This section highlights general findings observed with tensile-shear testing of FSSW samples. The detailed mechanism for tensile-shear testing of FSSW joints produced at various process parameters and examination of their weld structure is provided in subsections below.
3.1 Weld Structure Examination:
During FSSW, workpiece material experiences plastic deformation. In case of reinforced FSSW weld, distribution of reinforced particles in welded region altered the weld properties at micro as well as macro structural level. The flow of material during FSSW process is a complex phenomenon and with addition of reinforcement further increases its complications. Cross-sectional macrograph of FSSW welded specimen obtained at 900 rpm of tool rotation speed, 6 sec of pre-dwelling time and 3.0 mm of guiding hole diameter, is shown in Fig. 5, and it was observed that the structure of weld was not symmetrical about weld centre. There were three regions identified at joining inter-face as Unbonded Region (UBR), Bonded Region (BR) and Partly Bonded Region (PBR). The left side of keyhole in which all three regions were observed is as shown in Fig. 5 (a). Bonding region was formed due to proper stirring and mixing of workpieces, resulting in merging of interfaces. During FSSW tool plunging, tool pushed the workpiece material down and deformed it while stirring. The process of stirring and deforming caused flow of material around pin in spiral form and transport material upward. This process caused mixing of two interfaces around pin and created region known as bonded region. After this region, effect of process parameters and heat generated reduced and incomplete material flow caused partial bonding of interfaces. Partially bonded region generally started with a void and then partial diffusion of interfaces took place due to incomplete material flow. After partially bonded region, a space between both interfaces was observed, known as unbonded region. The partially bonded region bent upward due to plunging action known as hook. Therefore, hook is generally a partially bonded region, bent in a radius like a mountain [23]. The void defects were also obtained on either side of keyhole due to improper flow of material and stirring zone was also small. While, cross-sectional view of rest of the sample showed presence of perfectly bonded regions, visually looking good and defect free. The amount of heat input plasticized the material and its flow played a vital role in obtaining defect-free weld [24]. The bonding region of leading (left) side was different from trailing (right) side. The width of bonding region on leading side of keyhole was 731.885 micron as shown in Fig. 5 (b) while on trailing side of keyhole bonding region of width 1005.932 micron was observed. A hook formation starting from the right side with a void was also observed. Although, bonded region was less in this sample but sufficient improvement in weld strength was observed due to incorporation of reinforcements.
The macrograph, optical microscopy and SEM images along with EDS analysis of weld at 3.0 mm guiding hole diameter, 1700 rpm tool rotation speed and 6 sec pre-dwelling time is as shown in Fig. 6. A clear interface between stirring zone and thermo-mechanically affected zone was observed, as shown in Fig. 6 (b) and (c). SEM image of stirring zone of FSSW weld confirms homogeneous distribution of SiC particles in stirring zone which is clearly visible in Fig. 6 (e-f). It is apparent from these micrographs, that there was excellent bonding between reinforcement and matrix. Fig. 6 (g) shows the Electro Dispersive Spectroscopy (EDS) of FSSW weld and validates the presence of SiC particles in stirring zone of FSSW weld. The homogeneous distribution of SiC particles affect weld strength, because it restricts grain growth and act as hindrance for cracks to propagate easily. When tensile-shear load applied, it causes accumulation of plastic stress in region near to particle. When distance between particles is less, i.e., particles are homogeneously and closely placed, plastic flow of matrix gets trapped between reinforcements. The movement of dislocations restricted by particles, leads to increase in flow strength after yielding. When applied load exceeds critical value, stress releases and transferred to other particles. Therefore, the ductility and fracture load of the weld increases [25].
Whereas, Fig.7, shows SEM image of sample obtained at 1300 rpm, 6 sec pre-dwelling time and 3.5 mm guiding hole diameter, in which clusters of SiC particles were observed. This clustering was due to improper material flow and mixing of reinforcements. A cluster was considered as one single big particle, leading to inhomogeneity due to random placement of these particles. Therefore, distance between particles increases. When load is applied, large particles act as nucleation sites, which initiates stress and strain fields and helps in easy breaking of matrix-reinforcement interface. After, damaging interface, released stress fields moves fast to other particles placed at far distance without any significant disturbances and causes decrease in fracture load of weld. Therefore, clustered reinforcements are regarded as the site of initiation of premature failure [26]. The clustering can be avoided by choosing appropriate process parameters to increase heat input for inducing sufficient material flow and mixing of reinforcement in stirring zone. Hence, homogeneous distribution of SiC particles plays an important role in strengthening of the welded joint. The homogeneity of reinforcements was visible in most of the welds, studied in the research investigation.
3.2 Effect of process parameters on Tensile-Shear Load of FSSW weld:
Pictorial representation of tensile-shear testing of FSSW samples is shown in Fig. 8. After experimentation, FSSW welds were prepared for tensile-shear testing. The specimens were made equiaxed by adhering two aluminium pieces of same thickness at edges of workpieces to make weld sample aligned before mounting on UTM followed by gradually increasing tensile-shear load. The behaviour of tensile-shear load with variation in guiding hole diameter i.e., quantity of SiC particles, tool rotation speed and pre-dwelling time is presented in Fig. 9, Fig. 13 and Fig. 15 respectively.
The bahaviour of tensile-shear load with variation in guiding hole diameter i.e., quantity of SiC particles is presented in Fig. 9. It was observed that tensile-shear load significantly increased with increase in guiding hole diameter. The tensile shear load of FSSW weld produced without SiC particles was compared with welds produced with different quantities of SiC particles present in stirring zone, i.e., 25% (2.5 mm diameter), 36% (3.0 mm diameter) and 49% (3.5 mm diameter) of volume fraction. The maximum weld strength of 4952.07 N was obtained in sample produced with 49% volume fraction SiC particles, which was 29.78% higher than weld produced without SiC particles. When rotating tool plunges in workpiece material, SiC particles blend with workpiece material in stirring zone and surrounded the aluminium grains, which causes increased weld strength. This improvement in weld strength was because of reinforced SiC particles having superior mechanical and thermal properties which helps in reducing grain growth occurred due to increased heat input and causes dynamic recrystallization of aluminium grains. However, SiC particles, surrounding aluminium grains, resists grain growth as SiC particles do not expand with increasing heat input, known as Zenner pinning effect, shown in Fig. 10. It can be noted from Fig. 10 that grain boundaries are pinned by reinforcements present around them and their growth is restricted. Thus, refined microstructure was obtained which led to increased tensile-shear strength. While, grain boundaries without reinforcements grow freely and comparatively coarser grains are obtained.
Another possible reason for increase in tensile-shear strength with SiC particles is difference in thermal expansion coefficient of aluminium alloy and SiC particles. When FSSW joints cool down, strain fields formed around SiC particles, leading to formation of dislocations [27]. When FSSW weld subjected to tensile-shear loading, strain fields increase, which leads to piling up of dislocations and these piled up dislocations act as barrier on the way of crack propagation. Therefore, tensile-shear load increases as higher load required to break through piled-up dislocations and SiC particles. Apart from this, SiC particles act as shield that holds the applied forces and does not allow to damage aluminium grains and acts as hindrance for cracks to propagate, shown in Fig. 11. [28]. This continues until force becomes large enough to damage aluminium-SiC interface. It can be concluded that fracture of the joint starts with tearing of aluminium-SiC particles interface then propagate in weld region. However, as number of SiC particles increase, strain and stress field decrease as load is distributed to large number of particles and number of dislocations are blocked. This process limits the plastic zone around SiC particles and when this strain and stress fields increases to a critical value, fracture of particles take place [29].
In addition, a significant increase in strength of SiC-reinforced weld is associated with simultaneous increase in the elongation of the weld when subjected to tensile-shear load. Load-displacement curve of welds at varied reinforcement quantity is shown in Fig. 12. It is clear from Fig. 12 that, sample with reinforcements has elongated more as compared to sample without reinforcements due to grain refinement. As size of grains decreases, ductility of weld increases, which results in higher displacement obtained at higher tensile-shear load. All these factors aid in increasing strength of FSSW welds and shows dominating positive effect of SiC in improving tensile-shear load of the welds.
On the contrary, it was observed from Fig. 13 that tensile-shear load increased with increase in tool rotation speed and then decreased. As tool rotation speed increased from 900 rpm (B1) to 1700 rpm (B3), tensile-shear load is increased from 4110.51 N to 4600.22 N. Whereas, the further increase in tool rotation speed to 2100 rpm (B4), tensile-shear load decreased to 4498.17 N. The apparent reason for this can be explained in terms of elongation, shown in load versus displacement curve (Fig. 14). The elongation of weld first increased from B1 to B3, then reduced at B4. With decrease in size of grains greater number of grain boundaries were obtained and higher load was required to break through these boundaries. This increases ductility of weld, hence increases strength of weld. However, the phenomenon of obtaining smaller or larger grains depends on amount of heat input. With increase in rotational speed of tool, temperature and heat input to the workpiece increases and softens the workpiece which assist in easy material flow and mixing. During this process, refinement of grains in stir zone (SZ) takes place. Therefore, refined microstructure was obtained in SZ, which increase fracture load of joint. However, higher tool rotation speed led to higher heat input, which means weld will take more time to cool down. The slower cooling rate of weld gives more time to grains for dynamic recrystallisation, i.e., coarsening of grains due to grain growth. Therefore, fracture load of weld decreases. The strain rate also plays an important role for grain refinement with increased tool rotation speed. When rotational speed of tool increases up to 1700 rpm, strain rate increases, which has positive effect on weld strength. The increase in strain rate restricts the motion of dislocations which get activated due to mechanical and thermal activities [30]. However, with additional increase in rotational speed of tool, influence of strain rate was overpowered by heat input and its effect on grain modification and ultimately weld strength. This behaviour of tool rotation speed is similar to previous researches on influence of rotation speed of tool on weld strength [23].
The behaviour of tensile-shear load with pre-dwelling time at constant guiding hole diameter (3.0 mm) and tool rotation speed (1300 rpm) is shown in Fig. 15. Tensile-shear load dominantly increased with increase in pre-dwelling time. The increase in pre-dwelling time from 2 sec (C1) to 14 sec (C4) increased weld strength from 4191.51 N to 4544.02 N. This was because when pre-dwelling time increased, the tool came in contact with upper workpiece for more time, which increased frictional heat input. The upper workpiece material became soft and helped in easy plunging of tool and developed required material flow. Hence, adequate material flow and mixing of both workpiece material led to proper bonding of both workpieces. The increase in bonding of weld lead to increased elongation of weld when load was applied, shown in Fig. 16. Hence, higher tensile-shear load was required to break the weld. However, pre-dwelling time also has a positive effect on tool because when workpiece becomes softer with higher pre-dwelling time, tool can easily plunge into the workpiece without facing any sudden impact. Also, when tool plunges easily into workpiece, the microstructure of workpiece not altered significantly. Therefore, higher pre-dwelling time is preferable for obtaining sound welds. Researchers have used different methods and equipment to pre-heat the specimen such as butane torch [31], induction coil [32], resistance rod [33] etc. and observed improved weld strength after pre heating the upper workpiece. Same observations were obtained in present study when pre-heating was done using friction between workpiece and tool. However, the method employed for pre-heating in present work didn’t require any special equipment, making the process economically beneficial.
From aforementioned facts and experimental observations, it can be concluded that welds attained at lower tool rotation speed and pre-dwelling time were not strong enough as compared to other welds. The reason behind this can be the formation of insufficient bonding between two workpieces. When both tool rotation speed and pre-dwelling time were low the heat produced was not sufficient to form a good welded joint. As shown in Fig. 9, Fig. 13 and Fig. 15, maximum strength was obtained at 3.5 mm of guiding hole diameter, tool rotation speed of 1700 rpm and pre-dwelling time of 14 secs. Hence, it can be concluded that maximum weld strength can be obtained at combination of higher quantity of reinforcement and pre-dwelling time and intermediary tool rotation speed.