**4. Diffusion bonding**

*Magnesium - The Wonder Element for Engineering/Biomedical Applications*

*Weld produced by friction stir welding and the microstructure generated [26].*

*Hardness profiles across Mg-alloys welded using friction stir welding and TIG welding.*

*EPMA results of the FSW region of the sample welded by using a weld-pitch ratio of 1400/40 r/mm in the cross-section perpendicular to the tool transverse direction. Element distribution: (A) Al and (B) Mg [36].*

**88**

**Figure 7.**

**Figure 6.**

**Figure 5.**

Diffusion bonding is a non-conventional solid-state welding process that can join a variety of materials in the solid state below the melting point of the base materials. The core mechanism involves the interdiffusion of atoms between the faying surfaces at the interface [37]. The process is frequently completed by one of two methods; firstly, solid-state diffusion bonding in which the base metals to be joined are heated to approximately 60% of the melting temperature of the metals under the influence of static load. Melting at the interface of the faying surfaces is prevented, however, interdiffusion of the diffusing species leads to the formation of solid-state bond. The second method is referred to as a transient liquid phase (TLP) diffusion bonding in which an interlayer is placed between the metals to be joined. Interdiffusion between the interlayer and the base metals facilitates the formation of a eutectic reaction which transitions from liquid to solid by isothermal solidification as the composition of the eutectic liquid changes due to diffusion. Each method will be further discussed below. A schematic of the diffusion bonding process is shown in **Figure 8**.

#### **4.1 Solid-state diffusion bonding of magnesium alloys**

Solid state diffusion bonding occurs due to the migration of atoms at the interface of the materials to be joined. The faying surfaces do not undergo any metallurgical discontinuity and as a result, the mechanical and microstructural property

**Figure 8.** *Schematic of the diffusion bonding process.*

of the joint formed closely resembles those of the base materials [38]. Solid-state diffusion bonding goes through three distinct stages. The first stage includes; contact of the faying surfaces. In this stage, removal of surface roughness and irregularity play a significant role in ensuring good contact between the faying surfaces. Secondly, micro-plastic deformations at the interface, as the pressure inside the voids decreases and finally, grain boundary migration that fills the shrinkage and minimizes surface free energy. Surface preparation typically includes generation of a smooth surface and elimination of any oxides from the faying surfaces. Application of the pressure at the first stage ensures a good contact between the faying surfaces. Bonding pressure, temperature, and holding time are critical to achieving good bonds. Eq. (1) shows that the bonding temperature is proportional to the coefficient of the diffusing species. While the applied pressure ensures metal–metal contact and micro-plastic deformation of asperities at the interface of the faying surfaces.

$$\mathbf{D} = \mathbf{D}\_o \mathbf{e}^{-\frac{Q}{RT}} \tag{1}$$

**91**

**Figure 9.**

*the reaction layer at the bond interface;*

*Dissimilar Welding and Joining of Magnesium Alloys: Principles and Application*

during dissimilar bonding and maximizes joint strength.

Mg17Al12 where 43.95 wt.% of the compound is Al [41].

determine the kinetic parameters involved in the diffusion process.

magnesium alloys. Unfortunately, the dissimilar joining of magnesium alloys to other metals is significantly inhibited by the differences in the properties of the materials such as melting temperature, the coefficient of thermal expansion and thermal conductivity. Solid-state diffusion bonding limits intermetallic formations

**Figure 9** shows the SEM micrograph of a solid-state diffusion bond formed between titanium (Ti-6Al-4V) and magnesium (AZ31) for 60 minutes. The width of the reaction layer at the interface appears to be in the form of dispersed particles approximately 20 μm in length. EDS point analysis of the samples revealed the formation of several compounds at the bond interface. The results show that holding the sample at the bonding temperature, Al reacts with Ti and Mg leading to the precipitation of the TiAl3 and Ti2Mg3Al18 intermetallic compounds at the Ti interface. The Mg17Al12 intermetallic compounds appear to have formed at the Mg grain boundaries. The Mg17Al12 compound is believed to have been produced by a eutectic reaction between Al and Mg. The joint formation was attributed to metallurgical bonding driven leading to the formation of TiAl3 and Ti2Mg3Al18. It is expected that the differences in the melting temperature of the two alloys at the bonding temperature of 500°C will cause the Mg sample to plastically deform ensuring complete contact with the Ti sample. Hidetoshi Somekawa [39] demonstrated that diffusion bonding of superplastic magnesium sheets is heavily dependent on the grain size of the material. For samples with grain size in the range of 11–15 μm bond strength of approximately 92% of the base metal strength is possible. According to the Ti-Mg-Al ternary phase diagram the following phases; Mg2Al3, Mg17Al12, TiAl3 and TiAl2 are likely to form at the bonding temperature and pressure used in this study [6]. The Gibbs energy for the formation of TiAl3 is approximately 234 kJ/mol, TiAl2 is 237 kJ/mol [40]. The diffusion of Al to the bond interface from Ti-side to the Mg-side led to the formation of a compound having a stoichiometric composition of

The Mg17Al12 is an intermetallic compound with a Gibbs free energy of formation of −6 kJ/mol at a temperature range of 700–1000 K. The Gibbs energy of formation increased to −3.9 kJ/mol. From the Gibbs free energy data, the Mg17Al12 intermetallic compound is most negative and therefore is expected to form first at the Mg-interface. The Gibbs energy for the formation of the ternary compound Ti2Mg3Al18 was found to be approximately −15 kJ/mol [42]. The width of the reaction layer that forms at the interface is believed to be time dependent. As such, the layer thickness bears a direct relation with growth kinetics. The average thicknesses of the TiAl3 and Ti2Mg3Al18 layers and the total intermetallic layer was used to

*(A) Solid-state diffusion bonding of Ti and Mg for 60 minutes; (B) magnified region of the bond-line showing* 

*DOI: http://dx.doi.org/10.5772/intechopen.85111*

Where, *D* is the diffusion coefficient of the diffusing species, *D*o is the pre-exponential term, *R* is gas constant and *T* is the bonding temperature. While bonding/hold time can be directly related to the volume of the diffusing species using Fick's second law of diffusion. Increasing the hold time is expected to enhance diffusion and strength the bond.

Diffusion bonding has been studied extensively as a method for both similar and dissimilar joining of magnesium alloys. The high temperature used in fusion welding process leads to the formation of intermetallic compounds, which have been shown to be detrimental to the mechanical performance of the joints when joining

#### *Dissimilar Welding and Joining of Magnesium Alloys: Principles and Application DOI: http://dx.doi.org/10.5772/intechopen.85111*

magnesium alloys. Unfortunately, the dissimilar joining of magnesium alloys to other metals is significantly inhibited by the differences in the properties of the materials such as melting temperature, the coefficient of thermal expansion and thermal conductivity. Solid-state diffusion bonding limits intermetallic formations during dissimilar bonding and maximizes joint strength.

**Figure 9** shows the SEM micrograph of a solid-state diffusion bond formed between titanium (Ti-6Al-4V) and magnesium (AZ31) for 60 minutes. The width of the reaction layer at the interface appears to be in the form of dispersed particles approximately 20 μm in length. EDS point analysis of the samples revealed the formation of several compounds at the bond interface. The results show that holding the sample at the bonding temperature, Al reacts with Ti and Mg leading to the precipitation of the TiAl3 and Ti2Mg3Al18 intermetallic compounds at the Ti interface. The Mg17Al12 intermetallic compounds appear to have formed at the Mg grain boundaries. The Mg17Al12 compound is believed to have been produced by a eutectic reaction between Al and Mg. The joint formation was attributed to metallurgical bonding driven leading to the formation of TiAl3 and Ti2Mg3Al18. It is expected that the differences in the melting temperature of the two alloys at the bonding temperature of 500°C will cause the Mg sample to plastically deform ensuring complete contact with the Ti sample. Hidetoshi Somekawa [39] demonstrated that diffusion bonding of superplastic magnesium sheets is heavily dependent on the grain size of the material. For samples with grain size in the range of 11–15 μm bond strength of approximately 92% of the base metal strength is possible. According to the Ti-Mg-Al ternary phase diagram the following phases; Mg2Al3, Mg17Al12, TiAl3 and TiAl2 are likely to form at the bonding temperature and pressure used in this study [6]. The Gibbs energy for the formation of TiAl3 is approximately 234 kJ/mol, TiAl2 is 237 kJ/mol [40]. The diffusion of Al to the bond interface from Ti-side to the Mg-side led to the formation of a compound having a stoichiometric composition of Mg17Al12 where 43.95 wt.% of the compound is Al [41].

The Mg17Al12 is an intermetallic compound with a Gibbs free energy of formation of −6 kJ/mol at a temperature range of 700–1000 K. The Gibbs energy of formation increased to −3.9 kJ/mol. From the Gibbs free energy data, the Mg17Al12 intermetallic compound is most negative and therefore is expected to form first at the Mg-interface. The Gibbs energy for the formation of the ternary compound Ti2Mg3Al18 was found to be approximately −15 kJ/mol [42]. The width of the reaction layer that forms at the interface is believed to be time dependent. As such, the layer thickness bears a direct relation with growth kinetics. The average thicknesses of the TiAl3 and Ti2Mg3Al18 layers and the total intermetallic layer was used to determine the kinetic parameters involved in the diffusion process.

#### **Figure 9.**

*(A) Solid-state diffusion bonding of Ti and Mg for 60 minutes; (B) magnified region of the bond-line showing the reaction layer at the bond interface;*

*Magnesium - The Wonder Element for Engineering/Biomedical Applications*

of the joint formed closely resembles those of the base materials [38]. Solid-state diffusion bonding goes through three distinct stages. The first stage includes; contact of the faying surfaces. In this stage, removal of surface roughness and irregularity play a significant role in ensuring good contact between the faying surfaces. Secondly, micro-plastic deformations at the interface, as the pressure inside the voids decreases and finally, grain boundary migration that fills the shrinkage and minimizes surface free energy. Surface preparation typically includes generation of a smooth surface and elimination of any oxides from the faying surfaces. Application of the pressure at the first stage ensures a good contact between the faying surfaces. Bonding pressure, temperature, and holding time are critical to achieving good bonds. Eq. (1) shows that the bonding temperature is proportional to the coefficient of the diffusing species. While the applied pressure ensures metal–metal contact and

micro-plastic deformation of asperities at the interface of the faying surfaces.

Where, *D* is the diffusion coefficient of the diffusing species, *D*o is the pre-exponential term, *R* is gas constant and *T* is the bonding temperature. While bonding/hold time can be directly related to the volume of the diffusing species using Fick's second law of diffusion. Increasing the hold time is expected to enhance

Diffusion bonding has been studied extensively as a method for both similar and dissimilar joining of magnesium alloys. The high temperature used in fusion welding process leads to the formation of intermetallic compounds, which have been shown to be detrimental to the mechanical performance of the joints when joining

*RT* (1)

*<sup>D</sup>* <sup>=</sup> *Do <sup>e</sup>* −\_\_\_ *<sup>Q</sup>*

diffusion and strength the bond.

*Schematic of the diffusion bonding process.*

**90**

**Figure 8.**

The data collected shows that as the bonding time increases the width of the reaction layer. The width also increased as predicted by the parabolic law shown in Eq. (2).

$$\mathcal{K} = k \mathcal{t}^n \tag{2}$$

Where *k* is the rate factor, *t* the diffusion time, and *n* the time exponent. **Figure 10** shows the relationship between the bonding time and the thickness of the reaction layer formed at the interface and shows that as the bonding time increased the thickness of the reaction layer also increased. When the parabolic rate law is applied to the results, the rate coefficient *k* was calculated to be 15.7 × 10<sup>−</sup><sup>7</sup> m/s. The value for *n* was assumed to be 0.5 since the growth of the reaction layer was assumed to be controlled by inter-diffusion. When the calculated rate coefficient is substituted into the parabolic rate law shown in Eq. (1) the results show that the predicted thickness of the intermetallic layer was overestimated by the model.

The difference between the properties of the base metal and the bonded zone was highlighted by the micro-hardness measurements plotted in **Figure 11** as a function of bonding time. The micro-hardness values were measured across the joint starting at 500 μm from the joint center. The figure shows that the hardness of the Ti sample fluctuated between 390 VHN and 420 VHN up to 100 μm from the joint center as the bonding time was increased from 10 to 60 minutes. The hardness within the joint center was observed to decrease to between 190VHN at 10 minutes bonding time and 250 VHN at 60 minutes bonding time. The hardness of the Mg sample was found to be significantly lower than that of the Ti sample with a hardness ranging from 60 VHN at 10 minutes bonding time to 65 VHN. The variation of the hardness number across the interface was attributed to the differences between the properties of Ti and Mg. The hardness at the center of the bond is believed to have been caused by the formation of the reaction layer at the joint interface [43]. The reaction layer was shown to be made-up of TiAl3 and Ti2Mg3Al18 intermetallic compounds dispersed within the joint region.

#### **4.2 Transient liquid phase diffusion bonding**

The transient liquid phase diffusion bonding process occurs due to the formation of a eutectic liquid at the interface between the faying surfaces. The eutectic reaction may form due to interdiffusion between the base metals or an interlayer and base metals which leads to the formation of a eutectic composition. Alternatively, a

#### **Figure 10.**

*(A) Shows the relationship between the thickness of the reaction layer and the bonding time; (B) shows the predicted relationship between the thickness of the reaction layer and the bonding time according to the parabolic rate law.*

**93**

*Dissimilar Welding and Joining of Magnesium Alloys: Principles and Application*

thin interlayer having a eutectic composition and melts at the bonding temperature. The formation of the liquid displaces the surface oxides at the bond interface and reduces the bonding pressure requirement normally used to achieve surface contact during solid-state diffusion bonding. The advantage of the process is the potential for minimizing microstructural changes and intermetallic formations within the bond region and achieving joints with higher strength. However, TLP bonding can be a lengthy process that required four distinct stages (heating, dissolution and widen, isothermal solidification and homogenization) for completion [44]. Important parameters studied in TLP bonding include; interlayer composition and

During TLP bonding the thickness and composition of the interlayer used are critical to the volume of liquid formed during bonding and invariably affects the quality of the bonds produced [46]. The scientific literature shows that suitable interlayers must allow eutectic melting while limiting the volume of the liquid that forms in order to control the width of the interface and shorten the homogenization stage of bonding [3]. Additionally, the wettability of the base metals by the molten interlayer is critical to displacing surface oxides in order to form a joint [2]. When suitable interlayers are used, the formation of intermetallic compounds (IMC) can be prevented or significantly reduced. The interlayers can be used in the form of; thin foils, fine powder, compact powder or paste, electroplated and vapor deposited coatings [47].

Selection of the interlayer composition depends on the base materials being joined. The literature shows that materials such as Sn, Ag, Al, Cu, Ni, Cr, V, and Zn are frequently used as interlayers during diffusion bonding. Most commonly used interlayers when joining magnesium alloys include Al, Cu, and Ni since these materials' leads to a eutectic reaction with magnesium which lowers the bonding temperature and catalyzes diffusion. When Al interlayer was used by Sun et al. [3] to bond Mg alloy MgAZ31, brittle IMC, *CuMg*2 was formed at the joint interface. However, increasing bonding temperature slightly increased the strength of the bond, as a result of microstructure homogenization for short bonding time. When the bonding time was increased beyond 60 minutes grain coarsening occurred

thickness, bonding temperature and bonding time [45].

*Micro-hardness measurements across the joint region as a function of bonding time.*

*4.2.1 Interlayer thickness and composition*

**Figure 11.**

which resulted in a reduction in joint strength.

*DOI: http://dx.doi.org/10.5772/intechopen.85111*

*Dissimilar Welding and Joining of Magnesium Alloys: Principles and Application DOI: http://dx.doi.org/10.5772/intechopen.85111*

#### **Figure 11.**

*Magnesium - The Wonder Element for Engineering/Biomedical Applications*

*x* = *kt*

compounds dispersed within the joint region.

**4.2 Transient liquid phase diffusion bonding**

Eq. (2).

15.7 × 10<sup>−</sup><sup>7</sup>

mated by the model.

The data collected shows that as the bonding time increases the width of the reaction layer. The width also increased as predicted by the parabolic law shown in

Where *k* is the rate factor, *t* the diffusion time, and *n* the time exponent. **Figure 10** shows the relationship between the bonding time and the thickness of the reaction layer formed at the interface and shows that as the bonding time increased the thickness of the reaction layer also increased. When the parabolic rate law is applied to the results, the rate coefficient *k* was calculated to be

reaction layer was assumed to be controlled by inter-diffusion. When the calculated rate coefficient is substituted into the parabolic rate law shown in Eq. (1) the results show that the predicted thickness of the intermetallic layer was overesti-

The difference between the properties of the base metal and the bonded zone was highlighted by the micro-hardness measurements plotted in **Figure 11** as a function of bonding time. The micro-hardness values were measured across the joint starting at 500 μm from the joint center. The figure shows that the hardness of the Ti sample fluctuated between 390 VHN and 420 VHN up to 100 μm from the joint center as the bonding time was increased from 10 to 60 minutes. The hardness within the joint center was observed to decrease to between 190VHN at 10 minutes bonding time and 250 VHN at 60 minutes bonding time. The hardness of the Mg sample was found to be significantly lower than that of the Ti sample with a hardness ranging from 60 VHN at 10 minutes bonding time to 65 VHN. The variation of the hardness number across the interface was attributed to the differences between the properties of Ti and Mg. The hardness at the center of the bond is believed to have been caused by the formation of the reaction layer at the joint interface [43]. The reaction layer was shown to be made-up of TiAl3 and Ti2Mg3Al18 intermetallic

The transient liquid phase diffusion bonding process occurs due to the formation of a eutectic liquid at the interface between the faying surfaces. The eutectic reaction may form due to interdiffusion between the base metals or an interlayer and base metals which leads to the formation of a eutectic composition. Alternatively, a

*(A) Shows the relationship between the thickness of the reaction layer and the bonding time; (B) shows the predicted relationship between the thickness of the reaction layer and the bonding time according to the* 

m/s. The value for *n* was assumed to be 0.5 since the growth of the

*<sup>n</sup>* (2)

**92**

**Figure 10.**

*parabolic rate law.*

*Micro-hardness measurements across the joint region as a function of bonding time.*

thin interlayer having a eutectic composition and melts at the bonding temperature. The formation of the liquid displaces the surface oxides at the bond interface and reduces the bonding pressure requirement normally used to achieve surface contact during solid-state diffusion bonding. The advantage of the process is the potential for minimizing microstructural changes and intermetallic formations within the bond region and achieving joints with higher strength. However, TLP bonding can be a lengthy process that required four distinct stages (heating, dissolution and widen, isothermal solidification and homogenization) for completion [44]. Important parameters studied in TLP bonding include; interlayer composition and thickness, bonding temperature and bonding time [45].

#### *4.2.1 Interlayer thickness and composition*

During TLP bonding the thickness and composition of the interlayer used are critical to the volume of liquid formed during bonding and invariably affects the quality of the bonds produced [46]. The scientific literature shows that suitable interlayers must allow eutectic melting while limiting the volume of the liquid that forms in order to control the width of the interface and shorten the homogenization stage of bonding [3]. Additionally, the wettability of the base metals by the molten interlayer is critical to displacing surface oxides in order to form a joint [2]. When suitable interlayers are used, the formation of intermetallic compounds (IMC) can be prevented or significantly reduced. The interlayers can be used in the form of; thin foils, fine powder, compact powder or paste, electroplated and vapor deposited coatings [47].

Selection of the interlayer composition depends on the base materials being joined. The literature shows that materials such as Sn, Ag, Al, Cu, Ni, Cr, V, and Zn are frequently used as interlayers during diffusion bonding. Most commonly used interlayers when joining magnesium alloys include Al, Cu, and Ni since these materials' leads to a eutectic reaction with magnesium which lowers the bonding temperature and catalyzes diffusion. When Al interlayer was used by Sun et al. [3] to bond Mg alloy MgAZ31, brittle IMC, *CuMg*2 was formed at the joint interface. However, increasing bonding temperature slightly increased the strength of the bond, as a result of microstructure homogenization for short bonding time. When the bonding time was increased beyond 60 minutes grain coarsening occurred which resulted in a reduction in joint strength.

In another study, Jin and Khan [48] studied the use of Ni, as an interlayer while joining the same Mg alloy, and found that the hardness of the joint increased as the bonding time increased as a result of the formation of Mg-Ni IMCs. Research results presented by Alhazaa et al. [49] showed that an optimum bonding time of 20 minutes was attained when bonding Mg AZ31 using Sn coatings. As shown in **Figure 12**, as the bonding time increased the bond line completely disappears, which is an indication of the homogenization of the bond region. The EPMA map shown in **Figure 13**, confirms the diffusion of Sn away from the interface and the homogenization of the composition within the joint region. Nevertheless, the formation of IMC at the joint consequently reduced the bond strength (see **Figure 15**). In contrast, Sun et al. [3] used Cu interlayer and suggested a bonding time of 30 minutes was optimal. The study also showed that the volume of intermetallic compounds formed was inversely proportional to hold time.

The potential for the dissimilar joining of Mg alloys to other metals is significantly greater with TLP bonding. Zhang et al. [50] demonstrated that Ni-coating can be used as an interlayer to join Mg to Al however suitable optimization of the bonding parameters are required to prevent the formation of a *Mg*2 Ni IMCs reaction layer between the metals. Alhazaa et al. [51] observed that the application of Cu coatings and Sn interlayers while bonding Mg AZ31 to the Ti-6Al-4V alloy. Increasing the bonding time also decreased the thickness of the IMC *Mg*2*Cu* compounds and better bonds were formed between these dissimilar metal alloys

(see **Figure 14**). The use of nanoparticle-reinforced composite interlayer during TLP process has also gained interest recently as they reduce grain size within the bond region and increase joint strength [52]. Similarly, Atieh and Khan [53], showed that when Ti-6Al-4V and Mg-AZ31 alloys were bonded using Ni and Cu nanoparticles, the joint formation was enhanced by increasing the rate of isothermal solidification during the TLP bonding process. **Figure 15** shows a TLP bond that was made between

#### **Figure 12.**

*SEM micrograph of region bonded using Cu-coating and Sn-foil combination as interlayers for (A) 10 minutes; (B) 20 minutes; (C) 30 minutes; and (D) 50 minutes [2].*

**95**

**Figure 14.**

*(AZ31) with Sn interlayer and bonded for 30 minutes.*

**Figure 13.**

*Sn-foil interlayer at 20 minutes [2].*

*Dissimilar Welding and Joining of Magnesium Alloys: Principles and Application*

Mg and Al using Cu/Al2O3 interlayer and confirms that the addition of nanoparticles can prevent the formation of a continuous IMC layer at the joint interface which leads to an increase in strength. The composition of the joint region was evaluated using EDS and is shown in **Figure 15(B)**. A summary of recent interlayers studied and the impact of composition on the mechanical performance of the bonds is presented in

*XRD spectrum of the compounds formed at the interface for joint made during similar TLP bonding of Mg* 

*Electron probe micro-analysis (EPMA) micrographs of Mg, Cu, and Sn for bonds made using Cu coating and* 

**Table 1**. The strength coefficient for each bond was estimated using Eq. (3).

*DOI: http://dx.doi.org/10.5772/intechopen.85111*

*Dissimilar Welding and Joining of Magnesium Alloys: Principles and Application DOI: http://dx.doi.org/10.5772/intechopen.85111*

#### **Figure 13.**

*Magnesium - The Wonder Element for Engineering/Biomedical Applications*

In another study, Jin and Khan [48] studied the use of Ni, as an interlayer while joining the same Mg alloy, and found that the hardness of the joint increased as the bonding time increased as a result of the formation of Mg-Ni IMCs. Research results presented by Alhazaa et al. [49] showed that an optimum bonding time of 20 minutes was attained when bonding Mg AZ31 using Sn coatings. As shown in **Figure 12**, as the bonding time increased the bond line completely disappears, which is an indication of the homogenization of the bond region. The EPMA map shown in **Figure 13**, confirms the diffusion of Sn away from the interface and the homogenization of the composition within the joint region. Nevertheless, the formation of IMC at the joint consequently reduced the bond strength (see **Figure 15**). In contrast, Sun et al. [3] used Cu interlayer and suggested a bonding time of 30 minutes was optimal. The study also showed that the volume of intermetallic compounds formed was inversely proportional to

The potential for the dissimilar joining of Mg alloys to other metals is significantly

greater with TLP bonding. Zhang et al. [50] demonstrated that Ni-coating can be used as an interlayer to join Mg to Al however suitable optimization of the bonding parameters are required to prevent the formation of a *Mg*2 Ni IMCs reaction layer between the metals. Alhazaa et al. [51] observed that the application of Cu coatings and Sn interlayers while bonding Mg AZ31 to the Ti-6Al-4V alloy. Increasing the bonding time also decreased the thickness of the IMC *Mg*2*Cu* compounds and better

(see **Figure 14**). The use of nanoparticle-reinforced composite interlayer during TLP process has also gained interest recently as they reduce grain size within the bond region and increase joint strength [52]. Similarly, Atieh and Khan [53], showed that when Ti-6Al-4V and Mg-AZ31 alloys were bonded using Ni and Cu nanoparticles, the joint formation was enhanced by increasing the rate of isothermal solidification during the TLP bonding process. **Figure 15** shows a TLP bond that was made between

*SEM micrograph of region bonded using Cu-coating and Sn-foil combination as interlayers for (A) 10 minutes;* 

bonds were formed between these dissimilar metal alloys

**94**

**Figure 12.**

*(B) 20 minutes; (C) 30 minutes; and (D) 50 minutes [2].*

hold time.

*Electron probe micro-analysis (EPMA) micrographs of Mg, Cu, and Sn for bonds made using Cu coating and Sn-foil interlayer at 20 minutes [2].*

**Figure 14.**

*XRD spectrum of the compounds formed at the interface for joint made during similar TLP bonding of Mg (AZ31) with Sn interlayer and bonded for 30 minutes.*

Mg and Al using Cu/Al2O3 interlayer and confirms that the addition of nanoparticles can prevent the formation of a continuous IMC layer at the joint interface which leads to an increase in strength. The composition of the joint region was evaluated using EDS and is shown in **Figure 15(B)**. A summary of recent interlayers studied and the impact of composition on the mechanical performance of the bonds is presented in **Table 1**. The strength coefficient for each bond was estimated using Eq. (3).

$$m = \frac{\pi\_{bound}}{\pi\_b} \tag{3}$$

Where *m* represents the strength coefficient of the bond, τ*bond* is the strength of the bond and <sup>τ</sup>*b* is the strength of the base metal.
