Investigations on microstructure, hardness and tribological behaviour of AA7075-Al₂O₃ composites synthesized via stir casting route

2.1. Experimental details

In this work, Aluminium 7xxx alloy (AA7075) was selected as the matrix material and alumina (Al₂O₃) was used as the reinforcement with an average particle size of 30 µm received from microtroniks lab chemicals. Among the aluminium alloy series, AA7075 has more significance in the automotive and aerospace industries for the development of gears and shafts, bicycle frame, aircraft fittings, missile parts and defense applications due to its strength to weight ratio, high toughness and natural aging characteristics (Moustafa et al., 2022Moustafa, E.B., Mikhaylovskaya, A.V., Taha, M.A., Mosleh, A.O. (2022). Improvement of the microstructure and mechanical properties by hybridizing the surface of AA7075 by hexagonal boron nitride with carbide particles using the FSP process. J. Mater. Res. Technol. 17, 1986-1999. https://doi.org/10.1016/j.jmrt.2022.01.150.; Uros Stamenkovic et al., 2023Uros Stamenkovic., Svetlana Ivanov., Markovic., I., Milan Gorgievski, Božinović, K., Kovačević, A. (2023). The influence of the ageing temperature on different properties of the EN AW-7075 aluminium alloy. Rev. Metal. 59 (1), e238. https://doi.org/10.3989/revmetalm.238.). The chemical composition (wt.%) of AA7075 is Zn-5.4, Mg-2.42, Cu-1.42, Fe-0.42, Cr-0.21, Si-0.13, Mn-0.12, Ti-0.11 and Al- to balance. Four composite specimens were prepared by varying the wt.% of Al₂O₃ particles viz plain AA7075, AA7075-4 wt.% Al₂O₃, AA7075-8 wt.% Al₂O₃ and AA7075-12 wt.% Al₂O₃. The composite samples were produced through bottom pouring type stir casting method. At first, pure AA7075 ingot was melted in graphite crucible using an electric furnace at the temperature of 800 °C and it was brought to 650 °C. Similarly, Al₂O₃ particles were preheated at 250 °C to remove the moisture and also to improve the wettability of the Al₂O₃ particles with molten AA7075. The molten AA7075 was stirred with a motorized agitator at a speed of 200 rpm constantly. Subsequently, the preheated Al₂O₃ particles were added slowly in the vortex of the AA7075 pool. Then, the mixture was constantly stirred at 250 rpm for 20 mins to achieve a homogeneous distribution of particles in the AA7075 pool. After stirring, the slurry was poured into the preheated steel die immediately. Figure 1 shows the experimental plan of the present study.

The surface morphology of the proposed composites were taken by using scanning electron microscopy (Vega3, Tescan model). The Brinell hardness test was carried out using ASTM E18 standard as a reference to measure the hardness of the different fabricated composite specimens. To perform the tribological behaviour studies, the composite sample was selected based on the hardness value obtained. Often, high hardness material possess good wear resistance. Hence, the higher hardness of the composite specimen was taken fo to study the tribological properties such as the WR and the COF respectively. A pin on disc tribometer was used to conduct the wear test under dry sliding conditions. The test pins were prepared using ASTM G-99 standard as a guide with a dimension of 10 mm in diameter and 32 mm in length by wire-cut electric discharge machining. In general, three factors namely load (P), sliding distance (D) and sling velocity (V) are control the WR and COF (Allasi et al., 2023Allasi, H.L., Soosaimariyan, M.V., Chidambaranathan, V.S. (2023). Wear behaviour of a Cu-Ni-Sn hybrid composite reinforced with B4C prepared by powder metallurgy technique. J. Mech. Eng. 69 (5-6), 275–283. https://doi.org/10.5545/sv-jme.2022.423.). Hece, these factors are taken as the input parameters and are provided in Table 1. Based on the parameters selection, the experiments were executed using a L₉ (3³) orthogonal array (OA) as it is shown in Table 1. After the test, the WR and COF are measured by using Eq. (1) and (2) (Alagarsamy and Ravichandran, 2019bAlagarsamy, S.V., Ravichandran, M. (2019b). Investigations on tribological behaviour of AA7075-TiO₂ composites under dry sliding conditions. Ind. Lubr. Tribol. 71 (9), 1064-1071. https://doi.org/10.1108/ILT-01-2019-0003.).

where, ∆m- mass loss of test specimen (grams), ρ - density of the composite (g•mm^-3), D- sliding distance (m), F_T- tangential force (N) and F_N- normal force (N).

2.2. Optimization using TOPSIS method

This method was proposed by Hwang and Yoon in 1981 to solve multi criteria decision making (MCDM) problems (Magibalan et al., 2020Magibalan, S., Senthilkumar, P., Senthilkumar, C., Prabu, M. (2020). Optimization of wear parameters for aluminium 4% fly-ash composites. Indian J. Eng. Mater. Sci. 27, 458-464.). MCDM methods are used to solve the multi objective problems by the researchers to find out the optimum parameters from definite number of alternatives (Senthamarai et al., 2023Senthamarai, K., Deepak, D., Saravanan, H., Gopikrishnan, C., Alagarsamy, S.V., Chanakyan, C. (2023). Investigations on the wire EDM characteristics of Al matrix composite using TOPSIS method. Mater. Today: Proc. 74 (1), 110-116. https://doi.org/10.1016/j.matpr.2022.12.130.). Hence, in this study, TOPSIS approach was implemented to optimize the wear control parameters on estimated WR and COF. In the TOPSIS method, best solution is selected which should have the shortest distance to the positive ideal solution and the farthest to the negative ideal solution (Azadeh et al., 2011Azadeh, A., Kor, H., Hatefi, S.-M. (2011). A hybrid genetic algorithm-TOPSIS-computer simulation approach for optimum operator assignment in cellular manufacturing systems. J. Chin. Inst. Eng. 34 (1), 57-74. https://doi.org/10.1080/02533839.2011.552966.). To determine the optimum levels of parameters, the given steps to be implemented and are as follows,

Step1: Initially, the decision matrix D_max is formulated for the m alternatives and n attributes as follows,

where x_ij is the value of the optimal characteristics, where i = 1 − m is the number of results of each characteristic, and j = 1 − n is the number of characteristics to be optimized.

Step 2: The obtained values from decision matrix is normalized (r_ij) using Eq.(3):

where i = 1, 2,…, m and j = 1, 2,…, n. x _ij represents the actual value of the i ^th value of j ^th experiment.

Step 3: The weighted normalized matrix is determined by the product of weight assigned for each response and the normalized value using Eq. (4)

where w _j represents the weight of the j ^th attribute_, w_j = 0.5, i = 1, 2,…, m and j = 1, 2,.., n. Table 2 illustrated the weighted normalized matrix.

Step 4: Positive ideal solution (A⁺) and negative ideal solution (A^-) are calculated using Eq. (5) and Eq. (6)

where, J is associated with the benefit parameters and J’ - is associated with non-benefit parameters.

Step 5: The next step is to determine the separation measure for the positive ideal solution and negative ideal solution using Eq. (7) and Eq. (8):

Step 6: Finally, the relative closeness (C_i) of each alternative to the ideal solution is calculated using Eq. (9)

The calculated separation measures and the relative closeness are provided in Table 3.

Step 7: Rank the relative closeness in ascending order. The higher rank will give the optimum combination of parameters. Figure 2 shows the rank plot for the obtained relative closeness (C_i). It reveals that the Ex. No 4 has a higher relative closeness (0.982), which consists of a better combination of optimal control parameter to obtain the lowest WR and COF.

3.1. Surface morphology

The surface morphology of the stir casted AA7075 matrix composites with varying compositions of Al₂O₃ particles were observed using a SEM. Figure 3 (a-d) represents the SEM micrographs of unreinforced AA7075 and AA7075-Al₂O₃ composite specimens. From the Figs., the dark region represents the matrix alloy and white region represents the presence of the reinforcement (Al₂O₃ particles) content. In Fig. 3a, it can be ensure that there is no particles presence in the matrix. From Fig. 3(b-d), it was clearly reveal that a uniform dispersion of Al₂O₃ particles within the casted AA7075 matrix. It can also be clearly observed from the image that there was no formation of agglomerations or clusters of Al₂O₃ particles in the fabricated composites. This can be attributed to the utilization of appropriate stir casting parameters used to fabricate the composite specimen. However, the pores are formed in some regions in the AA7075-12 wt.% Al₂O₃ composite due to the gas entrapped in the molten metal.

3.2. Effect of Al₂O₃ particles on the hardness

Figure 4 represents the graphical data of the hardness measured at various locations in the composite samples. From Fig. 4, it can be noted that AA7075 - 8 wt.% Al₂O₃ composite shows better hardness and with minimum scatter in the measured points compared to the other three composites. With the addition of the hard reinforcement into the ductile matrix, the overall hardness of the composite is enhanced (Halverson et al., 1989Halverson, D.C., Pyzik, A.J., Aksay, I.A., Snowden, E. (1989). A processing of boron carbid -aluminium composites. J. Am. Ceram. Soc. 72 (5), 775-780. https://doi.org/10.1111/j.1151-2916.1989.tb06216.x.). It can be also observed that the presence of peaks and valleys are found in the values obtained for AA7075-12 wt.% Al₂O₃. The peak indicates that the reinforcing particles are agglomerated such that it gives maximum hardness and the valley in AA7075-12wt.% Al₂O₃ shows that absence of reinforcing particles at that location. This reveals that there exists inhomogeneities in the distribution of the reinforcing particles introduced in the AA7075-12wt.% Al₂O₃ and also indicates that the increase in content of the reinforcement above 8 wt.% leads to a poor distribution of the reinforcement particles in the matrix thus reducing the hardness (Sakthivelu et al., 2019Sakthivelu, S., Meignanamoorthy, M., Ravichandran, M., Sethusundaram, P.P. (2019). Tribological behaviour of AA7075-ZrSiO₄ composites synthesized by stir casting technique. Mech. Mech. Eng. 23, 198-201. https://doi.org/10.2478/mme-2019-0026.).

3.3. Effect of parameters on tribological behaviour

The main objective of this study was to determine the optimal levels of input parameters to obtain the lowest WR and COF under dry sliding condition. The mean of the relative closeness for each parameter levels are shown in Fig. 5. From the Fig, the x-axis represents the individual level of parameters such as load (P), sliding velocity (V) and sliding distance (D), and the y-axis represents the mean of relative closeness (C_i) value. In general, the optimum combination of parameters is identified by a higher relative closeness value. As it is seen in Fig. 5, it can be found that the optimum level of control parameters are the load at level 1 (P₁ = 10 N), the sliding velocity at level 1 (V₁ = 1 m•s^-1) and the sliding distance at level 2 (D₂ = 1000 m), respectively. From the optimum condition we can observe that an increase in the load causes faster wear loss at the lowest sliding velocity and sliding distance. Normally, increasing the load and sliding distance causes higher wear loss because of the higher penetration of the abrasive particles into the specimen which ploughs the material from the specimen surface (Sahin, 2007Sahin, Y. (2007). Tribological behaviour of metal matrix and its composite. Mater. Des. 28, 1348-1352. https://doi.org/10.1016/j.matdes.2006.01.032.). Here, the initial condition of load, sliding velocity and middle level of sliding distance provide the lowest WR and COF for the fabricated AA7075-Al₂O₃ composites.

Table 4 depicts the response table for the mean relative closeness (C_i) with respect to each level of control parameters. From the table, it has been revealed the influence of the control parameters for the lowest WR and COF. As it is seen in Table 4, it can be observed that the load (P) takes rank 1 which plays a dominant role in obtaining the lowest WR and COF, followed by the sliding distance (D) and the sliding velocity (V) respectively. By employing the ANOVA analysis, this observation was ensured. ANOVA is a statistical tool which is widly used to determine the influence of parameters on the response (Agarwal et al., 2023Agarwal, V., Goyal, D., Pabla, B.S., Vettivel, S.C., Haiter Lenin, A. (2023). Optimization and characterization studies of dissimilar friction stir welding parameters of brass and aluminum alloy 6063 using Taguchi method. Adv. Mater. Sci. Eng. 1-10. ID 8275323. https://doi.org/10.1155/2023/8275323.). Hence, in this study, an ANOVA was performed to find out the impact of control parameters namely ‘P’, ‘V’ and ‘D’ on the WR and COF of the AA7075-8 wt.% Al₂O₃ composite using a dry sliding wear test. The results were reported in Table 5 and also the graphical representation of parameters contribution is shown in Fig. 6. Based on the results in Fig. 6, it can be concluded that the load is the primary dominant factor with a contribution of P = 38.36%, followed by the sliding distance, D = 28.32% and the sliding velocity, V = 16.27% respectively. The similar results were previously observed by Dhanalakshmi et al. (2018)Dhanalakshmi, S., Mohanasundararaju, N., Venkatakrishnan, P.G., Karthik, V. (2018). Optimization of friction and wear behaviour of Al7075-Al₂O₃-B₄C metal matrix composites using Taguchi method. IOP Conf. Series: Mater. Sci. Eng. 314, 012025. https://doi.org/10.1088/1757-899X/314/1/012025. after performing dry sliding wear tests on Al7075-Al₂O₃-B₄C composites and they reported that the load was the primary impact factor affecting the tribological behaviour.

Figure 7 (a-c) shows the contour mapping of the WR for the tested AA7075 matrix composites reinforced with Al₂O₃ particles with respect to the control parameters, namely, load (P), sliding velocity (V) and sliding distance (D), respectively. The interaction effect of load versus sliding velocity on the WR is presented in Fig. 7a. As the load and sliding velocity increased simultaneously, there was an improvement in the WR. However, the lowest WR (0.0003 mm³/m) is achieved at the moderate load and initial sliding velocity conditions. Meanwhile, the maximum load of 20 N with medium sliding velocity of 2 m•s^-1 gave the highest WR (>0.008 mm³/m). Fig. 7b reveals the effect of load versus sliding distance on the WR of the tested composite. It was clearly observed that the lowest WR is attained for 1000 m of sliding distance and a load of 15 N. But, with an increase in the load at the initial sliding distance of 1000 m produce more WR. This is because the maximum load creates a high pressure on the composite pin thus leading to an increase in the WR. Fig. 7c represents the effect of the sliding velocity and sliding distance on the WR. It can be understood that a lower sliding velocity (1 m•s^-1) with moderate sliding distance (1000 m) produces a low WR. Similarly, the maximum WR was obtained for the lowest sliding distance (500 m) using the highest sliding velocity (3 m•s^-1).

The contour mapping of the COF for the AA7075-8 wt.% Al₂O₃ composites with respect to the wear control parameters. Namely, load (P), sliding velocity (V) and sliding distance (D) are illustrated in Fig. 8 (a-c). The interactive effect of load versus sliding distance on the COF of the tested composite is depicted in Fig. 8a. It was found that the lowest COF (<0.45) is obtained at the lowest applied load of 10 N and sliding velocity of 1 m•s^-1 respectively. At the same time, the COF gradually increased with an increase in the load and sliding velocity. However, the load of 10 N using 3 m•s^-1 as the sliding velocity produced greater COF (>0.70). Fig. 8b demonstrates the influence of the sliding velocity versus the sliding distance on the COF. It has been noticed that the lowest COF (<0.45) was obtained for the middle level of sliding distance (1000 m) using a load from 10 N to 15 N. Furthermore, the COF is increased with an increase in the load to a maximum level of 20 N and using sliding distance of 1000 m respectively. Figure 8c shows the effect of the sliding distance and the sliding velocity on the COF of the casted composite during the dry sliding wear tests. It clearly reveals that the lowest COF is obtained for the medium level of the sliding distance (1000 m) and using 2 m•s^-1 as sliding velocity. However, an increase in the sliding velocity produces more COF at the middle level of sliding distance.

3.4. Confirmation experiments

The final step in the tribological behaviour analysis of AA7075-8 wt.% Al₂O₃ composite was to validate the predicted result and, for this purpose, a confirmation experiment was conducted by considering the optimized parameters’ conditions as shown in Table 6. It was revealed that the relative closeness (C_i) values for the experimental and predicted control parameters are 0.982 and 0.9323 respectively. Based on the results, a very low percentage error of 5.06% was obtained between the experimental and predicted results which ensures the very good correlation.

3.5. Worn surface morphology

Figure 9a represents the worn surface morphology of the AA7075-Al₂O₃ composite for the initial parametric conditions. From this figure, we can observe the deep ploughs present in the surface. This is due to the impregnation of the counter disc particles at the loading conditions. This leads to the ploughing of a large volume of particles from the base AA7075 which results in a higher wear rate. Figure 9 (b-c) shows the worn out surface morphology for the confirmation experiment sample AA7075-8 wt.% Al₂O₃ to understand the abrasive wear mechanism. Figure 9b reveals the impregnation of abrasive particles into the sample, forming a deeper plough. The increase in the load leads to the formation of plough. It can be understood that the increase in the load increases the wear loss; besides the contact between the abrasive particles and specimen also increases which leads to an increase in the COF. As the sliding velocity increases, the formation of wear debris will be reduced due to the low interaction between the abrasive particles and the specimen surface. This phenomenon shows that the wear rate will decrease by increasing the sliding velocity (Dharmalingam et al., 2013Dharmalingam, S., Subramanian, R., Kok, M. (2013). Optimization of abrasive wear performance in aluminium hybrid metal matrix composites using Taguchi-grey relational analysis.Proc. Inst. Mech. Eng. J: J. Eng. Tribol. 227 (7), 749-760. https://doi.org/10.1177/13506501124679.). It can also be concluded that the worn surface for the highest load has a significant delamination, which was likely brought on by the development of a high temperature at the interfaces. This delamination is due to the high amount of metal deformation and increased material removal caused by the frictional heat buildup between the two surfaces which are in intense contact. Thus, while combining severe delamination with an increasing load, moderate adhesion wear was achieved. Figure 9c represents the formation of a tribo-layer due to the increase in temperature between the abrasive wheel and the composite pin. Because of the formation of a tribo-layer, the friction coefficient is minimized, which improves the wear resistance of the composites (Balaji et al., 2022Balaji, S., Maniarasan, P., Alagarsamy, S. V., Alswieleh, A.M., Mohanavel, V., Ravichandran, M., Jeon, B.H., Allasi, H.L. (2022). Optimization and prediction of tribological behaviour of Al-Fe-Si alloy based nano grains refined composites using Taguchi with response surface methodology. J. Nanomater. 1-12, ID 9733264. https://doi.org/10.1155/2022/9733264.). In addition to that, the incorporation of Al₂O₃ particles is the main reason for the enhancement of the wear resistance because of the high hardness of the composite compared to the base alloy.