Optimization of stirrer parameters by Taguchi method for a better ceramic particle stirring performance in the production of Aluminum Alloy Matrix Composite

Abstract

Aluminum alloy ceramic matrix composites have become widely used in a variety of sectors and will grow in the future with the growth of measures to reduce environmental pollution and the necessity to reduce the use of fossil fuels. Stir casting is the most cost-effective method for mass-producing Aluminum Alloy Matrix Composite; however, it still has certain challenges that need to optimize for its performance. The current gap is to anticipate, recognize, and strive to minimize casting flaws during the liquid phase. Taguchi method for the design of the experiment was used to optimize the stirrer parameters. The signal-to-noise (S/N) ratio (larger is better), hardness, and ultimate tensile strength were employed as response output variables. In this work, it was confirmed that the mechanical properties of aluminum alloys can be enhanced better if the method of adding SiC and SiO2 particles as ceramic reinforcements is improved. The contribution of three parameters in enhancing mechanical properties has been identified. The research proved that there is a significant relationship between the Stirrer Design and the mechanical properties. The optimal stirrer design (D5) for mixing ceramic reinforcements with melted aluminum alloy has been chosen from different designs. Mechanical properties under the effect of the most important stirring parameters have been explained. Finally, adding silicon carbide instead of silica sand enhances the mechanical properties even further.

Keywords:

• Aluminum Alloy matrix composite
• stir casting
• Stirrer parameter optimization
• stirring performance
• the design of experiment (DOE)
• Taguchi
• mechanical properties

1. Introduction

The interest in finding stronger, lighter, and more environmentally friendly materials is constantly increasing as the threats of energy scarcity and environmental pollution increase. Aluminum alloy ceramic particle composites with a wide range of physical and mechanical characteristics might respond to this trend. Stir casting (vortex technique) is well known in the commercial industry as a low-cost method of producing MMCs. It offers advantages such as flexibility and adaptability for large-scale manufacture, but it also has disadvantages such as poor wettability (P. Kumar et al., 2021). The homogeneous dispersion of the reinforcing particles in the matrix determines the performance of production and products, which is impacted by stir-casting parameters. In a wide range of chemical and industrial operations, the mixing process is crucial. Many factors influence mixing efficiency, but one of the most important is impeller design. Attaining homogeneity in mixing processes is a demanding task for efficient and economical use. Less stirring leads to non-uniform distribution of particles and in addition to wasting energy, excess stirring forms a clustering of particles at some place. All of the factors stated have an impact on mixing efficiency, either alone or in combination with others. However, the primary focus of this study will be on the stirrer design, as well as stirring time and speed, both of which have been their positive effects on the mechanical properties of one of the logical axioms. Figure 1 categorizes the stir casting parameters into three groups based on [(Hanizam et al., 2019; Hashim et al., 2002; Krishnan et al., 2021; A. Kumar et al., 2020; Mehta & Sutaria, 2020; Muhammad et al., 2021; Prabu et al., 2006; Rasyid et al., 2018, 2019; Ravikumar et al., 2018; Sahu & Sahu, 2017; Shanmughasundaram & Subramanian, 2012; Singh et al., 2020; Su et al., 2010; Vishnu Prasad & Jayadevan, 2016; Yang et al., 2017)]. Also, after releasing two review papers in 2021 (Muhammad & Jalal, 2022a; Muhammad et al., 2021) about stir casting parameters, thre the authors selected Stirrer Design, Stirring Time, and Stirring Speed as the main research variables to explore the relationship between them.

Figure 1. The three categories of stir casting.

Figure 2 summarized the flow chart of the research work used to achieve the goal.

Figure 2. Diagram of the research work.

2. Matrix material

Composites are divided into three categories based on the kind of matrix material used: polymer matrix composites (PMCs), ceramic matrix composites (CMCs), and metal matrix composites (MMCs). In industrial applications, MMCs are the most used form of composite (Kareem et al., 2021). Aluminum matrix composites (AMCs) use pure aluminum or an alloy as the matrix and are becoming more popular in industrial applications due to their outstanding mechanical and tribological properties. The properties of Al alloys might be greatly tailored by adding ceramic reinforcing particles by stir casting. AA 6063 ingot was chosen as a matrix alloy which was analyzed carefully using an X-MET8000 Handheld XRF analyzer and the compositions of the used martial are shown in Tables 1, 2, and 3.

Table 1. Composition of the AA 6063 ingot

Element	Al	Si	Fe	Zr	Mn	Mg	Zn	Ni	Pb	Sn	Ti
Average content	98.76	0.19	0.16	0.02	0.02	0.74	<0.00	<0.00	0.04	0.07	<0.01

Table 2. Chemical composition of Silicon Carbide (SiC)

Component	SiC	C	Si	SiO2	Fe2O3
Wt.%	97.2	0.28	0.23	1.73	0.56

Table 3. Chemical composition of silica sand*

Constituent	% Percentage
Sio2	90.04
Al2O3	3.75
Fe2O3	1.94
MgO	1.02
Cao	0.99
Loss on ignition	2.26

*Silica sand samples from the Mass company (Iraq) (10 − 40-micron size) were used.

2.1. Reinforcement materials

According to the percentage of reinforcing particles used in the manufacture of composites; Silicon carbide (SiC) looks to be one of the most popular reinforcements used at around 18% in the fabrication of composites. In addition, silica (SiO2) makes up around 8% of the ceramic reinforcements used (P. Kumar et al., 2021). The majority of studies have found that increasing the percentage of ceramic particle reinforcement in specific proportions while ensuring a homogeneous distribution will improve the mechanical properties. According to Alaneme and Aluko (2012) who studied the tensile and fracture behavior of AA6063 silicon carbide particle composites, the strength of Al alloy 6063 matrix composites is significantly boosted when 9- and 12-volume fractions of SiC are used as reinforcement. Finally, they declared Al (6063) alloy is a good candidate to serve as a matrix for the creation of SiC-reinforced aluminum, in general (K. K. Alaneme & Aluko, 2012). Alaneme and Bodunrin (2013) investigated the mechanical behavior of AA 6063/Al2O3p composites using a two-step stir-casting process. They confirm that the tensile strength, yield strength, and hardness increased with the volume percent alumina, and the strain to fracture and fracture toughness decreased (K. Alaneme & Bodunrin, 2013). Rozhbiany and Jalal (2020) explored ceramic reinforced were combined with the alloy Al 6063 at a consistent rate of 5 w%. According to microstructure observation, metal matrix composites exhibit strong grains and tight grain boundaries when compared to Al 6063 alloys (Rozhbiany & Jalal, 2019). Stir-and-vortex casting is generally used to produce aluminum and Silica (SiO2) reinforcement composites. Although there is a challenge, which is the poor wettability between molten aluminum and quartz particles (Joyson Abraham et al., 2016; Khelge et al., 2021).

3. Experimental procedure

Muhammad and Jalal (2022) used visualization and a numerical approach to select the optimal stirrer Design between four stirrer designs (Four blades Single-stage, Four blades Double stages, Four blades Multi-stages, and Helical stirrer). They confirmed that Four blades Multi-stages own less performance compared to others and Helical Stirrer is optimal (Muhammad & Jalal, 2022b). The experiments are designed to choose the best stirrer designs from among four previous designs. After cleaning Al6063 ingots, they were cut to proper sizes, weighed, and charged into a graphite crucible placed in the digital electric furnace. Silicon carbide and Silica particles with microsize were selected as the reinforcement. 5%wt volume percent was evaluated using charge calculations. Figure 3 shows the stir casting experimental setup used for fabricating MMC. It consists of an electrical-inducing furnace for melting. AA6063 ingots after about two hours began to melt at a high temperature of 750°C ± 30◦C. Stirring was done using a SEIWA MG-915 Radial Drill coupled with four distinct stirrer designs (four blades single, double, and multi-stage stirrer, and helical) as shown in Figure 4.

Figure 3. SEIWA MG-915 radial drill setup with stirrers and an electrical inducing furnace for melting.

Figure 4. Five different designs of stirrers, D1, D2, D3, D4 and D5.

The stirrers manufactured from 304 austenitic stainless steel is submerged in the graphite crucible to generate a vortex, which is the first step in the stirring procedure. Then, reinforcement particles add up at regular time intervals into the center of the vortex by the feeder specified in the stir casting setup. The stirring process is continued for a certain amount of time. In this work, three stirring times are conducted: 2, 4, and 6 minutes, respectively. The reinforced particles required to have SiC or SiO2 wt. 5 % volume percent were evaluated using charge calculations. The stirring operation speed was performed at (100, 240, and 540) RPM. Finally, the molten metal is poured into a cylindrical steel mold which is preheated for 5 min. The specimen is extracted from the mold after cooling at normal air temperature. The above process is repeated 9 times according to the Taguchi method L9 orthogonal for three factors for three levels.

4. Mechanical properties measurement

The following equation (1)(1) $\mathit{HV}=1.854\left(\mathit{F}/D^2\right)$(1) was used to determine the Vickers hardness values:

(1) $\mathit{HV}=1.854\left(\mathit{F}/D^2\right)$(1)

where H_V is the Vickers hardness of composite samples; F is the applied load in (N) and D² is the area of the indentation (measured in square millimeters). The test was performed at three separate points in the center and circumferences of the samples, and the average was recorded. Tensile tests were carried out at room temperature through the use of TERCO MT 3037 universal testing machine according to ASTM A370 standard small-size measurement for the tensile test sample of metals and related alloys.

5. Taguchi method for (DOE)

In the beginning, DOE is used to choose the top two of the three designs, D1, D2, and D3 from four stirrers as shown in Table 4

Table 4. Five different stirrer designs

Symbol	Stirrer Design	Images
D1	Four blades Single-stage
D2	Four blades Double stages
D3	Four blades Multi-stages
D4	Helical stirrer with a cylindrical shaft
D5	Helical stirrer with a helical shaft

Taguchi method L9 orthogonal array with 9 experimental runs for each ceramic reinforcement. It was used to optimize, namely, Stirrer Design, Stirring Speed, and Stirring Time, with factorial levels for each factor shown in Table 5. The “Larger is better” S/N ratio was used to anticipate the optimum values in this investigation since improved hardness and tensile strength of the composites were desirable. Also, regression equations and ANOVA were conducted using the MINITAB software program to investigate the contribution of the components, at a level of significance of 0.05 (St, 2022). Table 6 shows the results of hardness and Ultimate Tensile strength of nine runs for SiC and SiO2 reinforcement.

Table 5. Levels and control factors

Symbol	Control factor	Level 1	Level 2	Level 3
A	Stirrer Design	1	2	3
B	Stirring Time (Minutes)	2	4	6
C	Stirring Speed (RPM)	100	240	540

Table 6. Design matrix and experimental observations of 5 % SiC and SiO2 reinforcements

Runs	Stirrer Design D1, D2, and D3	Stirring Speed	Stirring Time	Vickers hardness (N/mm2)		Ultimate Tensile Strength Mpa
				SiC	SiO2	SiC	SiO2
1	1	100	2	44.7	31.80	161	122
2	1	240	4	47.4	32.84	163	126
3	1	540	6	49.7	33.62	165	127
4	2	100	4	42.0	31.80	161	119
5	2	240	6	45.8	33.76	162	124
6	2	540	2	44.7	32.32	156	125
7	3	100	6	40.5	31.58	147	122
8	3	240	2	36.0	30.28	145	118
9	3	540	4	41.0	31.06	148	121

6. Result and discussion

This study shows the effect of the Stirrer Design on the liquid, not the semi-solid casting approach, and without any improvement in poor wettability. The mechanical responses were analyzed based on Taguchi (DOE) approach, as illustrated in the following figures and tables generated from Minitab software reports:

6.1. First case: SiC—Hardness response

As illustrated in Table 7, at a 95% confidence level, the P-value of the Stirrer Design variable has more than 0.05 very lightly it is (0.058) indicating very faintly that it is not significant. The other variables (Stirring Speed and Stirring Time) have more than 0.05 (0.348, and 0.247), respectively, indicating that they are not significant but at varying degrees.

Table 7. ANOVA of SN ratios for hardness responses for SiC

Source	DF	Seq SS*	Adj SS	Adj MS	F	P
Stirrer Design	2	4.1396	4.1396	2.0698	16.28	0.058
Stirring Speed	2	0.4763	0.4763	0.2382	1.87	0.348
Stirring Time	2	0.7742	0.7742	0.3871	3.04	0.247
Residual Error	2	0.2543	0.2543	0.1272
Total	8	5.6445

*Seq SS.: Sequential sums of squares

As shown in Figure 5, the main effects plot for SN ratios the optimal condition is (Stirrer Design (3), Stirring Time (6 minutes), and Stirring Speed (540 RPM).

Figure 5. Main effects plot for SN ratios for hardness responses for SiC.

The percentage contribution of each variable in the overall variance is the Stirrer Design (73.3%) was the major contributing factor influencing the hardness of the composites followed by Stirring Time (13.7%) and finally Stirring Speed (8.4%), while the rest is Residual Error which is equal 4.5.

6.2. Second case: SiC—Tensile strength response

As illustrated in Table 8, at a 95% confidence level, the P-value of the Stirrer Design variable has less than 0.05 it is (0.011) indicating that it is significant. The other variables (Stirring Time and Stirring Speed) have more than 0.05 (00.162, and 0.975), respectively, indicating that they are not significant.

Table 8. ANOVA of SN ratios for tensile strength responses for SiC particles addition

Source	DF	Seq SS	Adj SS	Adj MS	F	P
Stirrer Design	2	1.41623	1.41623	0.708115	86.98	0.011
Stirring Speed	2	0.00041	0.00041	0.000206	0.03	0.975
Stirring Time	2	0.08426	0.08426	0.042130	5.17	0.162
Residual Error	2	0.01628	0.01628	0.008141
Total	8	1.51718

As shown in Figure 6, the main effects plot for SN ratios of the optimal condition is (Stirrer Design (1), Stirring Time (6 minutes), and Stirring Speed (240 RPM).

Figure 6. Main effects plot for SN ratios for tensile strength responses for SiC.

The percentage contribution of each variable in the overall variance is the Stirrer Design (93.3%) was the major contributing factor influencing the hardness of the composites followed by Stirring Time (5.6%), and finally, Stirring Speed is about (zero%) while the rest is Residual Error which equals 1.1 which means closely grouped data points around the fitted regression line.

6.3. Third case: Sio2- Hardness response

In the third case, the trend of the effect of factors when adding silica particles followed in the Table 9, at a 95% confidence level, the P-value of the variables (Stirrer Design and Stirring Time) has less than 0.05, (0.039 and 0.063), respectively indicating that they are significant but Stirring Speed has the least effect on hardness.

Table 9. ANOVA of SN ratios for hardness responses for Silica sand particles addition

Source	DF	Seq SS	Adj SS	Adj MS	F	P
Stirrer Design	2	0.43849	0.43849	0.219245	24.76	0.039
Stirring Speed	2	0.04676	0.04676	0.023382	2.64	0.275
Stirring Time	2	0.26533	0.26533	0.132663	14.98	0.063
Residual Error	2	0.01771	0.01771	0.008853
Total	8	0.76829

As shown in Figure 7, the main effects plot for SN ratios of the optimal condition is (Stirrer Design D1), Stirring Time (6 minutes), and Stirring Speed (540 RPM).

Figure 7. Main effects plot for SN ratios for hardness responses for Silica sand.

The impact contribution of factors on the Hardness; of the Stirrer Design (57.1%) has the major contribution followed by Stirring Speed (6.1%), and finally Stirring Time (34.5%).

6.4. Fourth case: Sio2 tensile strength response

In the fourth case, the trend of the effect of factors when adding silica particles followed in Table 10, at a 95% confidence level, the P-value of variables of Stirrer Design, Stirring Speed, Stirring Time are (0.306, 0.466, and 0.53), respectively, indicating that they are not significant. The closest value to significance is the Stirrer Design followed by the Stirring Speed and then Stirring Time.

Table 10. ANOVA of SN ratios for tensile strength for silica sand

Source	DF	Seq SS	Adj SS	Adj MS	F	P
Stirrer Design	2	0.16340	0.16340	0.08170	2.27	0.306
Stirring Speed	2	0.08276	0.08276	0.04138	1.15	0.466
Stirring Time	2	0.06401	0.06401	0.03200	0.89	0.530
Residual Error	2	0.07213	0.07213	0.03606
Total	8	0.38229

As shown in Figure 8, the Main effects plot for SN ratios, Data means the optimal condition is Stirrer Design (D1), Stirring Time (6 minutes), and Stirring Speed (540 RPM).

Figure 8. Main effects plot for SN ratios for tensile strength responses for silica sand.

It can be observed that the Stirrer Design (42.7%) has the major contribution followed by Stirring Time (21.6%), and finally Stirring Speed (16.7%). As indicated, the tensile strength of the composites not followed the same trend of Hardness properties. The findings of the S/N ratio and ANOVA indicate that the effect of Stirring Time is more than the effect of Stirring Speed.

In the first stage, the four-blade multi-stage stirrer (D3) was left and confirmed the study (Muhammad & Jalal, 2022b) result that stated in the liquid state the four blades single-stage (D1) and four blades double stages (D2) conduct higher performance than the four-blade multi-stages stirrer (D3) when the viscosity of melt Aluminium alloys like water. This fact is reversed when the viscosity of the matrix decreases and turns from a liquid to a slurry state (Muhammad & Jalal, 2022b).

In the second stage, the Taguchi method was repeated for the remaining factors the results were checked after adding the fourth stirrer (D4) (Helical stirrer) to compare it with the four blades single-stage (D1) and Four blades double stages (D2). Table 11 shows the results of hardness and Ultimate Tensile strength of nine runs for SiC and Silica sand reinforcement.

Table 11. Design matrix and experimental observations of 5 % SiC and SiO2 reinforcements

Runs	Stirrer Design D1, D2, and D4	Stirring Speed	Stirring Time	Vickers hardness (N/mm2)		Ultimate Tensile strength Mpa
				SiC	SiO2	SiC	SiO2
1	1	100	2	44.7	31.80	161	122
2	1	240	4	47.4	32.84	163	126
3	1	540	6	49.7	33.62	165	127
4	2	100	4	42	31.80	161	119
5	2	240	6	45.8	33.76	162	124
6	2	540	2	44.7	32.32	156	125
7	3	100	6	49.4	34.32	167	126
8	3	240	2	48.7	32.70	163	124
9	3	540	4	50.7	33.70	166	127

6.5. Fifth case: SiC—Hardness response

As explained in Table 12, the P-value belonging to the variables (Stirrer Design, Stirring Speed, and Stirring Time) have less than 0.05 (0.007, 0.022, and, 0.038), respectively, indicating that they are significant but at varying degrees.

Table 12. ANOVA of SN ratios for hardness responses for SiC particles addition

Source	DF	Seq SS	Adj SS	Adj MS	F	P
Stirrer Design	2	1.55007	1.55007	0.775035	136.36	0.007
Stirring Speed	2	0.49395	0.49395	0.246976	43.45	0.022
Stirring Time	2	0.28435	0.28435	0.142174	25.01	0.038
Residual Error	2	0.01137	0.01137	0.005684
Total	8	2.33974

As shown in Figure 9, the Main effects plot for SN ratios the optimal condition is (Stirrer Design (D4), Stirring Time 6 minutes, and Stirring Speed 540 RPM.

Figure 9. Main effects plot for SN ratios for hardness responses for SiC.

The contribution of variables to Hardness is depicted in the table 18. Stirrer Design (66.2%) is the most important factor, followed by Stirring Speed (20.8%) and Stirring Time (12.2%)

6.6. Sixth case: SiC—tensile strength response

The trend of factor effect on tensile strength behavior has shown in Table 13. The P-value of the Stirrer Design is 0.043 indicating that it is significant. The P-values of Stirring Time and Stirring Speed are (0.06 and 0.748), respectively, that which are less than 0.05 indicating that they are not significant; however, there is an overwhelming variation when the stirring time is extremely near to be significant.

Table 13. ANOVA of SN ratios for hardness responses for SiC particles addition

Source	DF	Seq SS	Adj SS	Adj MS	F	P
Stirrer Design	2	0.139912	0.139912	0.069956	22.07	0.043
Stirring Speed	2	0.002140	0.002140	0.001070	0.34	0.748
Stirring Time	2	0.100106	0.100106	0.050053	15.79	0.060
Residual Error	2	0.006339	0.006339	0.003169
Total	8	0.248497

As shown in Figure 10, the Main effects plot for SN ratios, Data means of the optimal condition is Stirrer Design (D4), Stirring Time, 6 minutes, and less Stirring Speed, 100 RPM for the best tensile strength.

Figure 10. Main effects plot for SN ratios for tensile strength responses for SiC.

The contribution of variables to tensile strength is shown in table 18. Stirrer Design (56.3%) is the most important factor, followed by Stirring Time (40.3 %), and Stirring Speed (0.9%). As indicated the tensile strength of the composites not followed the same trend of Hardness properties. The influence of Stirring Speed outweighs the effect of Stirring Time.

6.7. Seventh case: Sio2 hardness response

As illustrated in Table 14, at a 95% confidence level, the P-value of the Stirring Time variable has less than 0.05 it is (0.078) indicating that it is significant. The other variables (Stirrer Design and Stirring Speed) have more than 0.05 (00.182, and 0. 374), respectively, indicating that they are not significant.

Table 14. ANOVA of SN ratios for hardness responses for SiC

Source	DF	Seq SS	Adj SS	Adj MS	F	P
Stirrer Design	2	0.10978	0.10978	0.05489	4.50	0.182
Stirring Speed	2	0.04093	0.04093	0.02046	1.68	0.374
Stirring Time	2	0.28681	0.28681	0.14340	11.75	0.078
Residual Error	2	0.02440	0.02440	0.01220
Total	8	0.46191

As shown in Figure 11, the Main effects plot for SN ratios the optimal condition is (Stirrer Design (3), Stirring Time (6 minutes), and Stirring Speed (540 RPM).

Figure 11. Main effects plot for SN ratios for hardness responses for SiO2.

It can be observed that the Stirrer Design (62.1%) has the major contribution followed by Stirring Time (21.79%), and finally Stirring Speed (1.49%). As indicated, the tensile strength of the composites not followed the same trend of Hardness properties. The findings of the S/N ratio and ANOVA indicate that the effect of Stirring Time is more than the effect of Stirring Speed.

6.8. Eighth case: Sio2 tensile strength response

Table 15, illustrates the P-value of variables (Stirring Speed, Stirrer Design, and Stirring Time) have more than 0.05 values (0.257,0.359, and 0.549), respectively, indicating that they are not significant whereas the Stirring Speed has less than P-value.

Table 15. ANOVA of SN ratios for tensile strength responses for SiC

Source	DF	Seq SS	Adj SS	Adj MS	F	P
Stirrer Design	2	0.07394	0.07394	0.03697	1.78	0.359
Stirring Speed	2	0.12024	0.12024	0.06012	2.90	0.257
Stirring Time	2	0.03403	0.03403	0.01702	0.82	0.549
Residual Error	2	0.04150	0.04150	0.02075
Total	8	0.26972

As shown in Figure 12, the Main effects plot for SN ratios, Data means the optimal condition is Stirrer Design (D4), Stirring Time (6 minutes), and Stirring Speed (540 RPM).

Figure 12. Main effects plot for SN ratios for tensile strength responses for SiO2.

The contribution of variables to Hardness is Stirring Speed (44.6%) is the most important factor followed by Stirrer Design (27.4%), and Stirring Time (12.6%). As shown, the tensile strength of the composites did not follow the Tensile Strength trend. The S/N ratio and ANOVA results show that the influence of Stirring Speed outweighs the effect of Stirring Time.

As shown in Table 16, the Stirrer design is the most significant contribution in six of eight cases. This confirms what they found when studying the impact of stirrer geometry (Jebeen Moses & Joseph Sekhar, 2017; Krishnan et al., 2021; Mehta & Sutaria, 2020). Stirring Time is the most important factor in the seventh case: Sio2- Hardness response. In the eighth case (Sio2 Tensile Strength response), Stirring Speed is considered the most contributing. It was noticed that the type of ceramic has a clear impact. Sio2 reinforcement particles respond less to parameters than SiC particles, particularly for stirrer design.

Table 16. Summary of the most important contribution to the eight cases

Cases	Percentage Contribution for each parameter
	Stirrer Design	Stirring Speed	Stirring Time
First case: SiC—Hardness response	73.3	8.4	13.7
Second case: SiC—Tensile Strength response	93.4	0.0	5.6
Third case: Sio2- Hardness response	57.1	6.1	34.5
Fourth case: Sio2 Tensile Strength response	42.7	21.6	16.7
Fifth case: SiC—Hardness response	66.2	21.1	12.2
Sixth case: SiC—Tensile Strength response	56.3	40.3	0.9
Seventh case: Sio2- Hardness response	23.8	8.9	62.1
Eighth case: Sio2 Tensile Strength response	27.4	44.6	12.6
The participation rate for the eight cases	55.025	18.875	19.7875

7. Regression analyses

7.1. Regression equation for measuring hardness and tensile strength for SiC (D1, D2, and D3)

Hardness = 43.25–4.050 Stirrer Design + 0.00632 Stirring Speed + 0.883 Stirring Time

Tensile strength = 164.83–8.17 Stirrer Design—0.00018 Stirring Speed + 1.000 Stirring Tim

7.2. Regression equation for measuring hardness and tensile strength for SiO2 (D1, D2, and D3)

Hardness = 32.031–0.890 Stirrer Design + 0.00118 Stirring Speed + 0.380 Stirring Time

Tensile strength = 122.54–2.333 Stirrer Design + 0.00726 Stirring Speed + 0.667 Stirring Time

7.3. Regression equation for measuring hardness and tensile strength for SiC (D1, D2, and D4)

Hardness = 37.56 + 1.17 Stirrer Design + 0.00630 Stirring Speed + 0.567 Stirring Time

Tensile strength = 152.09 + 1.17 Stirrer Design—0.00145 Stirring Speed + 1.167 Stirring Time

7.4. Regression equation for measuring hardness and tensile strength for SiO2 (D1, D2, and D4)

Hardness = 30.199 + 0.410 Stirrer Design + 0.001157 Stirring Speed + 0.407 Stirring Time

Tensile strength = 119.28 + 0.333 Stirrer Design + 0.00853 Stirring Speed + 0.500 Stirring Time

When observing the regression equations for both ceramics cases, these conclusions have been reached:

1. The Stirrer Design has a significant impact on mechanical properties.

2. The response of mechanical properties to SiC is larger than Silica (SiO2).

3. The response of the tensile strength among mechanical properties to SiC ceramic is higher than the hardness.

4. Finally, there is no clear trend in the role of Stirring Speed and Stirring Time in detail on mechanical properties of Hardness and Tensile Strength. The following is an explanation of the reasons for the variation in the size of the contribution and the absence of a significant relationship between variables with mechanical properties:

   1. The difference in density between SiC and Silica. Higher than AA6063 as SiC or less than AA6063 like silica.

   2. The change in the physical properties such as viscosity of the matrix with the temperature change and the wt% of ceramics particles inside the matrix.

   3. In this work, the stir-casting system is not closed. There is a gap for unwanted substances to enter the mixture, especially at high speed.

To deal with the above data, the following have been stated:

The helical stirrer is optimal. It seems that this design performs the same function, whether the viscosity is high or low. The reasons for the differences in the value of significance and the size of the contribution are due to the high-speed stirring of the vortex, which is created and which pushes the particles to the crucible wall directly as shown in Figure 13. In the helical stirrer to reduce the vortex around the shaft, it is preferable to change the shaft design from cylindrical to helical as shown in Figure 14, which is called (D5).

Figure 13. The velocity contour of the helical case (Muhammad & Jalal, 2022b).

Figure 14. (D5) Helical stirrer with helical shaft.

The variations in the significance and ratio of the contribution are related to the long Stirring Time within poor wettability of ceramics particles that will lead to more clustering (agglomeration), segregation, and sedimentation. Also when particle size increases as noted in SiC particles which have a bigger size than Silica particles, drag force decreases, and particles have a greater inclination to fall to the bottom by gravity rather than being entrained by fluid flow. The following are other problems that have been identified: Poor wettability between the particles and matrix has been noted the particles of ceramic reinforcements stayed on the edges of the stirrer and the ceramic particles have been noted at bottom of the crucibles. Another reason associated with low mechanical properties is that some studies have revealed that SiC will react with liquid aluminum to form A14C 3 according to the reaction (1)

(2) $4\mathit{AI}+3\mathit{SiC}\mathit{�}\mathit{A}14\mathit{C}3+3\mathit{Si}$(2)

The main detrimental impact of this reaction is it produces A14C3 at the interface between the reinforcement and the matrix, which could result in a degradation of the reinforcement strength and the interracial strength (Lloyd et al., 1989).

8. Comparison between D4 and D5 stirrer designs

Tables 17 and 18 show the results of the mechanical properties of the (D4) Helical stirrer with a cylindrical shaft and (D5) Helical stirrer with a helical shaft Stirrer design parts (1 and 2). The results confirm the validity of the hypothesis that the D5 yields better mechanical properties.

Table 17. Results of mechanical properties of D4 and D5 Stirrer designs part (1)

Stirrer designs	Mechanical properties
	Tensile strength		Vickers hardness (N/mm2)
	5% SiC	5% Silica	5% SiC	5% Silica
D4 Helical stirrer with a cylindrical shaft	167	127	50.7	34.32
D5 Helical stirrer with a helical shaft	172	132	56.4	37.32

Table 18. Results of mechanical properties of D4 and D5 Stirrer designs part (2)

Stirrer designs	Mechanical properties
	Elongation(%)		Impact strength (J)
	5% SiC	5% Silica	5% SiC	5% Silica
D4 Helical stirrer with a cylindrical shaft	14	11	9.4	6.30
D5 Helical stirrer with a helical shaft	16	13	10.2	7.2

9. Microstructure analyses

9.1. Optical microstructure analysis

Microscopic examination of the composites was carried out by optical and scanning electron microscopy. According to the American Society for Testing Materials (ASTM) standard, the specimens were chemically etched for 50 seconds (Annual Book of ASTM G 67, 2004). The etchant solution consists of 90 ml of water, 4 ml of sulphuric acid (H2S04), 2 ml of chromium (III) oxide (Cr2O3), and 4 ml of Hydrofluoric acid (Hf). Then specimens were placed on the metallurgical microscope, model IMM 901 from Metkon. Images with 100 micron magnification were taken respectively as shown in Figures 15 A, B, and C. It has been shown that when helical stirrer (D5), stirring speed (540 PRM), and longest stirring duration (6 minutes) were used for both ceramic reinforcements, reinforcement particles are well dispersed in the matrix. Pointing out that silicon carbide has higher uniformity than silica. The factors that contributed to the weaker properties of silica when compared to silicon carbide are that it has a lower density, which increases its desire to float and decreases the homogeneous distribution of the particles. Another factor is that the shape and size of the silica particles differ, and it appears in the image that it is similar to the case of mesoporous structure.

Figure 15. The optical microstructure of A: AA 6063 as cast B: 95%AA 6063 + 5% SiC C: AA 6063 + SiC +5% Silica sand.

9.2. SEM MICROSTRUCTURE ANALYSIS

As shown in Figure 16, there is no evidence of granule agglomeration in the images captured by SEM, 100 microns magnification when using the D5 stirrer, which proves the validity of the uniformity of the particle dispersion and the enhancement of the mechanical properties.

Figure 16. SEM of A: 100% AA 6063 B: 95% AA 6063 + 5% SiC C: 95% AA 6063 + 5% Silica sand.

10. Conclusions

The following was concluded:

1. Aluminum alloy 6063 ceramic composites were successfully prepared by using the stir casting route although it should be noted mixing silica particles uniformly have more challenges than silicon carbide.

2. Properties of AAMCs can be tailored by optimizing the Stirrers design in addition to the stirring time, stirring speed, type of constituents, and their volume fraction.

3. When the surface tension cannot be broken by the slower Stirring Speed, most of the particles tend to float as the particles cluster on the surface of the melt; there is limited incorporation with the alloy matrix.

4. The mechanical properties are further improved when silicon carbide is added compared to Silica sand.

5. The parameter behavior varies according to the type of mechanical properties. The effect of Stirring Speed on Hardness is more than the Stirring Time effect. Stirring Time has a greater beneficial impact on tensile strength in both types of ceramics than Stirring Speed.

6. The long Stirring Time at the wrong Stirrer Design leads to fewer mechanical properties due to its bad role in increasing the probability of more deposition of ceramic particles on the edges of the stirrer and at the bottom of the crucible.

7. As a result of comparing the use of two different types of ceramics (silicon carbide and silica), we concluded that the effect of the parameters differs evidently. The silica ceramic reinforcements have fewer densities than silicon carbide so silica particles have been affected by Stirring Speed more than Stirring Time.

8. The Stirrer Design of D4 (Helical Stirrer) with a cylindrical shaft was optimal compared to (D1, D2, and D3) to get a larger hardness and tensile strength at Stirring Speed (240 RPM), Stirring Times (6 Minutes).

9. Stirrer Design (D4) at a high Stirring Speed forms the forceful vortex that allows oxide, gases, and impurities to be entrained in the melt, resulting in increased porosity and low mechanical properties. To reduce the vortex around the shaft, it is preferable to change the shaft design from cylindrical to helical which is called (D5).

10. Stirrer Design (D5) yields better mechanical properties than (D4) because it allows using the Stirring Speed of more than 540 RPM without creating a gap in the center of the vortex for the entry of undesired materials into the casting.

11. Finally, Taguchi was unable to establish a clear trend about the impact of stirring time and speed on the mechanical properties of hardness and tensile strength along with the stirrer effect.

Disclosure statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data Availability statement

The data that support the findings of this study are available from the corresponding author, [author initials], upon reasonable request.

Additional information

Funding

The authors received no direct funding for this research.

Notes on contributors

Farooq Muhammad

Farooq Muhammad is a Ph.D. student in production engineering in the mechanical & mechatronics engineering department of the University of Salahaddin in Erbil, Iraq. He completed his first degree in production and metallurgy from Technology University Baghdad - Iraq, and his MSc in the performance evaluation of manufacturing enterprises from the engineering and application science department of Aston University. He had been an engineer and expert for 20 years in the ministry of industry. He has published articles about measuring manufacturing performance and lean manufacturing.

Shawnam Jalal

Shawnam Jalal is a staff member with a professor degree in the College of Engineering - Mechanical Engineering Department, Salahaddin University -Erbil in Iraq. She has a Ph.D. in material engineering from Salahaddin University and MSc in metallurgy from Technology University - Baghdad - Iraq. Her portfolio of teaching for more than three decades includes undergraduate and postgraduate-level courses in the areas of metallurgy, material science, corrosion, casting, welding, and engineering mechanics. She has more than 31 papers published in mechanical engineering fields in many journals both local and international Journals, supervised several undergraduates and post-graduate students. And participating in dozens of workshops and conferences around the world. She has been registered in Iraqi Engineering Union since 1981.