Analysis of wire-cut electric discharge machining behaviour of AA6061 / C / ZrO2 hybrid nano composites using Taguchi’s method

This study focuses on the optimization of wire-cut electric discharge machining (WEDM) process for hybrid metal matrix nanocomposites. AA6061 alloy reinforced with graphene (C) and zirconia (ZrO2) was fabricated through the stir casting method with ultrasonic assistance for three different weight proportions: 92% AA6061 / 3% C / 5% ZrO2, 87% AA6061 / 3% C / 10% ZrO2, and 82% AA6061 / 3% C / 15% ZrO2. Microstructural examination confirmed the uniform distribution of reinforcements in the matrix alloy. The fabricated composite was machined by WEDM using an L27 orthogonal array designed by Taguchi’s method. Four electrical control factors of pulse on-time, pulse off-time, wire feedrate and peak current were considered, along with one reinforcement factor and three output responses: kerf width, surface finish and material removal rate to analyse the machining behaviour. Analysis of Variance was performed to determine the significant parameters. The results revealed that the most significant factor is pulse on-time among the five different factors. Optimized parameters were identified and examined through confirmation experiments resulting in improved machining characteristics.


Introduction
Aluminum alloys particularly AA6061 are utilised widespread in the aerospace and automotive industries owing to their excellent strength-to-weight ratio.AA6061 based metal matrix composites are preferred in automobile applications due to their lower thermal coefficient, enhanced stiffness, high corrosion resistance and reduced wear rate.Reinforcements such as Al 2 O 3 , SiC, ZrO 2 and W characterized by high hardness are commonly employed in composite materials to augment their strength.However, these composites, reinforced with highhardness materials, pose challenges in machining, necessitating specialized processes to preserve their properties [1].Jaraslaw Wozniak et al [2] investigated the impact of cooling conditions on AA6061 composites, revealing non-uniform hardness values, increased dislocation density, and altered grain size in cooled composites due to microstructure reconstruction.Michael Oluwatosin Bodunrin et al [3] reviewed reinforcement strategies for different combinations in synthesizing hybrid composites.Optimization of machining parameters is crucial for improving the machining behavior of AA6061 hybrid composites.Haja et al [4,5] explored the performance of conventional machining for aluminum alloys, incorporating different coatings on cutting tools to achieve better surface finish.However, the results were comparable and limited in conventional techniques.Jaroslaw Wozniak et al [6] analyzed self-lubricating aluminum composites processed through Spark Plasma Sintering techniques, revealing the dominance of abrasive mechanisms in composite wear.Haja et al [7] investigated the mechanical and machining characteristics of AA6061 reinforced with nano-sized ZrO 2 and C particles, demonstrating enhanced tailor-made mechanical and tribological properties.Pazhouhanfar et al [8] examined the microstructure and mechanical properties of AA6061reinforced with TiB 2 , with potential applications in aerospace and defense industries due to increased strength.Haja et al [9] studied the performance characteristics of Electrical Discharge Machining (EDM), showing comparable results with conventional and unconventional techniques in machining difficult-to-cut materials.The application of promising AA6061 composites, especially those reinforced with ZrO 2 and C hybrids, requires notable machining characteristics.Wire-cut Electric Discharge Machining (WEDM) is a prominent technique known for better machining characteristics in difficult-to-cut materials, providing improved dimensional accuracy and surface finish [10].While several researchers have optimized WEDM parameters for composite materials reinforced with micro-sized particles, there is a research gap in analyzing the machining characteristics of composites fabricated with nano-sized reinforcements.Therefore, this research aims to study the machining characteristics of the WEDM process through optimization using Taguchi's method.
Metal matrix composites machined through the Wire Electric Discharge Machining (WEDM) process exhibit improved surface finish and dimensional accuracy.The impact of material strength remains consistent in the WEDM process, with the primary material removal occurring through erosion, melting and vaporization [11].Analyzing process characteristics is essential for understanding cutting actions in terms of the input parameters' effects on the output variables [12].Previous research has optimized the WEDM process parameters for machining Boron carbide-reinforced AA6063 and AA6351 composite materials.The study focused on peak current, wire feed-rate, pulse off-time, and on-time pulse, revealing that electrical parameters significantly influenced material removal rates [13].However, machining zirconia-reinforced composite materials poses challenges due to their high hardness, and the machining behavior of ceramic-reinforced composite materials, particularly in the context of WEDM, is not extensively explored [14][15][16][17][18].
The literature study indicates a gap in understanding the machining behavior of ceramic-reinforced aluminum matrix composites, especially in wire-cut electric discharge machining techniques, concerning electrical process parameters.This study aims to analyze the wire-cut electric discharge machining behavior of zirconia and graphene-reinforced AA6061 nanocomposites using Taguchi's method.The objective is to optimize electrical process parameters and identify the predominant machining characteristics for this specific composite material.

Experimental analysis
2.1.Materials AA6061 alloy serves as the matrix material, while Graphene (C) and Zirconia (ZrO 2 ) act as nano-sized reinforcements in the fabrication of hybrid composites.The ultrasonic and mechanically assisted stir casting method was employed to create the composite material.The base matrix is an ingot of AA6061 with a diameter of 70 mm, and the reinforcements consist of nano-scale irregularly shaped ZrO 2 and graphene particles, each with an average size of 200 nm.The mechanical and chemical characteristics of AA6061 are highlighted in tables 1 and 2. Table 3 lists Zirconia's mechanical characteristics.

Composite fabrication
There is a wide variety of fabrication methods for synthesizing nanomaterials, including recent novel methodologies such as one-pot facile and silver nano-dot methods [19,20].The use of nano-scale reinforcing particles poses challenges in achieving effective particle dispersion due to the high surface area to volume ratio, leading to clustering and agglomeration in specific areas.This can have consequences on the end results,  including poor wettability, ultimately resulting in weaker mechanical properties.To address these challenges, the ultrasonic-aided stir casting technique is employed.In this technique, the ultrasonic probe assists in uniformly dispersing the particles throughout the casting, mitigating issues associated with poor dispersion.The casted sample is illustrated in figure 1.The fabricated aluminum metal matrix hybrid nanocomposite (AMMHNCs) with varying weight percentages is presented in table 4.

Wire-cut electric discharge machining
The design of experiments using Taguchi's method was employed to limit the number of experimental runs.
Table 5 provides information on the number of levels and parameters used for analyzing the machining performance.The L 27 orthogonal array (OA) presented in table 6 outlines the design of experiments with respect to the output variables, including kerf width, material removal rate, and surface finish, for three different specimens.Machining analysis was conducted using a five-hub CNC ELEKTRA, MAXICUT 434 model from Electronica Machine Tools Ltd, as depicted in figure 2. Figures 3, 4, and 5 illustrate the machined samples, with nine samples each for three different compositions: AMMHNCs1, AMMHNCs2, and AMMHNCs3.

Microstructural examination
Figures 6(a)) to (c) displays SEM images of AMMHNCs.The ultrasonication method ensures homogeneous particle dispersion across the matrix materials, as evidenced by the grain structure showcasing graphene and zirconia reinforcements distributed in dendritic patches.The matrix and the nano-scale reinforcing materials exhibit strong intermolecular connections due to the absence of porosity and agglomeration.The uniform dispersion resulting from the dispersion strengthening process prevents dislocation movements, thereby enhancing the machining characteristics of nanocomposites.

Machining analysis
The analysis was centered on material removal rate (MRR), kerf width (KW) and surface roughness (SR) aiming to identify the optimised parameters using the Signal-to-Noise (S/N) ratio.Based on the experimental data, the S/N ratio for variables at each level was computed, and the distinctive effects of the S/N data processing parameters were illustrated.The best values of output responses, determined through average performance factors were obtained by interpreting the analysis of variance and response graphs.By employing L 27 OA experiments, the impact of process factors on KW, MRR, and SR was examined.

Kerf width analysis
Figures 7(a)) to (c) illustrate the kerf width images obtained during the measurement of the composite specimens.It is evident that the kerf width for the chosen parameter set values is smooth and exhibits a narrow width.Figure 9 depicts the main effects plot for standard means of KW.It is apparent that the lowest kerf width is achieved with the parameters: 4 μs pulse on-time, 3 μs off-time pulse, peak current of 3 amps, reinforcement of 10 wt%, and wire feed rate of 4 m /min.Table 7 presents the ANOVA for the importance of pulse on-time, input peak current and off-time pulse.The p-value of 0.025 (<0.05) indicates that the most influential parameter in KW is the wire feed rate.The R 2 value for kerf width is 0.9958, representing the proportion of variance between the variables.The adjusted R 2 is 0.9834, and the predicted R 2 is 0.9544.

MRR analysis
Figure 10 depicts the main effects plot for standard means of material removal rate.It is evident that the highest MRR is achieved with the parameters: on-time of 2 microseconds, off-time pulse of 1 microsecond, peak current of 5 amps, wire feed rate of 3 m min −1 , and reinforcement of 15 wt%.The MRR increases with higher values of on-time and peak current.The rate of material removal rises as the pulse expands, owing to the increase in discharge energy.Similarly, lowering the off-time pulse results in more discharges over a given period, contributing to an increased MRR.Table 8 presents the ANOVA for the importance of on-time, input peak current, and off-time pulse.The reinforcement, wire feed rate, and interactions of the input peak current with other output variables are all incorporated in the error term.The p-value of 0.049 (<0.05) indicates that the most influential parameter in MRR is the pulse on time.The R 2 value for MRR is 0.9964, representing the proportion of variance between the variables.The adjusted R 2 is 0.9718, and the predicted R 2 is 0.9386.feed rate of 3 m min −1 , and reinforcement of 5 wt%.Increasing on-time, off-time pulse and wire feed rate, input peak current and reinforcement all contribute to improving Sr The energy of discharge increases with the expansion of the pulse, resulting in a larger discharge and a subsequently larger crater, leading to a lower surface finish.As the off-time pulse increases, the frequency of discharges decreases, contributing to lower surface roughness due to more consistent machining.Table 9 presents the ANOVA for the importance of on-time,

Influence of control parameters
Figure 12 depicts the main effect graph for S/N ratios to determine the overall quality of the process.It is evident that the parameters of on time of 4 microseconds, off time of 3 microseconds, input current of 3 amps, and reinforcement wt.% of 5 yields the optimum parameter levels against the three output variables.The response table S/N ratio statistics for each factor level are presented in table 10.The importance of each component of the   response is indicated by the ranks.Pulse on time demonstrates the greatest impact on responses, as supported by the rankings and delta values, followed by other process variables.Figures 13, 14, and 15 display the interaction between the control parameters for KW, MRR, and SR, respectively.In figure 13, the highest mean strength for the minimum kerf width is achieved with on time of 2 microseconds, off time of 1 microsecond, and input current of 1 amp.Similarly, in figure 14, the highest mean strength for the maximum MRR is observed with ontime of 2 microseconds, off-time of 1 microsecond and input current of 1 amp.Conversely, in figure 15, the highest mean strength for the minimum SR is achieved with on-time of 6 microseconds, off-time of 3 microseconds, and input current of 1 amp.It is noteworthy that the interaction between the control parameters differs for surface roughness compared to kerf width and material removal rate.

Confirmation experiments
Using the optimal parameter values, the estimated values for the output variables of kerf width, surface roughness and material removal rate are obtained.To select the best parameters, the results of the experiments are thoroughly examined.Based on the optimal parameters for achieving the lowest kerf width, highest MRR

Conclusion
In this investigation work, the impact of control factors on kerf width, surface roughness and material removal rate in WEDM of AA6061/C/ZrO2 hybrid nano composites using Taguchi's method were analyzed, with the following determinations: • The optimized parameters for achieving the minimum kerf width are: on-time of 4 microseconds, off-time of 3 microseconds, peak current of 3 amps, wire feed of 4 m/min and reinforcement of 10 wt %.
• For the maximum material removal rate, the optimized parameters are: on-time of 2 microseconds, off-time of 1 microseconds, peak current of 5 amps, wire feed of 3 m/min and reinforcement of 15 wt %.  • The optimized parameters for the minimum surface roughness are: on-time of 2 microseconds, off-time pulse of 1 microseconds, peak current of 3 amps, wire feed of 3 m/min and reinforcement of 5 wt %.
• The most significant factor based on ANOVA for kerf width is wire feed rate, while for surface roughness and material removal rate, it is pulse on-time.

Figure 7 .
Figure 7. Kerf width image of AMMHNC1 specimen for 9 input parameters (b) Kerf width image of AMMHNC2 specimen for 9 input parameters (c) Kerf width image of AMMHNC3 specimen for 9 input parameters.

Figure 8 .
Figure 8. (a)) Surface roughness profile of AMMHNC1 specimen for 9 input parameters (b) Surface roughness profile of AMMHNC2 specimen for 9 input parameters (c) Surface roughness profile of AMMHNC3 specimen for 9 input parameters.

Figure 9 .
Figure 9. Main effects plot (Data means) for Kerf width in mm.

Figure 10 .
Figure 10.Main effects graph for MRR in mm 3 /min.

Figure 11 .
Figure 11.Main effects graph for SR in microns.

Figure 12 .
Figure 12.Main effects graph for S/N ratios.

Figure 13 .
Figure 13.Interaction graph of Input parameters for Kerf width.

Figure 14 .
Figure 14.Interaction graph of Input parameters for Material Removal rate.

Table 5 .
Machining / control factor and its levels.

Table 6 .
Design of experimentand its results for AMMHNCs.

Table 10 .
Response table for S/N Ratios.