**4. Some self-healing assisting phenomena in titanium metal and alloys**

### **4.1 Phase transformation in titanium metal and alloys**

Phase transformation occurs whenever a materials system is not at equilibrium, or changes its microstate, as a result of external constraints such as pressure or temperature. In effect, these materials adopt different crystal structures favorable for the minimization of their free energy. In general, the microstructural features and the order in the system changes, leading to variations in most of the important properties. By so doing, phase transformation provides an effective way to modify the microstructure of solids. If it can be activated by a mechanical or other physical force, it becomes part of the deformation process and directly affects the properties of materials as well [22].

In CP titanium and titanium alloys, the most common equilibrium phases are those of α and β, phases. The transformation of high temperature phase can occur by martensitic or by a diffusion controlled nucleation and growth process depending on cooling rate and alloy composition. Their relationship was confirmed for Zirconium by Burger [23, 24] and later for titanium by [22]. This Burgers

relationship is closely obeyed for both the martensite transformation and the diffusional transformations (see **Figure 3**).

Thus, during phase transition from β to α, several slip systems operate within the α titanium alloys system. The most common slip direction is 〈11‾20〉. The <sup>→</sup> *a* direction slip occurs in one (0002) basal plane, three {10‾20} prism planes, and six {10‾11} pyramidal planes [21, 26–29]. They are therefore responsible for the four independent slip systems. The fifth slip system required for homogeneous deformation of polycrystals (Von-Mises criteria) is provided by the two basal slips that occur in the 〈112‾3〉 direction and in the {11‾22} plane. If the slip systems are unable to operate, twinning occurs in α titanium. The main twinning modes are {10‾12} and {11‾21} in tension and {1122} in tension and {1122} in compression loading [21, 30–32].

#### **4.2 Martensitic transformation in titanium metal and alloys**

Phase transformation provides an effective way to modify microstructure and property of solids. It becomes part of a deformation process, when it can be activated by mechanical or other physical forces. Titanium and its alloy undergo a series of stable and metastable allotropic transformations, depending on alloy elements and process. The exact transus point is dependent on the composition and processing treatment for the alloy. Apart from the stable phase(s), other metastable phases can emerge in a quenched alloy such as martensite with hexagonal structure, martensite with orthorhombic structure or the metastable β phase (see **Figure 4**).

Some striking characteristics distinguish martensite transformations from any other [34]. First, the martensite phase is either a substitutional or interstitial solid solution. Secondly, the transformation takes place in a very short time (i.e.) very rapid. This can only be measured by high speed cameras. The complexity inherent in its measurements is an added problem to its study. The third is that it is accompanied by a shape change (surface relief) of a definite value. This has been confirmed metallographic ally by scratch line test [34]. The fourth is that martensite crystal has a specific habit plane; interfacing between the parent phase and martensite phase

#### **Figure 3.**

*Schematic illustration of the Burgers' lattice correspondence model of bcc-hcp transformation. The two simultaneous shears are marked by the red ellipses [25].*

**129**

**Figure 5.**

*to (a) simple shear, (b) angular distortion.*

**Figure 4.**

*Self-Healing in Titanium Alloys: A Materials Science Perspective*

which lie along the shear plane during the transformation. This implies that an orientation relationship exist between the two phase lattices. The presence of lattice defects will necessarily exist in martensite crystal due to shearing. The list of alloy components where martensitic transformation can be exploited are not exhaustible

*A schematic illustration showing the lattice correspondence between the β and α″ phases, after Kym et al. [33].*

The recent innovative drive for Ti-alloys has drifted to many other areas such as the formation of SIM for electrical appliances, GUM metal for industrial applications, bio-implants resulting from its excellent compatibility with body tissues, better the mechanical and physical properties. There is an agreement among researchers that the formation is due to deformation by twinning (A twinning process is shown in **Figure 5** as culled adapted from [38]) of β-phase material to a type of martensite α″ with

*Schematic 2D representation of a collective displacement of the atoms during deformation twinning according* 

and still growing with research in the field of Titanium [35–37].

*DOI: http://dx.doi.org/10.5772/intechopen.92348*

*Self-Healing in Titanium Alloys: A Materials Science Perspective DOI: http://dx.doi.org/10.5772/intechopen.92348*

#### **Figure 4.**

*Advanced Functional Materials*

fusional transformations (see **Figure 3**).

relationship is closely obeyed for both the martensite transformation and the dif-

slip occurs in one (0002) basal plane, three {10‾20} prism planes, and six {10‾11} pyramidal planes [21, 26–29]. They are therefore responsible for the four independent slip systems. The fifth slip system required for homogeneous deformation of polycrystals (Von-Mises criteria) is provided by the two basal slips that occur in the 〈112‾3〉 direction and in the {11‾22} plane. If the slip systems are unable to operate, twinning occurs in α titanium. The main twinning modes are {10‾12} and {11‾21} in

α titanium alloys system. The most common slip direction is 〈11‾20〉. The <sup>→</sup>

tension and {1122} in tension and {1122} in compression loading [21, 30–32].

Phase transformation provides an effective way to modify microstructure and property of solids. It becomes part of a deformation process, when it can be activated by mechanical or other physical forces. Titanium and its alloy undergo a series of stable and metastable allotropic transformations, depending on alloy elements and process. The exact transus point is dependent on the composition and processing treatment for the alloy. Apart from the stable phase(s), other metastable phases can emerge in a quenched alloy such as martensite with hexagonal structure, martensite with orthorhombic structure or the metastable β phase (see **Figure 4**). Some striking characteristics distinguish martensite transformations from any other [34]. First, the martensite phase is either a substitutional or interstitial solid solution. Secondly, the transformation takes place in a very short time (i.e.) very rapid. This can only be measured by high speed cameras. The complexity inherent in its measurements is an added problem to its study. The third is that it is accompanied by a shape change (surface relief) of a definite value. This has been confirmed metallographic ally by scratch line test [34]. The fourth is that martensite crystal has a specific habit plane; interfacing between the parent phase and martensite phase

*Schematic illustration of the Burgers' lattice correspondence model of bcc-hcp transformation. The two* 

**4.2 Martensitic transformation in titanium metal and alloys**

Thus, during phase transition from β to α, several slip systems operate within the

*a* direction

**128**

**Figure 3.**

*simultaneous shears are marked by the red ellipses [25].*

*A schematic illustration showing the lattice correspondence between the β and α″ phases, after Kym et al. [33].*

which lie along the shear plane during the transformation. This implies that an orientation relationship exist between the two phase lattices. The presence of lattice defects will necessarily exist in martensite crystal due to shearing. The list of alloy components where martensitic transformation can be exploited are not exhaustible and still growing with research in the field of Titanium [35–37].

The recent innovative drive for Ti-alloys has drifted to many other areas such as the formation of SIM for electrical appliances, GUM metal for industrial applications, bio-implants resulting from its excellent compatibility with body tissues, better the mechanical and physical properties. There is an agreement among researchers that the formation is due to deformation by twinning (A twinning process is shown in **Figure 5** as culled adapted from [38]) of β-phase material to a type of martensite α″ with

#### **Figure 5.**

*Schematic 2D representation of a collective displacement of the atoms during deformation twinning according to (a) simple shear, (b) angular distortion.*

orthorhombic structure. It is also thought that over a narrow compositional average the β-phase material is said to transform to martensite under an applied stress.

#### **4.3 Shape memory effect and super-elasticity in titanium alloys**

Shape Memory Alloys (SMAs) are special materials with great potential in various engineering applications since they possess a number of unique characteristics, including superior energy dissipation capacity compared to normal metallic materials [39]. Other beneficial properties, apart from SMEs, including superelasticity, favorable damping ability and other important characteristics of shape memory alloys, allow it to be applied in a wide range of fields, including electronic, chemical, medical devices, electricity, aerospace, etc. [40].

The value and demand of SMAs was not positively understood for most engineering and technological applications until William Buehler and Frederick Wang discovered the shape memory effect (SME) in a nickel-titanium (NiTi) alloy in 1962 [41]. Thereafter, the use of SMAs, has expanded and the research interests and patents have become quite large. Examples of the possible beneficiaries of these materials abound in a variety of fields, such as automobile and mechanical engineering applications [42, 43], automotive [42], aerospace [44], mini actuators and micro-electromechanical systems (MEMS) [45], robotics [35], biomedical [36] and even in clothing/fashion industries [37]. Titanium (Ti) alloys are one of the most important SMAs and until now, development of new Ti-based SMAs is still one of the most important directions of metal intelligent materials. The Ti-Nb based [35–37, 45], Ti-Ta based [46], Ti-Mo based [46–48] and Ti-Zr based [49] SMAs are developed in recent years.

The interest in using shape memory alloys (SMAs) stems from the fact that they can "remember" their original shape. When subjected to an external force above a threshold, they exhibit stress-induced martensitic transformation from austenite into martensite through twinning, and can recover the apparent permanent strains, returning to the original form. An Illustration of the superelastic response in shape memory alloys (deformation at a temperature N austenite finish temperature Af). This important attribute exhibited by many titanium based alloy can be exploited to accelerate the self-healing process in metallic materials. In addition by adjusting the hysteresis width can allow materials scientists to precisely adjust temperature change ∆T during the self-healing process. One of the probable set back is the functional degradation in properties that manifested as a reduction in the superelastic strains (εSE) and accumulation of residual strains (εresidual).

#### **4.4 Diffusion in titanium metals and alloys**

It is well known that atoms in almost all metals and alloys crystallize or has a tendency to pack in dense structural arrangement at room temperature, due to the strong bond that bind atom together in a metallic substance. This architecture determines how fast or how slow a healing mechanism would autonomously respond when a metallic component fails. Therefore, it is rational to deduce that triggering autonomous self-healing should be easier if the rate of diffusion of the part is high enough to be transported to the point where it fails. The molecular diffusion or atomic transport of matter by diffusion is represented ideally by the net flux, J, of atoms per second per unit area of reference plane in opposite directions (±x) in the presence of a concentration gradient, dc/dx, as given by Fick's first law:

$$J = -D \left( dc / d\mathbf{x} \right) . \tag{1}$$

**131**

**Figure 6.**

*Self-Healing in Titanium Alloys: A Materials Science Perspective*

where *D* is the diffusion coefficient, given by:

*D* = *Do* exp.(−*Q*/R*T*). (2)

There have been several improvement in the study of diffusivity in the case of in titanium alloys, beginning with the work documented in German by Zwicker in 1974 [50]. On transformation from the α (hcp) to the β(bcc) phase, the diffusivity shows some changes. With respect to the widely used Ti-6Al-4V alloy. Liu and Welsch in 1987 studied the diffusivities of oxygen, aluminum, and vanadium in α and β titanium [33]. Zwicker observed in the plot that the self-diffusion of titanium in the β phase is about three orders of magnitude faster than the self-diffusion in the α phase [50] (see **Figure 6**). The diffusion rates of substitutional elements in the β phase can be either slower or faster than the self-diffusion of titanium [33].

*Temperature dependence of self- and temperature diffusivity in β type titanium alloy (as called from [33]).*

kJ/mol) and R (the gas constant, 8.314510 J/Kmol) are all constants, so the only variable is the temperature *T*, in Kelvin. In other words, In Dvs. 1/T forms a single straight line. In the case of Ti, higher temperature induces thermal diffusion needed to increase the kinetic energy needed to overcome the binding energy of the metallic substance. In addition to Vacant lattice sites or other in homogeneities within a metal, molecular transport can also be influenced by pressure and electrical and magnetic activations, chemical process and mechanical agitation of atoms.

/s), *Q* (the activation energy for diffusion in

*DOI: http://dx.doi.org/10.5772/intechopen.92348*

*Do* (the frequency factor in cm<sup>2</sup>

*Self-Healing in Titanium Alloys: A Materials Science Perspective DOI: http://dx.doi.org/10.5772/intechopen.92348*

where *D* is the diffusion coefficient, given by:

*Advanced Functional Materials*

developed in recent years.

orthorhombic structure. It is also thought that over a narrow compositional average the β-phase material is said to transform to martensite under an applied stress.

Shape Memory Alloys (SMAs) are special materials with great potential in various engineering applications since they possess a number of unique characteristics, including superior energy dissipation capacity compared to normal metallic materials [39]. Other beneficial properties, apart from SMEs, including superelasticity, favorable damping ability and other important characteristics of shape memory alloys, allow it to be applied in a wide range of fields, including electronic, chemi-

The value and demand of SMAs was not positively understood for most engineering and technological applications until William Buehler and Frederick Wang discovered the shape memory effect (SME) in a nickel-titanium (NiTi) alloy in 1962 [41]. Thereafter, the use of SMAs, has expanded and the research interests and patents have become quite large. Examples of the possible beneficiaries of these materials abound in a variety of fields, such as automobile and mechanical engineering applications [42, 43], automotive [42], aerospace [44], mini actuators and micro-electromechanical systems (MEMS) [45], robotics [35], biomedical [36] and even in clothing/fashion industries [37]. Titanium (Ti) alloys are one of the most important SMAs and until now, development of new Ti-based SMAs is still one of the most important directions of metal intelligent materials. The Ti-Nb based [35–37, 45], Ti-Ta based [46], Ti-Mo based [46–48] and Ti-Zr based [49] SMAs are

The interest in using shape memory alloys (SMAs) stems from the fact that they can "remember" their original shape. When subjected to an external force above a threshold, they exhibit stress-induced martensitic transformation from austenite into martensite through twinning, and can recover the apparent permanent strains, returning to the original form. An Illustration of the superelastic response in shape memory alloys (deformation at a temperature N austenite finish temperature Af). This important attribute exhibited by many titanium based alloy can be exploited to accelerate the self-healing process in metallic materials. In addition by adjusting the hysteresis width can allow materials scientists to precisely adjust temperature change ∆T during the self-healing process. One of the probable set back is the functional degradation in properties that manifested as a reduction in the superelastic

It is well known that atoms in almost all metals and alloys crystallize or has a tendency to pack in dense structural arrangement at room temperature, due to the strong bond that bind atom together in a metallic substance. This architecture determines how fast or how slow a healing mechanism would autonomously respond when a metallic component fails. Therefore, it is rational to deduce that triggering autonomous self-healing should be easier if the rate of diffusion of the part is high enough to be transported to the point where it fails. The molecular diffusion or atomic transport of matter by diffusion is represented ideally by the net flux, J, of atoms per second per unit area of reference plane in opposite directions (±x) in the presence of a concentration gradient, dc/dx, as given by Fick's

*J* = −*D*(*dc*/*dx*). (1)

strains (εSE) and accumulation of residual strains (εresidual).

**4.4 Diffusion in titanium metals and alloys**

**4.3 Shape memory effect and super-elasticity in titanium alloys**

cal, medical devices, electricity, aerospace, etc. [40].

**130**

first law:

$$D = Do \exp.\text{(-}Q/RT\text{)}.\tag{2}$$

*Do* (the frequency factor in cm<sup>2</sup> /s), *Q* (the activation energy for diffusion in kJ/mol) and R (the gas constant, 8.314510 J/Kmol) are all constants, so the only variable is the temperature *T*, in Kelvin. In other words, In Dvs. 1/T forms a single straight line. In the case of Ti, higher temperature induces thermal diffusion needed to increase the kinetic energy needed to overcome the binding energy of the metallic substance. In addition to Vacant lattice sites or other in homogeneities within a metal, molecular transport can also be influenced by pressure and electrical and magnetic activations, chemical process and mechanical agitation of atoms.

There have been several improvement in the study of diffusivity in the case of in titanium alloys, beginning with the work documented in German by Zwicker in 1974 [50]. On transformation from the α (hcp) to the β(bcc) phase, the diffusivity shows some changes. With respect to the widely used Ti-6Al-4V alloy. Liu and Welsch in 1987 studied the diffusivities of oxygen, aluminum, and vanadium in α and β titanium [33]. Zwicker observed in the plot that the self-diffusion of titanium in the β phase is about three orders of magnitude faster than the self-diffusion in the α phase [50] (see **Figure 6**). The diffusion rates of substitutional elements in the β phase can be either slower or faster than the self-diffusion of titanium [33].

**Figure 6.** *Temperature dependence of self- and temperature diffusivity in β type titanium alloy (as called from [33]).*

#### **Figure 7.**

*Temperature dependence of impurity diffusion coefficient in* α *titanium: Co, Fe, Ni, Mn, Cr and P in single crystal and Si, Al and Ti in polycrystal, culled from [33].*

Al and Mo are shown as examples of slow diffusing elements from the group of slow diffusing elements, Others includes the other alloying elements, such as, V and Sn, which are close to Al, and Nb lies in between Al and Mo. Element Fe is shown as an example in the group of fast diffusing elements in the figure. However, Ni is slightly faster, whereas Cr and Mn fall in between Fe and the βTi-self-diffusion line.

Subsequently systematic measurements were hitherto made for the diffusion of Fe [51] Ni, [52] Mn, [53]Cr [54] and P [55]. On the other hand, Raiszinen and Keinonen measured diffusivities of Al [56] and Si [56] in polycrystalline Ti by a nuclear-reaction method. A detailed analysis of the data is compiled in the form of Arrhenius plots in the review by [57] and presented in **Figure 7**. The findings has shown that transition metal elements and phosphorus exhibit fast diffusion, which are three to five orders of magnitude faster than the self-diffusion. While, measurements done on ultrahigh purity α titanium with respect to Fe, Ni, and Co impurities resulted in very low diffusivity rates for self-diffusion in titanium and about two orders of magnitude slower than Fe, Co, and Ni [58].

#### **5. Self-healing concepts in titanium based materials**

It is well known that research in the field metallic self-healing is still in its infancy stage. Self-healing metallic materials has received attention only in the past decade [13, 18]. While previous reviews on self-healing materials [59] have focused

**133**

*Self-Healing in Titanium Alloys: A Materials Science Perspective*

used to engineer self-healing in Ti-metals and its alloys.

on describing the various routes to obtain self-healing mostly in polymeric materials, the present chapter is directed toward physical or chemical mechanism can be

Self-healing coatings inspired by biological systems possess the ability to repair physical damage or recover functional performance with minimal or no intervention. When the kinetics are extremely fast, the phenomenon is controlled by the diffusion (mass transport) of the species that enters or leaves the surface of the material under consideration. Consequently, the composition of the system will also be changing. Analogous effects have been found by other workers in systems of biological interest, e.g., with processes involving membranes and enzymes. It is well known that the basic diffusion controlled modes, such as surface diffusion, Ds; grain boundary diffusion, Dgb; vacancy diffusion, Dv and pipe diffusion, Dp, are fundamental to determine the rate of atomic diffusion in polycrystalline metals. In general, surface diffusion occurs much faster than grain boundary diffusion, and grain boundary diffusion occurs much faster than lattice diffusion. Atomic diffusion and indeed electrochemicalinduced self-healing in polycrystalline materials is therefore often modelled using a combination of diffusion kinetics (see previous section). More details of the transformation modes in titanium have been discussed elsewhere [31–33]. For this, electrochemically induced self-healing are said to be a good strategy to be exploited in metals. For instance, a damage to oxide films, which normally protect the surfaces of Ti materials from corrosion, can be repaired by reoxidation in air. Recently, Gang Lu et al. [60] studied the oxidation of a polycrystalline titanium surface by oxygen and water and found that at 150 K O2 can oxidize Ti to Ti5, Ti3 and Ti2, while exposure of Ti to H2O at this temperatures only produces Ti2 species. At temperatures above 300 K, H2O can by both O2 and H2O slightly increases a further oxidize Ti2 to higher oxidation states. They observed rising temperature promotes the diffusion of oxygen into the bulk of the sample, which increasing overall oxidation. This is because the thickness of the oxide coating on Ti surface depends on both the duration of O2 exposure and on the sample temperature. At a given temperature, Ti oxidation by both O2

Additionally, a crack on the surface of a titanium component can also be healed, when the oxidation reaction products fill up the crack cap. Therefore, cracks developed due to operational related stress can be autonomously self-healed or repaired by re-oxidative reaction that occur in Ti-based materials. Although self-healing coatings are considered as an alternative route for efficient anti-corrosion protection, intense research and development effort are been done in the area of corrosion protection coatings of metals and alloys. However, in order to improve the equipment service prediction capabilities of infrastructure, the use of Ti-based materials in infrastructures are beneficial as it can act as a second line of safety assurance even after the coating has failed. In this context, autonomic healing materials respond without external intervention to environmental stimuli, and have great potential for advanced engineering systems [61]. However, the limitation of this self-healing approach is that the extent of oxidation depends on sample temperature. A recent study identified 550–600 K as maximum oxidation in Ti based alloy. Upon heating the oxidized Ti

, which is effective.

*DOI: http://dx.doi.org/10.5772/intechopen.92348*

**5.1 Electrochemically induced self-healing**

and H2O slightly increases as exposure increases.

above 850 K the titanium oxide layer is completely reduced to Tio

This mechanism is based on a thermoelastic displacive phase transformation design methodology. Certain strongly ordered intermetallic systems exhibit

**5.2 Thermally activated solid phase healing in titanium**

on describing the various routes to obtain self-healing mostly in polymeric materials, the present chapter is directed toward physical or chemical mechanism can be used to engineer self-healing in Ti-metals and its alloys.
