Examining the influence of Scandium interlayer and parameters on the mechanical behaviour in FSW of AA 5754-H111 plate through response surface approach

Friction stir welding (FSW) was conducted on 5754-H111 aluminium alloy using a response surface methodology to examine the impact of process parameters on the efficiency of the weld. The key FSW parameters were the percentage of scandium inclusion in the interlayer, tool rotational speed, and tool travel speed. The significant responses were identified as the tensile strength of the weld and the hardness in the weld zone. An Al-Mg-Sc alloy was employed as an interlayer material between the 5754 plates to enhance the strength in the weld zone. The addition of Scandium had a more pronounced effect on the hardness of the stir zone. The tensile strength was increased by raising the tool’s rotational speed. The results showed that the maximum tensile strength of the joints was 225 MPa and the hardness was highest at 118 HV1 in the weld stir zone. The introduction of Scandium improved the joint strength by producing fine grains and Al3Sc precipitates. The welding speed had the least impact on the tensile properties of the joint among all the variables investigated.


Introduction
Material joining is critical in several industries, including shipbuilding, automotive, and aerospace.Different thermal techniques, such as friction stir welding, arc, laser, and electron beam welding, can be employed in the joining method.FSW is one of the most successful and effective techniques for combining metals owing to its high joint strength, minimal heat input, and limited Heat Affected Zone (HAZ) volume compared to Fusion zone (FZ).The aluminium alloy 5754 is primarily utilised in the aerospace, marine, and automotive industries [1,2].The Welding Institute invented and patented friction stir welding in 1991 by Thomas, [3].It is a solid-state welding technique that relies on friction and stirring.This method produces welding heat by utilizing a rotating, non-consumable tool that consists of a precisely designed pin and shoulder.Other advantages over standard fusion welding include no distortions, less HAZ, and superior surface quality [4].Furthermore, the FSW technique produces prospective joints with minimal preparatory and minimal post-weld cleaning.
The material flow pattern and temperature dispersion are influenced by the tool geometry, welding parameters, and joint designs, which determine the microstructural evolution [5].The transverse and tool rotational speeds are the FSW process factors that most affect the mechanical qualities of the joints, according to various studies [6,7].These latter are primarily determined by grain size and dislocation density as a result of plastic deformation and recrystallisation events that occur during welding [8][9][10][11].Several difficulties that occur in fusion welding of Al alloys, notably the evaporation of low melting point elements in the fusion zone and the development of porosity and cracking, do not occur in the FSW of these materials.However, in the FSW of these strain-hardened alloys, the loss of strength in the weld zone occurs as well, though to a lesser amount than in fusion welds.Further, it is determined by the weld parameters, which occurs even when the best weld parameters are utilised to minimise grain size and enhance strength.
Several types of research have shown the purpose of post-weld heat treatment (PHWT) to recover the lost strength in the weld region of Al alloys.Several researchers have said that PWHT has boosted the weld zone's strength in FSWed 2000, 6000, and 7000 series Al alloy welds [12].Wahid et al (2018) studied the structure and mechanical behaviour of the 5754-H22 alloys after it was friction stir welded in water and air medium [13].They claim that during FSW, fine and equiaxed grains were seen in the stir zone regardless of cooling medium or rotational speed.Filippis et al (2017) optimised the process parameters of FSW on the 5754-H111 plates using the RSM approach [14].UTS values and thermal behaviour were the output responses that were measured.With a tool rotating speed of 500 rpm and a welding speed of 23 cm/min, they were able to retrieve 5% of the UTS value.Attallah et al (2007) have worked on the FSW of alloys in the 5000 series with a variety of process settings [15].The formation of onion rings in the FSW joints has been explained.Barlas et al (2012) have provided information on the effect of process factors on the tensile and fatigue behaviour of AA 5754 FSW joints [16].The joint's maximum tensile strength is 217 MPa, according to the researchers.Nidhi Sharma et al 2017 has evaluated the influence of varying tool pin profile used during FSW of 5754 Al alloy.The various tool pin profile used were cylindrical, taper, cylindrical cam, taper cam and square shape.FSW joint made using square tool pin profile facilitates good amount of mixing at nugget zone, which consequently increases the hardness [12].
The post-welding heat treatment technique was not used since the 5754 alloy isn't heat-treatable.The postweld heat treatment procedure, on the other hand, is expensive and not viable in some alloys.As a result, the reduction of strength in the weld region has to be maintained to a bare minimum while welding.For example, this can be accomplished by using a suitable high-strength material as an interlayer.While FSW does not require filler material, the inability to adjust the composition in the weld zone prevents this joining approach from replacing traditional fusion welding with filler material.Attempts to modify the chemistry of the nugget zone have been reported in the literature.For example, before joining, one of the contacting faces was coated with metallic particles.The main goal was to prevent brittle intermetallics from forming in the weld region of FSW of incompatible metals.Furthermore, to boost the strength of joints, one of the contacting faces was coated with a layer of strengthening particles inserted in the adjoining surfaces.
The application of interlayer on the FSW of 6061-T6 Al alloy joints has been investigated by G cam 2014 [17].By changing the chemistry and employing interlayer material, they were able to get a considerable increase in hardness in the stir zone.However, because HAZ was tempered Al alloy, it was not able to regain the strength loss, and joint performance did not improve.The use of a zinc interlayer as a third material to improve the mechanical properties of AA 2024-Cu FSW joints was investigated by Anbukkarasi 2019 [18].When compared to AA2024-Cu welds, the hard particles of Al-Mg-Zn phases improved strength and elongation.Fenoel 2020 conducted an experimental investigation on the influence of a zinc interlayer on lap joints of FSW on Al to Cu or brass [19].They discovered phase change between the joints and the interlayer material, as well as the material's shear lap tensile properties.
In the existing research, there is little information about the inclusion of a different high-strength interlayer in FSW of strain-hardened Al alloys.Furthermore, the Al-Mg-Sc alloy has never been employed as a highstrength interlayer.This was the driving force behind the current investigation.In 1971, Willey patented Scandium's effect on the structure and dynamics of aluminium and aluminium alloys [20].Because of the extremely uniform density of the dislocations in Al-Mg solid solution alloys, Scandium has a significant influence.Weld filler rods containing Scandium resulted in grain refinement in the FZ [21,22].The mechanical properties of weld joints improve when Scandium is used as a grain refiner in an aluminium alloy fusion weld [23].According to Lathabai (2003), the Al-Mg-Sc alloy showed much higher 0.2% proof stress, tensile strength, and micro-hardness than the cast Al-Mg alloy but somewhat poorer ductility [24].Munoz 2017 welded an Al-4.5Mg-0.26Scalloy in the H116 condition using TIG welding [25].The weld zone softened due to the lack of a cold working effect and the absence of Al 3 Sc precipitates induced by Sc dissolution.Nafeez Ahmed et al have improved the TIG weld joint strength of 5052-H32 alloy using 0.25 wt% Sc added Al-Mg filler rod [26].The Hall-Petch strengthening proves that when particle or grain size reduces, hardness rises, as measured by Hall and Petch [27,28].
Previous research looked into a variety of parameters in the FSW of aluminium alloy plates with a highstrength interlayer.However, the effect of Scandium interlayers is relatively unexplored.A systematic analysis was conducted to understand their impact on weld quality, mechanical properties, and microstructural changes.In this investigation, AA5754-H111 alloy plates were friction stir welded with an Al-Mg-Sc alloy interlayer between the contacting faces of the plates to see if the strength in the weld zone could be recovered by modifying the composition in this zone.Additionally, utilising the design of experiments, the impacts of Scandium inclusion level, FSW tool rotating speed, and tool travel speed on mechanical behaviour were thoroughly investigated using the Response Surface Methodology (RSM) technique.The purpose of this research is to look into the joint's mechanical behaviour utilising tensile tests on weld samples and Vickers hardness in the FZ.The Vickers hardness of the two plates' welded zones was tested to determine the variation in hardness distribution in stir zone form and microstructure under various welding procedures.While response surface methodology is not entirely new in FSW research, its application in conjunction with Scandium interlayers to tailor mechanical behaviour is a novel aspect of our work.

Materials and methods
The investigations were conducted on a non-heat treatable AA5754 H111 Al alloy with the composition shown in table 1.The as-received material was in the shape of cold-rolled sheets and 5 mm thick plates.Scandium is added to a conventional Al-5% Mg alloy to create the modified Al-Mg-Sc alloy, which is then utilised as an interlayer.Stir casting is used in the process of alteration.In a vertical stir cast furnace device, a predetermined amount of Al-5% Mg alloy, as well as Aluminium Scandium master alloy, is melted at 650 °C and then stirred.The molten material is swirled before being put into a mould of the necessary size.The average hardness of the Al-Mg-Sc alloy was 105 HV.
For welding experiments, 150 × 60 mm blanks were cut from the as-received plates, and their butting surfaces were milled with a milling machine before joining.Interlayer strips having dimensions of 2 mm width, 150 mm length, and 5 mm thickness were wire cut machined for the fabrication of joints.An FSW tool with a square-shaped pin with dimensions of 5 mm diagonally was utilised (the length of the tool pin is 4.7 mm).The tool's shoulder was 15 mm in diameter.The FSW tool was produced with H13 tool steel, which was machined to the required dimensions.
The welding experiments were carried out on a CNC vertical machining centre with an FSW head.To explore the influence of Sc in the interlayer, 17 joints were made with varied Sc wt% in the interlayer and two different FSW parameters with different factors.All of the joints were made using square shape pins and a 12.5kN axial load.The schematic representation of the FSW process is shown in figure 1. Electro-discharge machining (EDM) machine was used to remove a microstructure and tensile sample corresponding to ASTM E8/E8M-13a from each weld.Universal tensile test equipment was used to perform the tensile tests.Keller's reagent was utilised to etch metallography specimens.For microstructural characterisation, these metallographic specimens were subjected to detailed optical microscopy.The microstructure in the welds was also identified using a scanning electron microscope (SEM).Micro-hardness tests were also performed with a load of 500 g for 10 seconds to assess the hardness pattern of each weld.

Experimental design
The impact of FSW settings on weld zone hardness and tensile strength of joints was examined using the Design of Experiments (DOE) approach.The RSM technique was utilised to optimise the FSW process parameters (rotation tool and travel speeds) as well as the level of Sc wt% in the interlayer material.The RSM is focused on the investigation of a mathematical relationship among parameters and responses.Table 2 shows how the Box-  Behnken design (BBD) method was utilised to select the process parameters and levels.On the tensile strength and weld zone hardness, the impacts of welding speed, tool rotating speed, and the wt% of Sc in the interlayer were examined.The effects of the operating settings on the interlayer material were also examined.Table 3 shows that a series of 17 trials were conducted for the three variables, which had one centre point.Table 3 shows the three levels of consideration for each parameter.The purpose of the testing was to look into and develop a regression correlation between the friction stir welding process working parameters and the response surfaces, such as weld zone hardness and tensile strength.Design-Expert software was used to create the BBD design.

Development of empirical models
The second-order regression equation was used to create a mathematical model.The equations show how the FSW parameters and output responses are related.The parameter Sc wt % (A) was selected as the most important factor for both responses in the Analysis of Variance (ANOVA), as shown in table 4, and Sc wt % on the interlayer (A) was the important factor for microhardness with a p-value less than 0.05.
Figure 1 shows a scatter plot with the experimental result and predicted result on the x-axis and y-axis, respectively, to evaluate the accuracy of the empirical correlations.The scattered plots are quite close to the 45°l ine, indicating that the derived empirical connections are perfectly suited.

Impact of process parameter on tensile strength
In perturbation plots in figure 2(a), the influence of input process factors on FSW of 5754 H111 aluminium alloys is displayed.Sc wt % level in the interlayer, tool rotational speed (rpm), and weld speed (mm/min) are represented by the letters A, B, and C, respectively.The graphic shows a steep curve, indicating how much the process factors can influence the outcome.The curve of parameter B (Tool rotational speed) may be shown to vary significantly over the plot's reference points.The parametric curves for parameters A (Sc Wt %) and C (Tool feed) show fluctuations over the reference points as well, though to a lesser extent than parameter B.
Figure 3(a) depicts the rising action of the tensile strength beyond 800 rpm of tool rotating speed, which diminishes beyond 1000 rpm of input tool rotational speed.The tensile strength declines when the Sc level is increased from 0.25 wt% Sc to 0.75 wt% Sc.Figures 3(a), (b) depicts the impacts of tool rotational speed and Sc wt%.When the tool rotational speed is increased to 1000 rpm, the heat generated is sufficient to attain the maximum tensile strength of 225 MPa.The heat generated by lower tool rotational speeds is smaller, resulting in lower tensile strength levels.Also, during FSW welding, when the tensile strength reaches 225 MPa, the wt% of Sc interlayer is 0.25%.When these levels are used under given parameters to enable increased interaction among the interlayer and base alloys contact, improved tensile strength is attained.
Figures 3(c), (d) shows that the tensile strength of the welded connection is impacted more by the wt % of Sc level in the interlayer and less by the feed rate.The wt% of Scandium added interlayer affects heat dispersion and material fusion, respectively.Figures 3(e), (f) shows that the tensile strength is greatest when the feed rate is between 28 and 32 mm min −1 and that it peaks at 30 mm min −1 .When the wt% of Sc added interlayer is 0.25%, the tensile strength peaks but diminishes after 0.35%.Because of the improved welding technique and the addition of Sc in the interlayer, the structure of the weld zone has fine grain dispersion.

Impact of process parameter on micro-hardness
Figure 2(b) depicts the perturbation plot for micro-hardness on the weld zone of the FSW welded specimen.Curve A has the sharpest curvature, indicating that it has the most influence on the weld zone's micro-hardness.As demonstrated in figure 4(a), the increase in micro-hardness over 800 rpm of tool rotational speed continues until 1000 rpm, after which it begins to diminish.Micro-hardness rises over 0.25 up to 0.5 wt % Scandium, and then reduces below 0.65 wt % Scandium.During the FSW, when the micro-hardness achieves a high of 131 HV, the wt% of Scandium added interlayer is 0.75%.Figure 4(a) depicts the rise in micro-hardness in the FSW joint as a function of tool rotating speed and scandium wt %.The tool rotational speed and the wt % of scandium interlayer effect heat dispersal and fusion.As shown in figure 4(c), the micro-hardness is highest when the tool rotational speed is between 900 and 1100 rpm and peaks at 1000 rpm.
When the Scandium added interlayer's wt% is 0.25, the micro-hardness increases, and when it is 0.65%, it decreases.When the wt% of Scandium added interlayer is 0.50%, the micro-hardness reaches its maximum value.Scandium in the welding process helps to reduce grain size in the weld zone and enhances the microhardness of the weld metal significantly.This happens due to the additional Scandium, together with the magnesium in the aluminium alloys, which reacts together in the weld zone to provide increased strength and fewer flaws.The centre location of the weld is found to include a mixture of magnesium and Scandium, resulting in a peak micro-hardness (131HV).

Optimisation of FSW parameters
The results of the trials and optimisation procedure show that for maximum tensile strength and microhardness levels, the tool rotational speed should be 1000 rpm.The tool feed and wt% of Scandium added interlayer necessary to attain the maximal tensile strength levels are 30 mm/min and 0.25%, respectively.The tool feed and wt% of Scandium added interlayer necessary to attain the peak micro-hardness levels are 32 mm min −1 and 0.75%, respectively.Figure 5 depicts a numerical optimisation ramp that emphasises the importance of input parameters in obtaining the desired tensile strength and hardness.The tensile strength was set to the greatest value possible.The micro-hardness value has been adjusted to 80 HV, which is larger than the hardness of the base metal.The optimal settings were 0.25 wt% Sc in the interlayer material, 1051 rpm tool rotating speed, and 31.39 mm min −1 feed rate.The desirability value of 0.971 was achieved for the targeted results.

Microstructural analysis and fractography
The microstructure of all friction stir welded (FSW) joints was categorized into three zones: the weld zone (WZ), the heat-affected zone (HAZ), and the base metal zone (BMZ).An optical micrograph of the base metal AA5754-H111 and Al-2%Sc master alloy is shown in figures 6(a) and (b). Figure 6(c) shows the macrostructure of the FSW sample with the Al-Mg-Sc interlayer.As illustrated macroscopically in figures 6(d) and (e), sound weld joints with complete intermixing between both base metals and interlayer were obtained, covering the whole weld thickness.There were no weld flaws observed in the joints.As seen in figure 6(f), grain refinement has occurred in the dynamically recrystallized zone.A banded type morphology was seen in the weld zone of the joints, which is developed by the mixing of base alloy and interlayer materials.In these micrographs, the dark regions belong to the Al-Mg-Sc alloy, whereas the brighter regions correspond to the Al matrix.The microstructures of the FZ and the FZ/HAZ interface of the joints generated are shown in figure 6(f).These micrographs clearly show that all of the joints produced underwent grain refining in FZ.Furthermore, the inclusion of Scandium, resulted in a finer grain structure in the FZ of the joint made with an interlayer.Furthermore, as seen in figure 6(f), enhanced FSW parameters resulted in finer grain size in FZ.The existence of hardness variation in the FZ region caused all of the specimens to fail inside this region, as evidenced by the hardness plot of the weld (figure 8). Figure 6(g) shows the failure region of the tensile sample which is brittle mode in nature.Figure 6(h) shows the fractography of the tensile sample where the voids and facets can be seen.The absence of dimples confirms the failure mode is not ductile.The tensile test findings show that the weld made with a higher strength interlayer (0.75 wt % Sc) did not have a higher response than the weld made with a lower strength interlayer (0.25 wt% Sc), despite the FZ having a higher hardness.This is likely due to the variation of hardness in the FZ areas of joints created with different  scandium content in the interlayer.Overall, the microstructural analysis and fractography of the FSW joints indicate that the fine-grained microstructure of the WZ is responsible for the high tensile strength and microhardness of the joints.The presence of scandium in the interlayer material further improves the grain refinement of the WZ, which further improves the tensile strength and micro-hardness of the joints.However, increasing the scandium content in the interlayer does not affect the joint efficiency in FSW joints.Figure 7 shows the stress strain curve of the tensile tested specimen for sample 3.

Hardness plot
Figure 8 shows the hardness profile of all FSW joints measured in three places, namely the weld zone, HAZ, and BM areas.In this plot, a hardness gain was found in the weld made with an Al-Mg-Sc interlayer, with the nominal hardness occurring in the HAZ area.The hardness plot derived from the centre of the welds revealed that the interlayer joints had much greater hardness in the FZ than the base metal.This clearly shows that the beneficence of the Al-Mg-Sc interlayer enhances the hardness.It was also discovered that the hardness levels recorded inside the FZs of the joints with interlayer vary greatly.Resulting in banded morphology, certain hardness indentations strike Al-Mg-Sc alloy layers, whereas some hit AA5754 alloy layers (figures 6(d) and (e)).
However, a high-strength interlayer might boost the weld zone's strength, and a strength restoration in the HAZ was not possible because of the coarsening of strengthening particles in HAZ regions.As a result, for all of the joints created, the nominal hardness is in the HAZ region, showing that the applied interlayer will not influence the HAZ region.Hardness may be lower in the weld zone's shoulder portion of interlayer-produced joints, as no mixing of base metal and interlayer happens, as shown in figure 6(c).

Conclusion
This study investigated the effects of FSW parameters and scandium content on the tensile strength and microstructural characteristics of FSW joints between 5754-H111 Al alloy plates and Al-Mg-Sc alloy as interlayer material.The key findings are as follows: • Tensile strength is most affected by scandium content and tool rotational speed.
• Increasing tool rotational speed improves grain refinement and reduces defects, resulting in higher tensile strength.
• The optimal FSW parameters for achieving the target tensile strength and micro-hardness values are 0.25 wt% Sc in the interlayer, 1051 rpm tool rotating speed, and 31.39 mm/min feed rate.
• The micro-structural SEM investigation of the optimised FSW joint revealed improved bonding between the base metal grains and the interlayer in the weld zone.

Figure 2 .
Figure 2. Experimental versus predicted values of responses and Perturbation plot for (a) UTS and (b) micro-hardness.

Figure 3 .
Figure 3. 2D and 3D plots showing the influence of multiple parameters on tensile strength.

Figure 4 .
Figure 4. 2D and 3D plots showing the influence of multiple parameters on hardness.

Figure 6 .
Figure 6.(a) OM of base metal 5754-H111 (b)Al-2%Sc master alloy (c) Macroscopic view of FSW sample (d) OM of stir zone region (e) OM of Weld bottom region (f) Weld zone/HAZ interface (g) Tensile fracture sample (h) SEM Fractography image of the tensile sample.

Figure 8 .
Figure 8. Hardness profile of a cross-section of joints obtained from center regions.

Table 1 .
Composition of 5754 and interlayer material.

Table 2 .
Table 3 depicts the numerical results of the developed empirical relationship.The response tensile strength is represented by the second-order polynomial (regression) equation: Process parameters and levels.
The response micro-hardness is represented by the second-order polynomial (regression) equation:

Table 3 .
Box-Behnken design and experimental results.

Table 4 .
ANOVA for tensile strength and hardness.The chance of finding a significant outcome in an investigation (If the p-value is less than 0.05, the result is usually considered significant).