., Moura Paulo S., Seabra J.O. MoS2/Ti Low-Friction Coating for Gears.

[22] Martins R., Amaro R., Seabra J. Influence of Low Friction Coatings on The Scuffing Load Capacity and Efficiency of Gears. Tribology International 2008;41 234-243.

[23] Szczerek M., Michalczewski R., Piekoszewski W. The Problems of Application of PVD/ CVD Thin Hard Coatings for Heavy-Loaded Machine Components. In: proceedings of the ASME/STLE International Joint Tribology Conference, 2007, San Diego, USA.

[24] Michalczewski R., Piekoszewski W., Szczerek M., Tuszynski W. The Lubricant-Coating Interaction in Rolling and Sliding Contacts. Tribology International 42;2009 554-560.

[25] Michalczewski R., Piekoszewski W., Szczerek M., Tuszynski W. Scuffing Resistance of DLC Coated Gears Lubricated with Ecological Oil. Estonian Journal of Engineering

[26] CEC L-07-95: Load Carrying Capacity Test for Transmission Lubricants (FZG Test Rig).

Scuffing Shock Test. Tribologia 2009;2 259-274 (*in Polish*).

Tribology International 2006;39 1686-1697.

Ostfildern, Germany.

i maziva 2008;47-2 141-152.

1270-1280.

2009;15-4 367-373.


[16] Michaelis K., Höhn B.-R., Graswald C. Scuffing Tests for API GL-1 to GL-5 Gear Lubricants. In: proceedings of the 13th International Colloquium Tribology, 2002, Ostfildern, Germany.

**References**

(*in Polish*).

214 Tribology - Fundamentals and Advancements

(accessed 22 September 2012).

Southwest Research Institute; 1968.

tigations. ZEM 1983;3 325-332.

journal 1999;5-4 383-390.

Sliding Circular Contacts. Wear 1996;201 217-226.

Washington, USA.

[1] Pytko S., Sroda P. Classification and Evaluation of Machines for Investigation of

[2] Tomaszewski J., Drewniak J., editors. Scuffing of Gears. Gliwice: CMG KOMAG; 2007

[3] ARTEC Machine Systems. Failure by scuffing due to poor distribution of load. http:// www.artec-machine.com/images/failure\_due\_to\_poor\_distribution\_of\_load\_1.pdf

[4] Kratz S.H., Wedeven L.D., Black W.F., Carlisle D.J., Wedeven G.G. Simulation of Space Shuttle Gear Performance. In: proceedings of the III World Tribology Congress, 2005,

[5] Kuo W.F., Chiou Y.C., Lee R.T. A Study on Lubrication Mechanism and Wear Scar in

[6] Ku P.M., editor. Interdisciplinary Approach to Friction and Wear. Washington D.C.:

[7] Makowska M., Matuszewska A., Gradkowski M. Migration of Active Elements from a Lubricant to the Material of the Friction Pair. Tribologia 2011;4 163-176 (*in Polish*).

[8] Wachal A. Analysis of Boundary Layer Estimating Criteria in Lubricating Oils Inves‐

[9] Höhn B.-R., Oster P., Schedl U. Pitting Load Capacity Test on The FZG Gear Test Rig with Load-Spectra and One-Stage Investigations. Tribotest journal 1999;5-4 417-430.

[10] Szczerek M., Tuszynski W. A Method for Testing Lubricants under Conditions of Scuffing. Part I. Presentation of the Method. Tribotest journal 2002;8-4 273-284.

[11] Piekoszewski W., Szczerek M., Tuszynski W. The Action of Lubricants under Extreme Pressure Conditions in a Modified Four-Ball Tester. Wear 2001;249 188-193.

[12] Tuszynski W., Michalczewski R., Piekoszewski W., Szczerek M. Effect of Ageing Automotive Gear Oils on Scuffing and Pitting. Tribology International 2008;41 875-888.

[13] Bisht R.P.S., Singhal S. A Laboratory Technique for the Evaluation of Automotive Gear

[14] Method to Assess the Scuffing Load Capacity of Lubricants with High EP Performance Using an FZG Gear Test Rig. FVA Information Sheet No. 243 Status June 2000.

[15] Höhn B.-R., Michaelis K., Eberspächer C., Schlenk L. A Scuffing Load Capacity Test with the FZG Gear Test Rig for Gear Lubricants with High EP Performance. Tribotest

Oils of API GL-4 Level. Tribotest journal 1999;6-1 69-77.

Materials for Production of Gear Wheels. ZEM 1975; 1 39-58 (*in Polish*).


**Chapter 7**

**Introduction of the Ratio of the Hardness to**

Modeling the wear rate is a complex process. The several possibilities of chemical, physical and mechanical changes at the interface are the most probable reasons for this [1]. In this manner, it is reasonable to consider the wear rate as a stochastic process [2, 3], and indeed this approach was taken into account by Archard [4], when he formulated his well-known model. Since then, the majority of available models are based on his proposition, independent on the characteristics of mechanical system. Considering a sharp contact, both Torrance [5] and Yi-Ling & Zi-Shan [6] for sliding and rolling abrasion, respectively, modified Archard's equation based on elastic effects, and the ratio of the hardness (H) to the Young's modulus (E) was the main parameter of the models. In a tribological system with dissimilar materials, for example a ceramic abrading a metal, one material can experience the yielding and the other the brittle failure. This difference in the mechanical behaviors can be decisive for the final performance

The use of Young's modulus to model the wear resistance was applied for coatings [7]; it appeared in other modifications of Archard's equation [8], and even in empirical relationships between the wear rate and the mechanical properties [9]. Eventually, in all cited cases, the

A selection of parameters involving hardness and elastic moduli can be summarized (Table 1) [9-11]. All of them have physical meanings that possess some interest for abrasion resistance

defines the transition between elastic to plastic contact in a ball-on-plane system, applying the analytical solutions provided by Hertz in Contact Mechanics [12], where the reduced modulus

/E2

© 2013 Pintaude; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,

© 2013 Pintaude; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

distribution, and reproduction in any medium, provided the original work is properly cited.

, is proportional to the load that

elastic modulus was not the reduced one (Er), as will be treated here.

of materials. The last parameter presented in Table 1, H3

is already taken into account.

**the Reduced Elastic Modulus for Abrasion**

Additional information is available at the end of the chapter

Giuseppe Pintaude

**1. Introduction**

to wear.

http://dx.doi.org/10.5772/55470

### **Introduction of the Ratio of the Hardness to the Reduced Elastic Modulus for Abrasion**

Giuseppe Pintaude

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/55470

#### **1. Introduction**

Modeling the wear rate is a complex process. The several possibilities of chemical, physical and mechanical changes at the interface are the most probable reasons for this [1]. In this manner, it is reasonable to consider the wear rate as a stochastic process [2, 3], and indeed this approach was taken into account by Archard [4], when he formulated his well-known model. Since then, the majority of available models are based on his proposition, independent on the characteristics of mechanical system. Considering a sharp contact, both Torrance [5] and Yi-Ling & Zi-Shan [6] for sliding and rolling abrasion, respectively, modified Archard's equation based on elastic effects, and the ratio of the hardness (H) to the Young's modulus (E) was the main parameter of the models. In a tribological system with dissimilar materials, for example a ceramic abrading a metal, one material can experience the yielding and the other the brittle failure. This difference in the mechanical behaviors can be decisive for the final performance to wear.

The use of Young's modulus to model the wear resistance was applied for coatings [7]; it appeared in other modifications of Archard's equation [8], and even in empirical relationships between the wear rate and the mechanical properties [9]. Eventually, in all cited cases, the elastic modulus was not the reduced one (Er), as will be treated here.

A selection of parameters involving hardness and elastic moduli can be summarized (Table 1) [9-11]. All of them have physical meanings that possess some interest for abrasion resistance of materials. The last parameter presented in Table 1, H3 /E2 , is proportional to the load that defines the transition between elastic to plastic contact in a ball-on-plane system, applying the analytical solutions provided by Hertz in Contact Mechanics [12], where the reduced modulus is already taken into account.

© 2013 Pintaude; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Pintaude; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Ei

νi

is the Young's modulus of conical indenter;

E is the Young's modulus of tested material, and;

**Figure 1.** Elastic recovery during an indentation process. Symbology - a: indentation radius, he: elastic recovery; hP: fi‐ nal depth; h: the maximum depth; hC: contact depth and hS: deflected depth. Adapted from ISO 14577-1 [14]

and it is called as 'indentation modulus', using the symbol EIT. Exactly this term was used by references [5] and [6]. In this way, the mechanical properties of abrasive particle were discarded in both cases. It is notable that this aspect has not been ruled out by Stilwell and

A great difference between the Torrance's paper [5] and the Yi-Ling and Zi-Shan one [6] is with respect to the volume of wear. In the former, it was taken as directly proportional to h<sup>2</sup>

as more appropriate because it takes into account the elastic effects at a worn surface, so that the final formulation provided by [6] will be presented. Thus, an equation for wear rate, Q

*P*

) can be found in ISO/FDIS 14577-1 standard [14],

, following the symbology of Figure 1. The latter can be considered

Introduction of the Ratio of the Hardness to the Reduced Elastic Modulus for Abrasion

http://dx.doi.org/10.5772/55470

219

*<sup>W</sup> QC K <sup>H</sup>* <sup>=</sup> (3)

, and

is the Poisson's ratio of conical indenter;

ν is the Poisson's ratio of tested material.

A modern definition for the term E/(1-ν<sup>2</sup>

2

Tabor [13] in 1961.

(m3

where,

for the latter, related to hP

/m), can be written as:

W is the applied load;

C is a constant and;

H is the hardness of worn material;

**Table 1.** Parameters based on the hardness and elastic moduli, used as indicators of abrasion resistance and their physical meanings

Using some of the abovementioned solutions, a case study will be presented. For two tribo‐ logical pairs with known wear coefficients, the ratio of hardness to reduced modulus works well than the single property of the worn material (E). The expectation regarding the H/Er ratio is confirmed also by other aspects used to characterize abrasion, especially the cutting efficiency.

#### **2. Modelling abrasion with E/H ratio**

In 1980 Torrance [5] published a model for abrasive wear rate based on the elastic recovery after scratching, supposing that the abrasive particle has a conical geometry. This choice is suitable, because there is an analytical model to describe the changes at the recovered surface [13]. Some years later, another paper [6] adopted the same physical basis but here the appli‐ cation occurred to systems where the abrasive particles roll, instead of slide. The key similar‐ ities and differences of both manuscripts will be discussed below.

First of all, it is important to see the main definition presented in [13], because it was the basis for the referred models. This reference presents an equation that relates the reduced modulus with the amount of elastic recovery, he (indicated in Figure 1), considering a conical indenter:

$$h\_e = h \quad -h\_P = \frac{H \times \pi \times a}{E\_r} \tag{1}$$

In Equation 1 the term Er is the reduced modulus, defined as:

$$\frac{1}{E\_r} = \frac{1 - \upsilon\_i^2}{E\_i} + \frac{1 - \nu^2}{E} \tag{2}$$

where,

Er is the reduced modulus;

Ei is the Young's modulus of conical indenter;

νi is the Poisson's ratio of conical indenter;

**Parameter Physical meaning (taking into account a rigid-plastic material)**

**Table 1.** Parameters based on the hardness and elastic moduli, used as indicators of abrasion resistance and their

Using some of the abovementioned solutions, a case study will be presented. For two tribo‐ logical pairs with known wear coefficients, the ratio of hardness to reduced modulus works well than the single property of the worn material (E). The expectation regarding the H/Er ratio is confirmed also by other aspects used to characterize abrasion, especially the cutting

In 1980 Torrance [5] published a model for abrasive wear rate based on the elastic recovery after scratching, supposing that the abrasive particle has a conical geometry. This choice is suitable, because there is an analytical model to describe the changes at the recovered surface [13]. Some years later, another paper [6] adopted the same physical basis but here the appli‐ cation occurred to systems where the abrasive particles roll, instead of slide. The key similar‐

First of all, it is important to see the main definition presented in [13], because it was the basis for the referred models. This reference presents an equation that relates the reduced modulus with the amount of elastic recovery, he (indicated in Figure 1), considering a conical indenter:

*r*

n

=-= (1)

= + (2)

*E* ´ ´ p

(H/E)2 Transition on mechanical contact – elastic to plastic [9]

<sup>2</sup> Resistance to the plastic indentation [10] H3/E2 Resistance to the plastic indentation [11]

H/E Deformation relative to yielding [9]

**2. Modelling abrasion with E/H ratio**

ities and differences of both manuscripts will be discussed below.

In Equation 1 the term Er is the reduced modulus, defined as:

*e P*

*r i*

*H a hhh*

<sup>2</sup> <sup>2</sup> 1 1 <sup>1</sup> *<sup>i</sup>*

*v EE E* - -

H2/2E Modulus of resilience [9]

218 Tribology - Fundamentals and Advancements

H/Er

physical meanings

efficiency.

where,

Er is the reduced modulus;

E is the Young's modulus of tested material, and;

ν is the Poisson's ratio of tested material.

**Figure 1.** Elastic recovery during an indentation process. Symbology - a: indentation radius, he: elastic recovery; hP: fi‐ nal depth; h: the maximum depth; hC: contact depth and hS: deflected depth. Adapted from ISO 14577-1 [14]

A modern definition for the term E/(1-ν<sup>2</sup> ) can be found in ISO/FDIS 14577-1 standard [14], and it is called as 'indentation modulus', using the symbol EIT. Exactly this term was used by references [5] and [6]. In this way, the mechanical properties of abrasive particle were discarded in both cases. It is notable that this aspect has not been ruled out by Stilwell and Tabor [13] in 1961.

A great difference between the Torrance's paper [5] and the Yi-Ling and Zi-Shan one [6] is with respect to the volume of wear. In the former, it was taken as directly proportional to h<sup>2</sup> , and for the latter, related to hP 2 , following the symbology of Figure 1. The latter can be considered as more appropriate because it takes into account the elastic effects at a worn surface, so that the final formulation provided by [6] will be presented. Thus, an equation for wear rate, Q (m3 /m), can be written as:

$$Q = C \frac{W}{H} K\_p \tag{3}$$

where,

W is the applied load;

H is the hardness of worn material;

C is a constant and;

KP will be called here as partial wear coefficient, based on elastic effects during indentation, defined as (1 + *k* ×*H* / *E*)<sup>2</sup> , being k another constant.

In order to differentiate the use of elastic modulus to the reduced one, another symbology will be considered for the latter case, where KP' is introduced:

$$K\_p\text{ '} = \left(1 + k \times H / E\_r\right)^2\tag{4}$$

Figure 2 shows a linear relationship between the relative wear resistance β<sup>i</sup>

present the same behavior when they are abraded by hard particles (Figure 3).

The experimental results obtained in [19] are shown in Figure 5.

ness, considering pure metals and heat-treated steels. Adapted from reference [18]

where,

this curve included pure metals and heat-treated steels. These groups of materials do not

In other words, a pure metal with similar hardness of heat-treated steel presents a higher abrasion resistance. This implies that steels present a different slope on a wear resistance curve when it is put as a function of hardness. Using the ratio of the hardness to the Young's modulus, Torrance [5] put in the same curve these referred groups of materials, which present, a priori, different behaviors. This kind of result was also described in reference [19], but based on a different approach. This research made use of the abrasion factor (fab) definition (Equation 6 and Figure 4 [20]) to describe the wear resistance. It is a parameter related with the cutting efficiency, i.e., when this factor is equal to unity only the micro-cutting would be the wear mechanism, while for values correspondent to zero no removal of material would be detected.

**Figure 3.** Schematic representation of the wear resistance (Q' = 1/Q) produced by hard particles as a function of hard‐

2 1


2 *ab A A <sup>f</sup> <sup>A</sup>*

A1 is the cross section area relative to pileup produced by a single scratch and;

A2 is the cross section area relative to the groove produced by a single scratch.

partial wear coefficient, defined in [4] as β<sup>i</sup>

and the relative

221

http://dx.doi.org/10.5772/55470

\* (Equation 5). The experimental points used to build

Introduction of the Ratio of the Hardness to the Reduced Elastic Modulus for Abrasion

where,

k is a constant.

To define the constant C, Yi-Ling and Zi-Shan [6] made use of nine constants to fit the experimental results. The constant k of term KP varies with many variables of tribosystem. On the other hand, although Torrance [5] has not make use of constants to fit experimental results obtained by others [15-17], he chose a material as reference (a steel of 401 HV hardness), and all cases were then compared with this material (Equation 5). In this way, he only needed to calculate the constant k (=10), finding a very interesting result, as can be seen in Figure 2.

$$
\beta\_i \,^\* = \frac{H}{H\_{ref}} \frac{\left(1 + 10H\_{ref} / E\_{ref}\right)}{\left(1 + 10H / E\right)}\tag{5}
$$

**Figure 2.** Relative wear resistance (β<sup>i</sup> ) as a function of relative partial wear coefficient (β<sup>i</sup> \*), as defined in [5]. Experi‐ mental points derived from [15-17]

Figure 2 shows a linear relationship between the relative wear resistance β<sup>i</sup> and the relative partial wear coefficient, defined in [4] as β<sup>i</sup> \* (Equation 5). The experimental points used to build this curve included pure metals and heat-treated steels. These groups of materials do not present the same behavior when they are abraded by hard particles (Figure 3).

In other words, a pure metal with similar hardness of heat-treated steel presents a higher abrasion resistance. This implies that steels present a different slope on a wear resistance curve when it is put as a function of hardness. Using the ratio of the hardness to the Young's modulus, Torrance [5] put in the same curve these referred groups of materials, which present, a priori, different behaviors. This kind of result was also described in reference [19], but based on a different approach. This research made use of the abrasion factor (fab) definition (Equation 6 and Figure 4 [20]) to describe the wear resistance. It is a parameter related with the cutting efficiency, i.e., when this factor is equal to unity only the micro-cutting would be the wear mechanism, while for values correspondent to zero no removal of material would be detected. The experimental results obtained in [19] are shown in Figure 5.

**Figure 3.** Schematic representation of the wear resistance (Q' = 1/Q) produced by hard particles as a function of hard‐ ness, considering pure metals and heat-treated steels. Adapted from reference [18]

$$f\_{ab} = \frac{A2 - A1}{A2} \tag{6}$$

where,

KP will be called here as partial wear coefficient, based on elastic effects during indentation,

In order to differentiate the use of elastic modulus to the reduced one, another symbology will

( )

To define the constant C, Yi-Ling and Zi-Shan [6] made use of nine constants to fit the experimental results. The constant k of term KP varies with many variables of tribosystem. On the other hand, although Torrance [5] has not make use of constants to fit experimental results obtained by others [15-17], he chose a material as reference (a steel of 401 HV hardness), and all cases were then compared with this material (Equation 5). In this way, he only needed to calculate the constant k (=10), finding a very interesting result, as can be seen in Figure 2.

> ( ) ( )

*H H E H H E*

*ref ref*

) as a function of relative partial wear coefficient (β<sup>i</sup>

1 10 / \* 1 10 /

*ref*

<sup>+</sup> <sup>=</sup> <sup>+</sup>

*i*

b

<sup>2</sup> '1 / *K kHE P r* = +´ (4)

(5)

\*), as defined in [5]. Experi‐

, being k another constant.

be considered for the latter case, where KP' is introduced:

defined as (1 + *k* ×*H* / *E*)<sup>2</sup>

220 Tribology - Fundamentals and Advancements

where,

k is a constant.

**Figure 2.** Relative wear resistance (β<sup>i</sup>

mental points derived from [15-17]

A1 is the cross section area relative to pileup produced by a single scratch and;

A2 is the cross section area relative to the groove produced by a single scratch.

static (hardness test) and kinetic cases (scratch test), being the higher the pileup, the smaller

Introduction of the Ratio of the Hardness to the Reduced Elastic Modulus for Abrasion

http://dx.doi.org/10.5772/55470

223

**Figure 6.** Physical parameters extracted from a residual profile of a spherical indentation. Notation: *ac* is the indenta‐ tion radius at the contact, and *s* is the height associated to the indentation morphology. Pileup is associated to the

In reference [22] an equation that correlates the pileup formation with the H/E ratio, based on results obtained in scratch tests conducted with a Berkovich indenter (Figure 7) can be found:

Important evidence was detected in [23] that the application of Equation (7) was dependent

**Figure 7.** Relation between the pileup (hc/h) and the ln(E/H), obtained after scratch tests with Berkovich indenter.

*h H* <sup>=</sup> - (7)

0.41498 ln 0.14224 *Ch <sup>E</sup>*

the cutting efficiency.

hc/h ratio. Adapted from reference [21]

on the level of applied load.

Adapted from reference [22]

**Figure 4.** Abrasion factor definition (areas A1 and A2 defined in Equation 6). Real profile adapted from reference [20]

**Figure 5.** Abrasive wear resistance (Q') produced by flint particles as a function of wear debris hardness-to-fab ratio. Adapted from reference [19], which defines Q' as a non dimensional parameter

Again, in Figure 5 both pure metals and steels are put in the same curve, showing that the hardness alone, in some cases, is not a complete descriptor of the wear resistance. The similarity between the results presented in Figure 2 and Figure 5 opens a possibility to the abrasion factor to be described as a function of the hardness-to-elastic modulus ratio.

Although in reference [19] the elastic effects have not been incorporated to the model, there are results in the literature relating the pileup formation (Figure 6, [21]) with the mechanical properties [22]. In this case, it is possible to consider that the pileup formation (hc/h) works for static (hardness test) and kinetic cases (scratch test), being the higher the pileup, the smaller the cutting efficiency.

**Figure 6.** Physical parameters extracted from a residual profile of a spherical indentation. Notation: *ac* is the indenta‐ tion radius at the contact, and *s* is the height associated to the indentation morphology. Pileup is associated to the hc/h ratio. Adapted from reference [21]

**Figure 4.** Abrasion factor definition (areas A1 and A2 defined in Equation 6). Real profile adapted from reference [20]

**Figure 5.** Abrasive wear resistance (Q') produced by flint particles as a function of wear debris hardness-to-fab ratio.

Again, in Figure 5 both pure metals and steels are put in the same curve, showing that the hardness alone, in some cases, is not a complete descriptor of the wear resistance. The similarity between the results presented in Figure 2 and Figure 5 opens a possibility to the abrasion factor

Although in reference [19] the elastic effects have not been incorporated to the model, there are results in the literature relating the pileup formation (Figure 6, [21]) with the mechanical properties [22]. In this case, it is possible to consider that the pileup formation (hc/h) works for

Adapted from reference [19], which defines Q' as a non dimensional parameter

222 Tribology - Fundamentals and Advancements

to be described as a function of the hardness-to-elastic modulus ratio.

In reference [22] an equation that correlates the pileup formation with the H/E ratio, based on results obtained in scratch tests conducted with a Berkovich indenter (Figure 7) can be found:

$$
\left\langle \begin{matrix} \hbar\_{\text{C}} \\ \hbar \end{matrix} \right\rangle\_{\hbar} = 0.41498 \ln \left\langle \begin{matrix} -0.14224 \\ H \end{matrix} \right\rangle\_{H} - 0.14224 \tag{7}
$$

Important evidence was detected in [23] that the application of Equation (7) was dependent on the level of applied load.

**Figure 7.** Relation between the pileup (hc/h) and the ln(E/H), obtained after scratch tests with Berkovich indenter. Adapted from reference [22]

#### **3. Case study**

Two tribological pairs were studied, also considered in a previous investigation [24]. Their wear rates are known: glass abrading a quenched and tempered (Q&T) 52100 steel [25], and alumina wearing a hard metal [26]. The mechanical properties, determined from instrumented indentation testing, used to calculate the partial wear coefficients, KP and KP' (Equations 3 and 4), are presented in Table 2. To calculate the reduced modulus (Equation 2), Poisson's ratios were extracted from [27].

The resulting values of partial wear coefficients, using 8.25 as k factor, are presented in

Glass - Q&T steel 1.41 2.66 Alumina – Hard metal 1.41 1.98

Table 3 shows that the KP values were similar for the studied pairs, while the KP' did not follow this trend. The wear coefficients of tribological pairs (K values) determined using sliding abrasion tests were 0.014 and 0.008, for glass against steel [25] and alumina against hard metal [26], respectively. Visibly this difference in wear rates is only reflected by the KP' values, which

Another observation for the KP values is that the similarity presented in Table 3 is not affected by the variation of the k factor. On the contrary, KP' is strongly dependent on this factor, as

One can ask about the good agreement with experimental data found in [5] and [6] when they used only the Young's modulus in wear model. The key point for that is the superiority of abrasive hardness in relation to the worn surface. All experimental results in these cases were performed using abrasives much harder than the tested materials, such that the mechanical properties of them can be considered as unaltered along the tests. An investigation [29] showed significant changes in glass particles when they abraded steel surfaces, even for non heattreated ones. The same thing was demonstrated in [30] for different abrasives wearing WC-Co thermal sprayed coatings. When the abrasive particle is relatively soft to the abraded material, their mechanical properties play a key role during the wear process, and an extensive discus‐

In addition, in reference [24] these tribological pairs were separated using the difference in the plasticity index, δH, defined in [32] (Equation 8), being the smallest difference related to the calcite-fluorite, and the largest for glass-steel pair within the materials analyzed in [24]. This aspect seems to be important again when one observes Figure 8. When the difference is insignificant, as in the case of calcite-fluorite, the variation of KP' with k factor is minor, and on the other hand, for the case of the glass-steel pair, for a notable difference in the plasticity

*<sup>δ</sup><sup>H</sup>* =1−14.3(1−*<sup>ν</sup>* <sup>−</sup>2*<sup>ν</sup>* 2)*<sup>H</sup>*

A brief discussion of the k factor is instructive. Following [6], this factor is especially associated to the particle geometry. As stated, a hypothetical KP' of calcite-fluorite pair would be less affected by the variation on k factor. Therefore, one can imagine a small fragment of mineral

*<sup>E</sup>* (8)

**Table 3.** Partial wear coefficient values for selected tribological pairs

is a direct result of the ratio of hardness to the reduced modulus.

**Pair KP (equation 3) KP' (equation 4)**

Introduction of the Ratio of the Hardness to the Reduced Elastic Modulus for Abrasion

http://dx.doi.org/10.5772/55470

225

Table 3.

can be seen in Figure 8.

sion on it can be found in [31].

index, a great variation of KP' with the k factor occurs.

As seen in the previous item, due to the similarity of Figures 2 and 5, the partial wear coefficient can be well related to the abrasion factor (cutting efficiency), fab. An indirect way to know the fab factor is found in reference [28], which defined it as the ratio between the wear coefficient K (Archard's definition) and the ploughing fraction of friction coefficient (µp). The cutting efficiency values, following this definition, are 0.106 for glass abrading bearing steel and 0.079 for alumina wearing hard metal, a difference of 34%. As the constant KP' can vary along a broad range (Figure 8), we select a value so as to the difference is also 34%, and for this purpose it is 8.25.


**Table 2.** Mechanical properties of selected tribological pairs. \*Obs.: Q&T steel is a wire-drawing, which implies in a reduction of elastic modulus due to the work-hardening effect

An example of variation in KP' value with k factor is presented in Figure 8, for the tribological pairs considered in Table 2. Another pair was added (calcite-fluorite) in this figure, in order to help the discussion with their differences.

**Figure 8.** Variation of KP' wear coefficient with factor k for three cases

The resulting values of partial wear coefficients, using 8.25 as k factor, are presented in Table 3.


**Table 3.** Partial wear coefficient values for selected tribological pairs

**3. Case study**

224 Tribology - Fundamentals and Advancements

were extracted from [27].

Soda-lime glass 4.07 69

Alumina 19.6 376.1

reduction of elastic modulus due to the work-hardening effect

**Figure 8.** Variation of KP' wear coefficient with factor k for three cases

to help the discussion with their differences.

8.25.

Two tribological pairs were studied, also considered in a previous investigation [24]. Their wear rates are known: glass abrading a quenched and tempered (Q&T) 52100 steel [25], and alumina wearing a hard metal [26]. The mechanical properties, determined from instrumented indentation testing, used to calculate the partial wear coefficients, KP and KP' (Equations 3 and 4), are presented in Table 2. To calculate the reduced modulus (Equation 2), Poisson's ratios

As seen in the previous item, due to the similarity of Figures 2 and 5, the partial wear coefficient can be well related to the abrasion factor (cutting efficiency), fab. An indirect way to know the fab factor is found in reference [28], which defined it as the ratio between the wear coefficient K (Archard's definition) and the ploughing fraction of friction coefficient (µp). The cutting efficiency values, following this definition, are 0.106 for glass abrading bearing steel and 0.079 for alumina wearing hard metal, a difference of 34%. As the constant KP' can vary along a broad range (Figure 8), we select a value so as to the difference is also 34%, and for this purpose it is

**Material H, GPa E, GPa** *Er***, GPa H/E H/Er**

Q&T steel 5.5 180\* 0.023 0.076

Hard metal (grade K) 11 480 0.023 0.049

**Table 2.** Mechanical properties of selected tribological pairs. \*Obs.: Q&T steel is a wire-drawing, which implies in a

An example of variation in KP' value with k factor is presented in Figure 8, for the tribological pairs considered in Table 2. Another pair was added (calcite-fluorite) in this figure, in order

53.24

222.43

0.080 0.103

0.052 0.088

Table 3 shows that the KP values were similar for the studied pairs, while the KP' did not follow this trend. The wear coefficients of tribological pairs (K values) determined using sliding abrasion tests were 0.014 and 0.008, for glass against steel [25] and alumina against hard metal [26], respectively. Visibly this difference in wear rates is only reflected by the KP' values, which is a direct result of the ratio of hardness to the reduced modulus.

Another observation for the KP values is that the similarity presented in Table 3 is not affected by the variation of the k factor. On the contrary, KP' is strongly dependent on this factor, as can be seen in Figure 8.

One can ask about the good agreement with experimental data found in [5] and [6] when they used only the Young's modulus in wear model. The key point for that is the superiority of abrasive hardness in relation to the worn surface. All experimental results in these cases were performed using abrasives much harder than the tested materials, such that the mechanical properties of them can be considered as unaltered along the tests. An investigation [29] showed significant changes in glass particles when they abraded steel surfaces, even for non heattreated ones. The same thing was demonstrated in [30] for different abrasives wearing WC-Co thermal sprayed coatings. When the abrasive particle is relatively soft to the abraded material, their mechanical properties play a key role during the wear process, and an extensive discus‐ sion on it can be found in [31].

In addition, in reference [24] these tribological pairs were separated using the difference in the plasticity index, δH, defined in [32] (Equation 8), being the smallest difference related to the calcite-fluorite, and the largest for glass-steel pair within the materials analyzed in [24]. This aspect seems to be important again when one observes Figure 8. When the difference is insignificant, as in the case of calcite-fluorite, the variation of KP' with k factor is minor, and on the other hand, for the case of the glass-steel pair, for a notable difference in the plasticity index, a great variation of KP' with the k factor occurs.

$$\delta\_H = 1 - 14.3 \left\{ 1 - \nu - 2\nu^2 \right\} ^H \Big|\_{E} \tag{8}$$

A brief discussion of the k factor is instructive. Following [6], this factor is especially associated to the particle geometry. As stated, a hypothetical KP' of calcite-fluorite pair would be less affected by the variation on k factor. Therefore, one can imagine a small fragment of mineral used in Mohs scale [33] and, considering the previous assertive, conclude that its geometry would not be important. Probably, this is not the case. The k factor should be understood in a broad manner, i.e., all variables of a system can alter its value. Thus, slight changes in tribo‐ logical variables could bring higher alterations in the wear coefficient for the pair glass-steel than for the other described cases. At this moment, no experimental result is available to corroborate this hypothesis, but it is an interesting field to be explored further.

E Young's modulus

Er Reduced modulus

hC Contact depth

he Elastic recovery

hS Deflected depth

K Wear coefficient

hP Final depth

k Constant

Q Wear rate

W Applied load

βi

βi

νi

Q' Wear resistance (= 1/Q)

Relative wear resistance

*δH* Plasticity parameter

\* Relative partial wear coefficient

ν Poisson's ratio of the worn material

Poisson's ratio of the indenter

Young's modulus of indenter

*Eref* Young's modulus of a reference material

Introduction of the Ratio of the Hardness to the Reduced Elastic Modulus for Abrasion

http://dx.doi.org/10.5772/55470

227

fab Abrasion factor or cutting efficiency

*Hd* Vickers hardness of wear debris

h Maximum depth at applied load

*H* Vickers hardness of the worn material

*Href* Vickers hardness of a reference material

KP Partial wear coefficient, defined as (1 + *k* ×*H* / *E*)<sup>2</sup>

*s* Height associated to the indentation morphology

µp Ploughing fraction of friction coefficient, taken as 0.2

KP' Partial wear coefficient calculated with Er instead of E

Ei

Finally, a discussion concerning the Equation (7) and the values presented in Table 2 is valuable. If the values of H/E ratio for steel and alumina were applied in Equation (7), the results would be similar. A similar height of pileup for these materials obviously is not a reasonable result, taking into account the experimental values obtained for wear and friction coefficients for them. Nevertheless, applying the reduced modulus in the place of E in that equation, one can find fewer tendencies to form pileup after abrasion of steel by glass, which means a higher cutting efficiency in this case, meeting with the values described before. Consequently, it is more a case of successful application of reduced modulus to predict changes at the mechanical contact.

#### **4. Conclusions and final remarks**

The viability of the use the hardness-to-reduced modulus ratio to model the wear coefficient for abraded materials was demonstrated. Previous models were developed taking into account only the Young's modulus of worn surface, discarding the properties of abrasive material. These cases work only for pairs where the abrasive particle is harder than the abraded material and it was demonstrated that they fail when the abrasive hardness is relatively low.

In addition, other questions were discussed and they open some possibilities to carry out future research. First, the model of wear coefficient treated here involves higher requirements, because a constant is needed. This constant seems to affect more the wear coefficient of some pairs than the others, and the reason for that is not clear. Finally, an extensive work could be made exploring the relation between cutting efficiency (abrasion factor) and H/Er ratio, computing a large variation of the applied load and the abrasive (indenter) properties. Probably, an investigation based on these aspects should supply answers to the improvement of a wear model containing the H/Er ratio.

#### **Nomenclature**

A1 and A2 Cross section areas used in the definition of abrasion factor

*a* Indentation radius


E Young's modulus

used in Mohs scale [33] and, considering the previous assertive, conclude that its geometry would not be important. Probably, this is not the case. The k factor should be understood in a broad manner, i.e., all variables of a system can alter its value. Thus, slight changes in tribo‐ logical variables could bring higher alterations in the wear coefficient for the pair glass-steel than for the other described cases. At this moment, no experimental result is available to

Finally, a discussion concerning the Equation (7) and the values presented in Table 2 is valuable. If the values of H/E ratio for steel and alumina were applied in Equation (7), the results would be similar. A similar height of pileup for these materials obviously is not a reasonable result, taking into account the experimental values obtained for wear and friction coefficients for them. Nevertheless, applying the reduced modulus in the place of E in that equation, one can find fewer tendencies to form pileup after abrasion of steel by glass, which means a higher cutting efficiency in this case, meeting with the values described before. Consequently, it is more a case of successful application of reduced modulus to predict changes

The viability of the use the hardness-to-reduced modulus ratio to model the wear coefficient for abraded materials was demonstrated. Previous models were developed taking into account only the Young's modulus of worn surface, discarding the properties of abrasive material. These cases work only for pairs where the abrasive particle is harder than the abraded material

In addition, other questions were discussed and they open some possibilities to carry out future research. First, the model of wear coefficient treated here involves higher requirements, because a constant is needed. This constant seems to affect more the wear coefficient of some pairs than the others, and the reason for that is not clear. Finally, an extensive work could be made exploring the relation between cutting efficiency (abrasion factor) and H/Er ratio, computing a large variation of the applied load and the abrasive (indenter) properties. Probably, an investigation based on these aspects should supply answers to the improvement

and it was demonstrated that they fail when the abrasive hardness is relatively low.

A1 and A2 Cross section areas used in the definition of abrasion factor

corroborate this hypothesis, but it is an interesting field to be explored further.

at the mechanical contact.

226 Tribology - Fundamentals and Advancements

**4. Conclusions and final remarks**

of a wear model containing the H/Er ratio.

**Nomenclature**

*a* Indentation radius

C Constant

*ac* Indentation radius at the contact


#### **Acknowledgements**

Financial support for this study was provided by CNPq under project no. 306727/2011-0.

[10] Joslin D.L., Oliver W.C. A New Method for Analyzing Data from Continuous Depth-Sensing Microindentation Tests. Journal of Materials Research 1990; 5(1) 123-126.

Introduction of the Ratio of the Hardness to the Reduced Elastic Modulus for Abrasion

http://dx.doi.org/10.5772/55470

229

[11] Tsui T.Y., Pharr G.M., Oliver, W.C., Bhatia C.S., White R.L., Anders A., Brown I.G. Nanoindentation and Nanoscratching of Hard Carbon Coatings for Magnetic Disks.

[12] Brizmer V., Kligerman Y., Etsion I. The Effect of Contact Conditions and Material Properties on the Elasticity Terminus of a Spherical Contact. International Journal of

[13] Stilwell, N. A., Tabor, D. Elastic Recovery of Conical Indentations. Proceedings of the

[14] ISO - International Organization for Standardization. ISO/FDIS 14577-1 - Metallic Materials - Instrumented Indentation Test for Hardness and Material Parameter -

[15] Richardson R.C.D. The Maximum Hardness of Strained Surfaces and the Abrasive

[16] Richardson R. C. D. The Wear of Metals by Hard Abrasives, Wear 1967; 10(4)

[17] Richardson R.C.D. The Wear of Metals by Relatively Soft Abrasives, Wear 1968; 11(4)

[18] Murray M.J., Mutton P.J., Watson J.D. Abrasive Wear Mechanisms in Steels. In: Lu‐ dema K.C., Glaeser W.A., Rhee S.K. (eds.), WOM 1979: Proceedings of International Conference on Wear of Materials, 16-18 April 1979, Dearborn, MI. New York: Ameri‐

[19] Zum Gahr, K.H. Modelling of Two-Body Abrasive Wear. Wear 1987; 124(1) 87–103.

[20] Buttery T.C., Archard J.F. Grinding and abrasive wear. Proceedings of the Institution

[21] Pintaude G., Hoechele A.R., Cipriano G.L. Relation between Strain Hardening Expo‐ nent of Metals and Residual Profiles of Deep Spherical Indentation. Materials Science

[22] Jardret V., Zahouani H., Loubet J.L., Mathia T.G. Understanding and Quantification of Elastic and Plastic Deformation during a Scratch Test. Wear 1998; 218(1) 8-14.

[23] Masen M.A., de Rooij M.B., Schipper D.J., Adachi K., Kato K. Single asperity abrasion

[24] Pintaude G. An Overview of the Hardness Differential Required for Abrasion, Jour‐

of coated nodular cast iron. Tribology International 2007; 40(2)170–179.

MRS Proceedings 1994;356, p.767.

Physical Society 1961; 78(2) 169-179.

291-309.

245-275.

Solids and Structures 2006; 43(18-19) 5736–5749.

Part 1: Test method. Geneva, Switzerland, 2002.

Wear of Metals and Alloys, Wear, 1967; 10(5) 353-382.

can Society of Mechanical Engineers, pp. 257–265, 1979.

of Mechanical Engineers 1970; 185(1) 537-552.

and Technology 2012; 28(9-10) 1051-1054.

nal of Tribology 2010;132(3) paper 034502.

#### **Author details**

Giuseppe Pintaude\*

Address all correspondence to: giuseppepintaude@gmail.com

Mechanical Academic Department, Federal University of Technology – Paraná, Curitiba, Brazil

#### **References**


[10] Joslin D.L., Oliver W.C. A New Method for Analyzing Data from Continuous Depth-Sensing Microindentation Tests. Journal of Materials Research 1990; 5(1) 123-126.

**Acknowledgements**

228 Tribology - Fundamentals and Advancements

**Author details**

Giuseppe Pintaude\*

Brazil

**References**

021013.

1953;24(8) 981–988.

246(1-2) 1-11.

Financial support for this study was provided by CNPq under project no. 306727/2011-0.

Mechanical Academic Department, Federal University of Technology – Paraná, Curitiba,

[1] Bayer R.G., Comments on Engineering Needs and Wear Models. In: Ludema K.C., Bayer, R.G. (eds.) Tribological Modeling for Mechanical designers. Philadelphia:

[2] Silva Jr., C.R.A., Pintaude G., Al-Qureshi H. A., Krajnc M.A. An Application of Mean Square Calculus to Sliding Wear. Journal of Applied Mechanics 2010;77(2) paper

[3] Silva Jr., C.R.A., Pintaude G. Uncertainty Analysis on the Wear Coefficient of Arch‐

[4] Archard J.F. Contact and Rubbing of Flat Surfaces. Journal of Applied Physics

[6] Yi-Ling W., Zi-Shan W. An Analysis of the Influence of Plastic Indentation on Three-

[7] Leyland A., Matthews A. On the Significance of the H/E Ratio in Wear Control: a Nanocomposite Coating Approach to Optimised Tribological Behaviour. Wear 2000;

[8] Liu R., Li D.Y. Modification of Archard's Equation by Taking Account of elastic/

[9] Finkin E.F. Examination of abrasion Resistance Criteria for Some Ductile Metals.

pseudoelastic Properties of Materials. Wear 2001; 251(1-12) 956-964.

[5] Torrance A.A. The Correlation of Abrasive Wear Tests. Wear 1980;63(2) 359-370.

Address all correspondence to: giuseppepintaude@gmail.com

ASTM International; 1991. p. 3-11.

ard Model. Tribology International 2008;41(6) 473-481.

Body Abrasive Wear of Metals. Wear 1988; 122(2) 123-133.

Journal of Lubrication Technology 1974;96(2) 210-214.


[25] Pintaude G., Bernardes F.G., Santos M.M., Sinatora A., Albertin E. Mild and Severe Wear of Steels and Cast Irons in Sliding Abrasion. Wear 2009;267(1-4) 19-25.

**Chapter 8**

**Artificial Slip Surface: Potential Application in Lubricated**

For the last years, there has been a tremendous effort towards the development of Micro-Electro-Mechanical System (MEMS) for a wide variety of applications in aerospace, automo‐ tive, biomedical, computer, agricultural industries, electronic instrumentation, industrial process control, biotechnology, office equipment, and telecommunications. MEMS devices integrate chemical, physical, and even biological processes in micro-scale technology

Stiction (a subtraction of 'static friction') in micro-system technology has been a problem ever since the advent of surface micromachining in the eighties of the last century. As the overall size of the machine is reduced, the capillary and surface tension force of liquid become large, which induce stiction rendering the devices to fail or malfunction. In particular, stiction forces created between moving parts that come into contact with one another, either intentionally or accidentally, during operation are a common problem with micro-mechanical devices. Stiction-type failures occur when the interfacial attraction forces exceed restoring forces. Consequently, the surfaces of these parts either temporarily or permanently adhere to each

Several approaches to address the stiction between two opposing surfaces have been presented in the various literatures [1-4]. The basic approaches to prevent stiction include increasing surface roughness (topography) and/or lowering solid surface energy by coating with low surface energy materials. This includes self-assembled molecular (SAM) coatings, hermetic

> © 2013 Tauviqirrahman et al.; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2013 Tauviqirrahman et al.; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,

distribution, and reproduction in any medium, provided the original work is properly cited.

**MEMS**

D.J. Schipper

**1. Introduction**

**1.1. Background**

packages.

http://dx.doi.org/10.5772/55745

M. Tauviqirrahman, R. Ismail, J. Jamari and

Additional information is available at the end of the chapter

other, causing device malfunction or failure.

packaging and the use of reactive materials in the package [5].


**Chapter 8**

### **Artificial Slip Surface: Potential Application in Lubricated MEMS**

M. Tauviqirrahman, R. Ismail, J. Jamari and D.J. Schipper

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/55745

#### **1. Introduction**

[25] Pintaude G., Bernardes F.G., Santos M.M., Sinatora A., Albertin E. Mild and Severe Wear of Steels and Cast Irons in Sliding Abrasion. Wear 2009;267(1-4) 19-25.

[26] Pintaude G., Farias M.C.M., Kohnlein M., Tanaka D.K., Sinatora A. Abrasive Wear of Cutting Tools Used in the Wood Industry (In Portuguese). In: UFRN – Universidade Federal do Rio Grande do Norte (ed.) CONEM 2000: XV Brazilian Congress of Me‐ chanical Engineering, 7-11 August 2000, Natal, Brazil. Rio de Janeiro: Associação Bra‐

[27] Gercek H. Poisson's Ratio Values for Rocks. International Journal Rock Mechanics

[28] Jacobson S., Wallen P., Hogmark S. Fundamental Aspects of Abrasive Wear Studied

[29] Pintaude G., Tanaka D.K., Sinatora A. The Effects of Abrasive Particle Size on the Sliding Friction Coefficient of Steel using a Spiral Pin-on-Disk Apparatus. Wear 2003;

[30] Bozzi A. C.; De Mello J.D.B. Wear Resistance and Wear Mechanisms of WC-12%Co Thermal Sprayed Coatings in Three-Body Abrasion, Wear 1999;233-235, 575-587. [31] Pintaude, G. (2011). Characteristics of Abrasive Particles and Their Implications on Wear, New Tribological Ways, Ghrib, T. (Ed.), ISBN: 978-953-307-206-7, InTech, Available from: http://www.intechopen.com/books/new-tribological-ways/character‐

[32] Milman Y.V., Galanov B.A., Chugunova B.I. Plasticity Characteristic Obtained through Hardness Measurement, Acta Metallurgica et Materialia 1993; 41(9)

[33] Broz M.E., Cook R.F., Whitney D.L. Microhardness, Toughness, and Modulus of

by a New Numerical Simulation Model. Wear 1998;123(2) 207-223.

istics-of-abrasive-particles-and-their-implications-on-wear

Mohs Scale Minerals. American Mineralogist 2006; 91(1) 135–142.

sileira de Ciências Mecânicas, 2000.

and Mining Sciences 2007;44(1) 1–13.

255 (1-6) 55-59.

230 Tribology - Fundamentals and Advancements

2523-2532.

#### **1.1. Background**

For the last years, there has been a tremendous effort towards the development of Micro-Electro-Mechanical System (MEMS) for a wide variety of applications in aerospace, automo‐ tive, biomedical, computer, agricultural industries, electronic instrumentation, industrial process control, biotechnology, office equipment, and telecommunications. MEMS devices integrate chemical, physical, and even biological processes in micro-scale technology packages.

Stiction (a subtraction of 'static friction') in micro-system technology has been a problem ever since the advent of surface micromachining in the eighties of the last century. As the overall size of the machine is reduced, the capillary and surface tension force of liquid become large, which induce stiction rendering the devices to fail or malfunction. In particular, stiction forces created between moving parts that come into contact with one another, either intentionally or accidentally, during operation are a common problem with micro-mechanical devices. Stiction-type failures occur when the interfacial attraction forces exceed restoring forces. Consequently, the surfaces of these parts either temporarily or permanently adhere to each other, causing device malfunction or failure.

Several approaches to address the stiction between two opposing surfaces have been presented in the various literatures [1-4]. The basic approaches to prevent stiction include increasing surface roughness (topography) and/or lowering solid surface energy by coating with low surface energy materials. This includes self-assembled molecular (SAM) coatings, hermetic packaging and the use of reactive materials in the package [5].

© 2013 Tauviqirrahman et al.; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Tauviqirrahman et al.; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Other attractive technique to tackle the stiction problem is by inserting a lubricant into the region around the interacting devices to reduce the chance of stiction-type failures. As is wellknown, many MEMS devices include moving (sliding/rolling) surfaces and thus it is necessary to apply a lubricant between the contacting surfaces to reduce friction and wear. However, a significant barrier to the development of MEMS lubrication is the problem of achieving effective tribological performance of their moving parts. This is because the lubricant behavior is different at micro-scale compared to macro-scale. At the macroscopic level, it is well accepted that the boundary condition for a viscous fluid at a solid wall is no-slip, i.e. the fluid velocity matches the velocity of the solid boundary. While the no-slip boundary condition has been proven experimentally to be accurate for a number of macroscopic flows, it remains an assumption that is not based on physicals principles. At micro-scale level, certain phenomena must be taken into account when analyzing liquid flows such as a slip condition at solid wall boundaries.

of control over the hydrodynamic pressure in confined systems and be important in lubricated-MEMS. To prevent a stiction, in a controlled way, one is able to enhance, a hydrophobic/ hydrophilic behavior of surfaces. If one surface is hydrophobic (slip) and the other is hydro‐ philic (no-slip) the sliding velocity or displacement between the surfaces is accommodated by shear at the hydrophobic surface (the lubricant is kept in the contact by the hydrophilic surface). In this way wear of the surfaces is prevented and the surfaces are able to move because

Artificial Slip Surface: Potential Application in Lubricated MEMS

http://dx.doi.org/10.5772/55745

233

The slip situation, however, can be controlled to obtain a positive effect by surface technology. Coating and texturing technologies can be used to engineer large slip. In practice, a large slip can be made using super-hydrophobic surfaces. Such surfaces can be manufactured by grafting or by deposition of hydrophobic compounds on the initial surface at a certain zone. Superhydrophobic surfaces were originally inspired by the unique water-repellent properties of the lotus leaf. It is the combination of a very large contact angle and a low contact-angle hysteresis that defines a surface as super-hydrophobic. Implementing the slip property (hydrophobicity) on a surface in a wide range of application for the mechanical components is of great challenge by numerous authors recently. In published works [15-23], both experimentally and numerical‐ ly, slip surface is able to reduce friction force at the contacting surfaces and finally reduce energy consumption, increase component's life-time and reduce economic and environmental costs.

In classical liquid lubrication it is assumed that surfaces are fully wetted and no-slip occurs between the fluid and the solid boundary. In MEMS, this wetting is actually an unwanted process because it can encourage the occurrence of stiction and as a result micro-parts can not be moved [25]. It is expected that slip can reduce the friction and improve the load support. However, with respect to the engineered slip pattern, the choice of slip zone on a certain surface must be taken carefully in relation to such tribological performances. In other words, an inappropriate slip zone pattern on a certain surface or the election of inappropriate surface containing a slip situation may lead to the deterioration of the lubrication performance. How to control the boundary slip in the application of a lubricated-MEMS is one of the challenging tasks in the future. This chapter will explore the provision of a new lubrication model based on the continuum approach for moving parts in MEMS in order to improve the tribological performance of lubricated con‐ tacts. In MEMS, by lubrication, low friction force and high load support are the goals which want to be achieved. The artificial slip surface will be introduced as one of the solutions to improve the lubrication performance of MEMS so as MEMS with a longer life-time can be obtained. The term "artificial slip surface" is used to address a non-homogeneous engineered slip/no-slip

Full film lubrication of lubricated contact is often described by the Reynolds theory [26]. According to the classical Reynolds theory, no-slip boundary is assumed and the convergent

pattern, i.e. a surface consisting of a slip zone and a no-slip zone.

stiction is prevented.

**1.2. Problem statement**

**2. Research methods**

As a consequence of the MEMS technology revolutionary application to many areas, it is possible for scientists to observe the boundary slip on micro/nano-meter scale. A variety of techniques are now available that are capable of probing lubricant flow on micro-scales and are therefore suitable for the investigation of boundary conditions. There are three techniques so far for detecting the boundary slip: nano-particle image velocimetry (NPIV) [6], atomic force microscope (AFM) [7-9] and surface force apparatus (SFA) [10]. The NPIV technique is a direct observation method with a measurement precision depending on the size of the nano-particles but with poor moderate accuracy. The AFM and SFA are indirect observation techniques based on the assumption that boundary slip takes place precisely on the interface of liquid and solid. These methods need a high accuracy boundary slip model to infer the slip velocity. Boundary slip has been observed not only for a hydrophobic surface [6, 7, 10] but also for a hydrophilic surface [8, 9]. Therefore, the slip evidence has been generally accepted and for certain cases the no-slip boundary condition is not valid.

There is a large body of literature dealing with the analysis of lubricant slip flow based on the analytical and numerical solution of molecular dynamic simulations [11, 12], Lattice-Boltz‐ man [13, 14], and the Reynolds equation [15-23]. The accurate description of slip at the wall is very difficult and still remains a subject of intensive research. The so-called Navier slip model and the critical shear stress model are usually used to describe a boundary slip. In fact, nearly two hundred years ago Navier [24] proposed a general boundary condition that incorporates the possibility of fluid slip at a solid boundary. Navier's proposed boundary condition as‐ sumes that the velocity, *u*, at a solid surface is proportional to the shear stress at the surface. It reads: *u* = *b* (d*u*/d*z*) where *b* is the slip length and d*u*/d*z* is the shear rate. The slip length *b*, which is defined as the distance beyond the solid/liquid interface at which the liquid velocity extrapo‐ lates to the velocity of the solid, is used to quantify a boundary slip. If *b* = 0 then the generally assumed no-slip boundary condition is obtained. If *b* = finite, fluid slip occurs at the wall, but its effect depends upon the length scale of the flow. The Navier-slip boundary condition is the most widely used boundary condition with the methods based on the solution of continuum equations.

In micro-scales such as MEMS, the boundary condition will play a very important role in determining the lubricant flow behavior. Control of the boundary condition will allow a degree of control over the hydrodynamic pressure in confined systems and be important in lubricated-MEMS. To prevent a stiction, in a controlled way, one is able to enhance, a hydrophobic/ hydrophilic behavior of surfaces. If one surface is hydrophobic (slip) and the other is hydro‐ philic (no-slip) the sliding velocity or displacement between the surfaces is accommodated by shear at the hydrophobic surface (the lubricant is kept in the contact by the hydrophilic surface). In this way wear of the surfaces is prevented and the surfaces are able to move because stiction is prevented.

The slip situation, however, can be controlled to obtain a positive effect by surface technology. Coating and texturing technologies can be used to engineer large slip. In practice, a large slip can be made using super-hydrophobic surfaces. Such surfaces can be manufactured by grafting or by deposition of hydrophobic compounds on the initial surface at a certain zone. Superhydrophobic surfaces were originally inspired by the unique water-repellent properties of the lotus leaf. It is the combination of a very large contact angle and a low contact-angle hysteresis that defines a surface as super-hydrophobic. Implementing the slip property (hydrophobicity) on a surface in a wide range of application for the mechanical components is of great challenge by numerous authors recently. In published works [15-23], both experimentally and numerical‐ ly, slip surface is able to reduce friction force at the contacting surfaces and finally reduce energy consumption, increase component's life-time and reduce economic and environmental costs.

#### **1.2. Problem statement**

Other attractive technique to tackle the stiction problem is by inserting a lubricant into the region around the interacting devices to reduce the chance of stiction-type failures. As is wellknown, many MEMS devices include moving (sliding/rolling) surfaces and thus it is necessary to apply a lubricant between the contacting surfaces to reduce friction and wear. However, a significant barrier to the development of MEMS lubrication is the problem of achieving effective tribological performance of their moving parts. This is because the lubricant behavior is different at micro-scale compared to macro-scale. At the macroscopic level, it is well accepted that the boundary condition for a viscous fluid at a solid wall is no-slip, i.e. the fluid velocity matches the velocity of the solid boundary. While the no-slip boundary condition has been proven experimentally to be accurate for a number of macroscopic flows, it remains an assumption that is not based on physicals principles. At micro-scale level, certain phenomena must be taken into account when analyzing liquid flows such as a slip condition at solid wall

As a consequence of the MEMS technology revolutionary application to many areas, it is possible for scientists to observe the boundary slip on micro/nano-meter scale. A variety of techniques are now available that are capable of probing lubricant flow on micro-scales and are therefore suitable for the investigation of boundary conditions. There are three techniques so far for detecting the boundary slip: nano-particle image velocimetry (NPIV) [6], atomic force microscope (AFM) [7-9] and surface force apparatus (SFA) [10]. The NPIV technique is a direct observation method with a measurement precision depending on the size of the nano-particles but with poor moderate accuracy. The AFM and SFA are indirect observation techniques based on the assumption that boundary slip takes place precisely on the interface of liquid and solid. These methods need a high accuracy boundary slip model to infer the slip velocity. Boundary slip has been observed not only for a hydrophobic surface [6, 7, 10] but also for a hydrophilic surface [8, 9]. Therefore, the slip evidence has been generally accepted and for certain cases

There is a large body of literature dealing with the analysis of lubricant slip flow based on the analytical and numerical solution of molecular dynamic simulations [11, 12], Lattice-Boltz‐ man [13, 14], and the Reynolds equation [15-23]. The accurate description of slip at the wall is very difficult and still remains a subject of intensive research. The so-called Navier slip model and the critical shear stress model are usually used to describe a boundary slip. In fact, nearly two hundred years ago Navier [24] proposed a general boundary condition that incorporates the possibility of fluid slip at a solid boundary. Navier's proposed boundary condition as‐ sumes that the velocity, *u*, at a solid surface is proportional to the shear stress at the surface. It reads: *u* = *b* (d*u*/d*z*) where *b* is the slip length and d*u*/d*z* is the shear rate. The slip length *b*, which is defined as the distance beyond the solid/liquid interface at which the liquid velocity extrapo‐ lates to the velocity of the solid, is used to quantify a boundary slip. If *b* = 0 then the generally assumed no-slip boundary condition is obtained. If *b* = finite, fluid slip occurs at the wall, but its effect depends upon the length scale of the flow. The Navier-slip boundary condition is the most widely used boundary condition with the methods based on the solution of continuum equations.

In micro-scales such as MEMS, the boundary condition will play a very important role in determining the lubricant flow behavior. Control of the boundary condition will allow a degree

boundaries.

232 Tribology - Fundamentals and Advancements

the no-slip boundary condition is not valid.

In classical liquid lubrication it is assumed that surfaces are fully wetted and no-slip occurs between the fluid and the solid boundary. In MEMS, this wetting is actually an unwanted process because it can encourage the occurrence of stiction and as a result micro-parts can not be moved [25]. It is expected that slip can reduce the friction and improve the load support. However, with respect to the engineered slip pattern, the choice of slip zone on a certain surface must be taken carefully in relation to such tribological performances. In other words, an inappropriate slip zone pattern on a certain surface or the election of inappropriate surface containing a slip situation may lead to the deterioration of the lubrication performance. How to control the boundary slip in the application of a lubricated-MEMS is one of the challenging tasks in the future. This chapter will explore the provision of a new lubrication model based on the continuum approach for moving parts in MEMS in order to improve the tribological performance of lubricated con‐ tacts. In MEMS, by lubrication, low friction force and high load support are the goals which want to be achieved. The artificial slip surface will be introduced as one of the solutions to improve the lubrication performance of MEMS so as MEMS with a longer life-time can be obtained. The term "artificial slip surface" is used to address a non-homogeneous engineered slip/no-slip pattern, i.e. a surface consisting of a slip zone and a no-slip zone.

#### **2. Research methods**

Full film lubrication of lubricated contact is often described by the Reynolds theory [26]. According to the classical Reynolds theory, no-slip boundary is assumed and the convergent geometrical wedge is one of the most important conditions to generate hydrodynamic pressure. Therefore, the lubrication model for lubricated MEMS will be an extension of the classical lubrication theory. This means that modeling the lubricant through very narrow gap, normally modeled by assuming no-slip at the boundaries will be modified by introducing a boundary slip.

#### **2.1. Modified reynolds equation**

The classical Reynolds equation that is valid under no-slip condition can be generalized for taking into account slip conditions. It is then possible, for any film height distributions, to calculate the pressure distribution and the shear rate profile. The model of lubrication presented here is based on the fact that slip of the lubricant will exist at the interface of a lubricated sliding contact. Thus, a boundary slip is employed both on the moving and stationary surface, see Figure 1. The proposed lubrication model with slip leads to a modified Reynolds equation as presented in Eq. (1).

$$\begin{split} \frac{\partial}{\partial \mathbf{x}} \left( \frac{h^3}{12\mu} \frac{h^2 + 4h\mu \left( a\_t + a\_b \right) + 12\mu^2 a\_t a\_b}{h \left( h + \mu \left( a\_t + a\_b \right) \right)} \frac{\partial p}{\partial \mathbf{x}} \right) &= \frac{u\_w}{2} \frac{\partial}{\partial \mathbf{x}} \left( \frac{h^2 + 2h\mu a\_t}{h + \mu \left( a\_t + a\_b \right)} \right) \\ -u\_w \frac{a\_t \mu}{h + \mu \left( a\_t + a\_b \right)} \frac{\partial h}{\partial \mathbf{x}} + \frac{h}{2\mu} \frac{\partial p}{\partial \mathbf{x}} \frac{\partial h}{\partial \mathbf{x}} \frac{h\alpha\_t \mu + 2a\_t a\_b \mu^2}{h + \mu \left( a\_t + a\_b \right)} \end{split} \tag{1}$$

The physical meanings of the symbols in Eq. (1) are as follows: *h* the lubrication film thickness (gap) at location, *p* the lubrication film pressure, *μ* the lubricant viscosity, *α* the slip coefficient (subscripts *t* and *b* denote the top (stationary) and bottom (moving) surface, respectively) and *uw* the velocity of the moving surface. It can be seen in that if the slip coefficient *α* is set to zero (no-slip condition), Eq. (1) reduces to the classical Reynolds equation. It should be noted that the product of multiplication of the slip coefficient by the viscosity, *αμ*, is usually called as 'slip length' *b*.

Eq. (1) is derived by following the usual approach to deduce the Reynolds equation from the Navier-Stokes system by assuming classical assumptions except that boundary slip is applied both on the stationary surface and moving surface as depicted in Figure 1.

Eq. (1) can be derived by considering the equilibrium of an element of fluid.

$$\frac{\partial^2 \mu}{\partial \mathbf{z}^2} = \frac{1}{\mu} \frac{\partial p}{\partial \mathbf{x}} \tag{2}$$

2

*C*1 and *C*2 in this case are either constants or functions of *x* and can be solved by applying a boundary condition for *u*. The bottom and the top surfaces have the slip condition based on

**Figure 1.** Schematic of a lubricated sliding contact with artificial slip surface applied both on the stationary and the mov‐ ing surface. The boundary slip zones (*Sxt* and *Sxb* ) are located at the leading edge of the contact. (Note: *uw* is the sliding ve‐

*<sup>p</sup> u z Cz C*

*w*

*<sup>u</sup> at z u u*

*<sup>u</sup> at z h u*

*t*

<sup>ý</sup> ¶ <sup>ï</sup> = = - ¶ <sup>ï</sup>

 m

*t b t b*

¶ ¶ ++ ++ ++ è ø

 m

*h p h x h* m

*p p h h h*

¶ ¶ æ ö + +

*b*

 m

= =+ <sup>ï</sup> ¶ <sup>ï</sup>

1 2

Artificial Slip Surface: Potential Application in Lubricated MEMS

0,

,

<sup>2</sup> <sup>1</sup> 2

 

*x x h h u*

*u z z u*

2

m

=- + + ç ÷

2 2

modified Reynolds equation as stated in Eq. (1).

m

slip length model.

locity, *Lx* is contact length, *hi*

This gives

m

m *x* ¶ = ++

lubricant *hi*

*Sxb*

*h(x) z* 

*x* 

*Sxt*

1 2

and *ho* are inlet and outlet film thickness, respectively, *h*(*x*) is variable film thickness).

*Lx*

0

¶ ü

=

*z*

*z*

*z*

*z h*

=

( ) ( ) ( )

 

*w tt*

m

( ) (2)

 m

*t b*

 

*b t*

This velocity is used to compute the flow rate, *q* by integrating across the fluid film thickness, *h*. When *q* is differentiated to fulfill the continuity of flow, assuming *μ* is constant; this gives a

m

 m þ

*w*

¶ <sup>+</sup> - ¶ + + (6)

*h*

*t b*

 

 m

m

¶ (3)

*uw*

Artificial Slip Surface: Potential Application in Lubricated MEMS

5

235

http://dx.doi.org/10.5772/55745

*ho*

(4)

(5)

where *z* lies along the direction across the thickness of the film, and *u* is the velocity field. To obtain the velocity profile, Eq. (2) can be integrated twice.

**Figure 1.** Schematic of a lubricated sliding contact with artificial slip surface applied both on the stationary and the mov‐ ing surface. The boundary slip zones (*Sxt* and *Sxb* ) are located at the leading edge of the contact. (Note: *uw* is the sliding ve‐ locity, *Lx* is contact length, *hi* and *ho* are inlet and outlet film thickness, respectively, *h*(*x*) is variable film thickness).

$$
\mu = \frac{1}{2\mu} \frac{\partial p}{\partial \mathbf{x}} z^2 + \mathbf{C}\_1 z + \mathbf{C}\_2 \tag{3}
$$

*C*1 and *C*2 in this case are either constants or functions of *x* and can be solved by applying a boundary condition for *u*. The bottom and the top surfaces have the slip condition based on slip length model.

$$\begin{aligned} \text{at } z = 0, \ u &= u\_w + \alpha\_b \mu \frac{\partial u}{\partial z} \bigg|\_{z=0} \\ \text{at } z = h, \quad u &= -\alpha\_t \mu \frac{\partial u}{\partial z} \bigg|\_{z=h} \end{aligned} \tag{4}$$

This gives

geometrical wedge is one of the most important conditions to generate hydrodynamic pressure. Therefore, the lubrication model for lubricated MEMS will be an extension of the classical lubrication theory. This means that modeling the lubricant through very narrow gap, normally modeled by assuming no-slip at the boundaries will be modified by introducing a

The classical Reynolds equation that is valid under no-slip condition can be generalized for taking into account slip conditions. It is then possible, for any film height distributions, to calculate the pressure distribution and the shear rate profile. The model of lubrication presented here is based on the fact that slip of the lubricant will exist at the interface of a lubricated sliding contact. Thus, a boundary slip is employed both on the moving and stationary surface, see Figure 1. The proposed lubrication model with slip leads to a modified

( ( )) ( )

*t b tb t t b t b*

2

m

 m

*t b*

 

4 12 2

m

The physical meanings of the symbols in Eq. (1) are as follows: *h* the lubrication film thickness (gap) at location, *p* the lubrication film pressure, *μ* the lubricant viscosity, *α* the slip coefficient (subscripts *t* and *b* denote the top (stationary) and bottom (moving) surface, respectively) and *uw* the velocity of the moving surface. It can be seen in that if the slip coefficient *α* is set to zero (no-slip condition), Eq. (1) reduces to the classical Reynolds equation. It should be noted that the product of multiplication of the slip coefficient by the viscosity, *αμ*, is usually called as 'slip

Eq. (1) is derived by following the usual approach to deduce the Reynolds equation from the Navier-Stokes system by assuming classical assumptions except that boundary slip is applied

2

m

 

(1)

m

¶ ¶ <sup>=</sup> ¶ ¶ (2)

*w*

boundary slip.

**2.1. Modified reynolds equation**

234 Tribology - Fundamentals and Advancements

Reynolds equation as presented in Eq. (1).

*t b*

 

m 

*w*

m

> m


m

*u*

length' *b*.

( )

m

*h h x xx*

12 2

*x h h x x h*

 

*hh h p h*

2

+ + ¶ ¶¶ + +

m

( ) ( )

both on the stationary surface and moving surface as depicted in Figure 1.

Eq. (1) can be derived by considering the equilibrium of an element of fluid.

2 2 *u* 1 *p*

*z*

obtain the velocity profile, Eq. (2) can be integrated twice.

m*x*

where *z* lies along the direction across the thickness of the film, and *u* is the velocity field. To

*t t tb*

¶ ¶ ¶ +

3 2 2 2

 m 

*h h h p u hh*

¶ ¶ æ ö + ++ ¶ æ ö <sup>+</sup> ç ÷ <sup>=</sup> ç ÷ ¶ + + ¶ ¶ + + è ø è ø

$$u = \frac{1}{2} \frac{\partial p}{\partial x} z^2 - \left( \frac{u\_w}{h + \mu \left( a\_t + a\_b \right)} + \frac{h}{2\mu} \frac{\partial p}{\partial x} \frac{h + 2a\_t \mu}{h + \mu \left( a\_t + a\_b \right)} \right) z + u\_w \frac{h + a\_t \mu}{h + \mu \left( a\_t + a\_b \right)} \tag{5}$$

$$-\frac{\hbar}{2\mu} \frac{\partial p}{\partial \alpha} \frac{a\_b \mu (h + 2a\_t \mu)}{h + \mu \left(a\_t + a\_b\right)}\tag{6}$$

This velocity is used to compute the flow rate, *q* by integrating across the fluid film thickness, *h*. When *q* is differentiated to fulfill the continuity of flow, assuming *μ* is constant; this gives a modified Reynolds equation as stated in Eq. (1).

When full film lubrication is assumed, the entire load *w* is carried by the lubricant film and the calculation is simply an integration of the lubricant film pressure distribution over contact area, i.e.

$$\varphi w = \int\_0^{L\_x} p(\mathbf{x}) d\mathbf{x} \tag{7}$$

The parametric analysis is performed using a developed computer code to investigate the effect of various slip parameters on the lubrication performances (load support, friction force, and friction coefficient). A parametric study is conducted with the variation of slip parameters (slip zone and slip length) over a large range of values considering different performance param‐ eters. The design variables and the objective function are referred to as the optimization variables. The design variables are independent quantities which are varied in order to achieve the optimum design. The objective function is the dependent variable that is maximized, i.e. the load support. In the present study, the design variables are slip zones as indicated in Figure

Artificial Slip Surface: Potential Application in Lubricated MEMS

http://dx.doi.org/10.5772/55745

237

1. The algorithm used in the present study is depicted on Figure 2.

Figure 2. Flow chart for numerical method.

The behavior of traditional (no-slip) hydrodynamic lubrication between the opposing surfaces can be estimated by a classical form of the Reynolds equation. The derivation of the classical Reynolds equation is based on the assumption of no-slip between the lubricant and the surfaces. In the classical Reynolds lubrication, the mechanism to generate a pressure is due to the convergent wedge effect. An artificial slip surface presented here is designed to be able to carry the external load during lubrication even if the

In this chapter, there are two main investigations. At first, the study is conducted in order to validate the developed numerical scheme. It assures that the numerical method used can be employed for solving other hydrodynamic characteristics. The no-slip case of lubricated contact is of main interest due to the availability of the analytical solution. Secondly, the study will be extended to explore the effect of the slip zone of the artificial slip surface on pressure, load support, friction force, and friction coefficient. The comparison between the modified sliding contact containing the artifical slip surface and the traditional one is conducted in order

The modified Reynolds equation (Eq. (1)) is the governing equation for the fluid lubrication system containing a boundary slip. If

, is set to zero, Eq. (1) reduces to the classical Reynolds equation. In this section, in order to validate the

wedge effect is not present. This situation is very beneficial in designing lubricated-MEMS which exhibits parallel gaps.

to describe the benefit of the use of an artificial slip pattern quantitatively.

**3. Key results** 

**Figure 2.** Flow chart for numerical method.

**3.1. No-slip condition** 

the slip coefficient,

The friction force *f* generated by the lubrication system is due to the fluid viscous shear. It is calculated by integrating the shear stress over the surface area. These shear stresses are given by

$$\pi\left(\mathbf{x}, z\right) = \left(\mu \frac{\partial \mu}{\partial z}\right)\_{z=h} \tag{8}$$

The simulation results will be presented in dimensionless form, i.e. *P* = *pho* <sup>2</sup> / *μL <sup>x</sup>uw* for the dimensionless pressure, *W* =*who* <sup>2</sup> / (*uwμL <sup>x</sup>* 2 ) for the dimensionless load support (where *w* is the load per unit width), *F* = *f ho* / *μuw L <sup>x</sup>* for dimensionless friction force (where *f* is the friction force per unit width), and *m = F / W* for dimensionless friction coefficient. In the present study, the dimensionless slip length *A* is determined by normalizing the slip length *b* with the outlet film thickess *ho.* For slip analysis in the following computations, the dimensionless slip length *A* varies from 3 to 300, which are reasonable values of the slip length based on the results published in literature [17, 18, 20].

#### **2.2. Solution method**

The modified Reynolds equation, Eq. (1) is discretized over the flow using the finite volume method, and is solved using the tridiagonal matrix algorithm (TDMA), [27]. By employing the discretization scheme, the computed domain is divided into a number of control volumes using a grid with uniform mesh size. The grid independency is validated by various numbers of mesh sizes. An assumption is made that the boundary pressures are zero at both sides of the contact.

A numerical simulation is conducted to investigate the possible application so as a boundary slip can be beneficial to achieve a high load support and low friction force. In order to maximize the performance of lubrication, the boundary conditions (slip zones, *Sxt* and *Sxb*, see Figure 1) of the model are optimized through a parametric analysis. The object of optimization is to maximize the hydrodynamic load support. The load support satisfies two main functional purposes: (a) carry the applied external load, and (b) to minimize the contact between the opposing solids, and thus wear. The optimization analysis attempts to satisfy both functional requirements with a single design parameter, the slip zone.

The behavior of traditional (no-slip) hydrodynamic lubrication between the opposing surfaces can be estimated by a classical form of the Reynolds equation. The derivation of the classical Reynolds equation is based on the assumption of no-slip between the lubricant and the surfaces. In the classical Reynolds lubrication, the mechanism to generate a pressure is due to the convergent wedge effect. An artificial slip surface presented here is designed to be able to carry the external load during lubrication even if the

In this chapter, there are two main investigations. At first, the study is conducted in order to validate the developed numerical scheme. It assures that the numerical method used can be employed for solving other hydrodynamic characteristics. The no-slip case of lubricated contact is of main interest due to the availability of the analytical solution. Secondly, the study will be extended to explore the effect of the slip zone of the artificial slip surface on pressure, load support, friction force, and friction coefficient. The comparison between the modified sliding contact containing the artifical slip surface and the traditional one is conducted in order

The modified Reynolds equation (Eq. (1)) is the governing equation for the fluid lubrication system containing a boundary slip. If

, is set to zero, Eq. (1) reduces to the classical Reynolds equation. In this section, in order to validate the

wedge effect is not present. This situation is very beneficial in designing lubricated-MEMS which exhibits parallel gaps.

to describe the benefit of the use of an artificial slip pattern quantitatively.

The parametric analysis is performed using a developed computer code to investigate the effect of various slip parameters on the lubrication performances (load support, friction force, and friction coefficient). A parametric study is conducted with the variation of slip parameters (slip zone and slip length) over a large range of values considering different performance param‐ eters. The design variables and the objective function are referred to as the optimization variables. The design variables are independent quantities which are varied in order to achieve the optimum design. The objective function is the dependent variable that is maximized, i.e. the load support. In the present study, the design variables are slip zones as indicated in Figure 1. The algorithm used in the present study is depicted on Figure 2.

Figure 2. Flow chart for numerical method. **Figure 2.** Flow chart for numerical method.

**3. Key results** 

**3.1. No-slip condition** 

the slip coefficient,

When full film lubrication is assumed, the entire load *w* is carried by the lubricant film and the calculation is simply an integration of the lubricant film pressure distribution over contact

The friction force *f* generated by the lubrication system is due to the fluid viscous shear. It is calculated by integrating the shear stress over the surface area. These shear stresses are given

*u*

*z*

the load per unit width), *F* = *f ho* / *μuw L <sup>x</sup>* for dimensionless friction force (where *f* is the friction force per unit width), and *m = F / W* for dimensionless friction coefficient. In the present study, the dimensionless slip length *A* is determined by normalizing the slip length *b* with the outlet film thickess *ho.* For slip analysis in the following computations, the dimensionless slip length *A* varies from 3 to 300, which are reasonable values of the slip length based on the results

The modified Reynolds equation, Eq. (1) is discretized over the flow using the finite volume method, and is solved using the tridiagonal matrix algorithm (TDMA), [27]. By employing the discretization scheme, the computed domain is divided into a number of control volumes using a grid with uniform mesh size. The grid independency is validated by various numbers of mesh sizes. An assumption is made that the boundary pressures are zero at both sides of

A numerical simulation is conducted to investigate the possible application so as a boundary slip can be beneficial to achieve a high load support and low friction force. In order to maximize the performance of lubrication, the boundary conditions (slip zones, *Sxt* and *Sxb*, see Figure 1) of the model are optimized through a parametric analysis. The object of optimization is to maximize the hydrodynamic load support. The load support satisfies two main functional purposes: (a) carry the applied external load, and (b) to minimize the contact between the opposing solids, and thus wear. The optimization analysis attempts to satisfy both functional

requirements with a single design parameter, the slip zone.

 m

2

æ ö ¶ <sup>=</sup> ç ÷

*z h*

=

*w p x dx* <sup>=</sup> ò (7)

è ø ¶ (8)

) for the dimensionless load support (where *w* is

<sup>2</sup> / *μL <sup>x</sup>uw* for the

0 ( ) *x L*

( , )

t

*x z*

The simulation results will be presented in dimensionless form, i.e. *P* = *pho*

<sup>2</sup> / (*uwμL <sup>x</sup>*

area, i.e.

236 Tribology - Fundamentals and Advancements

by

dimensionless pressure, *W* =*who*

published in literature [17, 18, 20].

**2.2. Solution method**

the contact.

#### **3. Key results**

The behavior of traditional (no-slip) hydrodynamic lubrication between the opposing surfaces can be estimated by a classical form of the Reynolds equation. The derivation of the classical Reynolds equation is based on the assumption of no-slip between the lubricant and the surfaces. In the classical Reynolds lubrication, the mechanism to generate a pressure is due to the convergent wedge effect. An artificial slip surface presented here is designed to be able to carry the external load during lubrication even if the wedge effect is not present. This situation is very beneficial in designing lubricated-MEMS which exhibits parallel gaps.

In Figure 3 the numerical results obtained with TDMA as well as analytical results for the dimensionless pressure distribution along the bottom wall of the contact are shown alongside those obtained from the Reynolds approximation. The wedge ratio *h\** of 2.2 was considered based on the fact that a maximum load support for a no-slip contact occurs when *h\** = 2.2 [28]. In the present study the wedge ratio *h*\* is defined as the inlet film thickness over the outlet film

the analytical solution based on the work of Cameron [28] as follows:

(8)

*x*

*L*

for the pressure distribution where *K* = (*hi /ho*) – 1, and

(2 )

from Figure 3 that the maximum error is within 0.01% between the pressure obtained from the

0.00 0.20 0.40 0.60 0.80 1.00

**Dimensionless** *x* **coordinate,** *X* 

**Figure 3.** Normalized pressure distribution along the bottom wall of the linear wedge with no-slip boundary condi‐

The comparison between the dimensionless friction force *F* obtained with the numerical prediction and the analytical solution are presented in Table 1. Like the result of the pressure distribution, the predicted dimensionless friction force shows very good agreement with the analytical solution. In general, the numerical solution of the classical Reynolds equation is matched well with the analytical solution. It assures that the numerical method used is valid

**Table 1.** The comparison between the analytical solution performed by Cameron [28] and the numerical simulation

In MEMS, liquid lubrication has generally been omitted due to high hydrodynamic friction force that occurs in fluid film. Compared with a solid coating, stiction prevention using liquid lubrication is less practical. However, recent studies have demonstrated that it is possible for

and thus can be extended for analysing other hydrodynamic characteristics.

Analytical solution [28] 0.77 Numerical prediction [present study] 0.77

/*h0*, and sometimes quoted as slope incline ratio in other literature. It is observed

Numerical solution Analytical solution [24]

developed computer code containing a numerical scheme using finite volume method combined with tridiagonal matrix algorithm (TDMA), the classical Reynolds equation (no-slip condition) is solved numerically for calculating the pressure distribution, and finally the friction in a lubricated sliding contact as depicted in Figure 1 and Table 1, respectively. These results are compared with

In Figure 3 the numerical results obtained with TDMA as well as analytical result for the dimensionless pressure distribution along the bottom wall of the contact are shown alongside those obtained from the Reynolds approximation. The wedge ratio *h\** of 2.2 was considered based on the fact that a maximum load support for a no-slip contact occurs when *h\** = 2.2 [28]. In the present study the wedge ratio is defined as inlet film thickness over the outlet film thickness, *hi*/*h0*, sometimes quoted as slope incline ratio). It is observed from Figure 3 that the maximum error is within 0.01% between the pressure obtained from the analytical solution and the

Artificial Slip Surface: Potential Application in Lubricated MEMS

http://dx.doi.org/10.5772/55745

239

Figure 3. Normalized pressure distribution along the bottom wall of the linear wedge with no-slip boundary condition for an optimal wedge ratio

The comparison between the dimensionless friction force *F* obtained with the numerical prediction and the analytical solution are presented in Table 1. Like the result of the pressure distribution, the predicted dimensionless friction force shows very good agreement with the analytical solution. In general, the numerical solution of the classical Reynolds equation is matched well with the analytical solution. It assures that the numerical method used is valid and thus can be extended for analysing other

In MEMS, liquid lubrication has generally been omitted due to high hydrodynamic friction force that occurs in fluid film. Compared with a solid coating, stiction prevention using liquid lubrication is less practical. However, recent studies have demonstrated that it is possible for Newtonian liquids to slip along very smooth solid walls [20] and this result may make liquid

**Dimensionless friction force, F**

Analytical solution [28] 0.77 Numerical prediction [present study] 0.77 Table 1. The comparison between the analytical solution performed by Cameron [28] and the numerical simulation code.

**Dimensionless friction,** *F*

thickness, *hi*

analytical solution and the numerical result.

numerical result.

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35

hydrodynamic characteristics.

**3.2. Artificial slip surface** 

**Dimensionless pressure,** 

*h\** = 2.2.

tion for an optimal wedge ratio *h\** of 2.2.

code.

**3.2. Artificial slip surface**

*P*

 2 2 1

*L u <sup>K</sup> <sup>f</sup> hK K*

for the friction force per unit width.

*u L L L <sup>p</sup> <sup>h</sup> <sup>x</sup> K KK*

2 1 *w x x x*

4ln(1 ) 6

*x x <sup>K</sup>*

 

(9)

6

*o*

*x w o*

In this chapter, there are two main investigations. At first, the study is conducted in order to validate the developed numerical scheme. It assures that the numerical method used can be employed for solving other hydrodynamic characteristics. The no-slip case of lubricated contact is of main interest due to the availability of the analytical solution. Secondly, the study will be extended to explore the effect of the slip zone of the artificial slip surface on pressure, load support, friction force, and friction coefficient. The comparison between the modified sliding contact containing an artificial slip surface and the traditional one is conducted in order to describe the benefit of the use of an artificial slip pattern quantitatively.

#### **3.1. No-slip condition**

The modified Reynolds equation (Eq. (1)) is the governing equation for the fluid lubrication system containing a boundary slip. If the slip coefficient, *α*, is set to zero, Eq. (1) reduces to the classical Reynolds equation. In this section, in order to validate the developed computer code containing a numerical scheme using finite volume method combined with tridiagonal matrix algorithm (TDMA), the classical Reynolds equation (no-slip condition) is solved numerically for calculating the pressure distribution, and finally the friction in a lubricated sliding contact as depicted in Figure 1 and Table 1, respectively. These results are compared with the analytical solution based on the work of Cameron [28] as follows:

$$p = \frac{6u\_w \mu L\_x}{h\_o^2} \frac{K \frac{\chi}{L\_x} \left(1 - \frac{\chi}{L\_x}\right)}{\left(2 + K\right) \left(1 + K - K \frac{\chi}{L\_x}\right)^2} \tag{9}$$

for the pressure distribution where *K* = (*hi /ho*) – 1, and

$$f = \frac{L\_x \mu u\_w}{h\_o} \left( \frac{4 \ln(1+K)}{K} - \frac{6}{(2+K)} \right) \tag{10}$$

for the friction force per unit width.

developed computer code containing a numerical scheme using finite volume method combined with tridiagonal matrix algorithm (TDMA), the classical Reynolds equation (no-slip condition) is solved numerically for calculating the pressure distribution, and finally the friction in a lubricated sliding contact as depicted in Figure 1 and Table 1, respectively. These results are compared with

In Figure 3 the numerical results obtained with TDMA as well as analytical results for the dimensionless pressure distribution along the bottom wall of the contact are shown alongside those obtained from the Reynolds approximation. The wedge ratio *h\** of 2.2 was considered based on the fact that a maximum load support for a no-slip contact occurs when *h\** = 2.2 [28]. In the present study the wedge ratio *h*\* is defined as the inlet film thickness over the outlet film thickness, *hi* /*h0*, and sometimes quoted as slope incline ratio in other literature. It is observed from Figure 3 that the maximum error is within 0.01% between the pressure obtained from the analytical solution and the numerical result. 4ln(1 ) 6 (2 ) *x w o L u <sup>K</sup> <sup>f</sup> hK K* (9) for the friction force per unit width. In Figure 3 the numerical results obtained with TDMA as well as analytical result for the dimensionless pressure distribution along the bottom wall of the contact are shown alongside those obtained from the Reynolds approximation. The wedge ratio *h\** of 2.2 was considered based on the fact that a maximum load support for a no-slip contact occurs when *h\** = 2.2 [28]. In the present study the wedge ratio is defined as inlet film thickness over the outlet film thickness, *hi*/*h0*, sometimes quoted as slope incline ratio). It is observed from Figure 3 that the maximum error is within 0.01% between the pressure obtained from the analytical solution and the

the analytical solution based on the work of Cameron [28] as follows:

(8)

*x*

*L*

 2 2 1

*u L L L <sup>p</sup> <sup>h</sup> <sup>x</sup> K KK*

2 1 *w x x x*

*x x <sup>K</sup>*

 

6

*o*

numerical result.

Figure 3. Normalized pressure distribution along the bottom wall of the linear wedge with no-slip boundary condition for an optimal wedge ratio *h\** = 2.2. **Figure 3.** Normalized pressure distribution along the bottom wall of the linear wedge with no-slip boundary condi‐ tion for an optimal wedge ratio *h\** of 2.2.

The comparison between the dimensionless friction force *F* obtained with the numerical prediction and the analytical solution are presented in Table 1. Like the result of the pressure distribution, the predicted dimensionless friction force shows very good agreement with the analytical solution. In general, the numerical solution of the classical Reynolds equation is matched well with the analytical solution. It assures that the numerical method used is valid and thus can be extended for analysing other hydrodynamic characteristics. **Dimensionless friction,** *F* Analytical solution [28] 0.77 The comparison between the dimensionless friction force *F* obtained with the numerical prediction and the analytical solution are presented in Table 1. Like the result of the pressure distribution, the predicted dimensionless friction force shows very good agreement with the analytical solution. In general, the numerical solution of the classical Reynolds equation is matched well with the analytical solution. It assures that the numerical method used is valid and thus can be extended for analysing other hydrodynamic characteristics.


Numerical prediction [present study] 0.77

In MEMS, liquid lubrication has generally been omitted due to high hydrodynamic friction force that occurs in fluid film.

Compared with a solid coating, stiction prevention using liquid lubrication is less practical. However, recent studies have demonstrated that it is possible for Newtonian liquids to slip along very smooth solid walls [20] and this result may make liquid **Table 1.** The comparison between the analytical solution performed by Cameron [28] and the numerical simulation code.

#### **3.2. Artificial slip surface**

**3. Key results**

238 Tribology - Fundamentals and Advancements

**3.1. No-slip condition**

The behavior of traditional (no-slip) hydrodynamic lubrication between the opposing surfaces can be estimated by a classical form of the Reynolds equation. The derivation of the classical Reynolds equation is based on the assumption of no-slip between the lubricant and the surfaces. In the classical Reynolds lubrication, the mechanism to generate a pressure is due to the convergent wedge effect. An artificial slip surface presented here is designed to be able to carry the external load during lubrication even if the wedge effect is not present. This situation

In this chapter, there are two main investigations. At first, the study is conducted in order to validate the developed numerical scheme. It assures that the numerical method used can be employed for solving other hydrodynamic characteristics. The no-slip case of lubricated contact is of main interest due to the availability of the analytical solution. Secondly, the study will be extended to explore the effect of the slip zone of the artificial slip surface on pressure, load support, friction force, and friction coefficient. The comparison between the modified sliding contact containing an artificial slip surface and the traditional one is conducted in order

The modified Reynolds equation (Eq. (1)) is the governing equation for the fluid lubrication system containing a boundary slip. If the slip coefficient, *α*, is set to zero, Eq. (1) reduces to the classical Reynolds equation. In this section, in order to validate the developed computer code containing a numerical scheme using finite volume method combined with tridiagonal matrix algorithm (TDMA), the classical Reynolds equation (no-slip condition) is solved numerically for calculating the pressure distribution, and finally the friction in a lubricated sliding contact as depicted in Figure 1 and Table 1, respectively. These results are compared with the analytical

( )

*L u <sup>K</sup> <sup>f</sup> hK K*

æ ö <sup>+</sup> <sup>=</sup> ç ÷ -

2 1

*w x x x*

*u L L L <sup>p</sup> <sup>h</sup> <sup>x</sup> K KK*

è ø <sup>=</sup>

2 2

4ln(1 ) 6

1

æ ö ç ÷ -

æ ö + +- ç ÷ è ø

*x x <sup>K</sup>*

*x*

(9)

*L*

(2 )

è ø <sup>+</sup> (10)

is very beneficial in designing lubricated-MEMS which exhibits parallel gaps.

to describe the benefit of the use of an artificial slip pattern quantitatively.

solution based on the work of Cameron [28] as follows:

for the pressure distribution where *K* = (*hi /ho*) – 1, and

for the friction force per unit width.

6

*o*

*x w o*

m

m

In MEMS, liquid lubrication has generally been omitted due to high hydrodynamic friction force that occurs in fluid film. Compared with a solid coating, stiction prevention using liquid lubrication is less practical. However, recent studies have demonstrated that it is possible for Newtonian liquids to slip along very smooth solid walls [20] and this result may make liquid lubricants for MEMS devices feasible. The main advantage of a liquid lubricant over a solid lubricant is that they generally produce no-contact shear stresses. Unfortunately, a stictiontype failure due to a large shear and capillary forces occurs. In [15, 16], a lubrication model of low load contacts was proposed to reduce such stiction or friction. The idea behind that work was how to use a lubricant that does not wet one of the solid surfaces. It was found that a halfwetted bearing generates a significant friction reduction compared to a traditional bearing.

surface, or both of them. Besides that, the types of slip zone pattern become also great issue. Therefore,aseriesofsimulationswereconductedwithsuchboundariestofindthebestpossibility of slip boundary application in terms of load support. Investigations were made for four kinds of slip boundaries to find the best boundary slip in terms of tribological performance, i.e. (1) slip applied on both the stationary and moving surfaces is referred as 'condition 1', (2) slip applied on the stationary surface is referred as 'condition 2', (3) slip applied on the moving surface is referred as 'condition 3', and (4) no-slip condition applied on the both of surfaces is referred as

Artificial Slip Surface: Potential Application in Lubricated MEMS

http://dx.doi.org/10.5772/55745

241

Figure 4 presents the effect of the wedge ratio *h*\* on the dimensionless load support *W*. It is shown that the contact with homogeneous slip condition of 1, 2, and 3 have a negative effect, i.e. a reduced load support. The highest achievement of a load support *W* is obtained when the slip is applied on the stationary surface (condition 2). However, the value of the predicted load support is much lower than the conventional lubricated contact for all values of wedge ratio. It is only half of what the conventional Reynolds theory predicts for an optimal wedge ratio of a traditional slider contact. Fortunately, the direct trend of homogeneous slip to decrease the load support *W* is counterbalanced by the fact that such surface also reduces friction significantly. This is indicated in Figure 5 which shows the comparison between the dimensionless friction *F* for condition 2 and the condition 4 for the range of wedge ratio, *h\**. The reduced friction as an advantageous effect in the lubrication can be explained by the fact that the boundary slip tends to reduce the wall shear rate at a prescribed film thickness, and then the wall shear stress and friction. Therefore, in the following design for the maximal lubrication performance, the lubricant has a no-slip boundary condition at the moving solid surface but can slip along the stationary surface. In the next section, the geometry of the boundary slip zone making an artificial slip surface will be investigated in order to achieve a

> Figure 4. Dimensionless load support versus wedge ratio for several homogeneous boundary slip conditions. (Note: slip on stationary and moving surfaces (condition 1); slip on stationary surface (condition 2); slip on moving surface (condition 3); traditional no-slip (condition 4)). For the slip

> > condition 2 condition 4

Figure 5. Dimensionless friction force versus wedge ratio at boundary condition in which slip applied on the stationary surface (condition 2)

This section is intended to investigate the optimum slip zone of the artificial slip surface for several values of wedge ratios *h*\* with

In the following computations, as discussed in the previous section, for a high load support, it is considered that the artificial slip surface will take place on the stationary surface, whereas no-slip condition occurs on the moving surface. The parameter *SxD* (in which *SxD* = *S*x/*L*x) is introduced in order to completely define the dimensionless slip zone. A parametric analysis is conducted by varying the dimensionless slip zone *SxD* from zero (i.e. no-slip boundary) to one (i.e. homogeneous boundary slip). The effect of slip zone geometry on the load support is presented in Figure 6. It can be shown that the load support has a maximum value when *SxD* = 0.65 and *h*\* = 1 (i.e. parallel sliding surfaces). It should be noted that no hydrodynamic pressure can be built up in parallel sliding surfaces for traditional no-slip contact. From Figure 6, it is also shown that with the increase in wedge ratio *h*\*, a clear shift of the maximum load support toward to outlet can be observed in this work. In addition, the maximum load support decreases with the increase in *h\**. A first conclusion is that for the best artificial slip surface with respect to the load support performance, the

'condition 4'. Here, a homogeneous slip surface is employed for all slip conditions.

higher load support as well as a low friction force.

0.00 0.03 0.06 0.09 0.12 0.15 0.18

**Dimensionless load support,** 

*W*

cases, the dimensionless slip length *A* of 20 is assumed.

compared to the no-slip condition (condition 4).

respect to the load support.

0.0

0.2

0.4

**Dimensionless friction**

 **force,** 

*F*

0.6

0.8

1.0

1.2

**3.2.2. The optimum slip zone of the artificial slip surface** 

1 1.3 1.6 1.9 2.2 2.5 2.8

**Wedge ratio,** *h\**

1 1.3 1.6 1.9 2.2 2.5 2.8

**Figure 4.** Dimensionless load support versus wedge ratio for several homogeneous boundary slip conditions. (Note: slip on stationary and moving surfaces (condition 1); slip on stationary surface (condition 2); slip on moving surface (condition 3); traditional no-slip (condition 4)). For the slip cases, the dimensionless slip length *A* of 20 is assumed.

**Wedge ratio,** *h\**

condition 1 condition 2 condition 3 condition 4

slip zone must be employed on the leading edge of the parallel sliding contact.

In order to reduce stiction, two principal methods are available, chemical and physical modification of the surfaces. To generate wall slip, in the chemical approach, the chemical composition of the surface is altered. In the physical approach, the surfaces are roughened to decrease the effective contact area [29].

In practice, the slip zone of the artificial slip surface can be prepared from (super)hydrophobic surface which uses chemical properties as well as micro- and nano-structures in order to achieve a high level of friction force reduction. The main characteristic of (super)hydrophobic surfaces is the slip length. Extensive studies have confirmed that the chemical treatment of the surfaces generates a slip length in the order of 1 µm [30], while longer slip length up to 100 µm can be obtained through a combination of a hydrophobic surface with textured structure [20, 31, 32]. In the present study, it will be shown by the computational analysis that a longer slip length applied on the slip zone of the artificial slip surface leads to a greater friction force reduction in combination with an improved load support.

#### *3.2.1. Beneficial surface of slip*

Recently, the use of an engineered slip surface has become popular with respect to lubrication, since this type of surface enhancement would give a better tribological performance. The great challenge for an engineered slip surface from the perspective of a numerical simulation is choosing the optimal slip zone geometry with respect to the lubrication performance. Two engineered slip surface modes were used currently: homogeneous slip surface (i.e. slip applied over the whole surface) and artificial slip surface (i.e. surface consisting of slip zone and noslip zone). It can be noted that term "artificial slip surface" was sometimes also called as heterogeneous slip/no-slipsurface [17,18]andmixedslipsurface [19].The first studytomention using a homogeneous slip was dedicated by Spikes [15, 16] who numerically studied the effect of slip profiles on friction. The author pointed out that by introducing the half-wetted bearing having a homogeneous slip boundary on one of the surfaces, a reduced friction can be ob‐ tained. Subsequently, an experimental study was published in [20] confirming the finding of [15, 16]. However, in addition to the friction reduction, it was shown that a homogeneous slip surfaceusuallyhasanegative effect,i.e.thedecrease inthe loadsupport.Ifthe lubricatedcontact exhibits a perfect slip property, it was found that the fluid load support was only half of that without slip [115, 19, 21, 23]. Clearly, this is unwanted effect with respect to the lubrication. Therefore, to date, an artificial slip surface has become of great interest by some researchers [17-19, 21, 23] with the focus of how to balance the slip effect on the load support and friction.

The big question with respect to the tribological performance of lubricated-MEMS emerges in accordance with at which wall boundary slip must be applied, at the stationary surface, moving

surface, or both of them. Besides that, the types of slip zone pattern become also great issue. Therefore,aseriesofsimulationswereconductedwithsuchboundariestofindthebestpossibility of slip boundary application in terms of load support. Investigations were made for four kinds of slip boundaries to find the best boundary slip in terms of tribological performance, i.e. (1) slip applied on both the stationary and moving surfaces is referred as 'condition 1', (2) slip applied on the stationary surface is referred as 'condition 2', (3) slip applied on the moving surface is referred as 'condition 3', and (4) no-slip condition applied on the both of surfaces is referred as 'condition 4'. Here, a homogeneous slip surface is employed for all slip conditions.

Newtonian liquids to slip along very smooth solid walls [20] and this result may make liquid lubricants for MEMS devices feasible. The main advantage of a liquid lubricant over a solid lubricant is that they generally produce no-contact shear stresses. Unfortunately, a stictiontype failure due to a large shear and capillary forces occurs. In [15, 16], a lubrication model of low load contacts was proposed to reduce such stiction or friction. The idea behind that work was how to use a lubricant that does not wet one of the solid surfaces. It was found that a halfwetted bearing generates a significant friction reduction compared to a traditional bearing. In order to reduce stiction, two principal methods are available, chemical and physical modification of the surfaces. To generate wall slip, in the chemical approach, the chemical composition of the surface is altered. In the physical approach, the surfaces are roughened to

In practice, the slip zone of the artificial slip surface can be prepared from (super)hydrophobic surface which uses chemical properties as well as micro- and nano-structures in order to achieve a high level of friction force reduction. The main characteristic of (super)hydrophobic surfaces is the slip length. Extensive studies have confirmed that the chemical treatment of the surfaces generates a slip length in the order of 1 µm [30], while longer slip length up to 100 µm can be obtained through a combination of a hydrophobic surface with textured structure [20, 31, 32]. In the present study, it will be shown by the computational analysis that a longer slip length applied on the slip zone of the artificial slip surface leads to a greater friction force

Recently, the use of an engineered slip surface has become popular with respect to lubrication, since this type of surface enhancement would give a better tribological performance. The great challenge for an engineered slip surface from the perspective of a numerical simulation is choosing the optimal slip zone geometry with respect to the lubrication performance. Two engineered slip surface modes were used currently: homogeneous slip surface (i.e. slip applied over the whole surface) and artificial slip surface (i.e. surface consisting of slip zone and noslip zone). It can be noted that term "artificial slip surface" was sometimes also called as heterogeneous slip/no-slipsurface [17,18]andmixedslipsurface [19].The first studytomention using a homogeneous slip was dedicated by Spikes [15, 16] who numerically studied the effect of slip profiles on friction. The author pointed out that by introducing the half-wetted bearing having a homogeneous slip boundary on one of the surfaces, a reduced friction can be ob‐ tained. Subsequently, an experimental study was published in [20] confirming the finding of [15, 16]. However, in addition to the friction reduction, it was shown that a homogeneous slip surfaceusuallyhasanegative effect,i.e.thedecrease inthe loadsupport.Ifthe lubricatedcontact exhibits a perfect slip property, it was found that the fluid load support was only half of that without slip [115, 19, 21, 23]. Clearly, this is unwanted effect with respect to the lubrication. Therefore, to date, an artificial slip surface has become of great interest by some researchers [17-19, 21, 23] with the focus of how to balance the slip effect on the load support and friction. The big question with respect to the tribological performance of lubricated-MEMS emerges in accordance with at which wall boundary slip must be applied, at the stationary surface, moving

decrease the effective contact area [29].

240 Tribology - Fundamentals and Advancements

*3.2.1. Beneficial surface of slip*

reduction in combination with an improved load support.

Figure 4 presents the effect of the wedge ratio *h*\* on the dimensionless load support *W*. It is shown that the contact with homogeneous slip condition of 1, 2, and 3 have a negative effect, i.e. a reduced load support. The highest achievement of a load support *W* is obtained when the slip is applied on the stationary surface (condition 2). However, the value of the predicted load support is much lower than the conventional lubricated contact for all values of wedge ratio. It is only half of what the conventional Reynolds theory predicts for an optimal wedge ratio of a traditional slider contact. Fortunately, the direct trend of homogeneous slip to decrease the load support *W* is counterbalanced by the fact that such surface also reduces friction significantly. This is indicated in Figure 5 which shows the comparison between the dimensionless friction *F* for condition 2 and the condition 4 for the range of wedge ratio, *h\**. The reduced friction as an advantageous effect in the lubrication can be explained by the fact that the boundary slip tends to reduce the wall shear rate at a prescribed film thickness, and then the wall shear stress and friction. Therefore, in the following design for the maximal lubrication performance, the lubricant has a no-slip boundary condition at the moving solid surface but can slip along the stationary surface. In the next section, the geometry of the boundary slip zone making an artificial slip surface will be investigated in order to achieve a higher load support as well as a low friction force.

Figure 4. Dimensionless load support versus wedge ratio for several homogeneous boundary slip conditions. (Note: slip on stationary and moving surfaces (condition 1); slip on stationary surface (condition 2); slip on moving surface (condition 3); traditional no-slip (condition 4)). For the slip cases, the dimensionless slip length *A* of 20 is assumed. **Figure 4.** Dimensionless load support versus wedge ratio for several homogeneous boundary slip conditions. (Note: slip on stationary and moving surfaces (condition 1); slip on stationary surface (condition 2); slip on moving surface (condition 3); traditional no-slip (condition 4)). For the slip cases, the dimensionless slip length *A* of 20 is assumed.

**3.2.2. The optimum slip zone of the artificial slip surface** 

1 1.3 1.6 1.9 2.2 2.5 2.8

**Wedge ratio,** *h\**

slip zone must be employed on the leading edge of the parallel sliding contact.

compared to the no-slip condition (condition 4).

respect to the load support.

0.0

0.2

0.4

**Dimensionless friction**

 **force,** 

*F*

0.6

0.8

1.0

1.2

Figure 5. Dimensionless friction force versus wedge ratio at boundary condition in which slip applied on the stationary surface (condition 2)

condition 2 condition 4

This section is intended to investigate the optimum slip zone of the artificial slip surface for several values of wedge ratios *h*\* with

In the following computations, as discussed in the previous section, for a high load support, it is considered that the artificial slip surface will take place on the stationary surface, whereas no-slip condition occurs on the moving surface. The parameter *SxD* (in which *SxD* = *S*x/*L*x) is introduced in order to completely define the dimensionless slip zone. A parametric analysis is conducted by varying the dimensionless slip zone *SxD* from zero (i.e. no-slip boundary) to one (i.e. homogeneous boundary slip). The effect of slip zone geometry on the load support is presented in Figure 6. It can be shown that the load support has a maximum value when *SxD* = 0.65 and *h*\* = 1 (i.e. parallel sliding surfaces). It should be noted that no hydrodynamic pressure can be built up in parallel sliding surfaces for traditional no-slip contact. From Figure 6, it is also shown that with the increase in wedge ratio *h*\*, a clear shift of the maximum load support toward to outlet can be observed in this work. In addition, the maximum load support decreases with the increase in *h\**. A first conclusion is that for the best artificial slip surface with respect to the load support performance, the **Dimensionless load support,** 

*W*

0.00 0.03 0.06 0.09 0.12 0.15 0.18

1 1.3 1.6 1.9 2.2 2.5 2.8

**Wedge ratio,** *h\**

condition 1 condition 2 condition 3 condition 4

Figure 4. Dimensionless load support versus wedge ratio for several homogeneous boundary slip conditions. (Note: slip on stationary and moving surfaces (condition 1); slip on stationary surface (condition 2); slip on moving surface (condition 3); traditional no-slip (condition 4)). For the slip

This section is intended to investigate the optimum slip zone of the artificial slip surface for several values of wedge ratios *h*\* with

convergent wedge as predicted by the classical Reynolds assumption (*h*\*opt = 2.2 [28]), but at a parallel surface. The predicted maximum pressure for a parallel gap is over three times as large as the maximum pressure obtained from a no-slip contact when the wedge ratio *h*\* = 2.2. From this perspective, the well-chosen artificial slip surface pattern can be considered as a potential application which is possible for lubricated-MEMS based devices with respect to the load support. In this way, liquid lubricants on the modified opposing surfaces (i.e. artificial slip surface) can prevent the lubrication to break down during device operation to the point where

**Figure 6.** Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless load support *W* for

0.0 0.2 0.4 0.6 0.8 1.0

**Dimensionless slip zone,** *SxD*

Figure 6. Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless load support *W* for several wedge ratios *h*\*.

is over three times as large as the maximum pressure obtained from a no-slip contact when the wedge ratio *h*\*

Figure 7 shows the normalized representation of lubrication film pressure distributions as a function of wedge ratio which are predicted by the modified Reynolds equation (Eq. (1)). For slip configuration, the optimal slip zone *SxD* of 0.65, is employed. It can be shown that compared with the no-slip boundary condition, the artificial slip surface yields a positive fluid pressure. However, the performance improvement obtained through artificial slip surface is rather mild for the high wedge ratio. Obviously, the lower the base geometry wedge ratio (thus leads to the parallel sliding contact), the larger improvement that boundary slip can induce. In other words, using the artificial slip surface considered here, the maximum pressure occurs not at a convergent wedge as predicted by the classical Reynolds assumption (*h*\*opt = 2.2 [28]), but at a parallel surface. The predicted maximum pressure for a parallel gap

Artificial Slip Surface: Potential Application in Lubricated MEMS

http://dx.doi.org/10.5772/55745

243

perspective, the well-chosen artificial slip surface pattern can be considered as a potential application which is possible for lubricated-MEMS based devices with respect to the load support. In this way, liquid lubricants on the modified opposing surfaces (i.e. artificial slip surface) can prevent the lubrication to break down during device operation to the point where they no longer give

Figure 7. Lubrication film pressure distributions for several values of wedge ratio. The slip profiles are calculated for optimized dimensionless slip

Figure 8 shows the dimensionless surface friction force *F* at the bottom surface. It can be seen that the friction of the artificial slip surface becomes smaller than that of a traditional no-slip lubricated contact especially when *SxD* is larger than about 0.6. This agrees with the numerical analysis of Wu *et al*. [19] eventhough the slip model and numerical method used are different. One can remark

means that the optimized artificial slip surface of lubricated-MEMS can produce a lower friction force than a no-slip contact.

Figure 8 shows the dimensionless surface friction force *F* at the bottom surface. It can be seen that the friction of the artificial slip surface becomes smaller than that of a traditional no-slip lubricated contact especially when *SxD* is larger than about 0.6. This agrees with the numerical analysis of Wu *et al*. [19] eventhough the slip model and numerical method used are different.

surface gives the minimum dimensionless friction force of 0.65, while the no-slip lubricated

that the optimized artificial slip surface of lubricated-MEMS can produce a lower friction force

The hydrophobicity of a solid surface, as discussed in the previous section, is usually expressed in terms of a slip length, which quantifies the extent to which the fluid elements near the surface are affected by the surface energy and the surface geometry. The surface energy is an intrinsic property of a material that can be controlled by chemical treatment, such as etching approach and/or coat-on/cast approach. The surface roughness of a hydrophobic solid material can be

0 0.2 0.4 0.6 0.8 1

**Dimensionless** *x* **coordinate,** *X* 

force of 0.65, while the no-slip lubricated contact gives its dimensionless minimum friction of 0.77 at *h*\*

= 1) and *SxD* = 0.65, the artificial slip

optimal = 2.2, see Table 1. It means

= 1) and *SxD* = 0.65, the artificial slip surface gives the minimum dimensionless friction

= 2.2. From this

optimal = 2.2, see Table 1. It

they no longer give proper lubrication.

0

0.1 0.2 0.3 0.4 0.5 0.6 0.7

proper lubrication.

0.0

several wedge ratios *h*\*.

0.1

0.2

**Dimensionless load support,** 

*W*

0.3

0.4

h\* = 1 h\* = 1.4 h\* = 2.2 h\* = 3

than a no-slip contact.

**Dimensionless presure,**

*P*

One can remark that when there is no wedge effect (i.e. *h*\*

h\* = 1 h\* = 1.2 h\* = 2.2 h\* = 2.2 (no-slip) h\* = 3.2

contact gives its dimensionless minimum friction of 0.77 at *h*\*

*3.2.3. Effect of dimensionless slip length on lubrication performance*

zone *SxD* = 0.65 and dimensionless slip length *A* = 20.

that when there is no wedge effect (i.e. *h*\*

Figure 5. Dimensionless friction force versus wedge ratio at boundary condition in which slip applied on the stationary surface (condition 2) compared to the no-slip condition (condition 4). **Figure 5.** Dimensionless friction force versus wedge ratio at boundary condition in which slip applied on the station‐ ary surface (condition 2) compared to the no-slip condition (condition 4).

#### **3.2.2. The optimum slip zone of the artificial slip surface**  *3.2.2. The optimum slip zone of the artificial slip surface*

cases, the dimensionless slip length *A* of 20 is assumed.

respect to the load support. In the following computations, as discussed in the previous section, for a high load support, it is considered that the artificial slip This section is intended to investigate the optimum slip zone of the artificial slip surface for several values of wedge ratios *h*\* with respect to the load support.

surface will take place on the stationary surface, whereas no-slip condition occurs on the moving surface. The parameter *SxD* (in which *SxD* = *S*x/*L*x) is introduced in order to completely define the dimensionless slip zone. A parametric analysis is conducted by varying the dimensionless slip zone *SxD* from zero (i.e. no-slip boundary) to one (i.e. homogeneous boundary slip). The effect of slip zone geometry on the load support is presented in Figure 6. It can be shown that the load support has a maximum value when *SxD* = 0.65 and *h*\* = 1 (i.e. parallel sliding surfaces). It should be noted that no hydrodynamic pressure can be built up in parallel sliding surfaces for traditional no-slip contact. From Figure 6, it is also shown that with the increase in wedge ratio *h*\*, a clear shift of the maximum load support toward to outlet can be observed in this work. In addition, the maximum load support decreases with the increase in *h\**. A first conclusion is that for the best artificial slip surface with respect to the load support performance, the slip zone must be employed on the leading edge of the parallel sliding contact. In the following computations, as discussed in the previous section, for a high load support, it is considered that the artificial slip surface will take place on the stationary surface, whereas no-slip condition occurs on the moving surface. The parameter *SxD* (in which *SxD* = *S*x/*L*x) is introduced in order to completely define the dimensionless slip zone. A parametric analysis is conducted by varying the dimensionless slip zone *SxD* from zero (i.e. no-slip boundary) to one (i.e. homogeneous boundary slip). The effect of slip zone geometry on the load support is presented in Figure 6. It can be shown that the load support has a maximum value when *SxD* = 0.65 and *h*\* = 1 (i.e. parallel sliding surfaces). It should be noted that no hydrodynamic pressure can be built up in parallel sliding surfaces for traditional no-slip contact. From Figure 6, it is also shown that with the increase in wedge ratio *h*\*, a clear shift of the maximum load support toward to outlet can be observed in this work. In addition, the maximum load support decreases with the increase in *h\**. A first conclusion is that for the best artificial slip surface with respect to the load support performance, the slip zone must be employed on the leading edge of the parallel sliding contact.

Figure 7 shows the normalized representation of lubrication film pressure distributions as a function of wedge ratio which are predicted by the modified Reynolds equation (Eq. (1)). For slip configuration, the optimal slip zone *SxD* of 0.65, is employed. It can be shown that compared with the no-slip boundary condition, the artificial slip surface yields a positive fluid pressure. However, the performance improvement obtained through artificial slip surface is rather mild for the high wedge ratio. Obviously, the lower the base geometry wedge ratio (thus leads to the parallel sliding contact), the larger improvement that boundary slip can induce. In other words, using the artificial slip surface considered here, the maximum pressure occurs not at a

Figure 4. Dimensionless load support versus wedge ratio for several homogeneous boundary slip conditions. (Note: slip on stationary and moving surfaces (condition 1); slip on stationary surface (condition 2); slip on moving surface (condition 3); traditional no-slip (condition 4)). For the slip

> condition 2 condition 4

Figure 5. Dimensionless friction force versus wedge ratio at boundary condition in which slip applied on the stationary surface (condition 2)

This section is intended to investigate the optimum slip zone of the artificial slip surface for several values of wedge ratios *h*\* with

In the following computations, as discussed in the previous section, for a high load support, it is considered that the artificial slip surface will take place on the stationary surface, whereas no-slip condition occurs on the moving surface. The parameter *SxD* (in which *SxD* = *S*x/*L*x) is introduced in order to completely define the dimensionless slip zone. A parametric analysis is conducted by varying the dimensionless slip zone *SxD* from zero (i.e. no-slip boundary) to one (i.e. homogeneous boundary slip). The effect of slip zone geometry on the load support is presented in Figure 6. It can be shown that the load support has a maximum value when *SxD* = 0.65 and *h*\* = 1 (i.e. parallel sliding surfaces). It should be noted that no hydrodynamic pressure can be built up in parallel sliding surfaces for traditional no-slip contact. From Figure 6, it is also shown that with the increase in wedge ratio *h*\*, a clear shift of the maximum load support toward to outlet can be observed in this work. In addition, the maximum load support decreases with the increase in *h\**. A first conclusion is that for the best artificial slip surface with respect to the load support performance, the

cases, the dimensionless slip length *A* of 20 is assumed.

0.00 0.03 0.06 0.09 0.12 0.15 0.18

**Dimensionless load support,** 

*W*

compared to the no-slip condition (condition 4).

several values of wedge ratios *h*\* with respect to the load support.

ary surface (condition 2) compared to the no-slip condition (condition 4).

respect to the load support.

edge of the parallel sliding contact.

*3.2.2. The optimum slip zone of the artificial slip surface*

0.0

0.2

0.4

**Dimensionless friction**

 **force,** 

*F*

242 Tribology - Fundamentals and Advancements

0.6

0.8

1.0

1.2

**3.2.2. The optimum slip zone of the artificial slip surface** 

1 1.3 1.6 1.9 2.2 2.5 2.8

**Figure 5.** Dimensionless friction force versus wedge ratio at boundary condition in which slip applied on the station‐

This section is intended to investigate the optimum slip zone of the artificial slip surface for

In the following computations, as discussed in the previous section, for a high load support, it is considered that the artificial slip surface will take place on the stationary surface, whereas no-slip condition occurs on the moving surface. The parameter *SxD* (in which *SxD* = *S*x/*L*x) is introduced in order to completely define the dimensionless slip zone. A parametric analysis is conducted by varying the dimensionless slip zone *SxD* from zero (i.e. no-slip boundary) to one (i.e. homogeneous boundary slip). The effect of slip zone geometry on the load support is presented in Figure 6. It can be shown that the load support has a maximum value when *SxD* = 0.65 and *h*\* = 1 (i.e. parallel sliding surfaces). It should be noted that no hydrodynamic pressure can be built up in parallel sliding surfaces for traditional no-slip contact. From Figure 6, it is also shown that with the increase in wedge ratio *h*\*, a clear shift of the maximum load support toward to outlet can be observed in this work. In addition, the maximum load support decreases with the increase in *h\**. A first conclusion is that for the best artificial slip surface with respect to the load support performance, the slip zone must be employed on the leading

**Wedge ratio,** *h\**

1 1.3 1.6 1.9 2.2 2.5 2.8

**Wedge ratio,** *h\**

condition 1 condition 2 condition 3 condition 4

slip zone must be employed on the leading edge of the parallel sliding contact.

Figure 7 shows the normalized representation of lubrication film pressure distributions as a function of wedge ratio which are predicted by the modified Reynolds equation (Eq. (1)). For slip configuration, the optimal slip zone *SxD* of 0.65, is employed. It can be shown that compared with the no-slip boundary condition, the artificial slip surface yields a positive fluid pressure. However, the performance improvement obtained through artificial slip surface is rather mild for the high wedge ratio. Obviously, the lower the base geometry wedge ratio (thus leads to the parallel sliding contact), the larger improvement that boundary slip can induce. In other words, using the artificial slip surface considered here, the maximum pressure occurs not at a

Figure 6. Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless load support *W* for several wedge ratios *h*\*. Figure 7 shows the normalized representation of lubrication film pressure distributions as a function of wedge ratio which are **Figure 6.** Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless load support *W* for several wedge ratios *h*\*.

predicted by the modified Reynolds equation (Eq. (1)). For slip configuration, the optimal slip zone *SxD* of 0.65, is employed. It can

Figure 8 shows the dimensionless surface friction force *F* at the bottom surface. It can be seen that the friction of the artificial slip surface becomes smaller than that of a traditional no-slip lubricated contact especially when *SxD* is larger than about 0.6. This agrees with the numerical analysis of Wu *et al*. [19] eventhough the slip model and numerical method used are different. One can remark

means that the optimized artificial slip surface of lubricated-MEMS can produce a lower friction force than a no-slip contact.

= 1) and *SxD* = 0.65, the artificial slip surface gives the minimum dimensionless friction

= 2.2. From this

optimal = 2.2, see Table 1. It

convergent wedge as predicted by the classical Reynolds assumption (*h*\*opt = 2.2 [28]), but at a parallel surface. The predicted maximum pressure for a parallel gap is over three times as large as the maximum pressure obtained from a no-slip contact when the wedge ratio *h*\* = 2.2. From this perspective, the well-chosen artificial slip surface pattern can be considered as a potential application which is possible for lubricated-MEMS based devices with respect to the load support. In this way, liquid lubricants on the modified opposing surfaces (i.e. artificial slip surface) can prevent the lubrication to break down during device operation to the point where they no longer give proper lubrication. be shown that compared with the no-slip boundary condition, the artificial slip surface yields a positive fluid pressure. However, the performance improvement obtained through artificial slip surface is rather mild for the high wedge ratio. Obviously, the lower the base geometry wedge ratio (thus leads to the parallel sliding contact), the larger improvement that boundary slip can induce. In other words, using the artificial slip surface considered here, the maximum pressure occurs not at a convergent wedge as predicted by the classical Reynolds assumption (*h*\*opt = 2.2 [28]), but at a parallel surface. The predicted maximum pressure for a parallel gap is over three times as large as the maximum pressure obtained from a no-slip contact when the wedge ratio *h*\* perspective, the well-chosen artificial slip surface pattern can be considered as a potential application which is possible for lubricated-MEMS based devices with respect to the load support. In this way, liquid lubricants on the modified opposing surfaces (i.e. artificial slip surface) can prevent the lubrication to break down during device operation to the point where they no longer give proper lubrication.

Figure 8 shows the dimensionless surface friction force *F* at the bottom surface. It can be seen that the friction of the artificial slip surface becomes smaller than that of a traditional no-slip lubricated contact especially when *SxD* is larger than about 0.6. This agrees with the numerical analysis of Wu *et al*. [19] eventhough the slip model and numerical method used are different. One can remark that when there is no wedge effect (i.e. *h*\* = 1) and *SxD* = 0.65, the artificial slip surface gives the minimum dimensionless friction force of 0.65, while the no-slip lubricated contact gives its dimensionless minimum friction of 0.77 at *h*\* optimal = 2.2, see Table 1. It means that the optimized artificial slip surface of lubricated-MEMS can produce a lower friction force than a no-slip contact. 0.2 0.3 0.4 0.5 0.6 0.7 **Dimensionless presure,** *P* h\* = 1 h\* = 1.2 h\* = 2.2 h\* = 2.2 (no-slip) h\* = 3.2

#### *3.2.3. Effect of dimensionless slip length on lubrication performance* 0.1

0

that when there is no wedge effect (i.e. *h*\*

The hydrophobicity of a solid surface, as discussed in the previous section, is usually expressed in terms of a slip length, which quantifies the extent to which the fluid elements near the surface are affected by the surface energy and the surface geometry. The surface energy is an intrinsic property of a material that can be controlled by chemical treatment, such as etching approach and/or coat-on/cast approach. The surface roughness of a hydrophobic solid material can be Figure 7. Lubrication film pressure distributions for several values of wedge ratio. The slip profiles are calculated for optimized dimensionless slip zone *SxD* = 0.65 and dimensionless slip length *A* = 20. 0 0.2 0.4 0.6 0.8 1 **Dimensionless** *x* **coordinate,** *X* 

force of 0.65, while the no-slip lubricated contact gives its dimensionless minimum friction of 0.77 at *h*\*

proper lubrication.

0.0

0.1

0.2

**Dimensionless load support,** 

*W*

0.3

0.4

0.0

0.1

0.2

**Dimensionless load support,** 

*W*

0.3

0.4

h\* = 1 h\* = 1.4 h\* = 2.2 h\* = 3

Figure 6. Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless load support *W* for several wedge ratios *h*\*.

is over three times as large as the maximum pressure obtained from a no-slip contact when the wedge ratio *h*\*

0.0 0.2 0.4 0.6 0.8 1.0

**Dimensionless slip zone,** *SxD*

Figure 7 shows the normalized representation of lubrication film pressure distributions as a function of wedge ratio which are predicted by the modified Reynolds equation (Eq. (1)). For slip configuration, the optimal slip zone *SxD* of 0.65, is employed. It can be shown that compared with the no-slip boundary condition, the artificial slip surface yields a positive fluid pressure. However, the performance improvement obtained through artificial slip surface is rather mild for the high wedge ratio. Obviously, the lower the base geometry wedge ratio (thus leads to the parallel sliding contact), the larger improvement that boundary slip can induce. In other words, using the artificial slip surface considered here, the maximum pressure occurs not at a convergent wedge as predicted by the classical Reynolds assumption (*h*\*opt = 2.2 [28]), but at a parallel surface. The predicted maximum pressure for a parallel gap

perspective, the well-chosen artificial slip surface pattern can be considered as a potential application which is possible for

(i.e. artificial slip surface) can prevent the lubrication to break down during device operation to the point where they no longer give

Figure 8 shows the dimensionless surface friction force *F* at the bottom surface. It can be seen that the friction of the artificial slip

surface energy is an intrinsic property of a material that can be controlled by chemical treatment, such as etching approach and/or

the larger the slip length at the optimized slip zone of the artificial slip zone, the higher the load support. However, when dimensionless slip length is larger than 30, the dimensionless load support is not affected significantly with the increase in the dimensionless slip length. So, the increase of the load support is not infinitely large. It can be deduced that there is no fluid load support for a lubricated-MEMS when it contains no-slip condition (*A* = 0). It indicates that the absence of the wedge effect on the pressure generation at parallel lubricated sliding contact has been counterbalanced by the influence of the ariticial slip surface application. Again, this condition is very advantageous in engineering a lubricated-MEMS which demonstrates parallel gaps.

= 2.2. From this

However, when dimensionless slip length is larger than 30, the dimensionless load support is not affected significantly with the increase in the dimensionless slip length. So, the increase of the load support is not infinitely large. It can be deduced that there is no fluid load support for a lubricated-MEMS when it contains no-slip condition (*A* = 0). It indicates that the absence of the wedge effect on the pressure generation at parallel lubricated sliding contact has been counterbalanced by the influence of the ariticial slip surface application. Again, this condition is very advantageous in engineering a lubricated-MEMS which demonstrates parallel gaps.

values of wedge ratio *h\**. The slip profiles are evaluated for dimensionless slip length *A* = 20.

0

0.0

0.1

0.2

**Dimensionless load support,** 

*W*

0.3

0.4

0.2

0.4

h\* = 1 h\* = 1.4 h\* = 2.2 h\* = 3

0.6

**Dimensionless friction**

 **force,** 

*F*

0.8

1

1.2

**3.2.3. Effect of dimensionless slip length on lubrication performance** 

0 0.2 0.4 0.6 0.8 1

**Dimensionless slip zone,** *SxD*

slip length on the lubrication behavior is studied. The dimensionless slip length is varied from 3 to 300.

0 0.2 0.4 0.6 0.8 1

**Figure 9.** Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless load support *W* for

In Figure 10 the effect of slip zone of the artificial slip surface for several dimensionless slip length values on the dimensionless friction force is presented. As is well-known, the ability to control and manipulate friction force during sliding is extremely important key to prolong a life-time of lubricated-MEMS. Better understanding of the friction force phenomena at microscales is needed to provide designers and engineers the required tools and capabilities to

As can be seen in Figure 10 that the artificial slip surface leads to a reduction of the friction force for all dimensionless slip length. The friction force decreases with increasing the slip zone *Sx*. It can be seen that when the slip zone covers over the whole surface, i.e. homogeneous slip surface, the friction forces has a minimum value, especially for a high slip length. If the reduction in friction force is of only particular interest, the homogeneous slip surface (*SxD* = 1) is very beneficial. But if the performance is also related to the load support, homogeneous slip is not recommended because when *SxD* = 1, the predicted load support is very small for all wedge ratios, see Fig. 6. With respect to the influence of the dimensionless slip length, opposite to the hydrodynamic load support, the dimensionless friction force becomes smaller for higher dimensionless slip length. Therefore, the optimized artificial slip surface is a very promising

several values of dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1).

control friction force and predict failure of lubrication in MEMS.

**Dimensionless slip zone,** *SxD*

A = 3 A = 30 A = 300 A = 0 (no-slip)

Figure 8. Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction force *F* at bottom surface for several

The hydrophobicity of a solid surface, as discussed in the previous section, is usually expressed in terms of a slip length, which quantifies the extent to which the fluid elements near the surface are affected by the surface energy and the surface geometry. The surface energy is an intrinsic property of a material that can be controlled by chemical treatment, such as etching approach and/or coat-on/cast approach. The surface roughness of a hydrophobic solid material can be tuned in order to increment its hydrophobicity and obtain a super-hydrophobic solid surface [33, 34]. In this section, from the numerical point of view, the effect of

Artificial Slip Surface: Potential Application in Lubricated MEMS

http://dx.doi.org/10.5772/55745

245

Figure 9 shows the effect of slip zone of the artificial slip surface for several dimensionless slip length values on the dimensionless load support. As indicated in Figure 9, the increase in the slip length leads to an increase in the predicted load support. Generally, the larger the slip length at the optimized slip zone of the artificial slip zone, the higher the load support. However, when dimensionless slip length is larger than 30, the dimensionless load support is not affected significantly with the increase in the dimensionless slip length. So, the increase of the load support is not infinitely large. It can be deduced that there is no fluid load support for a lubricated-MEMS when it contains no-slip condition (*A* = 0). It indicates that the absence of the wedge effect on the pressure generation at parallel lubricated sliding contact has been counterbalanced by the influence of the ariticial slip surface application. Again, this condition is very advantageous in engineering a lubricated-MEMS which demonstrates parallel gaps.

optimal = 2.2, see Table 1. It

Figure 7. Lubrication film pressure distributions for several values of wedge ratio. The slip profiles are calculated for optimized dimensionless slip zone *SxD* = 0.65 and dimensionless slip length *A* = 20. **Figure 7.** Lubrication film pressure distributions for several values of wedge ratio. The slip profiles are calculated for optimized dimensionless slip zone *SxD* = 0.65 and dimensionless slip length *A* = 20.

Figure 8. Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction force *F* at bottom surface for several values of wedge ratio *h\**. The slip profiles are evaluated for dimensionless slip length *A* = 20. **Figure 8.** Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction force *F* at bot‐ tom surface for several values of wedge ratio *h\**. The slip profiles are evaluated for dimensionless slip length *A* = 20.

tuned in order to increment its hydrophobicity and obtain a super-hydrophobic solid surface [33, 34]. In this section, from the numerical point of view, the effect of slip length on the lubrication behavior is studied. The dimensionless slip length is varied from 3 to 300. **3.2.3. Effect of dimensionless slip length on lubrication performance**  The hydrophobicity of a solid surface, as discussed in the previous section, is usually expressed in terms of a slip length, which quantifies the extent to which the fluid elements near the surface are affected by the surface energy and the surface geometry. The

Figure 9 shows the effect of slip zone of the artificial slip surface for several dimensionless slip length values on the dimensionless load support. As indicated in Figure 9, the increase in the slip length leads to an increase in the predicted load support. Generally, the larger the slip length at the optimized slip zone of the artificial slip zone, the higher the load support. coat-on/cast approach. The surface roughness of a hydrophobic solid material can be tuned in order to increment its hydrophobicity and obtain a super-hydrophobic solid surface [33, 34]. In this section, from the numerical point of view, the effect of slip length on the lubrication behavior is studied. The dimensionless slip length is varied from 3 to 300. Figure 9 shows the effect of slip zone of the artificial slip surface for several dimensionless slip length values on the dimensionless load support. As indicated in Figure 9, the increase in the slip length leads to an increase in the predicted load support. Generally,

0 0.2 0.4 0.6 0.8 1

**Dimensionless slip zone,** *SxD*

A = 3 A = 30 A = 300 A = 0 (no-slip) Figure 8. Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction force *F* at bottom surface for several

coat-on/cast approach. The surface roughness of a hydrophobic solid material can be tuned in order to increment its

application. Again, this condition is very advantageous in engineering a lubricated-MEMS which demonstrates parallel gaps.

However, when dimensionless slip length is larger than 30, the dimensionless load support is not affected significantly with the increase in the dimensionless slip length. So, the increase of the load support is not infinitely large. It can be deduced that there is no fluid load support for a lubricated-MEMS when it contains no-slip condition (*A* = 0). It indicates that the absence of the wedge effect on the pressure generation at parallel lubricated sliding contact has been counterbalanced by the influence of the ariticial slip surface application. Again, this condition is very advantageous in engineering a lubricated-MEMS which demonstrates parallel gaps. hydrophobicity and obtain a super-hydrophobic solid surface [33, 34]. In this section, from the numerical point of view, the effect of slip length on the lubrication behavior is studied. The dimensionless slip length is varied from 3 to 300. Figure 9 shows the effect of slip zone of the artificial slip surface for several dimensionless slip length values on the dimensionless load support. As indicated in Figure 9, the increase in the slip length leads to an increase in the predicted load support. Generally, the larger the slip length at the optimized slip zone of the artificial slip zone, the higher the load support. However, when dimensionless slip length is larger than 30, the dimensionless load support is not affected significantly with the increase in the dimensionless slip length. So, the increase of the load support is not infinitely large. It can be deduced that there is no fluid load support for a lubricated-MEMS when it contains no-slip condition (*A* = 0). It indicates that the absence of the wedge effect on the pressure generation at parallel lubricated sliding contact has been counterbalanced by the influence of the ariticial slip surface

values of wedge ratio *h\**. The slip profiles are evaluated for dimensionless slip length *A* = 20.

0

0.2

0.4

h\* = 1 h\* = 1.4 h\* = 2.2 h\* = 3

0.6

**Dimensionless friction**

 **force,** 

*F*

0.8

1

1.2

**3.2.3. Effect of dimensionless slip length on lubrication performance** 

0 0.2 0.4 0.6 0.8 1

**Dimensionless slip zone,** *SxD*

**Figure 9.** Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless load support *W* for several values of dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1).

In Figure 10 the effect of slip zone of the artificial slip surface for several dimensionless slip length values on the dimensionless friction force is presented. As is well-known, the ability to control and manipulate friction force during sliding is extremely important key to prolong a life-time of lubricated-MEMS. Better understanding of the friction force phenomena at microscales is needed to provide designers and engineers the required tools and capabilities to control friction force and predict failure of lubrication in MEMS.

As can be seen in Figure 10 that the artificial slip surface leads to a reduction of the friction force for all dimensionless slip length. The friction force decreases with increasing the slip zone *Sx*. It can be seen that when the slip zone covers over the whole surface, i.e. homogeneous slip surface, the friction forces has a minimum value, especially for a high slip length. If the reduction in friction force is of only particular interest, the homogeneous slip surface (*SxD* = 1) is very beneficial. But if the performance is also related to the load support, homogeneous slip is not recommended because when *SxD* = 1, the predicted load support is very small for all wedge ratios, see Fig. 6. With respect to the influence of the dimensionless slip length, opposite to the hydrodynamic load support, the dimensionless friction force becomes smaller for higher dimensionless slip length. Therefore, the optimized artificial slip surface is a very promising

tuned in order to increment its hydrophobicity and obtain a super-hydrophobic solid surface [33, 34]. In this section, from the numerical point of view, the effect of slip length on the

**Figure 8.** Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction force *F* at bot‐ tom surface for several values of wedge ratio *h\**. The slip profiles are evaluated for dimensionless slip length *A* = 20.

**3.2.3. Effect of dimensionless slip length on lubrication performance** 

0 0.2 0.4 0.6 0.8 1

**Dimensionless slip zone,** *SxD*

values of wedge ratio *h\**. The slip profiles are evaluated for dimensionless slip length *A* = 20.

Figure 6. Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless load support *W* for several wedge ratios *h*\*.

is over three times as large as the maximum pressure obtained from a no-slip contact when the wedge ratio *h*\*

0.0 0.2 0.4 0.6 0.8 1.0

**Dimensionless slip zone,** *SxD*

proper lubrication.

0

0

0.0

0.1

0.2

**Dimensionless load support,** 

*W*

0.3

0.4

0.2

0.4

0.6

**Dimensionless friction**

 **force,** 

*F*

0.8

1

1.2

0.1 0.2 0.3 0.4 0.5 0.6 0.7

**Dimensionless presure,**

*P*

244 Tribology - Fundamentals and Advancements

0.0

0.1

0.2

**Dimensionless load support,** 

*W*

0.3

0.4

h\* = 1 h\* = 1.4 h\* = 2.2 h\* = 3

zone *SxD* = 0.65 and dimensionless slip length *A* = 20.

optimized dimensionless slip zone *SxD* = 0.65 and dimensionless slip length *A* = 20.

h\* = 1 h\* = 1.4 h\* = 2.2 h\* = 3

h\* = 1 h\* = 1.2 h\* = 2.2 h\* = 2.2 (no-slip) h\* = 3.2

that when there is no wedge effect (i.e. *h*\*

Figure 7 shows the normalized representation of lubrication film pressure distributions as a function of wedge ratio which are predicted by the modified Reynolds equation (Eq. (1)). For slip configuration, the optimal slip zone *SxD* of 0.65, is employed. It can be shown that compared with the no-slip boundary condition, the artificial slip surface yields a positive fluid pressure. However, the performance improvement obtained through artificial slip surface is rather mild for the high wedge ratio. Obviously, the lower the base geometry wedge ratio (thus leads to the parallel sliding contact), the larger improvement that boundary slip can induce. In other words, using the artificial slip surface considered here, the maximum pressure occurs not at a convergent wedge as predicted by the classical Reynolds assumption (*h*\*opt = 2.2 [28]), but at a parallel surface. The predicted maximum pressure for a parallel gap

perspective, the well-chosen artificial slip surface pattern can be considered as a potential application which is possible for lubricated-MEMS based devices with respect to the load support. In this way, liquid lubricants on the modified opposing surfaces (i.e. artificial slip surface) can prevent the lubrication to break down during device operation to the point where they no longer give

Figure 7. Lubrication film pressure distributions for several values of wedge ratio. The slip profiles are calculated for optimized dimensionless slip

Figure 8 shows the dimensionless surface friction force *F* at the bottom surface. It can be seen that the friction of the artificial slip surface becomes smaller than that of a traditional no-slip lubricated contact especially when *SxD* is larger than about 0.6. This agrees with the numerical analysis of Wu *et al*. [19] eventhough the slip model and numerical method used are different. One can remark

Figure 8. Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction force *F* at bottom surface for several

The hydrophobicity of a solid surface, as discussed in the previous section, is usually expressed in terms of a slip length, which quantifies the extent to which the fluid elements near the surface are affected by the surface energy and the surface geometry. The surface energy is an intrinsic property of a material that can be controlled by chemical treatment, such as etching approach and/or coat-on/cast approach. The surface roughness of a hydrophobic solid material can be tuned in order to increment its hydrophobicity and obtain a super-hydrophobic solid surface [33, 34]. In this section, from the numerical point of view, the effect of

Figure 9 shows the effect of slip zone of the artificial slip surface for several dimensionless slip length values on the dimensionless load support. As indicated in Figure 9, the increase in the slip length leads to an increase in the predicted load support. Generally, the larger the slip length at the optimized slip zone of the artificial slip zone, the higher the load support. However, when dimensionless slip length is larger than 30, the dimensionless load support is not affected significantly with the increase in the dimensionless slip length. So, the increase of the load support is not infinitely large. It can be deduced that there is no fluid load support for a lubricated-MEMS when it contains no-slip condition (*A* = 0). It indicates that the absence of the wedge effect on the pressure generation at parallel lubricated sliding contact has been counterbalanced by the influence of the ariticial slip surface application. Again, this condition is very advantageous in engineering a lubricated-MEMS which demonstrates parallel gaps.

means that the optimized artificial slip surface of lubricated-MEMS can produce a lower friction force than a no-slip contact.

force of 0.65, while the no-slip lubricated contact gives its dimensionless minimum friction of 0.77 at *h*\*

0 0.2 0.4 0.6 0.8 1

**Dimensionless** *x* **coordinate,** *X* 

**Figure 7.** Lubrication film pressure distributions for several values of wedge ratio. The slip profiles are calculated for

= 1) and *SxD* = 0.65, the artificial slip surface gives the minimum dimensionless friction

= 2.2. From this

optimal = 2.2, see Table 1. It

Figure 9 shows the effect of slip zone of the artificial slip surface for several dimensionless slip length values on the dimensionless load support. As indicated in Figure 9, the increase in the slip length leads to an increase in the predicted load support. Generally, the larger the slip length at the optimized slip zone of the artificial slip zone, the higher the load support.

slip length on the lubrication behavior is studied. The dimensionless slip length is varied from 3 to 300.

0 0.2 0.4 0.6 0.8 1

**Dimensionless slip zone,** *SxD*

A = 3 A = 30 A = 300 A = 0 (no-slip)

lubrication behavior is studied. The dimensionless slip length is varied from 3 to 300.

lubrication in MEMS.

way to increase the hydrodynamic performance and the stability of the lubricated MEMS system because it gives an advanced load support in combination with a reduced friction force. wedge ratios, see Fig. 6. With respect to the influence of the dimensionless slip length, opposite to the hydrodynamic load support, the dimensionless friction force becomes smaller for higher dimensionless slip length. Therefore, the optimized artificial slip surface is a very promising way to increase the hydrodynamic performance and the stability of the lubricated MEMS system because it gives an advanced load support in combination with a reduced friction force.

dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1).

Figure 9. Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless load support *W* for several values of

In Figure 10 the effect of slip zone of the artificial slip surface for several dimensionless slip length values on the dimensionless friction force is presented. As is well-known, the ability to control and manipulate friction force during sliding is extremely important key to prolong a life-time of lubricated-MEMS. Better understanding of the friction force phenomena at micro-scales is needed to provide designers and engineers the required tools and capabilities to control friction force and predict failure of

As can be seen in Figure 10 that the artificial slip surface leads to a reduction of the friction force for all dimensionless slip length.

force is of only particular interest, the homogeneous slip surface (*SxD* = 1) is very beneficial. But if the performance is also related to the load support, homogeneous slip is not recommended because when *SxD* = 1, the predicted load support is very small for all

Figure 10.Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction force *F* for several values of

The combined effect of slip zone parameter on load support and friction force can be better analyzed using the dimensionless

when there is no wedge effect, the load support has a maximum value. In addition, the friction

**Figure 11.** Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction coefficient *m* for several values of dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1).

0 0.2 0.4 0.6 0.8 1

**Dimensionless slip zone,** *S***xD**

dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1).

Figure 11.Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction coefficient *m* for several values of

A = 3 A = 30 A = 300

Artificial Slip Surface: Potential Application in Lubricated MEMS

http://dx.doi.org/10.5772/55745

247

Numerical results show that the hydrodynamics of a lubrication film confined between a moving no-slip surface and a stationary with an artificial slip surface differ significantly from that of a film confined between two no-slip surfaces. It is found that a homogeneous slip boundary on one surface produces a lower hydrodynamic pressure in a lubricated sliding contact at various conditions (slope incline, and slip length), resulting in a reduced load support which reduces the positive effect of slip on friction. However, if the surface is designed with an artificial slip pattern (the slip zone is applied on 0.65 of contact length), even when there is no wedge effect, the load support has a maximum value. In addition, the friction force can decrease significantly. Therefore, it is very beneficial to make one of the contacting surfaces in lubricated-MEMS with an artificial slip surface for achieving ideal

[1] Houston MR, Howe RT, Maboudian R. Effect of Hydrogen Termination on The Work of Adhesion Between Rough

[2] Maboudian R, Ashurst WR, Carraro C. Self-Assembled Monolayers as Anti-Stiction Coatings For MEMS: Characteristics and

[3] Tagawa M, Ikemura M, Nakayama Y, Ohmae N. Effect of Water Adsorption on Microtribological Properties of Hydrogenated

[4] Smallwood SA, Eapen KC, Patton ST, Zabinski JS. Performance Results of MEMS Coated with a Conformal DLC. Wear

[5] van Spengen WM, Puers R, De wolf I. On The Physics of Stiction and Its Impact on The Reliability of Microstructures. Journal

[6] Pit R, Hervet H, Leger L. Direct Experimental Evidence of Slip in Hexadecane: Solid Interfaces. Physical Review Letters

[7] Craig VSJ, Neto C, Williams DRM. Shear-Dependent Boundary Slip in an Aqueous Newtonian Liquid. Physical Review

[8] Bonaccurso E, Kappl M, Butt HJ. Hydrodynamic Force Measurements: Boundary Slip of Hydrophilic Surfaces and

[9] Bonaccurso E, Butt HJ, Craig VSJ. Surface Roughness and Hydrodynamic Boundary Slip of A Newtonian Fluid in A

[12] Cottin-Bizonne C, Barentin C, Charlaix E, Bocquet L, Barrat JL. Dynamics of Simple Liquids at Heterogeneous Surfaces: Molecular Dynamics Simulations and Hydrodynamic Description. European Physical Journal E 2004;15: 427– 438.

[10] Zhu YX, Granick S. Rate-Dependent Slip of Newtonian Liquid at Smooth Surfaces. Physical Review Letters 2001;87( 096105). [11] Priezjev NV, Darhuber AA, Troian SM. Slip Behaviour in Liquid Films on Surfaces of Patterned Wettability: Comparison

force can decrease significantly. Therefore, it is very beneficial to make one of the contacting

surfaces in lubricated-MEMS with an artificial slip surface for achieving ideal lubrication

lubrication performance, i.e. reduced friction coefficient and increased load support.

Polycrystalline Silicon Surfaces. Journal of Applied Physics 1997;81(8): 3474–3483.

Recent Developments. Sensors and Actuators A: Physical 2000;82(1–3): 219–223.

Diamond-Like Carbon Films. Tribology Letters 2004;17(3): 575–580.

of Adhesion Science and Technology 2003;17(4): 563–582.

Electrokinetic Effects. Physical Review Letters 2002;88(076103).

Completely Wetting System. Physical Review Letters 2003;90(144501).

1 Laboratory for Engineering Design and Tribology, Mechanical Engineering Department,

University of Diponegoro, Jl. Prof. H. Sudharto, Kampus UNDIP Tembalang, Semarang, Indo‐

2 Laboratory for Surface Technology and Tribology, University of Twente Drienerlolaan,

Between Continuum and Molecular Dynamics Simulations. Physical Review E 2005;71(041608).

and D.J. Schipper2\*

performance, i.e. reduced friction coefficient and increased load support.

**Author details**

nesia

Enschede, The Netherlands

M. Tauviqirrahman1,2, R. Ismail1,2, J. Jamari1

2006;260: 1179–1189.

2000;85: 980–983.

Letters 2001;87(054504).

**References** 

**4. Concluding remarks** 

**Dimensionless friction coefficient,** 

*m*

\*Address all correspondence to: d.j.schipper@utwente.nl

dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1). **Figure 10.** Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction force *F* for several values of dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1).

friction coefficient. In the present study the dimensionless friction coefficient *m* is defined as the ratio of the dimensionless friction coefficient *F* to the dimensionless load support *W*. Figure 11 shows the variation of the dimensionless friction coefficient *m* as a function of the dimensionless slip zone *SxD* for various slip lengths *A*. It appears that when the dimensionless slip zone is smaller than 0.2, the friction coefficient increases significantly. After *SxD* = 0.2, increasing the slip zone will be less significant to the reduction of the friction coefficient. It can also be deduced from Figure 11 that dimensionless friction coefficient *m* will decrease with the increase of the dimensionless slip length *A*, especially for small *SxD*. It implies that the minimum dimensionless friction coefficient is accord with the the highest dimensionless load support (when *SxD* = 0.65). The combined effect of slip zone parameter on load support and friction force can be better analyzed using the dimensionless friction coefficient. In the present study the dimensionless friction coefficient *m* is defined as the ratio of the dimensionless friction coefficient *F* to the dimensionless load support *W*. Figure 11 shows the variation of the dimensionless friction coefficient *m* as a function of the dimensionless slip zone *SxD* for various slip lengths *A*. It appears that when the dimensionless slip zone is smaller than 0.2, the friction coefficient increases significantly. After *SxD* = 0.2, increasing the slip zone will be less significant to the reduction of the friction coefficient. It can also be deduced from Figure 11 that dimensionless friction coefficient *m* will decrease with the increase of the dimensionless slip length *A*, especially for small *SxD*. It implies that the minimum dimensionless friction coefficient is accord with the the highest dimensionless load support (when *SxD* = 0.65).

#### **4. Concluding remarks**

Numerical results show that the hydrodynamics of a lubrication film confined between a moving no-slip surface and a stationary with an artificial slip surface differ significantly from that of a film confined between two no-slip surfaces. It is found that a homogeneous slip boundary on one surface produces a lower hydrodynamic pressure in a lubricated sliding contact at various conditions (slope incline, and slip length), resulting in a reduced load support which reduces the positive effect of slip on friction. However, if the surface is designed with an optimal artificial slip pattern (the slip zone is applied on 0.65 of contact length), even

Figure 11.Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction coefficient *m* for several values of dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1). **Figure 11.** Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction coefficient *m* for several values of dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1).

when there is no wedge effect, the load support has a maximum value. In addition, the friction force can decrease significantly. Therefore, it is very beneficial to make one of the contacting surfaces in lubricated-MEMS with an artificial slip surface for achieving ideal lubrication performance, i.e. reduced friction coefficient and increased load support. **4. Concluding remarks**  Numerical results show that the hydrodynamics of a lubrication film confined between a moving no-slip surface and a stationary with an artificial slip surface differ significantly from that of a film confined between two no-slip surfaces. It is found that a homogeneous slip boundary on one surface produces a lower hydrodynamic pressure in a lubricated sliding contact at various conditions (slope incline, and slip length), resulting in a reduced load support which reduces the positive effect of slip on friction. However, if the surface is designed with an artificial slip pattern (the slip zone is applied on 0.65 of contact length), even when there is no wedge effect, the load support has a maximum value. In addition, the friction force can decrease significantly. Therefore,

lubrication performance, i.e. reduced friction coefficient and increased load support.

Polycrystalline Silicon Surfaces. Journal of Applied Physics 1997;81(8): 3474–3483.

Diamond-Like Carbon Films. Tribology Letters 2004;17(3): 575–580.

of Adhesion Science and Technology 2003;17(4): 563–582.

it is very beneficial to make one of the contacting surfaces in lubricated-MEMS with an artificial slip surface for achieving ideal

[2] Maboudian R, Ashurst WR, Carraro C. Self-Assembled Monolayers as Anti-Stiction Coatings For MEMS: Characteristics and

[5] van Spengen WM, Puers R, De wolf I. On The Physics of Stiction and Its Impact on The Reliability of Microstructures. Journal

[9] Bonaccurso E, Butt HJ, Craig VSJ. Surface Roughness and Hydrodynamic Boundary Slip of A Newtonian Fluid in A

[12] Cottin-Bizonne C, Barentin C, Charlaix E, Bocquet L, Barrat JL. Dynamics of Simple Liquids at Heterogeneous Surfaces: Molecular Dynamics Simulations and Hydrodynamic Description. European Physical Journal E 2004;15: 427– 438.

#### **Author details** [1] Houston MR, Howe RT, Maboudian R. Effect of Hydrogen Termination on The Work of Adhesion Between Rough

**References** 

way to increase the hydrodynamic performance and the stability of the lubricated MEMS system because it gives an advanced load support in combination with a reduced friction force.

because it gives an advanced load support in combination with a reduced friction force.

dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1).

several values of dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1).

**Figure 10.** Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction force *F* for

The combined effect of slip zone parameter on load support and friction force can be better analyzed using the dimensionless friction coefficient. In the present study the dimensionless friction coefficient *m* is defined as the ratio of the dimensionless friction coefficient *F* to the dimensionless load support *W*. Figure 11 shows the variation of the dimensionless friction coefficient *m* as a function of the dimensionless slip zone *SxD* for various slip lengths *A*. It appears that when the dimensionless slip zone is smaller than 0.2, the friction coefficient increases significantly. After *SxD* = 0.2, increasing the slip zone will be less significant to the reduction of the friction coefficient. It can also be deduced from Figure 11 that dimensionless friction coefficient *m* will decrease with the increase of the dimensionless slip length *A*, especially for small *SxD*. It implies that the minimum dimensionless friction coefficient is accord

0.0 0.2 0.4 0.6 0.8 1.0

**Dimensionless slip zone,** *SxD*

coefficient is accord with the the highest dimensionless load support (when *SxD* = 0.65).

Numerical results show that the hydrodynamics of a lubrication film confined between a moving no-slip surface and a stationary with an artificial slip surface differ significantly from that of a film confined between two no-slip surfaces. It is found that a homogeneous slip boundary on one surface produces a lower hydrodynamic pressure in a lubricated sliding contact at various conditions (slope incline, and slip length), resulting in a reduced load support which reduces the positive effect of slip on friction. However, if the surface is designed with an optimal artificial slip pattern (the slip zone is applied on 0.65 of contact length), even

dimensionless slip length *A.* All profiles are calculated for parallel sliding surfaces (*h\** = 1).

lubrication in MEMS.

0.0

0.2

0.4

A = 3 A = 30 A = 300 A = 0 (no-slip)

with the the highest dimensionless load support (when *SxD* = 0.65).

**Dimensionless friction**

**4. Concluding remarks**

 **force,** 

*F*

246 Tribology - Fundamentals and Advancements

0.6

0.8

1.0

1.2

Figure 9. Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless load support *W* for several values of

In Figure 10 the effect of slip zone of the artificial slip surface for several dimensionless slip length values on the dimensionless friction force is presented. As is well-known, the ability to control and manipulate friction force during sliding is extremely important key to prolong a life-time of lubricated-MEMS. Better understanding of the friction force phenomena at micro-scales is needed to provide designers and engineers the required tools and capabilities to control friction force and predict failure of

As can be seen in Figure 10 that the artificial slip surface leads to a reduction of the friction force for all dimensionless slip length. The friction force decreases with increasing the slip zone *Sx*. It can be seen that when the slip zone covers over the whole surface, i.e. homogeneous slip surface, the friction forces has a minimum value, especially for a high slip length. If the reduction in friction force is of only particular interest, the homogeneous slip surface (*SxD* = 1) is very beneficial. But if the performance is also related to the load support, homogeneous slip is not recommended because when *SxD* = 1, the predicted load support is very small for all wedge ratios, see Fig. 6. With respect to the influence of the dimensionless slip length, opposite to the hydrodynamic load support, the dimensionless friction force becomes smaller for higher dimensionless slip length. Therefore, the optimized artificial slip surface is a very promising way to increase the hydrodynamic performance and the stability of the lubricated MEMS system

Figure 10.Effect of the dimensionless slip zone of the artificial slip surface *SxD* on the dimensionless friction force *F* for several values of

The combined effect of slip zone parameter on load support and friction force can be better analyzed using the dimensionless friction coefficient. In the present study the dimensionless friction coefficient *m* is defined as the ratio of the dimensionless friction coefficient *F* to the dimensionless load support *W*. Figure 11 shows the variation of the dimensionless friction coefficient *m* as a function of the dimensionless slip zone *SxD* for various slip lengths *A*. It appears that when the dimensionless slip zone is smaller than 0.2, the friction coefficient increases significantly. After *SxD* = 0.2, increasing the slip zone will be less significant to the reduction of the friction coefficient. It can also be deduced from Figure 11 that dimensionless friction coefficient *m* will decrease with the increase of the dimensionless slip length *A*, especially for small *SxD*. It implies that the minimum dimensionless friction

> M. Tauviqirrahman1,2, R. Ismail1,2, J. Jamari1 and D.J. Schipper2\* Recent Developments. Sensors and Actuators A: Physical 2000;82(1–3): 219–223. [3] Tagawa M, Ikemura M, Nakayama Y, Ohmae N. Effect of Water Adsorption on Microtribological Properties of Hydrogenated

> \*Address all correspondence to: d.j.schipper@utwente.nl [4] Smallwood SA, Eapen KC, Patton ST, Zabinski JS. Performance Results of MEMS Coated with a Conformal DLC. Wear 2006;260: 1179–1189.

> 1 Laboratory for Engineering Design and Tribology, Mechanical Engineering Department, University of Diponegoro, Jl. Prof. H. Sudharto, Kampus UNDIP Tembalang, Semarang, Indo‐ nesia [6] Pit R, Hervet H, Leger L. Direct Experimental Evidence of Slip in Hexadecane: Solid Interfaces. Physical Review Letters 2000;85: 980–983. [7] Craig VSJ, Neto C, Williams DRM. Shear-Dependent Boundary Slip in an Aqueous Newtonian Liquid. Physical Review Letters 2001;87(054504). [8] Bonaccurso E, Kappl M, Butt HJ. Hydrodynamic Force Measurements: Boundary Slip of Hydrophilic Surfaces and Electrokinetic Effects. Physical Review Letters 2002;88(076103).

> 2 Laboratory for Surface Technology and Tribology, University of Twente Drienerlolaan, Enschede, The Netherlands [10] Zhu YX, Granick S. Rate-Dependent Slip of Newtonian Liquid at Smooth Surfaces. Physical Review Letters 2001;87( 096105). [11] Priezjev NV, Darhuber AA, Troian SM. Slip Behaviour in Liquid Films on Surfaces of Patterned Wettability: Comparison Between Continuum and Molecular Dynamics Simulations. Physical Review E 2005;71(041608).

Completely Wetting System. Physical Review Letters 2003;90(144501).

#### **References**

[1] Houston MR, Howe RT, Maboudian R. Effect of Hydrogen Termination on The Work of Adhesion Between Rough Polycrystalline Silicon Surfaces. Journal of Applied Physics 1997;81(8): 3474–3483.

[14] Li BM, Kwok DY. Discrete Boltzmann Equation for Microfluidics. Physical Review

Artificial Slip Surface: Potential Application in Lubricated MEMS

http://dx.doi.org/10.5772/55745

249

[15] Spikes HA. The Half-Wetted Bearing. Part 1: Extended Reynolds Equation. Proceed‐ ings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribol‐

[16] Spikes HA. The Half-Wetted Bearing. Part 2: Potential Application in Low Load Con‐ tacts. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engi‐

[17] Salant RF, Fortier AE. Numerical Analysis of A Slider Bearing with A Heterogeneous

[18] Fortier AE, Salant RF. Numerical Analysis of A Journal Bearing with A Heterogene‐

[19] Wu CW, Ma GJ, Zhou P. Low Friction and High Load Support Capacity of Slider Bearing with A Mixed Slip Surface. ASME Journal of Tribology 2006;128: 904–907. [20] Choo JH, Glovnea RP, Forrest AK, Spikes HA. A Low Friction Bearing Based on Liq‐

[21] Tauviqirrahman M, Ismail R, Jamari, Schipper DJ. Effect of Boundary Slip on The Load Support in A Lubricated Sliding Contact. AIP Conference Proceedings

[22] Aurelian F, Patrick M, Mohamed H. Wall Slip Effects in (Elasto) Hydrodynamic Jour‐

[23] Tauviqirrahman M, Ismail R, Jamari, Schipper DJ. Wall Slip Effects in a Lubricated

[24] Navier CLMH. Mémoire Sur Les Lois Du Mouvement Des Fluides. Mémoires de

[25] Israelachvili J. Intermolecular and Surface Force, vol. 1, second ed. Academic Press,

[26] Reynolds O. On The Theory of Lubrication and Its Application to Mr. Beauchamp Tower's Experiments, Including An Experimental Determination of The Viscosity of Olive Oil. Philosophical Transactions of the Royal Society of London, Part I 1886;177:

[27] Patankar SV. Numerical Heat Transfer and Fluid Flow. Taylor & Francis, Levittown

[29] Kompvopoulos K. Surface Engineering and Microtribology for Microelectromechani‐

[28] Cameron A. The Principles of Lubrication. Longman Green and Co., ltd; 1966.

ous Slip/No-Slip Surface. ASME Journal of Tribology 2005;127: 820–825.

uid Slip at The Wall. ASME Journal of Tribology 2007;129: 611–620.

MEMS. International Journal of Energy Machinery 2011;4(1): 13–22.

l'Académie Royale des Sciences de l'Institut de France 1823:6: 389–440.

Slip/No-Slip Surface. Tribology Transactions 2004; 47: 328–334.

Letters 2003;90(124502).

ogy 2003; 217: 1–14.

neering Tribology 2003;217: 15–26.

2011;1415: 51-54. doi:10.1063/1.3667218.

London; 1995.

157–234.

1980:30–58.

cal System. Wear 1996;200: 305–327.

nal Bearing. Tribology International 2011;44: 868–877.


[14] Li BM, Kwok DY. Discrete Boltzmann Equation for Microfluidics. Physical Review Letters 2003;90(124502).

**References**

248 Tribology - Fundamentals and Advancements

Physics 1997;81(8): 3474–3483.

Letters 2004;17(3): 575–580.

2003;17(4): 563–582.

2002;88(076103).

2003;90(144501).

tors A: Physical 2000;82(1–3): 219–223.

[1] Houston MR, Howe RT, Maboudian R. Effect of Hydrogen Termination on The Work of Adhesion Between Rough Polycrystalline Silicon Surfaces. Journal of Applied

[2] Maboudian R, Ashurst WR, Carraro C. Self-Assembled Monolayers as Anti-Stiction Coatings For MEMS: Characteristics and Recent Developments. Sensors and Actua‐

[3] Tagawa M, Ikemura M, Nakayama Y, Ohmae N. Effect of Water Adsorption on Mi‐ crotribological Properties of Hydrogenated Diamond-Like Carbon Films. Tribology

[4] Smallwood SA, Eapen KC, Patton ST, Zabinski JS. Performance Results of MEMS

[5] van Spengen WM, Puers R, De wolf I. On The Physics of Stiction and Its Impact on The Reliability of Microstructures. Journal of Adhesion Science and Technology

[6] Pit R, Hervet H, Leger L. Direct Experimental Evidence of Slip in Hexadecane: Solid

[7] Craig VSJ, Neto C, Williams DRM. Shear-Dependent Boundary Slip in an Aqueous

[8] Bonaccurso E, Kappl M, Butt HJ. Hydrodynamic Force Measurements: Boundary Slip of Hydrophilic Surfaces and Electrokinetic Effects. Physical Review Letters

[9] Bonaccurso E, Butt HJ, Craig VSJ. Surface Roughness and Hydrodynamic Boundary Slip of A Newtonian Fluid in A Completely Wetting System. Physical Review Letters

[10] Zhu YX, Granick S. Rate-Dependent Slip of Newtonian Liquid at Smooth Surfaces.

[11] Priezjev NV, Darhuber AA, Troian SM. Slip Behaviour in Liquid Films on Surfaces of Patterned Wettability: Comparison Between Continuum and Molecular Dynamics

[12] Cottin-Bizonne C, Barentin C, Charlaix E, Bocquet L, Barrat JL. Dynamics of Simple Liquids at Heterogeneous Surfaces: Molecular Dynamics Simulations and Hydrody‐

[13] Harting J, Kunert C, Herrmann HJ. Lattice Boltzmann Simulations of Apparent Slip

namic Description. European Physical Journal E 2004;15: 427– 438.

in Hydrophobic Channels. Europhysic Letters 2006; 75: 328–334.

Coated with a Conformal DLC. Wear 2006;260: 1179–1189.

Interfaces. Physical Review Letters 2000;85: 980–983.

Physical Review Letters 2001;87( 096105).

Simulations. Physical Review E 2005;71(041608).

Newtonian Liquid. Physical Review Letters 2001;87(054504).


[30] Tretheway DC, Meinhart, CD. Apparent Fluid Slip at Hydrophobic Microchannel Walls. Physics of Fluids 2002;14: L9-12.

**Chapter 9**

**Friction and Wear of a Grease Lubricated Contact —**

Lubricating greases are colloid disperse systems with visco-elastical properties. These special lubricants have a wide range of application. More than 90% of the ball bearings are grease

The composition of greases consists of a base oil and a thickener (of course some additives can be found in commercial lubricants). The thickener forms a network which leads to a complex

Tribological contacts lubricated by greases are often working under mixed friction conditions. The situation inside the grease film determines the content of fluid friction. The situation inside

A special phenomenon is the structural degradation due to the effects of friction. It leads to a

Aims of this chapter are the description of the liquid friction process inside the grease film and of the procedure of structural degradation. Both phenomena have a strong influence on the

The presented work is based on a conception of friction and wear that investigates the process from an energy point of view. Some ideas differ from well-known definitions so that they need

> © 2013 Kuhn; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,

© 2013 Kuhn; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

distribution, and reproduction in any medium, provided the original work is properly cited.

lubricated but also gears and journal bearings are application examples for greases.

**An Energetic Approach**

http://dx.doi.org/10.5772/55837

rheological and tribological behaviour.

dependence on time of the grease behaviour.

the tribological gap influences the content of solid friction.

friction and wear behaviour of the whole tribo-system.

**2. General definitions and terminology**

to be introduced here for better understanding.

Additional information is available at the end of the chapter

Erik Kuhn

**1. Introduction**


### **Friction and Wear of a Grease Lubricated Contact — An Energetic Approach**

Erik Kuhn

[30] Tretheway DC, Meinhart, CD. Apparent Fluid Slip at Hydrophobic Microchannel

[31] Watanabe K, Yanuar, Udagawa H. Drag Reduction of Newtonian Fluid in a Circular Pipe with a Highly Water-Repellant Wall. Journal of Fluid Mechanics 1999;381: 225–

[32] Ou J, Perot B, Rothstein JP. Laminar Drag Reduction in Microchannels using Ultrahy‐

[33] Patankar NA. On the Modelling of Hydrophobic Contact Angles on Rough Surfaces.

[34] Lafuma A, Quere D. Superhydrophobic States. Nature Materials 2003;2: 457–460.

Walls. Physics of Fluids 2002;14: L9-12.

Langmuir 2003;19: 1249–1253.

drophobic Surfaces. Physics of Fluids 2004;16: 4635.

238.

250 Tribology - Fundamentals and Advancements

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/55837

#### **1. Introduction**

Lubricating greases are colloid disperse systems with visco-elastical properties. These special lubricants have a wide range of application. More than 90% of the ball bearings are grease lubricated but also gears and journal bearings are application examples for greases.

The composition of greases consists of a base oil and a thickener (of course some additives can be found in commercial lubricants). The thickener forms a network which leads to a complex rheological and tribological behaviour.

Tribological contacts lubricated by greases are often working under mixed friction conditions.

The situation inside the grease film determines the content of fluid friction. The situation inside the tribological gap influences the content of solid friction.

A special phenomenon is the structural degradation due to the effects of friction. It leads to a dependence on time of the grease behaviour.

Aims of this chapter are the description of the liquid friction process inside the grease film and of the procedure of structural degradation. Both phenomena have a strong influence on the friction and wear behaviour of the whole tribo-system.

#### **2. General definitions and terminology**

The presented work is based on a conception of friction and wear that investigates the process from an energy point of view. Some ideas differ from well-known definitions so that they need to be introduced here for better understanding.

A *tribological contact* describes the geometrical situation of rubbing bodies with interactions in the sense of the tribological process.

*Friction* is (only) an energy expenditure.

*Wear* is a production of irreversibility due to affected friction energy. It covers all elements of a tribo-system [1].

This chapter distinguishes between wear volume and volume of the removed material.

*Wear volume* is presented by the material area where irreversible friction effects lead to an excess of a critical energetic level. (in contrast to the removed material volume (loss of material)).

*Mixed friction* describes an energy expenditure in several states of friction that exists simulta‐ neously inside the same tribo-system (see also [2]). A direct contact of the solid bodies is not necessary for mixed friction of a lubricated couple (mix of solid and liquid friction).

*Lubricating greases* are colloid disperse systems with visco-elastical properties.

*The grease structure* is characterised by the geometry and the distribution of the thickener, the interactions between the thickener and the tribo-system and the ability of storing energy.

#### **3. Energy balance for a grease lubricated contact**

Grease lubricated contacts are often working in mixed friction. That means liquid friction (grease film) and solid friction (asperity deformation) have to be considered. An idea of the situation inside the lubricated contact is presented in the next figures.

**Figure 3.** Single contact and modelled contact situation inside the grease film

**Figure 2.** Modelled mixed friction contact of a grease lubricated gap

the Gaussian distribution.

r

x

 x x xx

expended energy is formed as a sum of different contents

Fig.3 the contact of two density areas inside the greases film is highlighted.

For the contact probability *F* illustrated in Fig.3 (left) can be obtained [3]

The grease topography comes from IR-microscopy and presents the density distribution. In

Investigations of the surface profile and of the density distribution of the grease lead to a discrete contact model. Random variables are asperity height *ξ*1,2, radius of the modelled asperity shape *ρ*1,2 and the observed density *δ*1,2. All random parameters can be described with

( ) ( ) ( ) ( ) ( ) ( ) <sup>2</sup> 1 22 2

 - <sup>=</sup> × ×× × ò ò ò òò ò (1)

An energy balance is created by investigation of a single grease lubricated contact. The

 r r

 d d

Friction and Wear of a Grease Lubricated Contact — An Energetic Approach

http://dx.doi.org/10.5772/55837

253

*W W* <sup>=</sup> <sup>=</sup> å (2)

r

 r

 d d

 r

1 2 12 1 1 2 1 1 2 2 0 0 <sup>1</sup> <sup>2</sup> <sup>1</sup> <sup>2</sup> (,, ) *z z ug ug <sup>g</sup> <sup>g</sup> uk uk <sup>k</sup> <sup>k</sup> Fzu f f dd f d f f d f d*

> *friction i n* 1.. *i*

 r r

**Figure 1.** Grease lubricated contact with two rough surfaces

The surface profile is modelled with a spherical shape of the asperities and the gap configu‐ ration can be illustrated with Fig.2.

The situation of a single contact in consideration of the simple idea of lubricant layers inside the grease film is presented in Fig.3.

**Figure 2.** Modelled mixed friction contact of a grease lubricated gap

A *tribological contact* describes the geometrical situation of rubbing bodies with interactions in

*Wear* is a production of irreversibility due to affected friction energy. It covers all elements of

*Wear volume* is presented by the material area where irreversible friction effects lead to an excess of a critical energetic level. (in contrast to the removed material volume (loss of material)).

*Mixed friction* describes an energy expenditure in several states of friction that exists simulta‐ neously inside the same tribo-system (see also [2]). A direct contact of the solid bodies is not

*The grease structure* is characterised by the geometry and the distribution of the thickener, the interactions between the thickener and the tribo-system and the ability of storing energy.

Grease lubricated contacts are often working in mixed friction. That means liquid friction (grease film) and solid friction (asperity deformation) have to be considered. An idea of the

The surface profile is modelled with a spherical shape of the asperities and the gap configu‐

The situation of a single contact in consideration of the simple idea of lubricant layers inside

This chapter distinguishes between wear volume and volume of the removed material.

necessary for mixed friction of a lubricated couple (mix of solid and liquid friction).

*Lubricating greases* are colloid disperse systems with visco-elastical properties.

**3. Energy balance for a grease lubricated contact**

**Figure 1.** Grease lubricated contact with two rough surfaces

ration can be illustrated with Fig.2.

the grease film is presented in Fig.3.

situation inside the lubricated contact is presented in the next figures.

the sense of the tribological process.

252 Tribology - Fundamentals and Advancements

a tribo-system [1].

*Friction* is (only) an energy expenditure.

**Figure 3.** Single contact and modelled contact situation inside the grease film

The grease topography comes from IR-microscopy and presents the density distribution. In Fig.3 the contact of two density areas inside the greases film is highlighted.

Investigations of the surface profile and of the density distribution of the grease lead to a discrete contact model. Random variables are asperity height *ξ*1,2, radius of the modelled asperity shape *ρ*1,2 and the observed density *δ*1,2. All random parameters can be described with the Gaussian distribution.

For the contact probability *F* illustrated in Fig.3 (left) can be obtained [3]

$$F(z,u,\rho) = \int\_0^z \int\_0^{z-\frac{\rho}{\tau\_2}} f\left(\xi\_1\right) f\left(\xi\_2\right) d\xi\_1 d\xi\_2 \cdot \int\_{uk1}^{u\cdot y 1} f\left(\rho\_1\right) d\rho\_1 \cdot \int\_{uk2}^{u\cdot y} f\left(\rho\_2\right) \cdot \int\_{\rho k1}^{\rho\_3 q} f\left(\delta\_1\right) d\delta\_1 \cdot \int\_{\rho k2}^{\rho\_3 q} f\left(\delta\_2\right) d\delta\_2 \tag{1}$$

An energy balance is created by investigation of a single grease lubricated contact. The expended energy is formed as a sum of different contents

$$\mathcal{W}\_{friction} = \sum\_{i} \mathcal{W}\_{i=1\ldots n} \tag{2}$$

The assumption of mixed friction leads to the consideration of two main contents of the friction energy.

$$\mathcal{W}\_{mixed-friction} = \mathcal{W}\_{solid} + \mathcal{W}\_{fluid} \tag{3}$$

**4. Liquid friction inside the grease film**

As mentioned before a micro single contact of a grease lubricated couple is observed and the stress situation is analysed. The idea of the process model developed here is illustrated in Fig.5.

Friction and Wear of a Grease Lubricated Contact — An Energetic Approach

http://dx.doi.org/10.5772/55837

255

**Figure 5.** Grease lubricated gap between two asperities (micro single contact). Shear stress is the most important stress mechanism of the grease. Normal force (+) leads to solidification effects. Normal force (-) leads to tensile stress.

In addition the possibility of a friction wave inside the grease film is observed. The idea is

**Figure 6.** Schematic representation of a friction wave caused by the contact of volume elements of the grease film

**4.1. Stress situation inside the gap geometry**

clarified in the Fig.6.

with different properties

A proposal is made with

$$\mathcal{W}\_{solid} = e\_{elast/plast} \cdot \mathcal{V}\_{elast/plast} \cdot \mathcal{n}\_{elast/plast} \tag{4}$$

$$\mathcal{W}\_{\text{liquid}} = \mathfrak{e}\_{\text{hro1},2,3} \cdot V\_{\text{sha1},2,3} \cdot \mathfrak{n}\_{\text{hro1},2,3} + \mathfrak{e}\_{\text{sáid}} \cdot \mathfrak{n}\_{\text{sáid}f} \cdot V\_{\text{sáid}f} + \mathfrak{e}\_{\text{treslele}} \cdot V\_{\text{treslele}} \cdot \mathfrak{n}\_{\text{treslele}} + \mathfrak{e}\_{\text{uure}} \cdot V\_{\text{uure}} \cdot \mathfrak{n}\_{\text{uure}} \tag{5}$$

In Eq.(4) and (5) the notation *energy density* for the selected mechanism multiplied by *stressed volume* and by the number of contacts with the same mechanism is used. For a single contact n=1 holds.The solid friction from Eq.(4) considers the deformation (elastically/plastically) of the stressed asperities. Information about this is given in [4].

The summands in Eq.(5) describe the shearing process in different gap situations (index *rheo*), solidification effects in the center of the observed gap (index *solidif*), tensile stress in the outlet of the contact (index *tensile*) and the formation of a friction wave (index *wave*).

**Figure 4.** Different stress situations for the grease inside the contact geometry (*erheo*1,2,3)

#### **4. Liquid friction inside the grease film**

#### **4.1. Stress situation inside the gap geometry**

The assumption of mixed friction leads to the consideration of two main contents of the friction

*W e V n e n V e V n eVn liquid rheo rheo rheo solidif solidif solidif tensile tensile tensile wave wave wave* 1,2,3 1,2,3 1,2,3 = × × + × × + × × +× × (5)

In Eq.(4) and (5) the notation *energy density* for the selected mechanism multiplied by *stressed volume* and by the number of contacts with the same mechanism is used. For a single contact n=1 holds.The solid friction from Eq.(4) considers the deformation (elastically/plastically) of

The summands in Eq.(5) describe the shearing process in different gap situations (index *rheo*), solidification effects in the center of the observed gap (index *solidif*), tensile stress in the outlet

of the contact (index *tensile*) and the formation of a friction wave (index *wave*).

**Figure 4.** Different stress situations for the grease inside the contact geometry (*erheo*1,2,3)

the stressed asperities. Information about this is given in [4].

*W WW mixed friction solid fluid* - = + (3)

*We V n solid elast plast elast plast elast plast* / // =× × (4)

energy.

A proposal is made with

254 Tribology - Fundamentals and Advancements

As mentioned before a micro single contact of a grease lubricated couple is observed and the stress situation is analysed. The idea of the process model developed here is illustrated in Fig.5.

**Figure 5.** Grease lubricated gap between two asperities (micro single contact). Shear stress is the most important stress mechanism of the grease. Normal force (+) leads to solidification effects. Normal force (-) leads to tensile stress.

In addition the possibility of a friction wave inside the grease film is observed. The idea is clarified in the Fig.6.

#### **4.2. Empirical proposals to quantify the friction energy**

The experimental work is focused on the quantification of the energy expenditure during the shear process of a lubricant. This requires an experimental procedure that simulates the liquid friction inside the grease film.

of fluid friction inside the grease film experiments within the linear visco-elastical range can

cos *elast*

 is the storage modulus [Pa], *γ* is the deformation [-] and *δ* is the phase different angle. Eq. (7) describes the energy per volume J/m3 to deform the sample elastically and is related to

**Figure 8.** Typical evolution of the storage modulus and the loss modulus during an amplitude sweep (oscillating

The energy expenditure expressed by the rheological energy density *erheo*−*elast* presents the liquid friction behaviour of the investigated grease samples. Equation (7) observes only the shear mechanism inside the tribological gap. To compare the greases the same deformation (oscillating amplitude) has to be used. Some greases with the same thickener type (Li-soap) and the same base oil (mineral oil) were observed by a variation of soap content and test

d

g

*G*


2

¢ <sup>×</sup> <sup>=</sup> (7)

http://dx.doi.org/10.5772/55837

257

Friction and Wear of a Grease Lubricated Contact — An Energetic Approach

To quantify the friction behaviour the following proposal [3] can be used

*rheo elast*

*e*

be analysed.

the thickener.

measurement)

temperature.

**4.3. Results from experimental work**

*G* ′

A proposal is made by the use of a rheometer [5], [6]. Shearing a grease sample in a rotating or oscillating test can be interpreted as a fluid friction experiment. All reaction measured by the rheometer is caused by the grease behaviour. The plate and cone in a rheometer configu‐ ration does not lead to the state of mixed friction. Although the test conditions are fare from real contact situation rheometer tests are helpful for fundamental investigations of fluid friction.

A picture of a cone-plate configuration is shown in Fig.7.

**Figure 7.** Evolution of the observed shear stress (left) during a rheometer experiment with a cone-plate system (right) in rotational modus

The energy per volume that is necessary to shear the grease during the experiment can be obtained from a shear test (Fig.7). The experimental conditions are constant shear rate and constant test temperature. To compare different grease samples the test time for each experi‐ ment has to be held constant.

An empirical proposal [7] is made with

$$
\varepsilon\_{\rm rheo} = \dot{\gamma}\_{\rm const} \cdot \int\_0^t \tau \left( \boldsymbol{\zeta} \,\right) d\boldsymbol{\zeta} \tag{6}
$$

with *erheo* the energy density for the shear process [J/m<sup>3</sup> ], t the test time [s], *γ*˙ the shear rate [1/ s] and *ς* the current time [s].

A more interesting experimental procedure is an oscillating rheometer test [Fig.8]. The grease behaviour can be observed in a wide range of the oscillating amplitude. For the investigation of fluid friction inside the grease film experiments within the linear visco-elastical range can be analysed.

To quantify the friction behaviour the following proposal [3] can be used

**4.2. Empirical proposals to quantify the friction energy**

A picture of a cone-plate configuration is shown in Fig.7.

friction inside the grease film.

256 Tribology - Fundamentals and Advancements

friction.

in rotational modus

ment has to be held constant.

s] and *ς* the current time [s].

An empirical proposal [7] is made with

with *erheo* the energy density for the shear process [J/m<sup>3</sup>

The experimental work is focused on the quantification of the energy expenditure during the shear process of a lubricant. This requires an experimental procedure that simulates the liquid

A proposal is made by the use of a rheometer [5], [6]. Shearing a grease sample in a rotating or oscillating test can be interpreted as a fluid friction experiment. All reaction measured by the rheometer is caused by the grease behaviour. The plate and cone in a rheometer configu‐ ration does not lead to the state of mixed friction. Although the test conditions are fare from real contact situation rheometer tests are helpful for fundamental investigations of fluid

**Figure 7.** Evolution of the observed shear stress (left) during a rheometer experiment with a cone-plate system (right)

The energy per volume that is necessary to shear the grease during the experiment can be obtained from a shear test (Fig.7). The experimental conditions are constant shear rate and constant test temperature. To compare different grease samples the test time for each experi‐

> ( ) <sup>0</sup> *t rheo const e d* = × g

ò & (6)

], t the test time [s], *γ*˙ the shear rate [1/

 tV V

A more interesting experimental procedure is an oscillating rheometer test [Fig.8]. The grease behaviour can be observed in a wide range of the oscillating amplitude. For the investigation

$$\left(e\_{rhco-elast} = \frac{G' \cdot \mathcal{V}\_{elast}}{\cos \delta}\right) \tag{7}$$

*G* ′ is the storage modulus [Pa], *γ* is the deformation [-] and *δ* is the phase different angle. Eq. (7) describes the energy per volume J/m3 to deform the sample elastically and is related to the thickener.

#### **4.3. Results from experimental work**

The energy expenditure expressed by the rheological energy density *erheo*−*elast* presents the liquid friction behaviour of the investigated grease samples. Equation (7) observes only the shear mechanism inside the tribological gap. To compare the greases the same deformation (oscillating amplitude) has to be used. Some greases with the same thickener type (Li-soap) and the same base oil (mineral oil) were observed by a variation of soap content and test temperature.

**Figure 9.** Energy densities from the linear visco-elastical range =20°*C* for different soap contents

An increase of the soap content leads to an increase of the liquid friction. Experiments with a temperature *ϑ* =50°*C* and the same grease samples deliver the results in Fig.10.

**Figure 11.** Critical energy level vs. soap content

Friction and Wear of a Grease Lubricated Contact — An Energetic Approach

http://dx.doi.org/10.5772/55837

259

**Figure 12.** Energy densities for the crossing point vs. soap content

**Figure 10.** Energy densities for a test temperature =50°*C* (same conditions as Fig.9)

Compared with the test temperature of *ϑ* =20°*C* lower values of the energy densities are obtained. This behaviour is in accordance with the experience that fluid friction is decreasing with an increasing temperature. A more or less linear correlation of energy density and soap content can be observed.

**Figure 11.** Critical energy level vs. soap content

**Figure 9.** Energy densities from the linear visco-elastical range =20°*C* for different soap contents

temperature *ϑ* =50°*C* and the same grease samples deliver the results in Fig.10.

**Figure 10.** Energy densities for a test temperature =50°*C* (same conditions as Fig.9)

content can be observed.

258 Tribology - Fundamentals and Advancements

An increase of the soap content leads to an increase of the liquid friction. Experiments with a

Compared with the test temperature of *ϑ* =20°*C* lower values of the energy densities are obtained. This behaviour is in accordance with the experience that fluid friction is decreasing with an increasing temperature. A more or less linear correlation of energy density and soap

**Figure 12.** Energy densities for the crossing point vs. soap content

### **5. Irreversible effects due to friction**

#### **5.1. Idea of the structural degradation of lubricating greases**

A typical curve obtained from rheometer experiments for constant shear rate and temperature (rotational mode) shows a strong dependence on time. The drop of shear stress versus stress time is an indirect expression of the structural degradation and well known from many papers [8],[9],[10].

To illustrate the friction effects AFM-investigations made by [11] are presented below (Fig. 13). The change of thickener structure caused by the liquid friction is evident. The geometry and distribution of the thickener is completely different to the initial situation and it can be assumed that the new grease structure shows a different tribological behaviour.

Volume elements inside the grease film are modelled to observe the tribological process. Because of the thickener distribution these volume elements have different properties as elasticity, density, level of accumulated energy, level of critical energy etc.. The consequence of the property distribution is a different tribological behaviour of the observed volume elements forming a lubricant layer. The contact situation of two assumed grease layer com‐ posed of different volume elements is presented in Fig. 14.

**Figure 14.** Liquid friction between two modelled grease layer. Different volume elements have different rheological

Friction and Wear of a Grease Lubricated Contact — An Energetic Approach

http://dx.doi.org/10.5772/55837

261

**Figure 15.** Modelled volume element inside the grease film. Left – unstressed, right – stressed by liquid friction

A contact model is developed to describe the energetic situation of critical exceedance.

used. Information about the conditions and definitions can obtained in [1] and [12].

<sup>0</sup> <sup>0</sup>

e*L*

0

<sup>1</sup> ( ) <sup>2</sup>

*<sup>m</sup> <sup>m</sup> EN u <sup>e</sup>* p*m*

® e

The degradation of the grease structure begins with a transition process within a volume element. Overstepping a critical energy density initiates an irreversible change of the structure.

To quantify the number of exceedance of critical energy level a stationary Gaussian process is

e

2 <sup>0</sup> 2 2

*u*

0

é ù =× × ¢ ë û <sup>×</sup> ò (8)

é ù =× × ë û (9)

0 <sup>1</sup> ( ) lim [ ( , ) ( )] *L EN u E A x h x dx*

and tribological properties.

**Figure 13.** Left- fresh grease sample, right- grease stressed in a rheometer [11]

Liquid friction between two modelled grease layer. Different volume elements have different rheological and tribological properties.

An energy stress is applied to the control volume if liquid friction take place (Fig.15). Due to this energy stress an energy accumulation process, a dissipation process and a transition process (overstep of a critical energy level) starts.

**5. Irreversible effects due to friction**

260 Tribology - Fundamentals and Advancements

[8],[9],[10].

**5.1. Idea of the structural degradation of lubricating greases**

posed of different volume elements is presented in Fig. 14.

**Figure 13.** Left- fresh grease sample, right- grease stressed in a rheometer [11]

rheological and tribological properties.

process (overstep of a critical energy level) starts.

A typical curve obtained from rheometer experiments for constant shear rate and temperature (rotational mode) shows a strong dependence on time. The drop of shear stress versus stress time is an indirect expression of the structural degradation and well known from many papers

To illustrate the friction effects AFM-investigations made by [11] are presented below (Fig. 13). The change of thickener structure caused by the liquid friction is evident. The geometry and distribution of the thickener is completely different to the initial situation and it can be

Volume elements inside the grease film are modelled to observe the tribological process. Because of the thickener distribution these volume elements have different properties as elasticity, density, level of accumulated energy, level of critical energy etc.. The consequence of the property distribution is a different tribological behaviour of the observed volume elements forming a lubricant layer. The contact situation of two assumed grease layer com‐

Liquid friction between two modelled grease layer. Different volume elements have different

An energy stress is applied to the control volume if liquid friction take place (Fig.15). Due to this energy stress an energy accumulation process, a dissipation process and a transition

assumed that the new grease structure shows a different tribological behaviour.

**Figure 14.** Liquid friction between two modelled grease layer. Different volume elements have different rheological and tribological properties.

**Figure 15.** Modelled volume element inside the grease film. Left – unstressed, right – stressed by liquid friction

The degradation of the grease structure begins with a transition process within a volume element. Overstepping a critical energy density initiates an irreversible change of the structure.

A contact model is developed to describe the energetic situation of critical exceedance.

To quantify the number of exceedance of critical energy level a stationary Gaussian process is used. Information about the conditions and definitions can obtained in [1] and [12].

$$E\left[N\_0(\mu)\right] = \lim\_{\varepsilon \to 0} \frac{1}{\varepsilon \cdot L} \cdot \int\_0^L E[A(\infty, \varepsilon) \cdot h'(\infty)] d\mathbf{x} \tag{8}$$

$$E\left[N\_0(\mu)\right] = \frac{1}{2\pi} \cdot \sqrt{\frac{m\_2}{m\_0} \cdot e^{\frac{\mu^2}{2m\_0}}}\tag{9}$$

**5.2. Thermodynamic investiagtions**

flow in lubricated sliding systems.

systematic study of wear and friction" [15].

*5.2.2. Entropy and structural degradation*

namic forces, and the *Onsager-reciprocity* [20].

lubricant layer 2.

promising tool to analyse the tribological processes.

Source of irreversible processes are thermodynamic forces *Xi*

ture, gradient of concentration...). These forces Xi evoke corresponding flows *Ii*

correlation between components wear and entropy flow.

produce entropy, entropy becomes a time base for wear" [13].

Friction process within a tribo-system is an irreversible process. It means that input of friction energy leads to irreversible effects. This approach interpreted friction and wear process as an

Friction and Wear of a Grease Lubricated Contact — An Energetic Approach

http://dx.doi.org/10.5772/55837

263

Different authors tried to find relations between the system behaviour expressed by mass loss (wear) and entropy flow/production [15]-[19]. *Abdel-Aal* [18] expressed the conjecture that relation between frictional heat generation and heat dissipation is related to wear transition. This author pointed out [19] that there exists a one to one relationship between entropy generation and mass loss. *Doelling et al.* [17] give the argumentation that there exists a strong

*Ling et al.* [13] give an experimental description of a correlation between wear and entropy

"Sliding wear is an irreversible degradation of surfaces induced by friction. On a microscopic scale irreversible physical interactions between the sliding surfaces – including plastic deformation of asperities, fracture, delamination, abrasive plowing, and corrosive wear, among others – creates friction resistance forces, dissipates power, and generates irreversible entropy. Since the physical interactions responsible for friction and wear monotonically

"The entropy production, in fact, enable one to bridge the atomic scale phenomena with the macro scale response" [14] in [13]. "In addition friction and wear, from the vantage point of thermodynamics irreversible transform mechanical energy into other forms through dissipa‐ tive processes. Therefore, entropy production is believed to be a propitious measure for a

All these investigations pointed out that the application of irreversible thermodynamics is a

The aim of this chapter is a description of relation between energetic situation of the tribosystem and the degradation of grease structure. Tribo-sytems as solid surface 1, solid surface 2 and lubricant are investigated but also subsystems as lubricant layer 1 against

flow, diffusion flow...) The generalisation of classical thermodynamic to describe irreversible effects leads to an investigation of local equilibrium. That means there exist macroscopic small system areas that are provided in equilibrium while the whole system is out of equilibrium. Two principles of thermodynamic are used: the linear dependence of flows and thermody‐

(*i* =1, 2...) (gradient of tempera‐

(*i* =1, 2, ...)(heat

*5.2.1. General aspects*

cause-effect-chain.

**Figure 16.** New model to describe the energetic situation fort he transition process inside the grease film [1]

Initial situation is described with Eq. (8) and the expectation of the number of overstepping *N*0 can be determined with Eq. (9). This proposal uses spectral moments *m*2;*m*0.

An example [1] was quantified by using density distribution from IR-microscopy to determine the parameter *m*2;*m*0. A mean value of the critical energy level was assumed.

**Figure 17.** Influence of the critical energy level of the number of oversteps [1]

#### **5.2. Thermodynamic investiagtions**

#### *5.2.1. General aspects*

Initial situation is described with Eq. (8) and the expectation of the number of overstepping

An example [1] was quantified by using density distribution from IR-microscopy to determine

*N*0 can be determined with Eq. (9). This proposal uses spectral moments *m*2;*m*0.

262 Tribology - Fundamentals and Advancements

**Figure 16.** New model to describe the energetic situation fort he transition process inside the grease film [1]

the parameter *m*2;*m*0. A mean value of the critical energy level was assumed.

**Figure 17.** Influence of the critical energy level of the number of oversteps [1]

Friction process within a tribo-system is an irreversible process. It means that input of friction energy leads to irreversible effects. This approach interpreted friction and wear process as an cause-effect-chain.

Different authors tried to find relations between the system behaviour expressed by mass loss (wear) and entropy flow/production [15]-[19]. *Abdel-Aal* [18] expressed the conjecture that relation between frictional heat generation and heat dissipation is related to wear transition. This author pointed out [19] that there exists a one to one relationship between entropy generation and mass loss. *Doelling et al.* [17] give the argumentation that there exists a strong correlation between components wear and entropy flow.

*Ling et al.* [13] give an experimental description of a correlation between wear and entropy flow in lubricated sliding systems.

"Sliding wear is an irreversible degradation of surfaces induced by friction. On a microscopic scale irreversible physical interactions between the sliding surfaces – including plastic deformation of asperities, fracture, delamination, abrasive plowing, and corrosive wear, among others – creates friction resistance forces, dissipates power, and generates irreversible entropy. Since the physical interactions responsible for friction and wear monotonically produce entropy, entropy becomes a time base for wear" [13].

"The entropy production, in fact, enable one to bridge the atomic scale phenomena with the macro scale response" [14] in [13]. "In addition friction and wear, from the vantage point of thermodynamics irreversible transform mechanical energy into other forms through dissipa‐ tive processes. Therefore, entropy production is believed to be a propitious measure for a systematic study of wear and friction" [15].

All these investigations pointed out that the application of irreversible thermodynamics is a promising tool to analyse the tribological processes.

#### *5.2.2. Entropy and structural degradation*

The aim of this chapter is a description of relation between energetic situation of the tribosystem and the degradation of grease structure. Tribo-sytems as solid surface 1, solid surface 2 and lubricant are investigated but also subsystems as lubricant layer 1 against lubricant layer 2.

Source of irreversible processes are thermodynamic forces *Xi* (*i* =1, 2...) (gradient of tempera‐ ture, gradient of concentration...). These forces Xi evoke corresponding flows *Ii* (*i* =1, 2, ...)(heat flow, diffusion flow...) The generalisation of classical thermodynamic to describe irreversible effects leads to an investigation of local equilibrium. That means there exist macroscopic small system areas that are provided in equilibrium while the whole system is out of equilibrium. Two principles of thermodynamic are used: the linear dependence of flows and thermody‐ namic forces, and the *Onsager-reciprocity* [20].

For the entropy generation can be written

$$\frac{dS}{dt} = \sum\_{i} I\_{i} \cdot \mathbf{X}\_{i}$$

The variation of entropy is influenced by two terms

$$dS = dS\_{out} + dS\_{in} \tag{10}$$

The transport processes modeled from the investigated tribo-system are presented in Fig. 19,

Friction and Wear of a Grease Lubricated Contact — An Energetic Approach

http://dx.doi.org/10.5772/55837

265

As a consequence of the modelled mechanisms for the entropy production can be written as

**Figure 18.** Tribo-subsystem (grease layer against grease layer) inside the general tribo-system

heat flow by conduction from layer to layer; V - velocities of the observed layers

**Figure 19.** Observed contact situation (section of a complete tribological system). Solid rubbing body and some mod‐ elled grease layers. Heat transport: Q1 - heat amount from solid friction transported into the solid material and into the grease layer; Q2 - heat amount from liquid friction flows into layer 1 and 2; Q3 - same as Q2; *QCOND* presents the

20 and 21.

Heat transfer across the boundary of the modelled system (subsystem) delivers the entropy *dSout*. Entropy related to mechanisms taking place inside the system is described with *dSin* (entropy production). The intention is to relate the entropy production inside the system with different irreversible effects caused by the friction process within the grease layer. It can be written

$$
\rho \cdot \frac{dS}{dt} + div\sigma = \Theta \tag{11}
$$

with *σ* - entropy flow density (*σ* =ℑ / *T* ; ℑ =heat flow density) and *Θ* - local entropy increase per unit time (*Θ* = −(ℑ, *gradT* )/ *T* 2) [20]

Written in differential form and related to time

$$\frac{dS}{dt} = \frac{dS\_{in}}{dt} - \frac{dQ\_{1-2}}{dt} + s\_e \cdot \frac{dm\_e}{dt} - s\_a \cdot \frac{dm\_a}{dt} \tag{12}$$

with *se* ⋅ *dme dt* - entropy transported into the system with mass transport; *sa* <sup>⋅</sup> *dma dt* - entropy transported out of the system by mass loss. *Sin* describes the entropy production inside the system and ( ± )*SQ*1−2 leads to a change of system entropy by heat transfer across the system boundaries.

Eq. (12) describes the entropy balance for an open thermodynamic system (with exchange of mass). As Abdel-Aal [18] pointed out only the entropy source strength, namely entropy created in the system, should be used as a basis for systematic description of the irreversible process (degradation of materials).

The process of structural degradation can be described as a process of energy accumulation, energy dissipation and transition of critical energy levels. Each of these mechanisms delivers a contribution to an entropy balance and will change the production term in Eq.(10) and (12). Entropy production (for the observed volume element) is determined by fluid friction and its effects. The process of solid friction (for a mixed friction contact) delivers only a heat portion into the modelled system. The tribo-subsystem can be illustrated with Fig.18.

The transport processes modeled from the investigated tribo-system are presented in Fig. 19, 20 and 21.

For the entropy generation can be written

264 Tribology - Fundamentals and Advancements

per unit time (*Θ* = −(ℑ, *gradT* )/ *T* 2) [20]

Written in differential form and related to time

The variation of entropy is influenced by two terms

*out in dS dS dS* = + (10)

× + =Q (11)


*dma*

*dt* - entropy

Heat transfer across the boundary of the modelled system (subsystem) delivers the entropy *dSout*. Entropy related to mechanisms taking place inside the system is described with *dSin* (entropy production). The intention is to relate the entropy production inside the system with different irreversible effects caused by the friction process within the grease layer. It can be

> *dS div dt* r

 s

with *σ* - entropy flow density (*σ* =ℑ / *T* ; ℑ =heat flow density) and *Θ* - local entropy increase

*in* 1 2 *e a e a*

*s s*

*dS dS dQ dm dm*

*dt dt dt dt dt*

*dt* - entropy transported into the system with mass transport; *sa* <sup>⋅</sup>

transported out of the system by mass loss. *Sin* describes the entropy production inside the system and ( ± )*SQ*1−2 leads to a change of system entropy by heat transfer across the system

Eq. (12) describes the entropy balance for an open thermodynamic system (with exchange of mass). As Abdel-Aal [18] pointed out only the entropy source strength, namely entropy created in the system, should be used as a basis for systematic description of the irreversible process

The process of structural degradation can be described as a process of energy accumulation, energy dissipation and transition of critical energy levels. Each of these mechanisms delivers a contribution to an entropy balance and will change the production term in Eq.(10) and (12). Entropy production (for the observed volume element) is determined by fluid friction and its effects. The process of solid friction (for a mixed friction contact) delivers only a heat portion

into the modelled system. The tribo-subsystem can be illustrated with Fig.18.

*dS dt* <sup>=</sup>∑ *i*

written

with *se* ⋅

boundaries.

*dme*

(degradation of materials).

*Ii* ⋅ *Xi*

As a consequence of the modelled mechanisms for the entropy production can be written as

**Figure 18.** Tribo-subsystem (grease layer against grease layer) inside the general tribo-system

**Figure 19.** Observed contact situation (section of a complete tribological system). Solid rubbing body and some mod‐ elled grease layers. Heat transport: Q1 - heat amount from solid friction transported into the solid material and into the grease layer; Q2 - heat amount from liquid friction flows into layer 1 and 2; Q3 - same as Q2; *QCOND* presents the heat flow by conduction from layer to layer; V - velocities of the observed layers

An assumption is made that heat flow gets a (−) that means for a balance that heat leaves the

*def R acc*

= (15)

Friction and Wear of a Grease Lubricated Contact — An Energetic Approach

http://dx.doi.org/10.5772/55837

267

<sup>×</sup> <sup>=</sup> (16)

*e V*

*T* × × V

*acc*

with *edef* the energy density used for deformation process, *ςR* describes the part of friction energy which is accumulated, *Vacc* the accumulation volume and *Tacc* the temperature of

2 '

g

cos *critic*

from oscillating rheometer measurements (see Eq.(7)). The temperature situation is assumed as *Tsolid* >*Tlayer*<sup>1</sup> >*Tlayer*2. Furthermore it is assumed that different temperature appears for

deformation) enters the first modelled grease layer. Part of this thermal load will be conducted into the next lubricant layer. Liquid friction between the modelled layers leads to a transport

An interesting description of the entropy production term for different processes at sliding

To link the energetic situation with the structural degradation the entropy production term has to be simplified. The friction energy *Wf* is observed with the temperature *Tf* . It can be

*f*

*T*

*V*

An interpretation delivers (*ρ<sup>a</sup>* ⋅ *sa*) as an entropy density leaving the system with the mass exchange. It means an increasing energetic release by entropy flow out of the system with the degraded structure leads to an increasing capacity to withstand stresses expressed by *eRheo* \* .

*a*

( ) ( ) \*

*Rrheo f a a e Q*

*e T s SS*

r

1 2


d

*trans <sup>G</sup> <sup>S</sup>*

In general it is conceivable that part of heat generated by solid friction (asperity

*acc*

*S*

observed volume element.

It can be proposed

accumulation process.

different mechanism. It means

*Tacc* ≠*Tdiss* ≠*Ttrans*.

of heat into the layers.

obtained

interfaces comes from [21].

**Figure 20.** Imagine of transport of thickener structure in and out of the observed volume element (mass transport)

**Figure 21.** Modelled entropy transport processes inside the observed system (entropy flow) [22]

As a consequence of the modelled mechanisms for the entropy production can be Written with

$$\frac{dS\_{\text{in}}}{dt} = \frac{dS\_{\text{acc}}}{dt} + \frac{dS\_{\text{diss}}}{dt} + \frac{dS\_{\text{trans}}}{dt} \tag{13}$$

It means the process of energy accumulation the process of energy dissipation and the transition of an critical energy level produce entropy. Any chemical potential is disregarded in this investigation. We may rewrite Eq.(12)

$$\frac{dS}{dt} = \left(\frac{dS\_{\text{acc}}}{dt} + \frac{dS\_{\text{diss}}}{dt} + \frac{dS\_{\text{trans}}}{dt}\right) - \frac{dQ\_{1-2}}{dt} + s\_e \cdot \frac{dm\_e}{dt} - s\_a \cdot \frac{dm\_a}{dt} \tag{14}$$

An assumption is made that heat flow gets a (−) that means for a balance that heat leaves the observed volume element.

It can be proposed

$$\mathcal{S}\_{\rm acc} = \frac{\mathcal{e}\_{\rm def} \cdot \mathcal{g}\_{\mathcal{R}} \cdot V\_{\rm acc}}{T\_{\rm acc}} \tag{15}$$

with *edef* the energy density used for deformation process, *ςR* describes the part of friction energy which is accumulated, *Vacc* the accumulation volume and *Tacc* the temperature of accumulation process.

$$\mathcal{S}\_{trans} = \frac{G \cdot \chi^2\_{critical}}{\cos \delta} \tag{16}$$

from oscillating rheometer measurements (see Eq.(7)). The temperature situation is assumed as *Tsolid* >*Tlayer*<sup>1</sup> >*Tlayer*2. Furthermore it is assumed that different temperature appears for different mechanism. It means

$$T\_{acc} \# T\_{diss} \# T\_{trans} \cdot$$

**Figure 20.** Imagine of transport of thickener structure in and out of the observed volume element (mass transport)

**Figure 21.** Modelled entropy transport processes inside the observed system (entropy flow) [22]

Written with

266 Tribology - Fundamentals and Advancements

in this investigation. We may rewrite Eq.(12)

As a consequence of the modelled mechanisms for the entropy production can be

*in acc diss trans dS dS dS dS dt dt dt dt*

It means the process of energy accumulation the process of energy dissipation and the transition of an critical energy level produce entropy. Any chemical potential is disregarded

*acc diss trans* 1 2 *e a*

*dS dS dS dS dQ dm dm*

= + + - +× -× ç ÷


*dt dt dt dt dt dt dt*

=+ + (13)

*e a*

*s s*

è ø (14)

In general it is conceivable that part of heat generated by solid friction (asperity

deformation) enters the first modelled grease layer. Part of this thermal load will be conducted into the next lubricant layer. Liquid friction between the modelled layers leads to a transport of heat into the layers.

An interesting description of the entropy production term for different processes at sliding interfaces comes from [21].

To link the energetic situation with the structural degradation the entropy production term has to be simplified. The friction energy *Wf* is observed with the temperature *Tf* . It can be obtained

$$\boldsymbol{e}\_{R\text{rebo}}^{\*} = \boldsymbol{T}\_{f} \cdot \left(\boldsymbol{\rho}\_{a} \cdot \boldsymbol{s}\_{a}\right) - \frac{\boldsymbol{T}\_{f}}{\boldsymbol{V}\_{a}} \left(\boldsymbol{\mathcal{S}}\_{e} - \boldsymbol{\mathcal{S}}\_{Q1-2}\right) \tag{17}$$

An interpretation delivers (*ρ<sup>a</sup>* ⋅ *sa*) as an entropy density leaving the system with the mass exchange. It means an increasing energetic release by entropy flow out of the system with the degraded structure leads to an increasing capacity to withstand stresses expressed by *eRheo* \* .

To use some experimental results some assumption were made. The specific entropy can be determined by

$$\mathbf{s} = \mathbf{c} \cdot \ln \left( \frac{T\_{str}}{T\_{unstr}} \right) \tag{18}$$

**•** A significant energetic release by the transport of degraded structure out of the system can

*prod Q*1 2 *e dS S S ms*

**•** No significant energetic release by the transport of degraded structure out of the system can

Some new definitions of general tribological subjects are made. A proposal for an energy balance of a grease lubricated contact is given and empirical proposals for quantification are presented. With the help of rheometer tests the fluid friction was investigated and results are illustrated. The degradation process of the grease structure is described with energetic parameters. An open thermodynamic system is created and described. The influence of the

Department of Mechanical Engineering and Production, Hamburg University of Applied

[1] Kuhn,E.: Tribology of lubricating greases. An energetical approach of the tribological

[2] Fleischer,G.: Criterions of mixed friction (in German). Tribologie und Schmierung‐

**•** An additional entropy source happens (for example the mentioned friction wave).

energy flow on the degradation process is presented too.

Address all correspondence to: erik.kuhn@haw-hamburg.de

process (in German). expert verlag, 2009

stechnik, Berlin Technik Verlag 1985

*dt* - = + +× & & & (20)

Friction and Wear of a Grease Lubricated Contact — An Energetic Approach

http://dx.doi.org/10.5772/55837

269

be observed.

be observed.

**6. Conclusions**

**Author details**

Sciences, Germany

**References**

Erik Kuhn

*dt* <sup>&</sup>lt;*S*˙ *prod* <sup>+</sup> *<sup>S</sup>*˙ *<sup>Q</sup>*1−<sup>2</sup> <sup>+</sup> *<sup>m</sup>*˙ <sup>⋅</sup> *se*

*dS*

with index *str* for the stressed layer and *unstr* for the unstressed layer. Test temperature in the rheometer was used for *Tunstr*. An increasing temperature with an increasing soap content during the friction process was assumed because all measurements show a direct correlation between soap content and fluid friction (see Fig. 10). To link the specific entropy leaving the system with the degradation process the parameter *eRheo* \* for the crossing point (rheometer tests) is used.

Fig. 22 presents the tendency of *eRheo* \* vs. *s*.

**Figure 22.** Correlation between *eRheo* \* (the crossing point) and specific entropy

Fig. 22 presents the assumed correlation between energetic stress and energetic release. Finally some conclusions were made

$$\frac{dS}{dt} \succ \dot{S}\_{prod} + \dot{S}\_{Q1-2} + \dot{m} \cdot \mathbf{s}\_e \tag{19}$$

**•** A significant energetic release by the transport of degraded structure out of the system can be observed.

$$\frac{dS}{dt} = \dot{S}\_{prod} + \dot{S}\_{Q1-2} + \dot{m} \cdot s\_e \tag{20}$$

**•** No significant energetic release by the transport of degraded structure out of the system can be observed.

*dS dt* <sup>&</sup>lt;*S*˙ *prod* <sup>+</sup> *<sup>S</sup>*˙ *<sup>Q</sup>*1−<sup>2</sup> <sup>+</sup> *<sup>m</sup>*˙ <sup>⋅</sup> *se*

(18)

\* for the crossing point (rheometer tests)

To use some experimental results some assumption were made. The specific entropy can be

ln *str unstr T*

\* (the crossing point) and specific entropy

Fig. 22 presents the assumed correlation between energetic stress and energetic release.

*prod Q*1 2 *e dS S S ms*

*dt* - > + +× & & & (19)

*T* æ ö = × ç ÷ è ø

with index *str* for the stressed layer and *unstr* for the unstressed layer. Test temperature in the rheometer was used for *Tunstr*. An increasing temperature with an increasing soap content during the friction process was assumed because all measurements show a direct correlation between soap content and fluid friction (see Fig. 10). To link the specific entropy leaving the

*s c*

system with the degradation process the parameter *eRheo*

Fig. 22 presents the tendency of *eRheo* \* vs. *s*.

**Figure 22.** Correlation between *eRheo*

Finally some conclusions were made

determined by

268 Tribology - Fundamentals and Advancements

is used.

**•** An additional entropy source happens (for example the mentioned friction wave).

#### **6. Conclusions**

Some new definitions of general tribological subjects are made. A proposal for an energy balance of a grease lubricated contact is given and empirical proposals for quantification are presented. With the help of rheometer tests the fluid friction was investigated and results are illustrated. The degradation process of the grease structure is described with energetic parameters. An open thermodynamic system is created and described. The influence of the energy flow on the degradation process is presented too.

#### **Author details**

Erik Kuhn

Address all correspondence to: erik.kuhn@haw-hamburg.de

Department of Mechanical Engineering and Production, Hamburg University of Applied Sciences, Germany

#### **References**


[3] Kuhn,E.: Investigation of the Structural Degradation of Lubricating Greases due to Tribological Stress. Intern. Colloquium Tribology Esslingen, 2012.

[20] Bryant, M.D.: Entropy and dissipative processes of friction and wear. SERBIATRIB

Friction and Wear of a Grease Lubricated Contact — An Energetic Approach

http://dx.doi.org/10.5772/55837

271

[21] Kuhn,E.: Experimental investigations of the structural degradation of lubricating

09, *11th Intern. Conf. on Tribology* 13.5.-15.5.2009, pp.3-8.

greases. GfT-conference Göttingen 2012


[20] Bryant, M.D.: Entropy and dissipative processes of friction and wear. SERBIATRIB 09, *11th Intern. Conf. on Tribology* 13.5.-15.5.2009, pp.3-8.

[3] Kuhn,E.: Investigation of the Structural Degradation of Lubricating Greases due to

[4] Fleischer,G.: Scientific findingy by Tross from a contemprorary point of view. (in

[5] Kuhn,E.: Inherent tribo-system response to optimise the process conditions. 8th Ar‐

[6] Kuhn,E.: Tribological stress and structural behaviour of lubricating greases. ECO‐

[7] Kuhn,E.: Energetics of the time dependent flow behaviour of greases. Applied Rheol‐

[9] Delgado,M.A.: Manufacture and flow process of lubricating greases. PhD thesis.*Uni‐*

[10] Åström, H.: Grease in elastohydrodynamic lubrication. PhD thesis, *Lulea University*,

[11] Franco, J.M. ; Delgado,M.A. and Valencia,C.: Combined oxidative-shear resistance of castor oil-based lubricating greases. *3rd Arnold Tross Colloquium. Hamburg 2007*. Proc.

[12] Dierich,P.: Modelling the influence of roughnes on the wear prediction. Habil-thesis.

[13] Ling,F.F.; Bryant,M.D. ang Doelling,K.L.: On irreversible thermodynamics for wear

[14] Buldum,A.; Ciraci,S.: Atomic-scale study of dry sliding friction. Phys. Rev. B55(4)

[15] Aghdam,A.B., Khonsari,M.M.: On the relation between wear and entropy in dry slid‐

[16] Doelling, K.L.; Ling, L.L.; Bryant, M.D. and Heilman, B.P.: An experimental study of the correlation between wear and entropy flow in machinery components. J*ournal of*

[17] Abdel-Aal, H.A.: On the interdependence between kinetics of friction- released ther‐ mal energy and the transition in wear mechanisms during sliding of metallic pairs.

[18] Abdel-Aal, H.A.: Wear and irreversible entropy generation in dry sliding.*Annals Du‐*

[8] Czarny,R.: Lubricating greases. *WNT Publisher,* Warsaw, 2004 (in Polish)

Tribological Stress. Intern. Colloquium Tribology Esslingen, 2012.

German) 3rd Arnold Tross Colloquium, Hamburg 2007

nold Tross Colloquium, Hamburg 2012

TRIB 7.-9.6.2011 Vienna

270 Tribology - Fundamentals and Advancements

ogy June 1997, p.118-122

*versity of Huelva,* Spain, 2005

(in German), Zittai 1986

(1997) 2606-2611.

prediction. Wear 253(2002)1165-1172

ing contact. Wear 270(2011) 781-790

*narea de Jos of Galati, Fascicle*,VIII,pp.34-44

[19] Basarov, I.P.: Thermodynamic. *D.V.W.,* Berlin 1964 (in German)

*Applies Physics .* 88(2000)5

*Wear* 253, (2002), pp. 11-12

1993

pp.18-59

[21] Kuhn,E.: Experimental investigations of the structural degradation of lubricating greases. GfT-conference Göttingen 2012

**Section 4**

**Sustainability of Tribosystems**

**Sustainability of Tribosystems**

**Chapter 10**

**Innovative "Green" Tribological Solutions for Clean**

Since its invention in the last quarter of the nineteenth century and during all the twentieth century, two-stroke engines penetrated in many industrial, automotive and handheld appli‐ cations where engines with high specific power, simple design, light overall weight and low cost are required. Presently, two-stroke engines are commonly used in motorcycles, scooters, chainsaw, agricultural machinery, railways grinding machines, outboard applications, etc. Usually, the moving parts of a two-stroke motor are lubricated either by using mixture of oil with fuel or by pumping oil from a separate tank. Both designs use total-loss lubrication method, with the oil being burnt in the combustion chamber. Therefore, the lubricating oil must meet specific requirements: it must have an optimal balance of light and heavy oil components to lubricate at high temperature, it must produce no deposits (carbon sooty and other) on moving parts, and it should be ash-less. In addition, the oil should provide good protection of moving parts at high speed under deceleration of engine with the throttle closed,

Also, two-stroke engine produce more contaminants than four-stroke engines, due to oil burning in the combustion chamber. Therefore, it is very important to reduce these contami‐

Most challenging issue of the European technological strategy resides in complete substitu‐ tion of fossil-based fuels and lubricating oils with renewable eco-friendly and high perform‐ ance materials. Esters and polyglycols were identified as alternative base oils because of their high biodegradability, low toxicity; low ash formation and absence of polymer compo‐

> © 2013 Fernández-Pérez et al.; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2013 Fernández-Pérez et al.; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,

distribution, and reproduction in any medium, provided the original work is properly cited.

**Small Engines**

http://dx.doi.org/10.5772/55836

**1. Introduction**

Xana Fernández-Pérez, Amaya Igartua,

Roman Nevshupa, Patricio Zabala, Borja Zabala, Rolf Luther, Flavia Gili and Claudio Genovesio

Additional information is available at the end of the chapter

when the engine usually suffers from oil-starvation.

nations to meet ecological requirements.

### **Innovative "Green" Tribological Solutions for Clean Small Engines**

Xana Fernández-Pérez, Amaya Igartua, Roman Nevshupa, Patricio Zabala, Borja Zabala, Rolf Luther, Flavia Gili and Claudio Genovesio

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/55836

#### **1. Introduction**

Since its invention in the last quarter of the nineteenth century and during all the twentieth century, two-stroke engines penetrated in many industrial, automotive and handheld appli‐ cations where engines with high specific power, simple design, light overall weight and low cost are required. Presently, two-stroke engines are commonly used in motorcycles, scooters, chainsaw, agricultural machinery, railways grinding machines, outboard applications, etc.

Usually, the moving parts of a two-stroke motor are lubricated either by using mixture of oil with fuel or by pumping oil from a separate tank. Both designs use total-loss lubrication method, with the oil being burnt in the combustion chamber. Therefore, the lubricating oil must meet specific requirements: it must have an optimal balance of light and heavy oil components to lubricate at high temperature, it must produce no deposits (carbon sooty and other) on moving parts, and it should be ash-less. In addition, the oil should provide good protection of moving parts at high speed under deceleration of engine with the throttle closed, when the engine usually suffers from oil-starvation.

Also, two-stroke engine produce more contaminants than four-stroke engines, due to oil burning in the combustion chamber. Therefore, it is very important to reduce these contami‐ nations to meet ecological requirements.

Most challenging issue of the European technological strategy resides in complete substitu‐ tion of fossil-based fuels and lubricating oils with renewable eco-friendly and high perform‐ ance materials. Esters and polyglycols were identified as alternative base oils because of their high biodegradability, low toxicity; low ash formation and absence of polymer compo‐

© 2013 Fernández-Pérez et al.; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Fernández-Pérez et al.; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

nents, in [1]. Synthetic esters are characterized by their polar structure, high wear resist‐ ance, good viscosity-temperature behaviour, miscibility with non-fossil fuels. Esther-base oils can be blended with various components like antifoam agents, oxidation inhibitors, pourpoint depressants, antirust agents, detergents, anti-wear agents, friction reducers, viscosity index improvers, etc., to create environmental friendly prototype engine oils and to meet the changing environmental requirements in low sulphur fuels and other alternative fuels and their application to engine oils.

of these prototype oils and selection of the oil with the best tribological performance (SEMO 10), a new improved formulation was developed based on the selected lubricating oil, designated SEMO 36. In addition, conventional mineral oil for two-stroke engines was used as reference oil. The additive package composition of the reference oil is different but it is ash-

Oil viscosity was characterized according to ASTM D-445-06 standard procedure in [3], and

Deposit forming tendency of the oils was characterized by the Coker test at 250 °C during 12 h. Some physical and rheological properties of the lubricating oils are shown in Table 1. Among the prototype lubricating oils, SEMO 10 has the lowest viscosity both at 40 and 100 °C, the

Unleaded petrol (E228) and bioethanol E85 (mixture of 85% of ethanol with 15% of gasoline) were selected to test miscibility of the lubricating oils with standard and alternative fuels. For this purpose two different lubricant/fuel ratios were used. Regarding to the miscibility method A (90% lubricant in fuel), SEMO 10 as well as SEMO 5 demonstrated good miscibility both with unleaded petrol and E85. Compared to this, the results for the 2% mixtures according (method B) differed. All tested lubricants proved to be perfectly miscible with EN228 fuel, whereas only SEMO 36 demonstrated to be fully miscible with E85. According to both miscibility methods the reference oil was only miscible with EN228. SEMO 36, when compared to its original prototype SEMO 10, has a much higher viscosity. Flash point for this lubricant

**Ref. Oil SEMO 4 SEMO 5 SEMO 10 SEMO 36**

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

277

EN228 Good Good Good Good not tested E85 Poor Poor Good Good Good

EN228 Good Good Good Good Good E85 Poor Poor Poor Poor Good

free as well as the other SEMO oils.

viscosity index was determined using ASTM D-2270-04 in [4].

highest flash point and the lowest deposit forming tendency.

is lower than for SEMO 10 but still higher than 200 °C.

Miscibility Method A (90%

Miscibility Method B (2% lubricant in fuel)

**Table 1.** Properties of the engine oils

lubricant in fuel)

\*Rating on base 10

Density, g/ml 0.877 0.915 0.917 0.935 0.999 Viscosity @ 40ºC,mm*<sup>2</sup>*/s 59.5 84.9 94 45.8 113.3 Viscosity @ 100ºC, mm*<sup>2</sup>*/s 8.6 12.5 13.2 8.0 18.3 Viscosity index 117 144 140 147 181 Flash point, ºC 120 204 190 260 218

Pour Point, ºC -21 -39 -33 -39 not tested Deposit forming \* 9 4 3 9 not tested

Low metal additives content and clean-burnt characteristics result in less engine fouling with much reduced ring stick and lower levels of dirt built-up on ring grooves, skirts and under crowns. Owing to the presence of polar ester groups in the molecule which have higher adhesion to metal surface, esters have much better lubricity than hydrocarbons. The perform‐ ance of the ester-based lubricating oils can be further improved by selecting a proper base oil and additive package.

Another important problem is related with performance of fuel injector system when bio fuels are used. Diesel injection nozzles consist of a body (usually in Ni-Cr steel) and needle valve (High speed steel, HSS), fitted together with very strict tolerances. The design of the nozzle, i.e., the number of orifices, their diameters, positions and drilling angles depend on specific engine application. The current trend is to use multi-hole nozzles with very small holes with diameter of only 0.10 - 0.14 mm in order to improve the fuel atomization and flow pattern.

Heat treatments are applied to the body and the needle to obtain the necessary hardness both on the surface and in the core of the parts and to face the following problems:


Adequate finishing of the orifice surfaces is very important also to optimize the erosion resistance.

The usage of new diesel blends characterized by different physical and chemical properties as compared with the traditional fuels could lead to modifications both in the choice of materials, geometry and positioning of orifices or their surface finishing to ensure the correct spray pattern. This work describes the results of our recent studies aimed at solving the problems related to the introduction of new eco-friendly oils and lubricants.

#### **2. New prototype engine oils**

#### **2.1. Oil characterization**

Three different synthetic ester base oils have been selected to formulate three prototype engine oils with the same additive composition. These oils are different mixtures of fully saturated polyglicol-ester and mono-ester types and designated as SEMO 4, SEMO 5 and SEMO 10. Same additive package has then been added to the three bases. After comparative characterization of these prototype oils and selection of the oil with the best tribological performance (SEMO 10), a new improved formulation was developed based on the selected lubricating oil, designated SEMO 36. In addition, conventional mineral oil for two-stroke engines was used as reference oil. The additive package composition of the reference oil is different but it is ashfree as well as the other SEMO oils.

Oil viscosity was characterized according to ASTM D-445-06 standard procedure in [3], and viscosity index was determined using ASTM D-2270-04 in [4].

Deposit forming tendency of the oils was characterized by the Coker test at 250 °C during 12 h. Some physical and rheological properties of the lubricating oils are shown in Table 1. Among the prototype lubricating oils, SEMO 10 has the lowest viscosity both at 40 and 100 °C, the highest flash point and the lowest deposit forming tendency.

Unleaded petrol (E228) and bioethanol E85 (mixture of 85% of ethanol with 15% of gasoline) were selected to test miscibility of the lubricating oils with standard and alternative fuels. For this purpose two different lubricant/fuel ratios were used. Regarding to the miscibility method A (90% lubricant in fuel), SEMO 10 as well as SEMO 5 demonstrated good miscibility both with unleaded petrol and E85. Compared to this, the results for the 2% mixtures according (method B) differed. All tested lubricants proved to be perfectly miscible with EN228 fuel, whereas only SEMO 36 demonstrated to be fully miscible with E85. According to both miscibility methods the reference oil was only miscible with EN228. SEMO 36, when compared to its original prototype SEMO 10, has a much higher viscosity. Flash point for this lubricant is lower than for SEMO 10 but still higher than 200 °C.


\*Rating on base 10

nents, in [1]. Synthetic esters are characterized by their polar structure, high wear resist‐ ance, good viscosity-temperature behaviour, miscibility with non-fossil fuels. Esther-base oils can be blended with various components like antifoam agents, oxidation inhibitors, pourpoint depressants, antirust agents, detergents, anti-wear agents, friction reducers, viscosity index improvers, etc., to create environmental friendly prototype engine oils and to meet the changing environmental requirements in low sulphur fuels and other alternative fuels and

Low metal additives content and clean-burnt characteristics result in less engine fouling with much reduced ring stick and lower levels of dirt built-up on ring grooves, skirts and under crowns. Owing to the presence of polar ester groups in the molecule which have higher adhesion to metal surface, esters have much better lubricity than hydrocarbons. The perform‐ ance of the ester-based lubricating oils can be further improved by selecting a proper base oil

Another important problem is related with performance of fuel injector system when bio fuels are used. Diesel injection nozzles consist of a body (usually in Ni-Cr steel) and needle valve (High speed steel, HSS), fitted together with very strict tolerances. The design of the nozzle, i.e., the number of orifices, their diameters, positions and drilling angles depend on specific engine application. The current trend is to use multi-hole nozzles with very small holes with diameter of only 0.10 - 0.14 mm in order to improve the fuel atomization and flow pattern.

Heat treatments are applied to the body and the needle to obtain the necessary hardness both

**•** fatigue failures at high stress areas due to repeated pulses of very high injection pressures;

Adequate finishing of the orifice surfaces is very important also to optimize the erosion

The usage of new diesel blends characterized by different physical and chemical properties as compared with the traditional fuels could lead to modifications both in the choice of materials, geometry and positioning of orifices or their surface finishing to ensure the correct spray pattern. This work describes the results of our recent studies aimed at solving the problems

Three different synthetic ester base oils have been selected to formulate three prototype engine oils with the same additive composition. These oils are different mixtures of fully saturated polyglicol-ester and mono-ester types and designated as SEMO 4, SEMO 5 and SEMO 10. Same additive package has then been added to the three bases. After comparative characterization

on the surface and in the core of the parts and to face the following problems:

related to the introduction of new eco-friendly oils and lubricants.

their application to engine oils.

276 Tribology - Fundamentals and Advancements

and additive package.

**•** thermal shocks.

**2. New prototype engine oils**

**2.1. Oil characterization**

resistance.

**Table 1.** Properties of the engine oils

Wettability of the surface of the cylinder liner by lubricating oil is important for corrosion- and wear-protection of the piston rings and cylinder liner at the start-up when the temperature of the components is low. In this work, the wetting characteristic of the tested oils was determined using the Sessile Drop method. The resulting contact angles of the drops of various oils on the honed surface of the cylinder liner are shown in Table 2. Same method could not be used to determine wettability of the piston ring because of the small width of the ring. Therefore, the following procedure for qualitative comparison of the wettability of the piston ring by different oils in [12] was applied: 1 µl of oil was placed on the circular flat surface of the phosphate cast iron piston ring and then, after 30 s, the extension of the oil drop along this surface was measured.

**Time (days) SEMO 4 SEMO 5 SEMO 10 REF** 29.4 41.3 25.2 27.6 36.2 68.4 53.5 38.7 52.0 79.6 69.6 33.4 61.1 81.2 75.7 51.2 Ultimate > 60% > 60% > 60% < 60%

> **Classification OECD 201**

SEMO 4 >100 not harmful \* >1000 not harmful \* SEMO 5 >100 not harmful \* >1000 not harmful \* SEMO 10 >100 not harmful \* >1000 not harmful \* SEMO 36 - - >1000 not harmful \*

EC50/EL50 is that concentration of test substance which results in a 50% reduction in either growth or growth rate

Tribological evaluation of lubricating oils was done using ball-on-disk configuration with reciprocating motion according to the standard procedure DIN 51834-2 in [9]. Ball and disk were made of 100Cr6 steel. The ball, 10 mm in diameter, performed reciprocating motion with a stroke of 1 mm and a friction frequency 50 Hz. Normal load was 50 N during short run-in period 45 s and 300 N during the test 60 min. The ball and the disk were immersed in the lubricating oil, which temperature during the test was constant and 50 °C. Friction force was measured as function of time. Friction coefficient was calculated as the ratio of the tangential

After test completion, diameter of the wear scar on the ball was measured using optical microscope, and, from this data, volume wear of the ball was calculated for each lubricating

Evolution of the friction coefficient in friction evaluation tests is shown in Figure 1. Oils with low additive content: SEMO 4, SEMO 5 and SEMO 10 showed an interval of frictional insta‐ bility after the run-in period. In the instability period, which lasted from 400 up to 800 s, there are some sharp peaks indicating damage of surface and seizure, probably due to microwelding. The reference lubricating oil had a less pronounced instability period without sharp

**EC50/EL50**

**with Daphnia Magna, mg/l**

Innovative "Green" Tribological Solutions for Clean Small Engines

**Classification OECD 202**

http://dx.doi.org/10.5772/55836

279

**Table 3.** Biodegradability of oils (% of biodegraded oil) in [12].

**with Alga, mg/l**

**Oil EC50/EL50**

\*With respect to aquatic organisms.

force to the normal force.

oils tested.

**Table 4.** Results of the toxicity tests in [12].

**2.2. Tribological evaluation according to DIN 51834-2**

relative to the control.


**Table 2.** Contact angle and oil spread distance

The contact angle for SEMO 36 oil on the honed cast iron was the highest among all the tested lubricating oils. The contact angles of SEMO 5 and SEMO 4 were very similar one to each other and only slightly lower than for SEMO 36. SEMO 10 had the lowest contact angle and the largest drop spread for all tested oils. The behaviour of the drop spread of the tested lubricating oils over the piston ring surface is similar to that of the contact angle, bearing in mind that large contact angle values correspond to small spread distances.

Biodegradability and toxicity of the lubricating oils were examined according to the recom‐ mendations of the Organization for Economic Co-operation and Development (OECD) in [5]. Biodegradability of lubricating oils was tested using OECD 301F Manometric Respirometry Method consisting of the measurement of oxygen uptake by a stirred solution of the test substance in a mineral medium, inoculated with micro-organisms in [6]. Toxicity of the lubricating oils was studied using "Alga, Growth Inhibition Test" OECD 201 in [7] and "Daph‐ nia Magna" 24 h Acute Immobilisation Test OECD 202 in [8]. In the "Alga, Growth Inhibition Test", selected green algae were exposed to various concentrations of the test oils over several generations under defined conditions. Results of biodegradability test are shown in Table 3. As expected, all synthetic ester base oils successfully passed the biodegradability test, while the reference mineral oil was not biodegradable according to the standard procedure OECD 301 Biodegradation of SEMO 5 and SEMO 10 exceeded 70%. In toxicity tests both with Alga and Daphnia Magna, the oils were classified as not harmful for aquatic organisms according to the standard procedures OECD 201 and 202 (see Table 4).


**Table 3.** Biodegradability of oils (% of biodegraded oil) in [12].


\*With respect to aquatic organisms.

Wettability of the surface of the cylinder liner by lubricating oil is important for corrosion- and wear-protection of the piston rings and cylinder liner at the start-up when the temperature of the components is low. In this work, the wetting characteristic of the tested oils was determined using the Sessile Drop method. The resulting contact angles of the drops of various oils on the honed surface of the cylinder liner are shown in Table 2. Same method could not be used to determine wettability of the piston ring because of the small width of the ring. Therefore, the following procedure for qualitative comparison of the wettability of the piston ring by different oils in [12] was applied: 1 µl of oil was placed on the circular flat surface of the phosphate cast iron piston ring and then, after 30 s, the extension of the oil drop along this surface was

**Oil Contact angle on honed cast iron, (º) Spread distance of the oil on the**

The contact angle for SEMO 36 oil on the honed cast iron was the highest among all the tested lubricating oils. The contact angles of SEMO 5 and SEMO 4 were very similar one to each other and only slightly lower than for SEMO 36. SEMO 10 had the lowest contact angle and the largest drop spread for all tested oils. The behaviour of the drop spread of the tested lubricating oils over the piston ring surface is similar to that of the contact angle, bearing in mind that

Biodegradability and toxicity of the lubricating oils were examined according to the recom‐ mendations of the Organization for Economic Co-operation and Development (OECD) in [5]. Biodegradability of lubricating oils was tested using OECD 301F Manometric Respirometry Method consisting of the measurement of oxygen uptake by a stirred solution of the test substance in a mineral medium, inoculated with micro-organisms in [6]. Toxicity of the lubricating oils was studied using "Alga, Growth Inhibition Test" OECD 201 in [7] and "Daph‐ nia Magna" 24 h Acute Immobilisation Test OECD 202 in [8]. In the "Alga, Growth Inhibition Test", selected green algae were exposed to various concentrations of the test oils over several generations under defined conditions. Results of biodegradability test are shown in Table 3. As expected, all synthetic ester base oils successfully passed the biodegradability test, while the reference mineral oil was not biodegradable according to the standard procedure OECD 301 Biodegradation of SEMO 5 and SEMO 10 exceeded 70%. In toxicity tests both with Alga and Daphnia Magna, the oils were classified as not harmful for aquatic organisms according

SEMO 4 46.1±3.1 5.33±0.04 SEMO 5 43.4±0.2 5.39±0.05 SEMO 10 33.1±1.1 7.01±0.12 SEMO 36 50.8±0.5 3.78±0.17

large contact angle values correspond to small spread distances.

to the standard procedures OECD 201 and 202 (see Table 4).

**piston ring, (mm)**

measured.

278 Tribology - Fundamentals and Advancements

**Table 2.** Contact angle and oil spread distance

EC50/EL50 is that concentration of test substance which results in a 50% reduction in either growth or growth rate relative to the control.

**Table 4.** Results of the toxicity tests in [12].

#### **2.2. Tribological evaluation according to DIN 51834-2**

Tribological evaluation of lubricating oils was done using ball-on-disk configuration with reciprocating motion according to the standard procedure DIN 51834-2 in [9]. Ball and disk were made of 100Cr6 steel. The ball, 10 mm in diameter, performed reciprocating motion with a stroke of 1 mm and a friction frequency 50 Hz. Normal load was 50 N during short run-in period 45 s and 300 N during the test 60 min. The ball and the disk were immersed in the lubricating oil, which temperature during the test was constant and 50 °C. Friction force was measured as function of time. Friction coefficient was calculated as the ratio of the tangential force to the normal force.

After test completion, diameter of the wear scar on the ball was measured using optical microscope, and, from this data, volume wear of the ball was calculated for each lubricating oils tested.

Evolution of the friction coefficient in friction evaluation tests is shown in Figure 1. Oils with low additive content: SEMO 4, SEMO 5 and SEMO 10 showed an interval of frictional insta‐ bility after the run-in period. In the instability period, which lasted from 400 up to 800 s, there are some sharp peaks indicating damage of surface and seizure, probably due to microwelding. The reference lubricating oil had a less pronounced instability period without sharp peaks, while SEMO 36 did not present any instability. Final values of friction coefficient after 60 min and the diameters of the wear scar on the ball are shown in Table 5.

*<sup>V</sup>* <sup>=</sup> *<sup>π</sup>*

*Ew* <sup>=</sup> *<sup>E</sup>* <sup>∆</sup> *<sup>m</sup>* =

and final time points of friction test time interval.

**Figure 2.** Friction coefficient and wear specific energy in [12].

in [12]:

10 lubricants.

where *R =5 mm* is the radius of the ball and *a* is the radius of the circular wear scar.

Wear specific energy, *Ew*, that is, the ratio of the dissipated energy, *E*, during friction per unit mass of worn material Δ*m*, is an important characteristic which shows the ability of a material to resist wearing. This is a complex parameter taking into account both friction, which characterizes energy supply to the material in the friction zone, and wear intensity. This parameter is considered a very useful tool to compare standard tribological evaluation and simulated tests in [10]. Wear specific energy was determined using the following equation

> *vmF <sup>N</sup> ∫ t i <sup>t</sup> <sup>f</sup> <sup>μ</sup> fr*(*t*)*dt* ∆ *m*

Where *vm* is mean sliding velocity obtained with a reciprocating frequency of 50 Hz and 1 mm stroke, *FN* is the normal load, *μfr* is the friction coefficient, *ti* and *tf* are respectively the initial

In this study, only ball wear was determined as specified by DIN 51834-2. So, the absolute value of wear specific energy could not be determined; since wear of the disk was not meas‐ ured. However, by using the ball mass loss in the denominator of eq. (2), the upper bound estimation of the wear specific energy can be determined. This upper bound can be used for qualitative comparison of anti-wear properties of the lubricating oils under constant friction conditions. These values, determined using eq. (2), are shown in Figure 2. SEMO 36 and the reference oil have much higher values of the wear specific energy, than other oils. Therefore, these lubricating oils improve contacting surfaces wear protection since much larger energy should be dissipated to produce the same wear as compared to SEMO 4, SEMO 5 and SEMO

<sup>3</sup> (*<sup>R</sup>* <sup>3</sup> - (*<sup>R</sup>* <sup>2</sup> <sup>+</sup> *<sup>a</sup>* 2) *<sup>R</sup>* <sup>2</sup> - *<sup>a</sup>* 2) (1)

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

(2)

281

**Figure 1.** Evolution of friction coefficient in time during tribological evaluation tests of the following oils: a) SEMO 4, b) SEMO 5, c) reference oil, d) SEMO 10, e) SEMO 36. Inset in graph e) shows the initial part of the plot together with the curve of the normal load in [12].


**Table 5.** Friction coefficient and wear of the ball in tribological evaluation test of oils (DIN 51834-2) in [12].

The volume of worn material of the ball was estimated geometrically on the basis of the diameter of the wear scar using the following equation (1):

$$V = \frac{\pi}{3} \left( R^{\frac{3}{4}} \text{--} \left( R^{\frac{2}{4}} + a^2 \right) \sqrt{R^{\frac{2}{4}} \text{--} a^2} \right) \tag{1}$$

where *R =5 mm* is the radius of the ball and *a* is the radius of the circular wear scar.

peaks, while SEMO 36 did not present any instability. Final values of friction coefficient after

**Figure 1.** Evolution of friction coefficient in time during tribological evaluation tests of the following oils: a) SEMO 4, b) SEMO 5, c) reference oil, d) SEMO 10, e) SEMO 36. Inset in graph e) shows the initial part of the plot together with

**Oil Final µfr Diameter of wear scar, µm Estimation of worn volume,**

SEMO 4 0.119 ± 3.40\*10-5 884 6.01\*10-3 SEMO 5 0.111 ± 4.34\*10-5 885 6.02\*10-3 REF 0.122 ± 2.73\*10-5 517 6.99\*10-4 SEMO 10 0.125 ± 3.81\*10-5 929 7.32\*10-3 SEMO 36 0.123 ± 1.54\*10-5 459 4.36\*10-4

diameter of the wear scar using the following equation (1):

**Table 5.** Friction coefficient and wear of the ball in tribological evaluation test of oils (DIN 51834-2) in [12].

The volume of worn material of the ball was estimated geometrically on the basis of the

**mm3**

the curve of the normal load in [12].

280 Tribology - Fundamentals and Advancements

60 min and the diameters of the wear scar on the ball are shown in Table 5.

Wear specific energy, *Ew*, that is, the ratio of the dissipated energy, *E*, during friction per unit mass of worn material Δ*m*, is an important characteristic which shows the ability of a material to resist wearing. This is a complex parameter taking into account both friction, which characterizes energy supply to the material in the friction zone, and wear intensity. This parameter is considered a very useful tool to compare standard tribological evaluation and simulated tests in [10]. Wear specific energy was determined using the following equation in [12]:

$$E\_{uv} = \frac{E}{\Delta m} = \frac{v\_m F\_N \int\_{t\_l}^{t\_f} \mu\_{f^c}(t)dt}{\Delta m} \tag{2}$$

Where *vm* is mean sliding velocity obtained with a reciprocating frequency of 50 Hz and 1 mm stroke, *FN* is the normal load, *μfr* is the friction coefficient, *ti* and *tf* are respectively the initial and final time points of friction test time interval.

In this study, only ball wear was determined as specified by DIN 51834-2. So, the absolute value of wear specific energy could not be determined; since wear of the disk was not meas‐ ured. However, by using the ball mass loss in the denominator of eq. (2), the upper bound estimation of the wear specific energy can be determined. This upper bound can be used for qualitative comparison of anti-wear properties of the lubricating oils under constant friction conditions. These values, determined using eq. (2), are shown in Figure 2. SEMO 36 and the reference oil have much higher values of the wear specific energy, than other oils. Therefore, these lubricating oils improve contacting surfaces wear protection since much larger energy should be dissipated to produce the same wear as compared to SEMO 4, SEMO 5 and SEMO 10 lubricants.

**Figure 2.** Friction coefficient and wear specific energy in [12].

#### **2.3. Piston ring/cylinder liner simulation**

Tribological simulation was performed using cast iron phosphated piston ring and cast iron cylinder liner using reciprocating motion configuration. The samples for the tests were cut from real engine parts (Minsel M165 two-stroke engine manufactured by Abamotor Energía) keeping original curved surfaces and surface finishing. The conformal contact between the piston ring and the cylinder counterpart was reproduced by placing a piston ring on a suitable frame, A, and fixing it by means of a clamp, B (Figure 3). Wear of the components was determined by weighting and geometry measurements.

Surface chemical composition of the friction zone of cylinder liner samples was characterized

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

283

Evolution of friction coefficient in time during friction between piston ring segment and a piece of the cylinder liner is shown in Figure 4. It is possible to highlight the increment of the coefficient of friction *μfr* for lubricants SEMO 4 and SEMO 5 overtaking the constant value reached by the lubricant of reference. In fact, for SEMO 4 and SEMO 5, friction coefficient gradually rose during the experiment (90 min) and did not stabilize. The growth behaviour

Initial friction coefficient was about 0.2 and the final one about 0.33 in both cases. SEMO 10 and SEMO 36 showed different behaviour. The initial values were 0.2 and 0.14 for SEMO 10

**Figure 4.** Evolution of friction coefficient in time during piston ring/cylinder liner simulation test. a) SEMO 4, b) SEMO 5, c) reference oil, d) SEMO 10, e) SEMO 36. Inset in the graph e) shows the initial part of the plot together with the

At the beginning, after a run-in period, friction coefficient increased and reached maximum. For SEMO 36 the maximum was reached usually between 100 and 200 s from the beginning of the test, while for SEMO 10 the period of increase was longer and the maximum was reached

using Energy Dispersion X-Ray Spectroscopy (EDS).

was almost linear in time.

curve of the normal load in [12].

and SEMO 36, correspondingly.

A - frame for fixing the piston ring segment, B – fixing clamp, C – base with oil bath for fixing cylinder liner sample.

**Figure 3.** Experimental set-up for piston ring/cylinder liner simulation.

The piston ring segments performed a reciprocating motion with a stroke of 1 mm and a friction frequency 40 Hz. Normal load was 50 N during short run-in period 45 s and 300 N during the test 90 min. During the test, the piston ring segment and the cylinder liner sample were immersed in the oil, which temperature was constant at 200 °C.

The mass change of the piston ring segments and cylinder liner sample was determined from weighting the components before and after friction tests. Since the mass change can be due to two competitive processes: (i) wear out and (ii) deposit formation from the oil at elevated temperature, estimation of wear out by weighting can give erroneous results. Indeed, after the tests the surface colour became yellowish and remained after dissolvent cleaning indicating some sparingly soluble deposits formed on the surface due to some chemical reaction. Therefore, in addition to the determination of the mass change, worn volume was calculated from surface geometry. Surface morphology of the friction zone was studied using white light confocal microscopy at three different zones along the wear track on the cylinder liner sample. The acquired 3D surface images were 0.5 mm wide in the direction of friction and each image contained 138 cross-section profiles of the wear track yielding totally 414 profiles for each sample. Firstly, the cross-section profiles were averaged for each sample and then among different samples tested using the same lubricating oil. Worn volume of the samples of cylinder liner was calculated as a product of a mean cross-section area of the groove and the total length of the groove. The cross-section area was determined by numerical integration of the crosssection profiles and then worn mass was calculated from the worn volume using the density of cast iron.

Surface chemical composition of the friction zone of cylinder liner samples was characterized using Energy Dispersion X-Ray Spectroscopy (EDS).

**2.3. Piston ring/cylinder liner simulation**

282 Tribology - Fundamentals and Advancements

determined by weighting and geometry measurements.

**Figure 3.** Experimental set-up for piston ring/cylinder liner simulation.

of cast iron.

immersed in the oil, which temperature was constant at 200 °C.

Tribological simulation was performed using cast iron phosphated piston ring and cast iron cylinder liner using reciprocating motion configuration. The samples for the tests were cut from real engine parts (Minsel M165 two-stroke engine manufactured by Abamotor Energía) keeping original curved surfaces and surface finishing. The conformal contact between the piston ring and the cylinder counterpart was reproduced by placing a piston ring on a suitable frame, A, and fixing it by means of a clamp, B (Figure 3). Wear of the components was

A - frame for fixing the piston ring segment, B – fixing clamp, C – base with oil bath for fixing cylinder liner sample.

The piston ring segments performed a reciprocating motion with a stroke of 1 mm and a friction frequency 40 Hz. Normal load was 50 N during short run-in period 45 s and 300 N during the test 90 min. During the test, the piston ring segment and the cylinder liner sample were

The mass change of the piston ring segments and cylinder liner sample was determined from weighting the components before and after friction tests. Since the mass change can be due to two competitive processes: (i) wear out and (ii) deposit formation from the oil at elevated temperature, estimation of wear out by weighting can give erroneous results. Indeed, after the tests the surface colour became yellowish and remained after dissolvent cleaning indicating some sparingly soluble deposits formed on the surface due to some chemical reaction. Therefore, in addition to the determination of the mass change, worn volume was calculated from surface geometry. Surface morphology of the friction zone was studied using white light confocal microscopy at three different zones along the wear track on the cylinder liner sample. The acquired 3D surface images were 0.5 mm wide in the direction of friction and each image contained 138 cross-section profiles of the wear track yielding totally 414 profiles for each sample. Firstly, the cross-section profiles were averaged for each sample and then among different samples tested using the same lubricating oil. Worn volume of the samples of cylinder liner was calculated as a product of a mean cross-section area of the groove and the total length of the groove. The cross-section area was determined by numerical integration of the crosssection profiles and then worn mass was calculated from the worn volume using the density

Evolution of friction coefficient in time during friction between piston ring segment and a piece of the cylinder liner is shown in Figure 4. It is possible to highlight the increment of the coefficient of friction *μfr* for lubricants SEMO 4 and SEMO 5 overtaking the constant value reached by the lubricant of reference. In fact, for SEMO 4 and SEMO 5, friction coefficient gradually rose during the experiment (90 min) and did not stabilize. The growth behaviour was almost linear in time.

Initial friction coefficient was about 0.2 and the final one about 0.33 in both cases. SEMO 10 and SEMO 36 showed different behaviour. The initial values were 0.2 and 0.14 for SEMO 10 and SEMO 36, correspondingly.

**Figure 4.** Evolution of friction coefficient in time during piston ring/cylinder liner simulation test. a) SEMO 4, b) SEMO 5, c) reference oil, d) SEMO 10, e) SEMO 36. Inset in the graph e) shows the initial part of the plot together with the curve of the normal load in [12].

At the beginning, after a run-in period, friction coefficient increased and reached maximum. For SEMO 36 the maximum was reached usually between 100 and 200 s from the beginning of the test, while for SEMO 10 the period of increase was longer and the maximum was reached after 700 to 1700 s from the beginning of the test. After reaching the maximum, friction coefficient decreased slowly and stabilized at 0.14 and 0.11 for SEMO 10 and SEMO 36, correspondingly. The friction coefficient of lubricant SEMO 10 showed a slow decline until reaching a constant value lower than the reference one. Friction coefficient for the improved lubricant SEMO 36 levelled out rapidly at a very low value and showed less scatter, probably due to some sort of surface deposition on the contact surfaces.

Figure 6 shows images of the friction zone of the piston ring segments after friction simulation tests with different lubricating oils. Wear and damage of the surface as function of the oil used was similar to that in the cylinder liner. In tests with SEMO 4 and SEMO 5, the material in the friction zone was heavily damaged. The wear can be classified to be of the adhesive type with intensive plastic deformation and edging. When the reference oil and SEMO 10 oil were used in the tests, the damage of the material was less pronounced than for SEMO 4 and SEMO 5, but the wear in all cases was of the adhesive type. Only small damage was observed on the piston ring segments when using SEMO 36. In this case, only summits of the circular grooves of the piston ring presented some wear and deformation. From the point of view of hydrody‐ namic lubrication these results may seem to be surprising, since, with the same additive composition, higher wear rate occurs for thinner oil (SEMO 10 in our case) than more viscous oils (SEMO 4 and SEMO 5). Therefore, these results lead to the following conclusions: 1) the lubrication regime should be of a boundary type and 2) surface protection against wear for SEMO 10 and SEMO 36 oils seems to be resulting from the formation of surface layer as a result of adsorption of oil components or tribochemical reactions between the oil components and

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

285

**Figure 6.** Optical images of the friction zone of the piston ring segments after friction simulation tests using different

Results of the mass change measurements of the components are shown in Table 6. Worn mass calculated from the worn volume is plotted vs. measured mass change in Figure 7 (dots). The experimental data are fitted by linear function with two adjusted parameters: slope and intersect (dashed line in Figure 7). The solid line is a linear fit with a fixed slope 1 and adjusted intersect. Coefficients of determination for these linear regressions are 0.983 and 0.949, correspondingly, indicating statistically significant linear relationship between the mass

lubricants. The scale of each image is the same and shown by a scale bar in [12].

the base material.

The averaged cross-section profiles of each liner sample tested are reported in Figure 5. Different scales of magnitude are used for better visualization of the mean contact surface profile. It is possible to notice very good performance of the lubricant SEMO 10 and its improvement in the lubricant SEMO 36. Samples tested using SEMO 4 and SEMO 5 had deep grooves with the maximum depth 22 to 25µm. The samples tested using the reference oil and SEMO 10 had less deep grooves with a maximum depth of 4 to 5µm. Surface of the samples tested using SEMO 36 oil had some thin scratches in the direction of friction while grooves had not been formed.

**Figure 5.** Average cross-section profile of the friction zone of cylinder liner samples tested using different lubricants in [12].

Figure 6 shows images of the friction zone of the piston ring segments after friction simulation tests with different lubricating oils. Wear and damage of the surface as function of the oil used was similar to that in the cylinder liner. In tests with SEMO 4 and SEMO 5, the material in the friction zone was heavily damaged. The wear can be classified to be of the adhesive type with intensive plastic deformation and edging. When the reference oil and SEMO 10 oil were used in the tests, the damage of the material was less pronounced than for SEMO 4 and SEMO 5, but the wear in all cases was of the adhesive type. Only small damage was observed on the piston ring segments when using SEMO 36. In this case, only summits of the circular grooves of the piston ring presented some wear and deformation. From the point of view of hydrody‐ namic lubrication these results may seem to be surprising, since, with the same additive composition, higher wear rate occurs for thinner oil (SEMO 10 in our case) than more viscous oils (SEMO 4 and SEMO 5). Therefore, these results lead to the following conclusions: 1) the lubrication regime should be of a boundary type and 2) surface protection against wear for SEMO 10 and SEMO 36 oils seems to be resulting from the formation of surface layer as a result of adsorption of oil components or tribochemical reactions between the oil components and the base material.

after 700 to 1700 s from the beginning of the test. After reaching the maximum, friction coefficient decreased slowly and stabilized at 0.14 and 0.11 for SEMO 10 and SEMO 36, correspondingly. The friction coefficient of lubricant SEMO 10 showed a slow decline until reaching a constant value lower than the reference one. Friction coefficient for the improved lubricant SEMO 36 levelled out rapidly at a very low value and showed less scatter, probably

The averaged cross-section profiles of each liner sample tested are reported in Figure 5. Different scales of magnitude are used for better visualization of the mean contact surface profile. It is possible to notice very good performance of the lubricant SEMO 10 and its improvement in the lubricant SEMO 36. Samples tested using SEMO 4 and SEMO 5 had deep grooves with the maximum depth 22 to 25µm. The samples tested using the reference oil and SEMO 10 had less deep grooves with a maximum depth of 4 to 5µm. Surface of the samples tested using SEMO 36 oil had some thin scratches in the direction of friction while grooves had

**Figure 5.** Average cross-section profile of the friction zone of cylinder liner samples tested using different lubricants in

due to some sort of surface deposition on the contact surfaces.

not been formed.

284 Tribology - Fundamentals and Advancements

[12].

**Figure 6.** Optical images of the friction zone of the piston ring segments after friction simulation tests using different lubricants. The scale of each image is the same and shown by a scale bar in [12].

Results of the mass change measurements of the components are shown in Table 6. Worn mass calculated from the worn volume is plotted vs. measured mass change in Figure 7 (dots). The experimental data are fitted by linear function with two adjusted parameters: slope and intersect (dashed line in Figure 7). The solid line is a linear fit with a fixed slope 1 and adjusted intersect. Coefficients of determination for these linear regressions are 0.983 and 0.949, correspondingly, indicating statistically significant linear relationship between the mass change and worn mass determined from the geometry of the groove. Therefore, the deposit formation has not much influence on the mass change and the last can be used as a measure of the components wear out in these tests. The upper bound of the wear specific energy was determined in accordance with eq. (2), using the cylinder liner mass change in the denominator of eq. (2).

energy was similar to that in the simulation test: 0.14 GJ/g in the ball-on-disk at 200 °C vs. 0.18 GJ/g in the piston ring/cylinder liner simulation test. Although these values are only upper bound estimations of the real values, they are close to one another. According to the structuralenergetic approach in [10], this means that the dominating wear mechanism in both cases is the same. Then, a significant decrease in the wear specific energy from 3.97 to 0.14 GJ/g with temperature increase from 50 to 200 °C implies changing in dominating wear mechanism at higher temperature. It can be stated that, under the applied experimental conditions, the chemical compositions of the base oil and the additives had greater influence on the tribological

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

287

performance of the lubricants than their rheological properties.

**Figure 8.** Friction coefficient and wear specific energy in friction simulation tests in [12].

**•** friction zone of the cylinder tested using SEMO 36 lubricant,

**•** reference cylinder not immersed neither heated in lubricating oil.

Surface chemical composition of the friction zone of cylinder liner samples was characterized using Energy Dispersive X-Ray Spectroscopy (EDS). Table 7 shows surface chemical compo‐

**Fe O C Si Mn**

Friction zone 61.5±1.9 11.5±0.7 23.5±3.2 2.83±0.68 0.615±0.085 Untouched zone 84.0±3.6 7.36±0.46 3.41±3.31 4.19±0.07 1.11±0.17 Reference sample 92.5±0.3 2.95±0.46 0.1 3.59±0.52 0.935±0.255

**Table 7.** Surface chemical composition (at.%) of the cylinder liner samples tested with SEMO 36 lubricating oil

**2.4. Surface characterization**

sition for three different surfaces:

**•** untouched surface of the same cylinder, and


**Table 6.** Results of friction simulation tests in [12].

**Figure 7.** Mass wear determined from the geometry of the groove vs. mass change of the cylinder liner samples. The dashed line is a linear regression of experimental data with two adjusted parameters: slope and intercept. The solid line is a linear regression with a fixed slope 1 in [12].

Final friction coefficient and wear specific energy are shown in Figure 8. SEMO 36 oil showed the best antifriction and wear resistance characteristics among all tested lubricants. Friction coefficient was almost a half of that for the reference oil, while specific wear energy was 7.8 times higher than for the reference oil. In comparison with the ball-on-disk tests, wear specific energy for SEMO 36 lubricant was much lower in the tribological simulation test; however, oil temperature in these two tests was different. When the ball-on-disk evaluation tests were performed at the same temperature as in the simulation test (200 °C), the value of wear specific energy was similar to that in the simulation test: 0.14 GJ/g in the ball-on-disk at 200 °C vs. 0.18 GJ/g in the piston ring/cylinder liner simulation test. Although these values are only upper bound estimations of the real values, they are close to one another. According to the structuralenergetic approach in [10], this means that the dominating wear mechanism in both cases is the same. Then, a significant decrease in the wear specific energy from 3.97 to 0.14 GJ/g with temperature increase from 50 to 200 °C implies changing in dominating wear mechanism at higher temperature. It can be stated that, under the applied experimental conditions, the chemical compositions of the base oil and the additives had greater influence on the tribological performance of the lubricants than their rheological properties.

**Figure 8.** Friction coefficient and wear specific energy in friction simulation tests in [12].

#### **2.4. Surface characterization**

change and worn mass determined from the geometry of the groove. Therefore, the deposit formation has not much influence on the mass change and the last can be used as a measure of the components wear out in these tests. The upper bound of the wear specific energy was determined in accordance with eq. (2), using the cylinder liner mass change in the denominator

**Oil Final µfr Mass change, mg Worn volume (cylinder.) mm3 cylinder segment**

**Figure 7.** Mass wear determined from the geometry of the groove vs. mass change of the cylinder liner samples. The dashed line is a linear regression of experimental data with two adjusted parameters: slope and intercept. The solid

Final friction coefficient and wear specific energy are shown in Figure 8. SEMO 36 oil showed the best antifriction and wear resistance characteristics among all tested lubricants. Friction coefficient was almost a half of that for the reference oil, while specific wear energy was 7.8 times higher than for the reference oil. In comparison with the ball-on-disk tests, wear specific energy for SEMO 36 lubricant was much lower in the tribological simulation test; however, oil temperature in these two tests was different. When the ball-on-disk evaluation tests were performed at the same temperature as in the simulation test (200 °C), the value of wear specific

SEMO 4 0.34 -3.95 -1.41 -1.12\*10-2 SEMO 5 0.32 -3.68 -3.22 -9.8\*10-3 REF 0.20 -1.1 -0.64 -1.74\*10-3 SEMO 10 0.14 -0.94 -1.25 -2.3\*10-3

SEMO 36 0.11 -0.09 1.23 0

**Table 6.** Results of friction simulation tests in [12].

line is a linear regression with a fixed slope 1 in [12].

of eq. (2).

286 Tribology - Fundamentals and Advancements

Surface chemical composition of the friction zone of cylinder liner samples was characterized using Energy Dispersive X-Ray Spectroscopy (EDS). Table 7 shows surface chemical compo‐ sition for three different surfaces:



**Table 7.** Surface chemical composition (at.%) of the cylinder liner samples tested with SEMO 36 lubricating oil

Silicon and manganese were alloying elements of the base material and did not show important variations in their concentration, whereas the most important variation was in the carbon and oxygen content. There was no significant difference for other elements since the oils had no metal-containing additives. Figure 9 shows surface concentration of four elements relative to iron. After the test, during which a cylinder was immersed in the SEMO 36 oil and heated at 200 °C, carbon and oxygen concentrations on untouched surfaces were slightly higher than on the reference sample, e.g., the sample not immersed into the oil. However, carbon and oxygen concentrations drastically increased on the surface of the friction zone, on which carbon was each forth atom. Also, in contrast to the untouched surface and the reference surface, on the surface of the friction zone, carbon concentration was higher, than the oxygen one. One can infer from these data that friction induced tribochemical reactions between oil components and base material to form surface layer enriched with carbon and oxygen. This surface layer or sliding lacquer may protect mating surfaces from adhesion and/or damage yielding lower friction and wear in [10].

**Test step Speed, rpm Time, min Power**

**Table 8.** Experimental conditions for scuffing tests of real two-stroke engines

Petrol 2% 2% 2% E10 2% - - E20 2% - - E85 2% - 2%

**Table 9.** Oil – petrol combinations tested in a real two-stroke engine scuffing test

**Fuel Reference oil New developed oils**

rol (a,d), bioethanol E10 (b,e) and bioethanol E20 (c,f) in [12].

 2000 5 0 0 4000 20 50 2.4 4000 20 75 3.2 2000 5 0 0

5 4500 90 100 full load

**Figure 10.** Macro images of two-stroke engine components after scuffing test using a mixture of mineral oil with pet‐

**SEMO 10 SEMO 36**

6 2000 5 0 0

**% HP**

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

289

**Figure 9.** Surface concentration of elements relative to iron in [12].

#### **2.5. Experimental evaluation in real two-stroke engines (Minsel M165)**

After previous simulation tribological test the performance of the oils was evaluated in real two-stroke engines (Minsel M165) with a swept volume of 158 cm³, a stroke of 54 mm, compression ratio 7,1:1, power (ISO 1585) 3.53/4.8 kW/HP, maximum torque 120 Nm and 4500 rpm rotation speed. Scuffing tests were performed using various lubricating oil – petrol mixtures in order to evaluate the lubricating performance of the lubricants under extreme load conditions. The test conditions applied are shown in Table 8, and the tested oil-fuel composi‐ tions are shown in Table 9.

Figure 10 shows the photographs of the engine components after scuffing tests, in which the reference mineral oil was used in a mixture with pure petrol and bioethanol. Increase in the bioethanol content in the fuel led to decrease in carbon soot deposition on the engine cylinder and piston. Also, when bioethanol was used, the surface was less damaged under extreme working conditions.


**Table 8.** Experimental conditions for scuffing tests of real two-stroke engines

Silicon and manganese were alloying elements of the base material and did not show important variations in their concentration, whereas the most important variation was in the carbon and oxygen content. There was no significant difference for other elements since the oils had no metal-containing additives. Figure 9 shows surface concentration of four elements relative to iron. After the test, during which a cylinder was immersed in the SEMO 36 oil and heated at 200 °C, carbon and oxygen concentrations on untouched surfaces were slightly higher than on the reference sample, e.g., the sample not immersed into the oil. However, carbon and oxygen concentrations drastically increased on the surface of the friction zone, on which carbon was each forth atom. Also, in contrast to the untouched surface and the reference surface, on the surface of the friction zone, carbon concentration was higher, than the oxygen one. One can infer from these data that friction induced tribochemical reactions between oil components and base material to form surface layer enriched with carbon and oxygen. This surface layer or sliding lacquer may protect mating surfaces from adhesion and/or damage yielding lower

friction and wear in [10].

288 Tribology - Fundamentals and Advancements

tions are shown in Table 9.

working conditions.

**Figure 9.** Surface concentration of elements relative to iron in [12].

**2.5. Experimental evaluation in real two-stroke engines (Minsel M165)**

After previous simulation tribological test the performance of the oils was evaluated in real two-stroke engines (Minsel M165) with a swept volume of 158 cm³, a stroke of 54 mm, compression ratio 7,1:1, power (ISO 1585) 3.53/4.8 kW/HP, maximum torque 120 Nm and 4500 rpm rotation speed. Scuffing tests were performed using various lubricating oil – petrol mixtures in order to evaluate the lubricating performance of the lubricants under extreme load conditions. The test conditions applied are shown in Table 8, and the tested oil-fuel composi‐

Figure 10 shows the photographs of the engine components after scuffing tests, in which the reference mineral oil was used in a mixture with pure petrol and bioethanol. Increase in the bioethanol content in the fuel led to decrease in carbon soot deposition on the engine cylinder and piston. Also, when bioethanol was used, the surface was less damaged under extreme


**Table 9.** Oil – petrol combinations tested in a real two-stroke engine scuffing test

**Figure 10.** Macro images of two-stroke engine components after scuffing test using a mixture of mineral oil with pet‐ rol (a,d), bioethanol E10 (b,e) and bioethanol E20 (c,f) in [12].

**Figure 11.** Macro images of two-stroke engine components after scuffing test using mixture of SEMO 10 lubricating oil with petrol: a) piston, b) cylinder, c) exhaust side, d) intake in [12].

Figures 11 show the photographs of the engine components after scuffing tests using a SEMO 10 – petrol mixture. Some seizure between compression piston ring and cylinder was observed when using a mixture of SEMO 10 with petrol. Several vertical abrasion marks were formed in the exhaust zone of the cylinder, where the temperature was higher. However, the piston and cylinder were quite clean with only some carbon soot deposits in the exhaust zone. The state of the cylinder head was quite healthy and clean in the intake zone, the carbon residues were considered normal.

Figure 12 shows the pictures of the engine components after scuffing test using SEMO 36 lubricating oil with petrol and bioethanol fuels. When using a mixture of SEMO 36 with bioethanol E85 or petrol, no scuffing or seizure was observed. Only light scratches were found on the cylinder surface, which were more pronounced when using petrol. In this case, carbon soot deposits formed intensively on the top part of the piston. The piston and cylinder were very clean, when using bioethanol.

**Oil Type and % Power**

E85 + 2% Ref. oil (not miscible)

E85 + 2% SEMO 36

**2.6. Life cycle**

in Table 11.

(miscible)

**(kW)**

**Consumption (g/kWh)**

**Figure 12.** Macro images of two-stroke engine components after scuffing test using mixture of SEMO 36 lubricating

SH3 Limit Normative 5.36 161 603 Petrol/Ref. Oil 2% 5.46 397 1.469 139.8 333.2 E10 + 2% Ref. oil 5.44 385 1.573 124.1 314.8 E20 + 2% Ref. oil 5.5 382 2.29 128 251.5

oil with bioethanol E85 (a, b, c) and petrol (d, e, f): a), b), d), e) piston, c), f) cylinder in [12].

**NOx (g/kWh)**

4.8 427 2.29 109.8 43.11

4.3 478 0.689 119.5 32.93

**Table 10.** Emission of gases from two-stroke engine tested with different lubricating oil– petrol combinations in [12].

The lifecycle analysis for a 2-stroke engine fed by petrol and E85 was carried out using the model M 165 Minsel engine running in a tiller during 1000 h, which characteristics are shown

**CHx (g/kWh)**

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

291

**CO (g/kWh)**

In addition, gaseous emissions from the engine were analyzed for various fuel-oil mixtures with different proportions of bioethanol to petrol: 20%, 30% and 85%. The gas emissions were measured using the Directive CE 2002/88, Portable, SH3 modality as reference limits. The differences in power and consumption were negligible when using bioethanol E10 and E20. When compared with the petrol, the NOx emissions showed an increasing trend and the emissions of CO and CH diminished in tests with bioethanol and reference oil. When using E85, the reference mineral oil was not miscible, but the new developed oil SEMO 36 was totally miscible. When using bioethanol E85, a considerable reduction in engine power was observed yielding value 13% to 22% less than in the tests with petrol. At the same time fuel consumption increased slightly between 7% and 20%, and gaseous emissions were considerably reduced (see Table 10). When using SEMO 36 the reduction in NOx emission was the most significant as compared with other gases and was probably due to the lower temperature generated.

**Figure 12.** Macro images of two-stroke engine components after scuffing test using mixture of SEMO 36 lubricating oil with bioethanol E85 (a, b, c) and petrol (d, e, f): a), b), d), e) piston, c), f) cylinder in [12].


**Table 10.** Emission of gases from two-stroke engine tested with different lubricating oil– petrol combinations in [12].

#### **2.6. Life cycle**

**Figure 11.** Macro images of two-stroke engine components after scuffing test using mixture of SEMO 10 lubricating

Figures 11 show the photographs of the engine components after scuffing tests using a SEMO 10 – petrol mixture. Some seizure between compression piston ring and cylinder was observed when using a mixture of SEMO 10 with petrol. Several vertical abrasion marks were formed in the exhaust zone of the cylinder, where the temperature was higher. However, the piston and cylinder were quite clean with only some carbon soot deposits in the exhaust zone. The state of the cylinder head was quite healthy and clean in the intake zone, the carbon residues

Figure 12 shows the pictures of the engine components after scuffing test using SEMO 36 lubricating oil with petrol and bioethanol fuels. When using a mixture of SEMO 36 with bioethanol E85 or petrol, no scuffing or seizure was observed. Only light scratches were found on the cylinder surface, which were more pronounced when using petrol. In this case, carbon soot deposits formed intensively on the top part of the piston. The piston and cylinder were

In addition, gaseous emissions from the engine were analyzed for various fuel-oil mixtures with different proportions of bioethanol to petrol: 20%, 30% and 85%. The gas emissions were measured using the Directive CE 2002/88, Portable, SH3 modality as reference limits. The differences in power and consumption were negligible when using bioethanol E10 and E20. When compared with the petrol, the NOx emissions showed an increasing trend and the emissions of CO and CH diminished in tests with bioethanol and reference oil. When using E85, the reference mineral oil was not miscible, but the new developed oil SEMO 36 was totally miscible. When using bioethanol E85, a considerable reduction in engine power was observed yielding value 13% to 22% less than in the tests with petrol. At the same time fuel consumption increased slightly between 7% and 20%, and gaseous emissions were considerably reduced (see Table 10). When using SEMO 36 the reduction in NOx emission was the most significant as compared with other gases and was probably due to the lower

oil with petrol: a) piston, b) cylinder, c) exhaust side, d) intake in [12].

were considered normal.

290 Tribology - Fundamentals and Advancements

temperature generated.

very clean, when using bioethanol.

The lifecycle analysis for a 2-stroke engine fed by petrol and E85 was carried out using the model M 165 Minsel engine running in a tiller during 1000 h, which characteristics are shown in Table 11.


**Table 11.** Characteristics of the engine used in life-cycle analysis

Two fuel + oil pairs named as "Cleanengine systems" were compared with the Conventional system for the same engine working in the same application. In the alternative Cleanengine system I the engine was fed by a mixture of bioethanol E20 and mineral oil. In the alternative Cleanengine system II, the engine was fed by bioethanol and newly developed advanced and biodegradable lubricating oil SEMO 36. The fuel and oil consumption for the conventional and two alternative systems is shown in Table 12.


**3. Nozzles for future engines**

**3.1. Tribological evaluation**

Compared to conventional liquid hydrocarbon fuels, bio-fuels exhibit considerable differences intheirphysicalpropertieswhichsignificantlyinfluenceontheinjectorflowaswellasonprimary and secondary spray break-up processes. As a consequence, spray mixture formation of biofuels is considered to be largely different compared to conventional fuels under engine operat‐ ingconditionswithsevereconsequencesonthecombustionandemissioncharacteristics.Hence, injection and combustion system optimization as well as optimization of the injector configura‐ tion (number of nozzle holes, diameter, spray targeting, etc.) for bio-fuels requires a detailed knowledge of how the fuel properties influence the injector flow and spray atomization characteristics. Optimization of the nozzles materials and design is an important task which will open new markets and enlarge the number of potential customers for eco-friendly applications.

**Figure 13.** Results of the life-cycle and environmental impact analysis for the conventional and two alternative sys‐

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

293

tems: a) components of the environment impact, b) total environmental impact in [12].

Different metal-doped DLC coatings were developed by Physical Vapour Deposition method (PVD). Friction and wear tests were carried out using SRV tribometer with "cylinder-on-disc" configuration in lubricated conditions. The coatings were deposited on steel cylinders and

**Table 12.** Parameters of the conventional and alternative systems used in the life-cycle analysis

The Eco-indicator 99 Methodology was used for the Impact Assessment method. The compo‐ nents of the environmental impact are shown in Figure 13 a), while the total environmental impact is shown in Figure 13 b). Almost all components of the environmental impact as well as the total environmental impact were higher for fossil fuel. However, the climate change was more affected by the renewable system.

The global environmental impact evaluated by Lifecycle Assessment tools for the Cleanengine system I and II using bioethanol was lower than for the reference system using petrol. The comparison between two alternative systems Cleanengine I and Cleanengine II showed that the last one had slightly higher environmental impact due to higher fuel and lubricant consumption that can be related to the lower calorific value of the ethanol compared to the petrol. While the reduction of the environmental impact is attributed to the reduction in emissions, the use of a biodegradable nontoxic lubricant will further reduce the environmental impact of the Cleanengine II system.

**Figure 13.** Results of the life-cycle and environmental impact analysis for the conventional and two alternative sys‐ tems: a) components of the environment impact, b) total environmental impact in [12].

#### **3. Nozzles for future engines**

Model of machine Tiller 3002 Machine weight 90-110 kg

292 Tribology - Fundamentals and Advancements

Engine weight 12.8 kg Engine life 1000 h Scuffing test results OK Engine power 3 kW

Engine model M165 Minsel 2-stroke

**Table 11.** Characteristics of the engine used in life-cycle analysis

two alternative systems is shown in Table 12.

more affected by the renewable system.

impact of the Cleanengine II system.

Petrol 900 kg

Mineral oil 36 kg

**Table 12.** Parameters of the conventional and alternative systems used in the life-cycle analysis

Fuel consumption per functional unit

Oil consumption per functional unit

Emissions Directive 97/68/CE and later 2002/88/CE and 2004/26/CE

Two fuel + oil pairs named as "Cleanengine systems" were compared with the Conventional system for the same engine working in the same application. In the alternative Cleanengine system I the engine was fed by a mixture of bioethanol E20 and mineral oil. In the alternative Cleanengine system II, the engine was fed by bioethanol and newly developed advanced and biodegradable lubricating oil SEMO 36. The fuel and oil consumption for the conventional and

**Conventional Cleanengine**

**system (I)**

Mineral oil 23 kg

BioE20 1123 kg

The Eco-indicator 99 Methodology was used for the Impact Assessment method. The compo‐ nents of the environmental impact are shown in Figure 13 a), while the total environmental impact is shown in Figure 13 b). Almost all components of the environmental impact as well as the total environmental impact were higher for fossil fuel. However, the climate change was

The global environmental impact evaluated by Lifecycle Assessment tools for the Cleanengine system I and II using bioethanol was lower than for the reference system using petrol. The comparison between two alternative systems Cleanengine I and Cleanengine II showed that the last one had slightly higher environmental impact due to higher fuel and lubricant consumption that can be related to the lower calorific value of the ethanol compared to the petrol. While the reduction of the environmental impact is attributed to the reduction in emissions, the use of a biodegradable nontoxic lubricant will further reduce the environmental

**Cleanengine system (II)**

BioE85 1405 kg

SEMO 36 29 kg

> Compared to conventional liquid hydrocarbon fuels, bio-fuels exhibit considerable differences intheirphysicalpropertieswhichsignificantlyinfluenceontheinjectorflowaswellasonprimary and secondary spray break-up processes. As a consequence, spray mixture formation of biofuels is considered to be largely different compared to conventional fuels under engine operat‐ ingconditionswithsevereconsequencesonthecombustionandemissioncharacteristics.Hence, injection and combustion system optimization as well as optimization of the injector configura‐ tion (number of nozzle holes, diameter, spray targeting, etc.) for bio-fuels requires a detailed knowledge of how the fuel properties influence the injector flow and spray atomization characteristics. Optimization of the nozzles materials and design is an important task which will open new markets and enlarge the number of potential customers for eco-friendly applications.

#### **3.1. Tribological evaluation**

Different metal-doped DLC coatings were developed by Physical Vapour Deposition method (PVD). Friction and wear tests were carried out using SRV tribometer with "cylinder-on-disc" configuration in lubricated conditions. The coatings were deposited on steel cylinders and disks. The cylinder, 15 mm in diameter, performed reciprocating motion with a stroke of 2 mm and a friction frequency 50 Hz. Normal load was 50 N during short run-in period 30 s and 200 N during the test 60 min. The cylinder and the disk were immersed in fluids, which temper‐ ature during the test was constant and 25 °C.

**3.2. Corrosion characterization**

immersion.

scan rate of 0.5 mV/s.

the Nyquist plots.

0

1000

2000

3000

**Zre (Ohm)**

4000

5000

Corrosion resistance of different materials and coatings used for nozzles fabrication (Cr and Ti DLC) was characterized using electrochemical impedance spectroscopy and potentiody‐ namic polarization techniques in order to determine the kinetics parameters and the corrosion

Base nozzles material, uncoated steel X82WMo, was also characterized under corrosion conditions and compared with DLC coated samples of the same material. The electrolyte used in these tests was K2SO4 0.2M. Cr DLC coating offered excellent corrosion protection. The coating did not exhibit any pores or defects, protecting effectively the substrate during

Open-circuit potential (OCP) was measured during 2200 s in order to analyze the samples tendency with the exposure time. After that, an electrochemical impedance spectroscopy was performed in a frequencies range from 10 k to 10 mHz. Once impedance measurements finished, a potentiodynamic potential swept was applied from OCP-0.2 V to OCP+0.6 V at a

Coated nozzles had more positive potential than the reference ones. For all surfaces, OCP was stable after first 2200 s of immersion. The difference between three nozzles regarding impe‐ dance results was very notable. Cr and Ti DLC coated samples had a semicircle Nyquist diagrams implying that the electrolyte did not reach the substrate during the immersion in the dissolution. The coating acted as an effective protective barrier. Uncoated nozzle had lower corrosion resistance. Two time constants could be clearly distinguished from two maxima in

0 500 1000 1500 2000 2500 3000

Table 13 shows the parameters obtained from equivalent circuit simulation of the experimental

data and Figure 16 shows the equivalent circuits used in the simulation process.

**Figure 15.** Nyquist diagrams. Impedance data of coated and uncoated nozzle in K2SO4.

**Zim (Ohm)**

Ref. Uncoated

Cr DLC Ti DLC

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

295

mechanisms of these materials in NaCl 0.5M or K2SO4 0.2M in [12].

Both Cr- and Ti DLC coatings had good friction and wear behaviour and they could be a good alternative to improve tribological properties of the actual uncoated nozzles.

**Figure 14.** Average cross-section profile of the friction zone of discs samples (uncoated reference, Cr and Ti DLC) test‐ ed using different fuels, AGIP and B50. Different scales of magnitude are used for better visualization of the mean contact surface profile.

Surface morphology of the friction zone was studied using white light confocal microscopy. The averaged cross-section profiles for each sample tested are shown in Figure 14. It is possible to notice very good performance of the coatings, them had deeper grooves with a maximum depth of 5.45 µm. Cr DLC tested against AGIP fuel had better performance than Ti DLC. Two different scales of magnitude, Z, are used for better visualization of the mean contact surface profile. From 5 to -5µm for coated discs (Cr DLC and Ti DLC lubricated with AGIP ref and B50) and from 16 to -16µm for uncoated ref samples.

#### **3.2. Corrosion characterization**

disks. The cylinder, 15 mm in diameter, performed reciprocating motion with a stroke of 2 mm and a friction frequency 50 Hz. Normal load was 50 N during short run-in period 30 s and 200 N during the test 60 min. The cylinder and the disk were immersed in fluids, which temper‐

Both Cr- and Ti DLC coatings had good friction and wear behaviour and they could be a good

**Figure 14.** Average cross-section profile of the friction zone of discs samples (uncoated reference, Cr and Ti DLC) test‐ ed using different fuels, AGIP and B50. Different scales of magnitude are used for better visualization of the mean

Surface morphology of the friction zone was studied using white light confocal microscopy. The averaged cross-section profiles for each sample tested are shown in Figure 14. It is possible to notice very good performance of the coatings, them had deeper grooves with a maximum depth of 5.45 µm. Cr DLC tested against AGIP fuel had better performance than Ti DLC. Two different scales of magnitude, Z, are used for better visualization of the mean contact surface profile. From 5 to -5µm for coated discs (Cr DLC and Ti DLC lubricated with AGIP ref and

alternative to improve tribological properties of the actual uncoated nozzles.

ature during the test was constant and 25 °C.

294 Tribology - Fundamentals and Advancements

contact surface profile.

B50) and from 16 to -16µm for uncoated ref samples.

Corrosion resistance of different materials and coatings used for nozzles fabrication (Cr and Ti DLC) was characterized using electrochemical impedance spectroscopy and potentiody‐ namic polarization techniques in order to determine the kinetics parameters and the corrosion mechanisms of these materials in NaCl 0.5M or K2SO4 0.2M in [12].

Base nozzles material, uncoated steel X82WMo, was also characterized under corrosion conditions and compared with DLC coated samples of the same material. The electrolyte used in these tests was K2SO4 0.2M. Cr DLC coating offered excellent corrosion protection. The coating did not exhibit any pores or defects, protecting effectively the substrate during immersion.

Open-circuit potential (OCP) was measured during 2200 s in order to analyze the samples tendency with the exposure time. After that, an electrochemical impedance spectroscopy was performed in a frequencies range from 10 k to 10 mHz. Once impedance measurements finished, a potentiodynamic potential swept was applied from OCP-0.2 V to OCP+0.6 V at a scan rate of 0.5 mV/s.

Coated nozzles had more positive potential than the reference ones. For all surfaces, OCP was stable after first 2200 s of immersion. The difference between three nozzles regarding impe‐ dance results was very notable. Cr and Ti DLC coated samples had a semicircle Nyquist diagrams implying that the electrolyte did not reach the substrate during the immersion in the dissolution. The coating acted as an effective protective barrier. Uncoated nozzle had lower corrosion resistance. Two time constants could be clearly distinguished from two maxima in the Nyquist plots.

**Figure 15.** Nyquist diagrams. Impedance data of coated and uncoated nozzle in K2SO4.

Table 13 shows the parameters obtained from equivalent circuit simulation of the experimental data and Figure 16 shows the equivalent circuits used in the simulation process.


branch. Cr DLC and Ti DLC notably improved substrate corrosion behaviour reducing its

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

297

The injectors were tested in the Minsel M-430 engine manufactured by Abamotor Energía, SL.

**Table 15.** Characteristics of the engine used in engine tests to evaluate the different alternative nozzles (Cr DLC and Ti

During the test the engine worked for 50 hours at full load (3000 rpm). Biodiesel B30 was used as a fuel, which was a mixture of FAME (100% Biodiesel) with diesel B at a rate of 30%.

corrosion current by several orders of magnitude (see table 14).

**Samples Ecorr (V) Icorr (μA/cm2)**

**Table 14.** Corrosion current of coated and uncoated nozzles calculated using Tafel approach

**3.3. Experimental evaluation in real four-stroke engines (Minsel M430)**

The parameters of the engine and test conditions are shown in Table 15.

Bore 85 mm Stroke 75 mm Displacement 426 c.c. Compresion ratio 19,3: 1 Power NB 8,4 / 8,7 Cv Rpm 3000

Dry weight 45 Kg

DLC).

Torque 23Nm / 2000 RPM

**Figure 18.** The engine on test bench and the tested nozzles installed on the engine

Nozzle Ref. Uncoated -0.533 18.5 Nozzle Cr DLC -0.039 0.003 Nozzle Ti DLC -0.331 0.18

**Table 13.** Equivalent circuit parameters of coated and uncoated nozzles

**Figure 16.** Equivalent circuits used for the experimental data simulation. Circuit A) for Nozzle Cr DLC; circuit B) for un‐ coated nozzle and circuit C) for Ti DLC coating.

Polarization curves for the coated nozzle are shown in Figure 17.

**Figure 17.** Polarization curves on coated and uncoated nozzles immersed in K2SO4

Cr DLC coating had passive behaviour and low corrosion current of the order of 10-9A for potentials near to OCP. Coating Ti DLC also had passive behaviour in a wide zone of the anodic branch. Cr DLC and Ti DLC notably improved substrate corrosion behaviour reducing its corrosion current by several orders of magnitude (see table 14).


**Table 14.** Corrosion current of coated and uncoated nozzles calculated using Tafel approach

**Samples (Nozzle appl.)** **OCP (V)**

Nozzle Uncoated -0.556 0.035

296 Tribology - Fundamentals and Advancements

Nozzle Cr DLC -0.022 12780

Nozzle Ti-DLC -0.354 10.38

coated nozzle and circuit C) for Ti DLC coating.





**log I(A)**



0

**R1 (kΩ/cm2)**

**Table 13.** Equivalent circuit parameters of coated and uncoated nozzles

Polarization curves for the coated nozzle are shown in Figure 17.

**Figure 17.** Polarization curves on coated and uncoated nozzles immersed in K2SO4

**CPE1 (μF/cm2)**

Yo=58.45 n=0.82

Yo=0.75

Yo=0.34

**Figure 16.** Equivalent circuits used for the experimental data simulation. Circuit A) for Nozzle Cr DLC; circuit B) for un‐


Cr DLC coating had passive behaviour and low corrosion current of the order of 10-9A for potentials near to OCP. Coating Ti DLC also had passive behaviour in a wide zone of the anodic

**R2 (kΩ/cm2)**

1.10

n=0.94 - - -

n=0.639 - - Yo=0.02

**CPE2 (μF/cm2)**

Yo=140.7 n=0.94

**E(V vs AG/AgCl)**

Ref Cr DLC Ti DLC

**ΖΟ**

B=3.23

**(μΩ−1·σ 1/2; σ 1/2)**

#### **3.3. Experimental evaluation in real four-stroke engines (Minsel M430)**

The injectors were tested in the Minsel M-430 engine manufactured by Abamotor Energía, SL. The parameters of the engine and test conditions are shown in Table 15.


**Table 15.** Characteristics of the engine used in engine tests to evaluate the different alternative nozzles (Cr DLC and Ti DLC).

During the test the engine worked for 50 hours at full load (3000 rpm). Biodiesel B30 was used as a fuel, which was a mixture of FAME (100% Biodiesel) with diesel B at a rate of 30%.

**Figure 18.** The engine on test bench and the tested nozzles installed on the engine

#### **3.4. Nozzle characterization after test in the engine**

Scanning electron microscope (SEM) and energy dispersion X-Ray spectroscopy (EDS) were used for characterization of the nozzles geometry after the engine tests. Cr DLC coating had better behaviour than Ti DLC.

The microanalysis showed that for the all coatings the deposited layer on the needle persisted after the test, with the exception of the tip where the Ti DLC layer has been detached

Additionally, the spray holes geometries of the nozzle body were analysed after endurance test with two different fluids: reference standard fuel and 30% biodiesel.

Figure 19 shows the scanning electron microscope images (EDS) of the nozzle body tip before the engine test (real part and its corresponding silicon model for orifices internal characteristics analysis), whereas Figures 20 and 21 show the nozzles after the tests with standard diesel fuel and B30 fuel, correspondingly. Though large quantities of carbonaceous deposits could be observed on free surfaces for both fuels, no deposits were found on internal surface of spray holes.

**Figure 20.** Images of the nozzle after endurance test with standard diesel fuel

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

299

**Figure 19.** Nozzle body tip and silicon model with labelled holes

**Figure 20.** Images of the nozzle after endurance test with standard diesel fuel

**3.4. Nozzle characterization after test in the engine**

**Figure 19.** Nozzle body tip and silicon model with labelled holes

better behaviour than Ti DLC.

298 Tribology - Fundamentals and Advancements

holes.

Scanning electron microscope (SEM) and energy dispersion X-Ray spectroscopy (EDS) were used for characterization of the nozzles geometry after the engine tests. Cr DLC coating had

The microanalysis showed that for the all coatings the deposited layer on the needle persisted

Additionally, the spray holes geometries of the nozzle body were analysed after endurance

Figure 19 shows the scanning electron microscope images (EDS) of the nozzle body tip before the engine test (real part and its corresponding silicon model for orifices internal characteristics analysis), whereas Figures 20 and 21 show the nozzles after the tests with standard diesel fuel and B30 fuel, correspondingly. Though large quantities of carbonaceous deposits could be observed on free surfaces for both fuels, no deposits were found on internal surface of spray

after the test, with the exception of the tip where the Ti DLC layer has been detached

test with two different fluids: reference standard fuel and 30% biodiesel.

Our findings indicated that, in addition to the rheological properties of the lubricating oil, deposit build-up was an important factor controlling the tribological performance of the oil both in simulation experiments and real two-stroke engines. Two kinds of deposits: carbon soot and transparent sliding lacquer were observed on the engine components after tests. Build-up of a transparent sliding lacquer was especially important in the case of SEMO 36 oil and it was related with considerable reduction both in wear rate and friction coefficient. For SEMO 36, surface chemical analysis of the friction zone showed important changes in surface chemical composition, which was especially marked by increase in carbon and oxygen content. It is evident that formation of the sliding deposits stemmed from tribochemical reactions between the oil components and base material (cast iron and steel). The chemical state of carbon and oxygen atoms on the surface of friction zone should be further investigated for better

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

301

Tests in real two-stroke engines were performed using mixtures of the developed lubricant with petrol or bioethanol. In both cases, no seizure between piston ring and cylinder liner was observed. When using bioethanol, the engine components were clean without important

Engine power slightly decreased and fuel consumption slightly increased - on a volumetric basis when bioethanol E85 was blended with the newly developed lubricating oil SEMO 36. However, these results might be related with lower calorific value of ethanol as compared with petrol. Besides, the new lubricating oil improved scuffing resistance in combination with miscible lubricants and significantly reduced the environmental impact. In addition to low toxicity and high biodegradability, emissions of CO, NOx and hydrocarbons from engines lubricated with the newly developed lubricants were lower than with traditional mineral oil

Concluding, a new generation of lubricating oils for two-stroke engines have been developed combining low friction, good protection against wear and scuffing, no ash residue, low carbon soot or other deposit formation. These lubricating oils are compatible with bioethanol E85.

Application of Cr DLC coating on injection nozzles significantly increased the corrosion

Though Ti-DLC coating also improved substrate corrosion resistance, its performance in

Deposit chemical composition and the nozzle performance did not significantly vary in

The authors acknowledge financial support of the European Commission, the project Clea‐ nEngine "Advanced technologies for highly efficient Clean Engines working with alternative fuels and lubes" under contract TST5-CT2006-031241, and the Spanish Minister of Science and

and much below the limits established for portable applications.

endurance tests when standard diesel was substituted by B30 blend.

resistance and improved behaviour in engine test.

engine test was worse than for Cr DLC coating.

**Acknowledgements**

understanding of these mechanisms.

carbon soot deposits.

**Figure 21.** Images of the nozzle after endurance test with B30 fuel

Finally, nozzle deposits were analyzed by Thermal Gravimetric Analysis (TGA), which showed no big difference in deposits composition for the nozzles operated with standard diesel and B30 blend.

#### **4. Conclusions**

Fully formulated prototype lubricants based on synthetic esters had low toxicity for aqueous organisms (algae and Daphnia Magna) and high biodegradability evaluated by the Mano‐ metric Respirometry Method.

Among three developed prototype lubricating oils, SEMO 10 had the best tribological per‐ formance which was comparable with that of the reference mineral oil. Further improvement of the tribological properties of this lubricating oil was achieved by additive re-formulation. The developed lubricant, SEMO 36, exceeded the reference mineral oil in tribological per‐ formance.

Our findings indicated that, in addition to the rheological properties of the lubricating oil, deposit build-up was an important factor controlling the tribological performance of the oil both in simulation experiments and real two-stroke engines. Two kinds of deposits: carbon soot and transparent sliding lacquer were observed on the engine components after tests. Build-up of a transparent sliding lacquer was especially important in the case of SEMO 36 oil and it was related with considerable reduction both in wear rate and friction coefficient. For SEMO 36, surface chemical analysis of the friction zone showed important changes in surface chemical composition, which was especially marked by increase in carbon and oxygen content. It is evident that formation of the sliding deposits stemmed from tribochemical reactions between the oil components and base material (cast iron and steel). The chemical state of carbon and oxygen atoms on the surface of friction zone should be further investigated for better understanding of these mechanisms.

Tests in real two-stroke engines were performed using mixtures of the developed lubricant with petrol or bioethanol. In both cases, no seizure between piston ring and cylinder liner was observed. When using bioethanol, the engine components were clean without important carbon soot deposits.

Engine power slightly decreased and fuel consumption slightly increased - on a volumetric basis when bioethanol E85 was blended with the newly developed lubricating oil SEMO 36. However, these results might be related with lower calorific value of ethanol as compared with petrol. Besides, the new lubricating oil improved scuffing resistance in combination with miscible lubricants and significantly reduced the environmental impact. In addition to low toxicity and high biodegradability, emissions of CO, NOx and hydrocarbons from engines lubricated with the newly developed lubricants were lower than with traditional mineral oil and much below the limits established for portable applications.

Concluding, a new generation of lubricating oils for two-stroke engines have been developed combining low friction, good protection against wear and scuffing, no ash residue, low carbon soot or other deposit formation. These lubricating oils are compatible with bioethanol E85.

Application of Cr DLC coating on injection nozzles significantly increased the corrosion resistance and improved behaviour in engine test.

Though Ti-DLC coating also improved substrate corrosion resistance, its performance in engine test was worse than for Cr DLC coating.

Deposit chemical composition and the nozzle performance did not significantly vary in endurance tests when standard diesel was substituted by B30 blend.

#### **Acknowledgements**

**Figure 21.** Images of the nozzle after endurance test with B30 fuel

and B30 blend.

**4. Conclusions**

formance.

metric Respirometry Method.

300 Tribology - Fundamentals and Advancements

Finally, nozzle deposits were analyzed by Thermal Gravimetric Analysis (TGA), which showed no big difference in deposits composition for the nozzles operated with standard diesel

Fully formulated prototype lubricants based on synthetic esters had low toxicity for aqueous organisms (algae and Daphnia Magna) and high biodegradability evaluated by the Mano‐

Among three developed prototype lubricating oils, SEMO 10 had the best tribological per‐ formance which was comparable with that of the reference mineral oil. Further improvement of the tribological properties of this lubricating oil was achieved by additive re-formulation. The developed lubricant, SEMO 36, exceeded the reference mineral oil in tribological per‐

The authors acknowledge financial support of the European Commission, the project Clea‐ nEngine "Advanced technologies for highly efficient Clean Engines working with alternative fuels and lubes" under contract TST5-CT2006-031241, and the Spanish Minister of Science and Innovation, for co-financing the project under contracts ENE 2008-00652-E/ALT "Tecnologías Avanzadas para motores limpios altamente eficientes, trabajando con combustibles y lubri‐ cantes renovables", RYC-2009-0412 and BIA2011-25653. Also, the authors acknowledge support received from other partners who participated in the projects: ARIZONA Chemical, OBR, GUASCOR Power, BAM, AVL and INSTITUTO MOTORI.

[6] OECD 301F: Manometric Respirometry Test. OECD guidelines for testing of chemi‐

Innovative "Green" Tribological Solutions for Clean Small Engines

http://dx.doi.org/10.5772/55836

303

[7] OECD 201: Alga, Growth Inhibition Test. OECD guidelines for testing of chemicals;

[8] OECD 202: Daphnia sp. Acute Immobilisation Test. OECD guidelines for testing of

[9] DIN 51834-2. Tribological test in the translatory oscillation apparatus. Part 2: Stand‐ ard Test Method for Measuring the Friction and Wear Properties of EP Lubricating

[10] Kostetsky, BI; (1992) *The structural-energetic concept in the theory of friction and wear*

[11] Nevshupa, RA ; (2009) The *role of athermal mechanisms in the activation of tribodesorption and triboluminisence in miniature and lightly loaded friction units*. Journal of Friction and

[12] Igartua, A; Nevshupa, R; Fernández-Pérez, X; Conte, M; Zabala, R; Bernaola, J; Zaba‐ la, P; Luther, R; Rausch, R; (2011) *Alternative Eco-Friendly Lubes for Clean Two-Stroke*

[13] Martínez, L; Nevshupa, R; Álvarez, L; Huttel, Y; Méndez, J; Román, E; Mozas, E; Valdés, JR ; Jimenez, MA; Gachon, Y; Heau, C; Faverjon, F; (2009) *Application of dia‐ mond-like carbon coatings to elastomers frictional surfaces*.Tribology International, v. 42,

[14] Bayón, R; Nevshupa, R; Zubizarreta, C; Ruiz de Gopegui, U; Barriga, J; Igartua, A; (2010) *Characterisation of tribocorrosion behaviour of multilayer PVD coatings*. Analytical

[15] Bayón, R; Zubizarreta, C; Nevshupa, R; Rodriguez, JC; Fernández-Pérez, X; Ruiz de Gopegui, U; Igartua, A; (2011) *Rolling-sliding, scuffing and tribocorrosion behaviour of PVD multilayer coatings for gears application*. Industrial Lubrication and Tribology. V.

[16] Alajbegovic, A; Meister, G; Greif, D; Basara, B; (2001) *Three Phase Cavitating Flows in High Pressure Swirl Injectors*. 4th Int. Conf. on Multiphase Flow – ICMF'01, May 27 –

[17] Von Berg, E; Alajbegovic, A; Tatschl R; Krüger, C; Michels, U; (2001) *Multiphase Mod‐ eling of Diesel Sprays with the Eulerian*/Eulerian Approach (DaimlerChrysler AG),

[18] Von Berg, E; Alajbegovic, A; Greif, D; Poredos, A; Tatschl, R; Winklhofer, E (2002); *Break-up Model for Diesel Jets based on Locally Resolved Flow Field in the Injection Hole*,

cals; 1992.

chemicals; 2004.

Wear; 30:118-126.

pp. 584-590.

63/1. P. 17–26.

Oils Using the SRV Test Machine; 2004.

(synergism and self-organization). Wear; 159:1-15.

*Engines*. Tribology International, 44, 727-736.

and Bioanalytical Chemistry.V. 396. P. 2855-2862.

June 1, 2001, New Orleans, Louisiana, U.S.A.

ILASS-Europe 2001, Sept. 2-6, 2001, Zürich, Switzerland

ILASS-Europe 2002, Sept. 9-11, 2002, Zaragoza, Spain

2006.

We appreciate the useful help of Olatz Areitioaurtena and Raquel Bayón on performing Biodegradability, toxicity and corrosion characterizations and tests.

#### **Author details**

Xana Fernández-Pérez1 , Amaya Igartua1 , Roman Nevshupa2 , Patricio Zabala3 , Borja Zabala3 , Rolf Luther4 , Flavia Gili5 and Claudio Genovesio6

1 Fundación Tekniker, Avda, Otaola, Eibar, Spain

2 Instituto de Ciencias de la Construcción Eduardo Torroja (IETcc), c/ Serrano Galvache, Ma‐ drid, Spain


6 FIRAD S.p.A, Fabbrica Italiana Ricambi Apparati Diesel, Bagnolo, ItaIy

#### **References**


[6] OECD 301F: Manometric Respirometry Test. OECD guidelines for testing of chemi‐ cals; 1992.

Innovation, for co-financing the project under contracts ENE 2008-00652-E/ALT "Tecnologías Avanzadas para motores limpios altamente eficientes, trabajando con combustibles y lubri‐ cantes renovables", RYC-2009-0412 and BIA2011-25653. Also, the authors acknowledge support received from other partners who participated in the projects: ARIZONA Chemical,

We appreciate the useful help of Olatz Areitioaurtena and Raquel Bayón on performing

, Roman Nevshupa2

2 Instituto de Ciencias de la Construcción Eduardo Torroja (IETcc), c/ Serrano Galvache, Ma‐

[1] Igartua, A; Barriga, J; Aranzabe, A; (2005) *Biodegradable Lubricants*. Virtual Tribology

[2] Woydt, M; Skopp, A; (2005) *Ash-free and bionotox engine oils*. In: Biodegradable lubri‐ cants, eds. A. Igartua, J. Barriga, A. Aranzabe, Radom: Virtual Tribology Institute, In‐

[3] ASTM D-445-06: Standard Test Method for Kinematic Viscosity of Transparent and

[4] ASTM D-2270-04: Standard Practice for Calculating Viscosity Index from Kinematic

, Patricio Zabala3

, Borja Zabala3

,

OBR, GUASCOR Power, BAM, AVL and INSTITUTO MOTORI.

Biodegradability, toxicity and corrosion characterizations and tests.

, Amaya Igartua1

and Claudio Genovesio6

6 FIRAD S.p.A, Fabbrica Italiana Ricambi Apparati Diesel, Bagnolo, ItaIy

Opaque Liquids (and Calculation of Dynamic Viscosity).

[5] OECD guidelines for testing of chemicals: Section 3; 2003. 12 p.

**Author details**

Rolf Luther4

drid, Spain

**References**

Xana Fernández-Pérez1

302 Tribology - Fundamentals and Advancements

, Flavia Gili5

5 CRF StradaTorino Orbassano, Italy

1 Fundación Tekniker, Avda, Otaola, Eibar, Spain

3 Abamotor Energía, SL, B. Astola, Abadiano, Spain

Institute Edition, ISBN 83-70204-418-X.

stitute of Terotechnology; p. IV.6-IV.9.

Viscosity at 40 and 100°C.

4 Fuchs Europe GmbH, Mannheim D, Germany


**Chapter 11**

**A Sensor System for Online Oil Condition Monitoring of**

A web-based oil diagnosis system for continuous online lubricant condition monitoring is presented. The new approach utilizes sensor detection of chemical aging of engineering oils and their additives or first traces of wear debris by precision measurement of the electrical

The application potential of the sensor system is discussed on the example of the early identification of critical operating conditions for premature white etching cracks failures of rolling bearings in industrial gearboxes. Causative vibration loading is revealed prior to any component damage. Large roller bearings in wind energy gearboxes unusually often fail prematurely, i.e. clearly before the nominal *L*10 life. The failure is characterized by axial raceway cracks, from which branching and spreading crack systems, partly decorated by white etching microstructure, develop into the depth by corrosion fatigue. High localized friction coeffi‐ cients, resulting from the specific vibration caused mixed friction operating conditions, initiate tensile stress induced cleavage-like brittle spontaneous surface cracking. The basic idea of the novel failure detection condition monitoring system is the early identification of chemical

The sensor effectively controls the proper operation conditions of, e.g., bearings and gears in gearboxes. The online diagnostics system measures components of the specific complex impedance of oils. For instance, metal abrasion due to wear debris, broken oil molecules, forming acids or oil soaps, result in an increase of the electrical conductivity, which directly correlates with the degree of contamination of the oil. For additivated lubricants, the stage of degradation of the additives can also be derived from changes in the dielectric constant. The determination of the reduction in the oil quality by contaminations and the quasi continuous evaluation of wear and chemical aging follow the holistic approach of a real-time monitoring

> © 2013 Mauntz et al.; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2013 Mauntz et al.; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

properties. The basic concept and physical background are introduced in detail.

aging of the lubricant and its additives under the influence of vibration loading.

**Operating Components**

Manfred R. Mauntz, Jürgen Gegner, Ulrich Kuipers and Stefan Klingau

http://dx.doi.org/10.5772/55737

**1. Introduction**

Additional information is available at the end of the chapter

### **A Sensor System for Online Oil Condition Monitoring of Operating Components**

Manfred R. Mauntz, Jürgen Gegner, Ulrich Kuipers and Stefan Klingau

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/55737

#### **1. Introduction**

A web-based oil diagnosis system for continuous online lubricant condition monitoring is presented. The new approach utilizes sensor detection of chemical aging of engineering oils and their additives or first traces of wear debris by precision measurement of the electrical properties. The basic concept and physical background are introduced in detail.

The application potential of the sensor system is discussed on the example of the early identification of critical operating conditions for premature white etching cracks failures of rolling bearings in industrial gearboxes. Causative vibration loading is revealed prior to any component damage. Large roller bearings in wind energy gearboxes unusually often fail prematurely, i.e. clearly before the nominal *L*10 life. The failure is characterized by axial raceway cracks, from which branching and spreading crack systems, partly decorated by white etching microstructure, develop into the depth by corrosion fatigue. High localized friction coeffi‐ cients, resulting from the specific vibration caused mixed friction operating conditions, initiate tensile stress induced cleavage-like brittle spontaneous surface cracking. The basic idea of the novel failure detection condition monitoring system is the early identification of chemical aging of the lubricant and its additives under the influence of vibration loading.

The sensor effectively controls the proper operation conditions of, e.g., bearings and gears in gearboxes. The online diagnostics system measures components of the specific complex impedance of oils. For instance, metal abrasion due to wear debris, broken oil molecules, forming acids or oil soaps, result in an increase of the electrical conductivity, which directly correlates with the degree of contamination of the oil. For additivated lubricants, the stage of degradation of the additives can also be derived from changes in the dielectric constant. The determination of the reduction in the oil quality by contaminations and the quasi continuous evaluation of wear and chemical aging follow the holistic approach of a real-time monitoring

© 2013 Mauntz et al.; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Mauntz et al.; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

of an alteration in the condition of the oil-machine system. The measuring signals can be transmitted online to a web-based monitoring system via LAN, WLAN or serial interfaces of the sensor. Control of the relevant damage mechanisms, e.g. tribiological wear or oil aging, during proper operation below certain tolerance limits then allows preventive, conditionoriented maintenance to be carried out, if necessary, long before regular overhauling. Outage durations are reduced and the life of components and machines is increased.

#### **2. Basic sensor concept and physical principles**

#### **2.1. Basic sensor concept**

The basic sensor concept of the novel engineering oil monitoring system is based on the measurement of complex oil impedance components *X*, particularly the specific electrical conductivity κ and the relative permittivity εr. Due to their temperature dependence, the oil temperature *T* is also recorded [1-3]. Two or more electrodes, between which the oil flows, serve as a basic sensor. Resistance and capacity are measured independently of each other. Zero-mean periodic quantities are used to prevent polarization effects. Figure 1 shows the sensor in its triple plate design.

measurement, respectively. Oil is an electrical non-conductor. High resistance of the basic sensor and resulting low measurement currents provide best interference sensitivity to interspersed electromagnetic fields. Due to the very small currents, moreover, sufficient interference suppression is achieved. To prevent polarization effects, zero-mean alternating current voltages are measured as test signals. However, no capacitive current components may be measured simultaneously during a conductivity measurement because the capacitive current is much higher than its ohmic components. Thus, rather high requirements are set on analog sensor electronic systems, which are met with the reported measurement procedure. The conductivities of the insulating construction elements and insulation of electrical feed‐ through are about the same size as for the pure oils to be analyzed. The developed basic sensors and precise sensor electronic system ensure that the conductivity of feedthroughs and substrates may not be included into the test results. The active basic sensor unit consists of two or several basic sensor plates which are fixed to metal pins of a glass/metal feedthrough in a constant distant from each other. The plates of the basic sensor are arranged in the middle of the measuring chamber, allowing for an adequate incident flow of the flowing medium. A special alignment of the sensor housing parts is thus not necessary in this design. The extension characteristics of the sensor housing materials and the glass/metal feedthrough pins are exactly adjusted to the material characteristics of the used feedthrough glass. The compression

A Sensor System for Online Oil Condition Monitoring of Operating Components

http://dx.doi.org/10.5772/55737

307

**Figure 2.** Temperature dependence of the electrical conductivity of sample oil.

The ion mobility and thus the electrical conductivity κ depend upon the internal friction of the oil and therefore also on its temperature. The oil conductivity increases with temperature.

Already for about 3 °C alteration in temperature, the conductivity changes by about 25%. The electrical conductivity κ is a temperature function that depends on oil impurities rather than on the oil itself. The type of pollution and its temperature dependence cannot be assumed to

Figure 2 shows the dependence of the conductivity κ on the temperature change Δ*T*.

strength is above 10 MPa

**2.2. Temperature compensation**

**Figure 1.** Sensor in triple plate design.

Oils are electrical non-conductors. The electrical residual conductivity of pure oils lies in the range below 1 pS/m. For comparison, the electrical conductivity of the electrical non-conductor distilled water is larger by six orders of magnitude.

Broken oil molecules, acids, abrasive (metallic) wear, ions, oil soaps, etc., cause an increase of the oil conductivity κ. It rises with increasing ion concentration and mobility. The electrical conductivity of almost all impurities is high compared to the extremely low corresponding property of original pure oils.

The basic sensor represents an electrode arrangement, in which the measured oil is used as electrical conductor and as dielectric material for conductivity and relative permittivity

**Figure 2.** Temperature dependence of the electrical conductivity of sample oil.

measurement, respectively. Oil is an electrical non-conductor. High resistance of the basic sensor and resulting low measurement currents provide best interference sensitivity to interspersed electromagnetic fields. Due to the very small currents, moreover, sufficient interference suppression is achieved. To prevent polarization effects, zero-mean alternating current voltages are measured as test signals. However, no capacitive current components may be measured simultaneously during a conductivity measurement because the capacitive current is much higher than its ohmic components. Thus, rather high requirements are set on analog sensor electronic systems, which are met with the reported measurement procedure.

The conductivities of the insulating construction elements and insulation of electrical feed‐ through are about the same size as for the pure oils to be analyzed. The developed basic sensors and precise sensor electronic system ensure that the conductivity of feedthroughs and substrates may not be included into the test results. The active basic sensor unit consists of two or several basic sensor plates which are fixed to metal pins of a glass/metal feedthrough in a constant distant from each other. The plates of the basic sensor are arranged in the middle of the measuring chamber, allowing for an adequate incident flow of the flowing medium. A special alignment of the sensor housing parts is thus not necessary in this design. The extension characteristics of the sensor housing materials and the glass/metal feedthrough pins are exactly adjusted to the material characteristics of the used feedthrough glass. The compression strength is above 10 MPa

#### **2.2. Temperature compensation**

of an alteration in the condition of the oil-machine system. The measuring signals can be transmitted online to a web-based monitoring system via LAN, WLAN or serial interfaces of the sensor. Control of the relevant damage mechanisms, e.g. tribiological wear or oil aging, during proper operation below certain tolerance limits then allows preventive, conditionoriented maintenance to be carried out, if necessary, long before regular overhauling. Outage

The basic sensor concept of the novel engineering oil monitoring system is based on the measurement of complex oil impedance components *X*, particularly the specific electrical conductivity κ and the relative permittivity εr. Due to their temperature dependence, the oil temperature *T* is also recorded [1-3]. Two or more electrodes, between which the oil flows, serve as a basic sensor. Resistance and capacity are measured independently of each other. Zero-mean periodic quantities are used to prevent polarization effects. Figure 1 shows the

Oils are electrical non-conductors. The electrical residual conductivity of pure oils lies in the range below 1 pS/m. For comparison, the electrical conductivity of the electrical non-conductor

Broken oil molecules, acids, abrasive (metallic) wear, ions, oil soaps, etc., cause an increase of the oil conductivity κ. It rises with increasing ion concentration and mobility. The electrical conductivity of almost all impurities is high compared to the extremely low corresponding

The basic sensor represents an electrode arrangement, in which the measured oil is used as electrical conductor and as dielectric material for conductivity and relative permittivity

durations are reduced and the life of components and machines is increased.

**2. Basic sensor concept and physical principles**

**2.1. Basic sensor concept**

306 Tribology - Fundamentals and Advancements

sensor in its triple plate design.

**Figure 1.** Sensor in triple plate design.

property of original pure oils.

distilled water is larger by six orders of magnitude.

The ion mobility and thus the electrical conductivity κ depend upon the internal friction of the oil and therefore also on its temperature. The oil conductivity increases with temperature. Figure 2 shows the dependence of the conductivity κ on the temperature change Δ*T*.

Already for about 3 °C alteration in temperature, the conductivity changes by about 25%. The electrical conductivity κ is a temperature function that depends on oil impurities rather than on the oil itself. The type of pollution and its temperature dependence cannot be assumed to be known. To improve the comparability of measurements, a self-learning adaptive temper‐ ature compensation algorithm is implemented. An integral alteration of the oil quality can then be assessed by the temperature compensated conductivity value, whereas the type of con‐ tamination is not determinable. The relative permittivity is measured with the same basic sensor arrangement as used for the electrical conductivity.

The electrical conductivity and relative permittivity are to be measured with respect to a reference temperature *T*R as close as possible to the operating temperature of the oil. These parameters can be evaluated by means of temperature-dependent approximating polyno‐ mials, as demonstrated below exemplarily for the electrical conductivity:

$$\kappa\_{\rm R} = \kappa\_{\rm R,0} + \left\{ a \Delta T\_{\rm C} + b \Delta T\_{\rm C}^2 + c \Delta T\_{\rm C}^3 \right\} \times \kappa\_{\rm M} \tag{1}$$

Here, κR and κR,0 denote the approximate and previously calculated (old) electrical conductivity of the oil at the reference temperature *T*R, respectively. *T*<sup>C</sup> stands for the current temperature of the oil and κ<sup>M</sup> is the electrical conductivity measured without temperature compensation. Moreover, *a*, *b*, und *c* are the coefficients of the approximating polynomial to be adaptively determined. The temperature difference is defined as follows:

$$
\Delta T = T\_R - T\_{\odot} \tag{2}
$$

**2.3. Calculation and linear approximation of relative permittivity and conductivity**

<sup>ε</sup>r=εr,add *<sup>f</sup>* <sup>×</sup>εr,oil

volume fraction *f* of the addition, 1−*f* becomes the volume fraction of the oil.

The permittivity of the addition and the oil, respectively, is denoted εr,add and εr,oil. With the

**Figure 4.** Electrical permittivity εr measured as a function of the water content and model fit according to Eq. (3).

**3. Premature failures of rolling bearings and correlation with oil aging**

Bearings in industrial, e.g. wind turbine, gearboxes unusually often suffer from a significantly shorter life than calculated by white etching cracks [5, 6]. Figure 5 shows the light-optical micrograph of a typical metallographic microsection [5]. The overrolling direction from left to right indicates surface initiation and top-down propagation of the extended crack system. These early failures are characterized by mostly axial raceway cracks, revealing vertical semi to fully circular cleavage-like lenticular brittle spontaneous incipient cracks in preparatively opened original fracture faces [5, 6]. Occasionally, pock-like spallings are associated with the

1- *<sup>f</sup>* (3)

http://dx.doi.org/10.5772/55737

309

A Sensor System for Online Oil Condition Monitoring of Operating Components

In a series of experiments on the non-additivated lubricating oil FVA03, fresh demineralized water was added to a volume of 3.01%. The oil conductivity data measured as a function of the water content are found to follow a linear relationship in good approximation. The theoretical course of the relative permittivity is calculated for dilute solutions according to different mixing rules by truncating a Taylor series expansion of the model equations after the linear term. The model of Lichtenecker is evaluated in Figure 4. Lichtenecker developed the formula of Eq. (3) for calculating the dielectric constant of a homogeneous mixture εr [4]:

The oil temperature *T*<sup>C</sup> is measured for this temperature compensation. The use of a polyno‐ mial of the third order in Eq. (1) ensures good approximation while keeping the computational effort for the applied microcomputer reasonably low. Figure 3 shows the measured values of the electrical conductivity κ after temperature compensation.

**Figure 3.** Measured conductivity values after temperature compensation.

#### **2.3. Calculation and linear approximation of relative permittivity and conductivity**

be known. To improve the comparability of measurements, a self-learning adaptive temper‐ ature compensation algorithm is implemented. An integral alteration of the oil quality can then be assessed by the temperature compensated conductivity value, whereas the type of con‐ tamination is not determinable. The relative permittivity is measured with the same basic

The electrical conductivity and relative permittivity are to be measured with respect to a reference temperature *T*R as close as possible to the operating temperature of the oil. These parameters can be evaluated by means of temperature-dependent approximating polyno‐

Here, κR and κR,0 denote the approximate and previously calculated (old) electrical conductivity of the oil at the reference temperature *T*R, respectively. *T*<sup>C</sup> stands for the current temperature of the oil and κ<sup>M</sup> is the electrical conductivity measured without temperature compensation. Moreover, *a*, *b*, und *c* are the coefficients of the approximating polynomial to be adaptively

The oil temperature *T*<sup>C</sup> is measured for this temperature compensation. The use of a polyno‐ mial of the third order in Eq. (1) ensures good approximation while keeping the computational effort for the applied microcomputer reasonably low. Figure 3 shows the measured values of

<sup>2</sup> <sup>+</sup> *<sup>c</sup>*Δ*T*<sup>C</sup> 3

) ×*κ*<sup>M</sup> (1)

Δ*T* =*TR* - *TC* (2)

sensor arrangement as used for the electrical conductivity.

308 Tribology - Fundamentals and Advancements

mials, as demonstrated below exemplarily for the electrical conductivity:

*κ*<sup>R</sup> =*κ*R,0 + (*a*Δ*T*<sup>C</sup> + *b*Δ*T*<sup>C</sup>

determined. The temperature difference is defined as follows:

the electrical conductivity κ after temperature compensation.

**Figure 3.** Measured conductivity values after temperature compensation.

In a series of experiments on the non-additivated lubricating oil FVA03, fresh demineralized water was added to a volume of 3.01%. The oil conductivity data measured as a function of the water content are found to follow a linear relationship in good approximation. The theoretical course of the relative permittivity is calculated for dilute solutions according to different mixing rules by truncating a Taylor series expansion of the model equations after the linear term. The model of Lichtenecker is evaluated in Figure 4. Lichtenecker developed the formula of Eq. (3) for calculating the dielectric constant of a homogeneous mixture εr [4]:

$$
\varepsilon\_{\mathbf{r}} = \varepsilon\_{\mathbf{r}, \text{add}}^f \times \varepsilon\_{\mathbf{r}, \text{ol}}^{1 \cdot f} \tag{3}
$$

The permittivity of the addition and the oil, respectively, is denoted εr,add and εr,oil. With the volume fraction *f* of the addition, 1−*f* becomes the volume fraction of the oil.

**Figure 4.** Electrical permittivity εr measured as a function of the water content and model fit according to Eq. (3).

#### **3. Premature failures of rolling bearings and correlation with oil aging**

Bearings in industrial, e.g. wind turbine, gearboxes unusually often suffer from a significantly shorter life than calculated by white etching cracks [5, 6]. Figure 5 shows the light-optical micrograph of a typical metallographic microsection [5]. The overrolling direction from left to right indicates surface initiation and top-down propagation of the extended crack system.

These early failures are characterized by mostly axial raceway cracks, revealing vertical semi to fully circular cleavage-like lenticular brittle spontaneous incipient cracks in preparatively opened original fracture faces [5, 6]. Occasionally, pock-like spallings are associated with the

XRD based material response analyses of run rolling bearings, suffering from white etching cracks on still largely undamaged raceways, reveal the causative vibration loading [5]. It is further reported that lubricant aging occurs under the influence of vibrations [7, 8]. An example of an infrared spectrum of used oil from rig test run of a roller bearing is provided in Figure 7 [7]. The verified O‒H and C=O oxidation bands indicate operational acidification of the oil, also reflected in the dissolution of MnS inclusion lines on the raceway (cf. Figure 8), as a result of polycondensation reactions towards resinification and beginning lacquer formation. It is this aging of the lubricating oil and its additives, which can be detected at an early stage by

A Sensor System for Online Oil Condition Monitoring of Operating Components

http://dx.doi.org/10.5772/55737

311

**Figure 7.** Oxidation peaks in the infrared spectrum of a used non-additivated aliphatic lubricating oil run under vibra‐

The mentioned crack initiation by tribochemical reactions is also found on lateral surfaces of rollers. In Figure 8 [5], a scanning electron microscope (SEM) image, taken in the secondary electron imaging mode, is shown on the left. Residues of manganese and sulphur, detected in the crack-like defect by energy dispersive X-ray spectroscopy (Figure 8, on the right), indicate

On a bearing and gear test rig, the new sensor based oil quality monitoring system is applied. Various load cycles are run and speeds and torques are measured. The results of the trial are

the causative tribo chemical dissolution of nonmetallic MnS inclusions [5, 7, 8].

**4. Trial of the oil sensor system on a bearing and gear test rig**

described, evaluated and discussed in the following sections.

The speed-related power *P*(*n*) of the test rig is given as follows:

**4.1. Loss of power and trial run characteristics**

tion loading in a rolling bearing rig test (λ is the wavelength).

the new sensor so that a gearbox operating at critical conditions is identified.

**Figure 5.** Radial microsection of a branching and spreading white etching crack system.

surface cracks, as shown exemplarily in Figure 6 [6]. The developing deep crack systems are open to the raceway, from which oil penetrates and promotes further corrosion fatigue crack growth as well as local secondary microstructural changes in the form of crack path decorating white etching constituents. It is evident from fractography and X-ray diffraction (XRD) residual stress analyses that the cleavage-like incipient cracks are caused by frictional tangential tensile stresses [5, 6], which occur in subregions of the contact area in specific, vibrationally induced mixed friction operating conditions [5–8].

**Figure 6.** Inner ring raceway with typical axial cracks and few associated pock-like spallings.

XRD based material response analyses of run rolling bearings, suffering from white etching cracks on still largely undamaged raceways, reveal the causative vibration loading [5]. It is further reported that lubricant aging occurs under the influence of vibrations [7, 8]. An example of an infrared spectrum of used oil from rig test run of a roller bearing is provided in Figure 7 [7]. The verified O‒H and C=O oxidation bands indicate operational acidification of the oil, also reflected in the dissolution of MnS inclusion lines on the raceway (cf. Figure 8), as a result of polycondensation reactions towards resinification and beginning lacquer formation. It is this aging of the lubricating oil and its additives, which can be detected at an early stage by the new sensor so that a gearbox operating at critical conditions is identified.

**Figure 7.** Oxidation peaks in the infrared spectrum of a used non-additivated aliphatic lubricating oil run under vibra‐ tion loading in a rolling bearing rig test (λ is the wavelength).

The mentioned crack initiation by tribochemical reactions is also found on lateral surfaces of rollers. In Figure 8 [5], a scanning electron microscope (SEM) image, taken in the secondary electron imaging mode, is shown on the left. Residues of manganese and sulphur, detected in the crack-like defect by energy dispersive X-ray spectroscopy (Figure 8, on the right), indicate the causative tribo chemical dissolution of nonmetallic MnS inclusions [5, 7, 8].

#### **4. Trial of the oil sensor system on a bearing and gear test rig**

On a bearing and gear test rig, the new sensor based oil quality monitoring system is applied. Various load cycles are run and speeds and torques are measured. The results of the trial are described, evaluated and discussed in the following sections.

#### **4.1. Loss of power and trial run characteristics**

surface cracks, as shown exemplarily in Figure 6 [6]. The developing deep crack systems are open to the raceway, from which oil penetrates and promotes further corrosion fatigue crack growth as well as local secondary microstructural changes in the form of crack path decorating white etching constituents. It is evident from fractography and X-ray diffraction (XRD) residual stress analyses that the cleavage-like incipient cracks are caused by frictional tangential tensile stresses [5, 6], which occur in subregions of the contact area in specific, vibrationally induced

**Figure 5.** Radial microsection of a branching and spreading white etching crack system.

**Figure 6.** Inner ring raceway with typical axial cracks and few associated pock-like spallings.

mixed friction operating conditions [5–8].

310 Tribology - Fundamentals and Advancements

The speed-related power *P*(*n*) of the test rig is given as follows:

**Figure 8.** SEM image of a crack on a bearing roller with elemental mapping of Mn and S.

$$P = M \cdot \omega \text{ with } \omega = 2\pi \cdot n \tag{4}$$

After switching to a higher load, the power loss increases abruptly before the bearings run in. Towards the end of the trial, the bearings reveal indication of advanced deterioration. A wide lubrication gap and vibrations when overrolling spalling results in higher oscillation ampli‐ tudes, which leads to the automatic shutdown of the test rig eventually. The measuring results

A Sensor System for Online Oil Condition Monitoring of Operating Components

http://dx.doi.org/10.5772/55737

313

Figure 10 shows the test readings of the conductivity measurement of the lubricating oil. The current bearing wear and the deteriorating oil condition in the conducted trial are reflected in

New oil from the storage container exhibits a conductivity κ of 2312 pS/m. After filling into the trial gear and before the start-up of the test rig, a conductivity of 2791 pS/m is measured. This increase can be attributed to existing residual impurities in the gear. During the trial run, the conductivity κ of the gear oil increases to 16868 pS/m. Besides changes in temperature, conductivity increase is caused, e.g., by wear debris and removed material from spalling, impurities, broken oil molecules or forming oil soap. As described above, the temperature dependence of the electrical conductivity of the used gear oil is compensated and the oil conductivity measured in the gear trial is converted into the relevant conductivity value at 40 °C. Figure 11 shows the development of the temperature compensated oil conductivity with

In the case of an initially low load, the electrical conductivity increases linearly with running time. It is to be assumed that the low bearing wear in this area also increases proportional to

During the necessary intermediate shutdown (interruption) and run-up of the drive machine to 330 Nm, the conductivity is virtually constant. After switching over to the higher load, the

the change of the electrical conductivity plotted vs. running time in the diagram.

obtained with the oil sensor system are presented in the following.

**Figure 10.** Measurement of the electrical conductivity κ vs. running time *t*.

running time during the gear trial.

the time.

**4.2. Conductivity of the lubricating oil**

Here, *M* denotes the torque, ω and *n* respectively stand for the angular velocity and rotational speed. The implemented power loss ∆*P* is derived from the transmission ratio *NÜ*:

$$
\Delta P = \omega\_1 \cdot \left(M\_1 \cdot N\_{\bullet} \cdot M\_2\right) \tag{5}
$$

*M*1 and *M*<sup>2</sup> indicate the torque of the drive and the load, respectively. In the first trial on the test rig, the rolling bearing is intentionally damaged. The time-dependent power loss in the gear, as derived from the measuring signal characteristics, is represented in Figure 9.

**Figure 9.** Calculated power loss Δ*P*.

After switching to a higher load, the power loss increases abruptly before the bearings run in. Towards the end of the trial, the bearings reveal indication of advanced deterioration. A wide lubrication gap and vibrations when overrolling spalling results in higher oscillation ampli‐ tudes, which leads to the automatic shutdown of the test rig eventually. The measuring results obtained with the oil sensor system are presented in the following.

#### **4.2. Conductivity of the lubricating oil**

*P* =*M* ⋅*ω with ω* =2*π* ⋅*n* (4)

∆*P* =*ω*<sup>1</sup> ⋅ (*M*<sup>1</sup> - *N*<sup>Ü</sup> ⋅*M*2) (5)

Here, *M* denotes the torque, ω and *n* respectively stand for the angular velocity and rotational

*M*1 and *M*<sup>2</sup> indicate the torque of the drive and the load, respectively. In the first trial on the test rig, the rolling bearing is intentionally damaged. The time-dependent power loss in the

gear, as derived from the measuring signal characteristics, is represented in Figure 9.

**Figure 9.** Calculated power loss Δ*P*.

312 Tribology - Fundamentals and Advancements

speed. The implemented power loss ∆*P* is derived from the transmission ratio *NÜ*:

**Figure 8.** SEM image of a crack on a bearing roller with elemental mapping of Mn and S.

Figure 10 shows the test readings of the conductivity measurement of the lubricating oil. The current bearing wear and the deteriorating oil condition in the conducted trial are reflected in the change of the electrical conductivity plotted vs. running time in the diagram.

**Figure 10.** Measurement of the electrical conductivity κ vs. running time *t*.

New oil from the storage container exhibits a conductivity κ of 2312 pS/m. After filling into the trial gear and before the start-up of the test rig, a conductivity of 2791 pS/m is measured. This increase can be attributed to existing residual impurities in the gear. During the trial run, the conductivity κ of the gear oil increases to 16868 pS/m. Besides changes in temperature, conductivity increase is caused, e.g., by wear debris and removed material from spalling, impurities, broken oil molecules or forming oil soap. As described above, the temperature dependence of the electrical conductivity of the used gear oil is compensated and the oil conductivity measured in the gear trial is converted into the relevant conductivity value at 40 °C. Figure 11 shows the development of the temperature compensated oil conductivity with running time during the gear trial.

In the case of an initially low load, the electrical conductivity increases linearly with running time. It is to be assumed that the low bearing wear in this area also increases proportional to the time.

During the necessary intermediate shutdown (interruption) and run-up of the drive machine to 330 Nm, the conductivity is virtually constant. After switching over to the higher load, the

When starting up at 2000 revolutions per minute and a torque of 150 Nm, a relatively constant change in conductivity from 0.6 to 0.8 pS/(m×3 min), equivalent to 3.3 to 4.4 fS/(m×s), occurs. In the case of higher load (330 Nm, 3000 min–1), the change in conductivity rises up to 3.8 pS/ (m×3 min), i.e. 21.1 fS/(m×s). After the intermediate load increase, the effect on the change of the oil conductivity appears stronger. This may be attributed to the time-dependent formation of impurities and changes in bearing stressing as can be expected during the development of spalling. Figure 13 shows the inner ring of the failed planet bearing with massive damage of

A Sensor System for Online Oil Condition Monitoring of Operating Components

http://dx.doi.org/10.5772/55737

315

The connection between the change in conductivity κ and the loss of power ∆*P* in the gear is also evaluated. Figure 14 represents this progress graphically. Both the increasing change in

the raceway at the end of the trial.

into account.

**Figure 13.** Heavily spalled inner ring raceway of the tested cylindrical roller bearing.

oil conductivity and the gear power loss correlate with the bearing wear.

**Figure 14.** Alteration of the electrical conductivity as function of power loss of the gearbox.

In the diagram of Figure 14, a trend line is drawn as polynomial of the third order. The higher the increases in conductivity and loss of power in the gear, the stronger the bearing wear occurs. Exceptional changes to the system, e.g. switching of the load conditions, are not taken

**Figure 11.** Time curve of the temperature compensated oil conductivity derived from Figure 10.

oil conductivity increases strongly. Here, the bearing run-in (shakedown) is shown as reduc‐ tion in the conductivity increase. More than about 30 minutes prior to the final forced shut‐ down of the trial run by an oscillation sensor, the conductivity remains almost constant followed by a temporary rise directly before disconnection. After switching off the test gear, the oil conductivity decreases strongly. This clearly emphasizes the influence of the additives. During the loading stages, more impurities per time unit are produced than bound to additives. After shutting down the test rig, no further oil contamination occurs while the effect of additives still continues.

The variations in electrical conductivity are depicted in Figure 12. In this diagram, the curve follows averages respectively calculated over 3 minutes.

**Figure 12.** Alteration of the electrical conductivity, expressed as ∆κ40/∆*t,* vs. running time *t*.

When starting up at 2000 revolutions per minute and a torque of 150 Nm, a relatively constant change in conductivity from 0.6 to 0.8 pS/(m×3 min), equivalent to 3.3 to 4.4 fS/(m×s), occurs. In the case of higher load (330 Nm, 3000 min–1), the change in conductivity rises up to 3.8 pS/ (m×3 min), i.e. 21.1 fS/(m×s). After the intermediate load increase, the effect on the change of the oil conductivity appears stronger. This may be attributed to the time-dependent formation of impurities and changes in bearing stressing as can be expected during the development of spalling. Figure 13 shows the inner ring of the failed planet bearing with massive damage of the raceway at the end of the trial.

**Figure 13.** Heavily spalled inner ring raceway of the tested cylindrical roller bearing.

oil conductivity increases strongly. Here, the bearing run-in (shakedown) is shown as reduc‐ tion in the conductivity increase. More than about 30 minutes prior to the final forced shut‐ down of the trial run by an oscillation sensor, the conductivity remains almost constant followed by a temporary rise directly before disconnection. After switching off the test gear, the oil conductivity decreases strongly. This clearly emphasizes the influence of the additives. During the loading stages, more impurities per time unit are produced than bound to additives. After shutting down the test rig, no further oil contamination occurs while the effect of

**Figure 11.** Time curve of the temperature compensated oil conductivity derived from Figure 10.

The variations in electrical conductivity are depicted in Figure 12. In this diagram, the curve

additives still continues.

314 Tribology - Fundamentals and Advancements

follows averages respectively calculated over 3 minutes.

**Figure 12.** Alteration of the electrical conductivity, expressed as ∆κ40/∆*t,* vs. running time *t*.

The connection between the change in conductivity κ and the loss of power ∆*P* in the gear is also evaluated. Figure 14 represents this progress graphically. Both the increasing change in oil conductivity and the gear power loss correlate with the bearing wear.

**Figure 14.** Alteration of the electrical conductivity as function of power loss of the gearbox.

In the diagram of Figure 14, a trend line is drawn as polynomial of the third order. The higher the increases in conductivity and loss of power in the gear, the stronger the bearing wear occurs. Exceptional changes to the system, e.g. switching of the load conditions, are not taken into account.

#### **4.3. Relative permittivity of the lubricating oil**

In addition to the electrical conductivity, the relative permittivity εr of the oil is measured. In the case of oils not enriched with additives, the water content can be determined that way. There are good prospects that the dwindling efficacy of the additives can be detected by means of the dielectric constant measurement. Figure 15 shows the time development of the relative permittivity during the trial run. Due to the dependence on temperature, this development is also depicted in the diagram.

Then the measurement signal increases further with increasing pollution, water entry, etc. Figure 17 schematically shows the temperature compensated time curve of the permittivity of additivated oil continuously contaminated by the addition of wear debris, water or oil acids from chemical aging. Once the additives are consumed, the vanishing shielding effect results

A Sensor System for Online Oil Condition Monitoring of Operating Components

http://dx.doi.org/10.5772/55737

317

**Figure 16.** Time curve of the temperature compensated relative permittivity εr.

The most commercially available particle counters only detect particles as small as 4 µm. In a very early stage of wear of bearings, gears, hydraulic cylinders, etc., however, particularly smaller particles are produced. A preventive maintenance lowing, rather than rigid inspection intervals, therefore requires recognition of even the smallest particles. These particles are far more common in the oils of functioning machines than larger ones. Oil aging can be involved

in a characteristic re-increase.

**Figure 17.** Temperature compensated permittivity.

in the failure, for instance, of rolling bearings [7].

**Figure 15.** Time curves of the relative permittivity εr and the temperature *T* as a function of the running time *t*.

The change of the relative permittivity could be caused by a combination of the effects of a chemical reaction of additives, water evaporation from the oil and the temperature dependence of the cell constants as well as the relative permittivity itself. During the trial run, the temper‐ ature increases from 42 to 52.9 °C. The temperature dependence justifies the developed adaptive, self-learning temperature compensation technique. Figure 16 shows the temperature compensated time development of the relative permittivity during the gear trial.

#### **5. Approach for condition monitoring of additivated lubricating oils**

A direct connection between the electrical conductivity and the degree of contamination of oils is found. An increase of the electrical conductivity of the oil in operation can thus be interpreted as increasing wear or contamination of the lubricant. The aging of the oil is also evident in the degradation of additives. The used additives reveal high conductivity compared with the oil.

The consumption of the additives is reflected in a reduction of the electrical conductivity and permittivity of the oil. The gradient, i.e. the time derivative, of the conductivity or the dielectric constant progression respectively represents a measure of the additive degradation and consumption. The full additive degradation is indicated by the slope of zero (bathtub curve).

**Figure 16.** Time curve of the temperature compensated relative permittivity εr.

**4.3. Relative permittivity of the lubricating oil**

also depicted in the diagram.

316 Tribology - Fundamentals and Advancements

In addition to the electrical conductivity, the relative permittivity εr of the oil is measured. In the case of oils not enriched with additives, the water content can be determined that way. There are good prospects that the dwindling efficacy of the additives can be detected by means of the dielectric constant measurement. Figure 15 shows the time development of the relative permittivity during the trial run. Due to the dependence on temperature, this development is

**Figure 15.** Time curves of the relative permittivity εr and the temperature *T* as a function of the running time *t*.

compensated time development of the relative permittivity during the gear trial.

**5. Approach for condition monitoring of additivated lubricating oils**

A direct connection between the electrical conductivity and the degree of contamination of oils is found. An increase of the electrical conductivity of the oil in operation can thus be interpreted as increasing wear or contamination of the lubricant. The aging of the oil is also evident in the degradation of additives. The used additives reveal high conductivity compared with the oil.

The consumption of the additives is reflected in a reduction of the electrical conductivity and permittivity of the oil. The gradient, i.e. the time derivative, of the conductivity or the dielectric constant progression respectively represents a measure of the additive degradation and consumption. The full additive degradation is indicated by the slope of zero (bathtub curve).

The change of the relative permittivity could be caused by a combination of the effects of a chemical reaction of additives, water evaporation from the oil and the temperature dependence of the cell constants as well as the relative permittivity itself. During the trial run, the temper‐ ature increases from 42 to 52.9 °C. The temperature dependence justifies the developed adaptive, self-learning temperature compensation technique. Figure 16 shows the temperature

Then the measurement signal increases further with increasing pollution, water entry, etc. Figure 17 schematically shows the temperature compensated time curve of the permittivity of additivated oil continuously contaminated by the addition of wear debris, water or oil acids from chemical aging. Once the additives are consumed, the vanishing shielding effect results in a characteristic re-increase.

**Figure 17.** Temperature compensated permittivity.

The most commercially available particle counters only detect particles as small as 4 µm. In a very early stage of wear of bearings, gears, hydraulic cylinders, etc., however, particularly smaller particles are produced. A preventive maintenance lowing, rather than rigid inspection intervals, therefore requires recognition of even the smallest particles. These particles are far more common in the oils of functioning machines than larger ones. Oil aging can be involved in the failure, for instance, of rolling bearings [7].

#### **6. Web-based decentralized lubricant quality monitoring system**

The integration into a suitable communication structure and the realization of an online monitoring system offers an interesting practice-oriented utilization of the oil sensor system. This is briefly discussed below.

Preferred areas of application of the sensor system are energy production and automated technical plants that are operated locally, like e.g. wind turbines, generators, hydraulic systems or gearboxes. Plant employers are interested in continuous automated in vivo examination of the oil quality rather than interrupting the operation for regular sampling. Online oil status monitoring significantly improves the economic and ecological efficiency by increasing operating safety, reducing down times or adjusting oil change intervals to actual requirements. Once the oil condition monitoring sensors are installed on the plants, the measuring data can be displayed and evaluated elsewhere. A flexible decentralized monitoring system also enables the analysis of measuring signals and monitoring of the plants by external providers. A userorientated service ensuring the quantitative evaluation of changes in the oil-machine system, including the recommendation of resulting preventive maintenance measures, relieves plant operators, increases reliability and saves costs.

**Figure 19.** WearSens®

**7. Conclusions**

sensor system with communication unit.

can also be deduced from recorded changes in the relative permittivity.

industrial gearboxes, of which vibrational contact loading is the root cause.

in the condition of the oil-machine system.

The online diagnostics system measures components of the specific complex impedance of oils. For instance, metal abrasion due to bearing wear at the tribological contact, broken oil molecules, acids or oil soap cause an increase in electrical conductivity that directly correlates with the degree of pollution of the oil. The dielectrical properties of the oils are especially determined by the water content, which, in the case of products that are not enriched with additives, becomes accessible by an additional accurate measurement of the relative permit‐ tivity. In the case of oils enriched with additives, statements on the degradation of additives

A Sensor System for Online Oil Condition Monitoring of Operating Components

http://dx.doi.org/10.5772/55737

319

Indication of damage and wear is measured as an integral factor of, e.g., the degree of pollution, oil aging and acidification, water content and the decomposition state of additives or abrasion of the bearings. It provides informative data on lubricant aging and material loading as well as the wear of the bearings and gears for the online operative monitoring of components of machines. Additional loading, for instance, by vibration induced mixed friction in rollingsliding contact (rolling bearings, gears, cams, etc.) causes specific faster oil aging, e.g., in the course of premature component failures. Verified in roller bearing vibration rig tests, the oil suffers from significant acidification by polycondensation reactions and incipient resinifica‐ tion, as proven by infrared spectroscopy of used lubricant. The application potential of the sensor is discussed on the example of the prevention of early rolling bearing failures in

For an efficient machine utilization and targeted damage prevention, the new electrical online oil condition monitoring system offers the prospect to carry out timely preventative mainte‐ nance on demand rather than in rigid inspection intervals. The determination of impurities or reduction in the quality of the lubricants and the quasi continuous evaluation of the bearing and gear wear and oil aging meet the holistic approach of a real-time monitoring of a change

The measuring signals can be transmitted to a web-based condition monitoring system via LAN, WLAN or serial interfaces of the sensor system. The monitoring of the tribological wear mechanisms during proper operation below the tolerance limits of the components then allows

In a web-based decentralized online oil condition monitoring system, the sensor signals are preferably transferred through the Internet to a database server and recorded on an HTML page as user interface [8]. Figure 18 shows the displayed measured data.

**Figure 18.** Displayed measured data.

Following authentication, a simple web browser permits access via the wired or wireless LAN. In case of alarm signals, an immediate automated generation of warning messages, for instance by e-mail or SMS, is possible from any computer with Internet connection. Figure 19 shows the new sensor system with communication unit [9].

**Figure 19.** WearSens® sensor system with communication unit.

#### **7. Conclusions**

**6. Web-based decentralized lubricant quality monitoring system**

This is briefly discussed below.

318 Tribology - Fundamentals and Advancements

operators, increases reliability and saves costs.

**Figure 18.** Displayed measured data.

the new sensor system with communication unit [9].

The integration into a suitable communication structure and the realization of an online monitoring system offers an interesting practice-oriented utilization of the oil sensor system.

Preferred areas of application of the sensor system are energy production and automated technical plants that are operated locally, like e.g. wind turbines, generators, hydraulic systems or gearboxes. Plant employers are interested in continuous automated in vivo examination of the oil quality rather than interrupting the operation for regular sampling. Online oil status monitoring significantly improves the economic and ecological efficiency by increasing operating safety, reducing down times or adjusting oil change intervals to actual requirements. Once the oil condition monitoring sensors are installed on the plants, the measuring data can be displayed and evaluated elsewhere. A flexible decentralized monitoring system also enables the analysis of measuring signals and monitoring of the plants by external providers. A userorientated service ensuring the quantitative evaluation of changes in the oil-machine system, including the recommendation of resulting preventive maintenance measures, relieves plant

In a web-based decentralized online oil condition monitoring system, the sensor signals are preferably transferred through the Internet to a database server and recorded on an HTML

Following authentication, a simple web browser permits access via the wired or wireless LAN. In case of alarm signals, an immediate automated generation of warning messages, for instance by e-mail or SMS, is possible from any computer with Internet connection. Figure 19 shows

page as user interface [8]. Figure 18 shows the displayed measured data.

The online diagnostics system measures components of the specific complex impedance of oils. For instance, metal abrasion due to bearing wear at the tribological contact, broken oil molecules, acids or oil soap cause an increase in electrical conductivity that directly correlates with the degree of pollution of the oil. The dielectrical properties of the oils are especially determined by the water content, which, in the case of products that are not enriched with additives, becomes accessible by an additional accurate measurement of the relative permit‐ tivity. In the case of oils enriched with additives, statements on the degradation of additives can also be deduced from recorded changes in the relative permittivity.

Indication of damage and wear is measured as an integral factor of, e.g., the degree of pollution, oil aging and acidification, water content and the decomposition state of additives or abrasion of the bearings. It provides informative data on lubricant aging and material loading as well as the wear of the bearings and gears for the online operative monitoring of components of machines. Additional loading, for instance, by vibration induced mixed friction in rollingsliding contact (rolling bearings, gears, cams, etc.) causes specific faster oil aging, e.g., in the course of premature component failures. Verified in roller bearing vibration rig tests, the oil suffers from significant acidification by polycondensation reactions and incipient resinifica‐ tion, as proven by infrared spectroscopy of used lubricant. The application potential of the sensor is discussed on the example of the prevention of early rolling bearing failures in industrial gearboxes, of which vibrational contact loading is the root cause.

For an efficient machine utilization and targeted damage prevention, the new electrical online oil condition monitoring system offers the prospect to carry out timely preventative mainte‐ nance on demand rather than in rigid inspection intervals. The determination of impurities or reduction in the quality of the lubricants and the quasi continuous evaluation of the bearing and gear wear and oil aging meet the holistic approach of a real-time monitoring of a change in the condition of the oil-machine system.

The measuring signals can be transmitted to a web-based condition monitoring system via LAN, WLAN or serial interfaces of the sensor system. The monitoring of the tribological wear mechanisms during proper operation below the tolerance limits of the components then allows preventive, condition-oriented maintenance to be carried out, if necessary, long before regular overhauling, thus reducing outages caused by wear while simultaneously increasing the overall lifetime of the oil-machine system.

[6] Nierlich, W., Gegner, J. Einführung der Normalspannungshypothese für Mischrei‐ bung im Wälz-Gleitkontakt, Düsseldorf: VDI Reports 2147, VDI Wissensforum; 2011.

A Sensor System for Online Oil Condition Monitoring of Operating Components

http://dx.doi.org/10.5772/55737

321

[7] Gegner, J., Nierlich, W. Operational Residual Stress Formation in Vibration-Loaded

[8] Nierlich, W., Gegner, J. Material Response Bearing Testing under Vibration Loading, In: J. M. Beswick (ed.), Advances in Rolling Contact Fatigue Strength Testing and Re‐ lated Substitute Technologies, STP 1548, ASTM International, West Conshohocken,

[9] Gegner, J., Kuipers U., Mauntz, M. New Electric Online Oil Condition Monitoring Sensor – an Innovation in Early Failure Detection of Industrial Gears, 4th Internation‐ al Multi-Conference on Engineering and Technological Innovation, 19.-22-07.2011, Orlando, Florida, USA 2011, Proceedings Volume I, International Institute of Infor‐

[10] Gegner, J., Kuipers U., Mauntz, M. High-precision online sensor condition monitor‐ ing of industrial oils in service for the early detection of contamination and chemical aging, Sensor+Test Conferences 07.-09.06.2011, Nürnberg, AMA Service GmbH,

Rolling Contact, Advances in X-ray Analysis, Vol. 52; 2008, pp. 722-731.

matics and Systemics, Winter Garten, Florida, USA, 2011, pp. 238-242.

pp. 277-290, in German.

Pennsylvania, USA, 2012.

Wunstorf, 2011, pp. 702-709.

On a bearing and gear rig test, various load cycles are run and the functionality of the intro‐ duced electric online condition monitoring sensor system is tested successfully. The evaluation of the experiment is presented.

#### **Author details**

Manfred R. Mauntz1\*, Jürgen Gegner2 , Ulrich Kuipers3 and Stefan Klingau1


#### **References**


[6] Nierlich, W., Gegner, J. Einführung der Normalspannungshypothese für Mischrei‐ bung im Wälz-Gleitkontakt, Düsseldorf: VDI Reports 2147, VDI Wissensforum; 2011. pp. 277-290, in German.

preventive, condition-oriented maintenance to be carried out, if necessary, long before regular overhauling, thus reducing outages caused by wear while simultaneously increasing the

On a bearing and gear rig test, various load cycles are run and the functionality of the intro‐ duced electric online condition monitoring sensor system is tested successfully. The evaluation

, Ulrich Kuipers3

[1] Gegner, J., Kuipers, U., Mauntz, M. Ölsensorsystem zur Echtzeit-Zustandsüberwa‐ chung von technischen Anlagen und Maschinen, Technisches Messen 77; 2010. pp.

[2] Kuipers, U., Mauntz, M. Ölsensorsystem – Sensorsystem zur Messung von Kompo‐ nenten der komplexen Impedanz elektrisch gering leitender und nichtleitender Flu‐ ide, dessen Realisierung und Anwendung, German Patent Application N° 10 2008 047 366.9, Applicant: cmc Instruments GmbH, German Patent Office, Munich, Filing

[3] Kuipers, U., Mauntz, M. Verfahren, Schaltungsanordnung, Sensor zur Messung physikalischer Größen in Fluiden sowie deren Verwendung, European Patent Appli‐ cation N° EP 09000244, European Patent Office, Munich; 09.01.2009, in German.

[4] Lichtenecker, K., Rother, K. Die Herleitung der logarithmischen Mischungsgesetzes aus allgemeinen Prinzipien der stationären Strömung, Physikalische Zeitschrift, 1931,

[5] Gegner, J. Tribological Aspects of Rolling Bearing Failures, In: C.-H. Kuo (ed.), Tri‐ bology – Lubricants and Lubrication, Rijeka: InTech; 2011. Chap. 2, pp. 33-94.

and Stefan Klingau1

overall lifetime of the oil-machine system.

of the experiment is presented.

320 Tribology - Fundamentals and Advancements

Manfred R. Mauntz1\*, Jürgen Gegner2

\*Address all correspondence to: mrm@cmc-instruments.de

3 South Westphalia University of Applied Sciences, Hagen, Germany

1 cmc Instruments GmbH, Eschborn, Germany

2 University of Siegen, Siegen, Germany

date: 15.09.2008, in German.

32, pp. 255-260.

**Author details**

**References**

283-292.


### *Edited by Jürgen Gegner*

As the subject of tribology comprises lubrication, friction and wear of contact components highly relevant to practical applications, it challenges scientists from chemistry, physics and materials engineering around the world on todays sophisticated experimental and theoretical foundation to complex interdisciplinary research. Recent results and developments are preferably presented and evaluated in the context of established knowledge. Consisting of eleven chapters divided into the four parts of Lubrication and Properties of Lubricants, Boundary Lubrication Applications, Testing and Modeling, and Sustainability of Tribosystems, this textbook therefore merges basic concepts with new findings and approaches. Tribology Fundamentals and Advancements, supported by competent authors, aims to convey current research trends in the light of the state of the art to students, scientists and practitioners and help them solve their problems.

Photo by Vladimir\_Timofeev / iStock

Tribology - Fundamentals and Advancements

Tribology

Fundamentals and Advancements