**2. Experimental procedures**

**1. Introduction**

80 Contact and Fracture Mechanics

Materials with high contact damage resistance are extensively required in aerospace and aircraft, vehicle industry, microelectro-mechanical systems and devices, cutting tools and bulletproof vests [1, 2]. An approach for improved resistance to surface contact damage is to design surface gradations in composition, microstructure and elastic and/or plastic properties [2–4]. Such design provide effective means to enhance materials contact damage resistance through redistribution of thermal and/or mechanical stresses, elimination of interface-induced stress concentrations and reduction in the local crack driving force [5–8]. Nature is a master in the design of sophisticated hierarchical structured materials which provide excellent damage resistance [9]. A typical example is the material structural design principle found in a fish armor [10]. In response to predatory threats, fish are protected by armor scales consisting of four distinct reinforcing layers of organic/inorganic nanocomposites with hardness and modulus decreasing gradually from the outer to the inner layers. The juxtaposition of multiple reinforcing composite layers and the gradations, both in microstructure and mechanical properties within and between material layers, provides a more compliant protective mechanism than the monolithic counterpart [10].

Inspired from the material structural design principle discovered in natural/biological systems, materials scientists have generated enormous interest in replicating natural/biological structures with excellent damage resistance than their conventional counterparts. Over the past 2 decades, significant progress has been made in synthesis and fabrication of materials with graded properties over multiple length scales. Elastically graded materials (EGMs), where the materials have gradient in elastic modulus as a function of depth beneath the surface, were synthesized by controlled infiltration of aluminosilicate or oxynitride glass into polycrystalline ceramic matrix, which offered superior resistance to contact damage than either constituent ceramic matrix or glass [11–13]. Plastically graded materials (PGMs) were produced by increasing or decreasing the grain size within the nanocrystalline or microcrystalline range to create a linear gradation of yield strength as a function of depth below the material surface according to classical Hall–Petch effect [3]. The benefit of the gradient effect on the stress–strain and deformation response under normal indentation have been demon-

strated by analytical [14], computational [15–17] and experimental studies [18, 19].

deformation and damage resistance of MHSed Ti.

In our recent work [20], we extended the EGM/PGM concept to design a multilayered hierarchical structure (MHS) on Ti. By the application of Surface Mechanical Attrition Treatment (SMAT) [21] to cryorolled Ti, a three-layered structure formed consisting of an outer amorphous/nanocrystallite (A/NC) layer, an inner nanograined (NG) layer and ultrafine-grained (UFG) core [20]. Nanoindentation through the cross-section of the multilayered hierarchical structured (MHSed) Ti revealed a gradual decrease in hardness and modulus within and between each successive structural layer [20]. These properties correlate with the microstructure characteristics and the design principle found in natural systems, such as fish armor [10]. The work hardening of the MHSed Ti was improved largely by such structural design [20]. Moreover, the gradations in structure and properties, pore and crack-free nature and the inherently damage tolerant top A/NC layer of MHSed Ti are expected to benefit the contact The MHSed Ti was produced by the following experimental procedures. A commercial Ti plate (Grade 2) with 36 mm in thickness was cryogenically rolled to 5 mm with per reduction of ∼2 mm. The detailed microstructure characterizations of the cryorolled Ti have been given elsewhere [22]. The cryorolled workpiece then was cut parallel to the rolling direction (RD) to a rectangular bar with dimensions of 5 × 5 × 90 mm<sup>3</sup> . Subsequently, one lateral surface of the rectangular bar was subjected to SMAT. The SMAT process was performed in a low vacuum condition using hardened stainless steel balls (8 mm in diameter) at a vibration frequency of 50 Hz for 60 min. The detailed MHS process can be found in [20]. The production of monolithic NG Ti has been given in [22].

**Figure 1a** shows a schematic illustration of the nanoindentation and contact load-bearing testing. Nanoindentation experiments were carried out at ambient temperature using an UMIS indentation system with a Berkovich diamond tip at a strain rate of 5 × 10−2 s−1 and a maximum load of 20 mN. Before testing, the cross-sectional surface was polished to 0.5 μm diamond suspension finish. The values of the nanoindentation hardness and modulus quoted here were the average of 10 measurements on the cross-sectional surface. Before Vickers microhardness and load-bearing testing, artifacts on the surface caused by MHS process were carefully removed by polishing to 0.5 μm diamond suspension finish (removal thickness < 2 μm). Vickers microhardness testing was conducted using a microhardness tester (FM 700) under a load of 0.5, 1, 3, 5, 10 N on the MHSed surface at more than 10 points and the average values were reported here. Load-bearing testing was conducted with a spherical tungsten carbide (WC) indenter with diameter of 1.5 mm in ambient conditions. The WC indenter had an elastic modulus of 640 GPa and a Poisson's ratio of 0.26. The indenter came into contact with the specimen surface and was loaded to a maximum load of 1000 N at a loading rate of 1000 N/s.

The upper right inset taken from the bright matrix region exhibited a broad diffuse halo in a selected-area diffraction pattern (SAD), which is typical of a fully amorphous phase. The SAD pattern in the lower left inset taken from the interface between the bright and dark phases clearly demonstrates the presence of a nanocrystalline phase together with the amorphous phase. The NG layer (∼60 μm thick), situated beneath the A/NC layer, consisted of nanograins (**Figure 2c**). The corresponding SAD pattern shows a ring pattern, demonstrating the nanostructure has random crystallographic orientations. The size of the nanograins was in the range from 5 to 80 nm with an average size of ∼40 nm (**Figure 2e**). The UFG core is composed of ultrafine equiaxed

Improving Contact Load-Bearing Resistance of Ultrafine-Grained Materials…

http://dx.doi.org/10.5772/intechopen.72197

83

Nanoindentation was used to measure the elastic and plastic mechanical properties spatially through the cross-section of the MHSed Ti. The elastic mismatch and delamination, commonly existed in other multilayered systems produced by deposition or coating, are two critical factors controlling crack confinement [23]. In contrast, due to the gradients in both the strain and strain rate induced by the cryorolling and SMAT process from the top surface to the inner core, the reported MHSed Ti is free from elastic mismatch and delamination between layers.

100nm

**Figure 2.** Microstructural characteristic of the MHSed Ti: (a) SEM cross-sectional image. (b) TEM bright-field (BF) image of the microstructure situated 20 μm below the top surface. The upper right and lower left insets are the selected-area diffraction (SAD) patterns of the bright region the dark region, respectively; (c) TEM BF image of the microstructure located 60 μm below the top surface. The inset shows the corresponding SAD pattern; (d) TEM BF image of the innermost core. The inset shows the corresponding SAD pattern; (e and f) histogram of the grain size distribution in the NG layer

longitudinal length. 300 grains were statistical measured from several TEM dark-field (DF) images.

\_\_\_\_\_\_\_

*dT* × *dL*, *dT* is the transverse length of the grain and *dL*

is the

100nm

f

50nm

e

grains with a grain size distribution of 50–250 nm (**Figure 2d** and **f**).

b

a e

A/NC

NG

UFG

20µm

(e) and UFG core (f). The grain size *D* was defined by *D* = √

c

d

**3.2. Mechanical gradations**

**Figure 1.** (a) Schematic illustration of the surface load-bearing and cross-sectional nanoindentation testing. The black ball represents the load-bearing testing indenter and the black triangle represents the indent of nanoindentation testing. (b) Finite element mesh of the ball indentation model.

Microstructural and damage observations were conducted using a field-emission gun scanning electron microscope (SEM) Zeiss Supra 55VP operated at 10 kV and a transmission electron microscope (TEM) Jeol JEM 2100 operated at 200 kV.

A two-dimensional axisymmetric model was developed to simulate the ball indentation using Abaqus v6.10. The specimens were modeled as isotropic, elastic–perfectly plastic following the large-deformation theory. The finite element mesh contained 42,163 four-node bilinear axisymmetric quadrilateral elements (CAX4R), with a refined mesh in the indentation region (**Figure 1b**). Mesh convergence was verified by comparing load–depth curves and stress contours using models with element number ranging from 11,183 to 42,163. A user subroutine UMAT was developed to take into account the gradation of material properties, in which the discrete and gradient model parameters were assigned to elements based on their distance from the surface (see later for further details). The ball indenter was modeled as a rigid body, and the contact between the indenter and specimen was assumed frictionless. A maximum load of 1000 N was applied to indent the samples at a loading rate of 1000 N/s. The finite element analysis (FEA) of the microindentation adopted a similar methodology to the axisymmetric nanoindentation simulations. The Vickers indenter was modeled to be a conical rigid indenter with an apex angle of 70.3° and tip radius of 3.4 μm, which approximates a Vickers indenter tip.
