Numéro 
Matériaux & Techniques
Volume 108, Numéro 1, 2020



Numéro d'article  101  
Nombre de pages  7  
DOI  https://doi.org/10.1051/mattech/2020013  
Publié en ligne  7 avril 2020 
Regular Article
Optimization of manufacturing coppergraphite composite for electrical contact applications using grey relational analysis
Department of Metallurgical Engineering, University of Babylon,
Hilla
51001, Iraq
^{*} email: alaahus30@gmail.com
Received:
12
March
2019
Accepted:
5
March
2020
Coppergraphite composite is one of the most important copper based composites, which is used, widely in electrical applications due to its excellent conductivity and high wear resistant. In this work, an attempt has been made to improve the multiperformance characteristics of copper graphite composite prepared by powder metallurgy and increase its expected life by optimizing the manufacturing process of this composite using statistical method. The experiments were carried out under various conditions of compacting pressure, sintering temperature and graphite content based on L9 orthogonal array. The experimental results indicated that multiperformance characteristics of the prepared composite such as electrical conductivity, densification and wear rate are influenced significantly by the studied parameters and the optimal process parameters level for optimum multiperformance characteristics was obtained in a compaction pressure of 750 MPa, sintering temperature of 950 °C and 10 Vol.% of graphite content.
Key words: coppergraphite composite / grey relational analysis / wear rate / densification / electrical conductivity
© SCF, 2020
1 Introduction
Copper alloys and composite are versatile material and extensively used in the electrical applications due to its unique combination of properties as high electrical conductivity, high melting point, and high corrosion resistance [1,2]. The main problem in the application of pure copper is its low wear resistance, which limits its usability. The useful lifetime of copper as an electrical sliding contact material is shortened due to wear. This material wears out significantly in the contact applications by the friction effect that arises during sliding with the counter surface material, which reduces the efficiency of the electrical machine [3]. So developing a robust solution combining high wear resistance, low friction, and low electrical loss is considered an important demand for the electromechanical industries. In such an application, selflubricating can be considered an optimum solution in the reduction of wear and friction. In this case, graphite as a reinforcement element is an excellent solution for copper due to its high electrical conductivity and the ability of selflubricating [4]. The lamellar crystal structure of graphite is a perfect solid lubricant due to the low shearing force of the VanderWaals layers and the weak bonding between them [5]. As a result, several copper basedcomposite reinforced with graphite particles, which is mainly produced by the powder metallurgy route, are commercially used as electrical sliding contacts as industrial bearings and brushes [6].
Many attempts have been made by researchers to improve the wear behaviour, electrical conductivity, and the densification of coppergraphite composite prepared by powder metallurgy route [7–9]. There is still a demand to produce a highquality coppergraphite composite with a longer expected life and improved multiple performance characteristics. The complicated multiobjective optimization problems can be solved efficiently using grey relational analysis (GRA) as proved by previous works [10–12], by converting these problems to a single objective optimization problem. Therefore, this study aims at improving the properties of coppergraphite composite (wear, electrical conductivity and densification) by optimization the process parameters (compacting pressure, sintering temperature, graphite content) using GRA.
2 Experimental details
2.1 Materials and processing
Coppergraphite composite was fabricated by powder metallurgy (P/M) route, copper powder (impurities <0.05 wt%, measured d_{50} = 45 μm) and graphite powder (impurities <0.05 wt%, measured d_{50} = 5 μm) were supplied by Lemandou Ltd. co. China. The powders were mixed for 1 hour by a ball mill type (STGQM15⁄2) at 350 rev/min. Uniaxial compacting was carried out using double action steel die via electrohydraulic compacting machine type (CT340CT440, USA). Cylindrical specimens of 10 mm diameter and 10 mm height were prepared. Hightemperature tube furnace was used for sintering the green compacts with a vacuum pressure of 10^{−4} torr at different sintering temperature with a holding time of 1 hour.
2.2 Testing procedure
All the prepared specimens were tested for electrical resistivity, sinterability and wear. The electrical resistivity tests were carried out using the precision ohmmeter type: (AT512). The sinterability of coppergraphite specimens were evaluated by densification parameter (DP) calculated according to equation (1): (1) The density of sintered specimens was measured by the Archimedes principle, the theoretical density was calculated by the inverse rule of mixture. A positive value of DP refers to shrinkage whereas the negative value signifies a volumetric expansion of the samples after the sintering process.
Wear tests were carried out on the samples using wear tester device of type: (MT4003, version 10.0) with a ballonplate configuration. The specimens grounded with SiC papers and then cleaned and dried in a vacuum furnace to remove all traces. The test was performed at room temperature without any lubricant at an applied load of 15 N, 6 mm diameter Al_{2}O_{3} ball, 500 rpm rotational speed for a period of 20 minutes. The specimens were weighed before and after the wear test using the sensitive electric balance model (M254A) with ±0.0001 g accuracy. The wear rate of the composites was calculated based on volume loss according to equation (2): (2) where: weight loss = weight before the test−weight after the test.
2.3 Design of experiments
The essential performance measures during the manufacturing process of the coppergraphite composite are electrical conductivity, wear rate and densification. They highly depend on the manufacturing process parameters as compacting pressure, sintering temperature and graphite content. The orthogonal arrays of Taguchi method are a very special design that can be used to study the entire parameters with a few experiments [13]. In this study, Taguchi’s L9 orthogonal array was adopted to design the experiments. Table 1 demonstrates the factors and their levels and Figure 1 shows the interrelationship between these factors and the responses.
The orthogonal array and the results of electrical conductivity, densification parameter and wear rate are shown in Table 2. The electrical conductivity results were obtained in the range of (33–45) MS/m while densification parameter and wear rate were obtained between (0.21–0.42) and (0.17–0.43) 10^{−4} mm^{3}/m respectively.
The controlled factors and their levels.
Fig. 1 Interrelationship between the factors and responses. 
Process variables and the results of responses.
3 Results and discussion
GRA is a method proposed by Deng in 1989 and used to measure the degree of relationships for every parameter in the system [14,15]. Many complicated multiobjective optimization problems can be converted to the single objective problem by performing GRA associated with the Taguchi method. In GRA, the grey relational coefficient (GRC) for each conducted experiment is calculated and the average of these GRCs is called grey relational grade (GRD). In this work, Taguchi’s method combined with GRA has been used to optimize the manufacturing process of coppergraphite composite for multiperformance characteristics, namely electrical conductivity, densification and wear rate. GRA method can determine the optimal level of controlled multiple factors simultaneously by applying the steps illustrated in Figure 2.
Step 1: calculate the signal to noise (S/N) ratio to estimate the quality and evaluate the parameters. Larger S/N ratio value signifies a better performance of parameter. The S/N ratio can be defined by equations (3) and (4). Equation (3) represents the higherthebetter type of S/N ratio and can be used when a maximization of response is intended, while equation (4) represents the smallerthebetter type. During the manufacturing process of a copper graphite composite, the densification parameter and electrical conductivity had been considered as higherthebetter type, whereas the wear rate is the smallerthebetter type. These considerations were set according to higher multiperformance characteristics of interest. (3) (4) where, n = number of repetitions; y_{ij} = the response, i = 1, 2... n; n is the number of the experiment in the orthogonal array; j = 1, 2, ..., m. m is the number of the process response. In the present work wear rate, electrical conductivity and densification parameter are selected, then m = 3.
Step 2: the second step of GRA is to normalize S/N ratio, in the range between 0 and 1, using formulation (5) and formulation (6) to maximize and minimize the quality characteristics respectively. (5) (6) Higher normalized S/N ratio results indicate a better performance of the parameter and the best normalized result should be equal to 1. The results of S/N ratio and normalized S/N ratio is shown in Table 3.
Step 3: computation of deviation using the formula, Δ = 1−x_{ij}, where x_{ij} = normalized S/N ratio.
Step 4: After processing the results, GRC for the j^{th} response in the i^{th} experiment can be expressed as follows: (7) where, Δ is the deviation of the response; Δmin is the smallest value of Δ; Δmax is the largest value of Δ; γ is the distinguishing factor (0.5 is widely accepted).
Step 5: after calculating GRC, GRD can be calculated by the following formula: (8)
where m is the responses number. The GRC and GRG for all experiments have been calculated. The higher value of GRG is near to the product quality for optimum process parameters [16]. The highest GRG value is assigned an order of 1 and ranking is done in decreasing order. Table 4 shows deviation, GRC, GRG and order for all experiments. Figure 3 shows the GRG for all the conducted experiments.
It is observed from Table 4 and Figure 3 that experiment 8 has the best multiperformance characteristics due to its highest GRG. However, the relative importance for each level of the parameters needs to be determined to obtain an optimum solution with more accuracy [17]. Therefore, the average of the GRG for each parameter level was calculated as shown in Table 5. The procedure of calculation is to group the GRGs by factor level for each column in the orthogonal array and estimate their average. The following is an example of c calculating the average GRG for the factor A at level 1: (9) According to the results of Table 5, it’s observed that the optimal process parameter level combination during manufacturing copper graphite composite are A3 B3 C1. The values of the optimum manufacturing process parameters are compacting pressure 750 MPa, sintering temperature 950 °C and graphite content 10 Vol.%. The delta value, which is represents the difference between max and min values of the average GRG, can indicate the influential factors that affect the multiple performance characteristics. According to Table 5, the graphite content (0.2769) has the highest effect on multiple performance characteristics followed by compacting pressure (0.1535) and sintering temperature (0.1509) as it is ordered in the last column of the table. Confirmation tests have been conducted to validate the predicted results of GRA by applying the optimal process parameters experimentally. Table 6 shows the results of the confirmation test, which confirm the match between the experimental results and predicted results. The sample produced by optimal conditions was tested by xray diffraction analysis. As can be seen in Figure 4, it is clear that all peaks belong to copper and graphite elements and no other component were detected. The copper remains not alloyed, while the graphite is completely preserved in the bulk and this guarantees the high electrical conductivity and high wear resistance in the produced composite.
Analysis of variances (ANOVA) was performed to identify the significant factors affecting the GRD. The results of the ANOVA are presented in Table 7. It is observed that the graphite content is the most influential parameter affecting the multiple performance characteristics with contribution percentage of (60.97%) followed by compacting pressure (18.75%) and sintering temperature (18.62%).
The results indicated that a low graphite content is preferable during the manufacturing of coppergraphite composite. The selflubricant ability of the graphite particles can lead to the decrease of friction and subsequently a decrease in wear rate can be noticed, but the present of high graphite content can have negative effects on the densification parameter and electrical conductivity due to the poor wetting between the graphite and copper particles as it is indicated by previous work [18]. The optimal graphite content of (10 Vol.%) was also recommended by previous work due to its most suited properties for electrical contact applications [19].
The results observed that high compact pressure and sintering temperature are preferable on the multiperformance characteristics. The high compact pressure results in large contact area between the particles, subsequently, lower spaces between the particles and this increases the green density. The high sintering temperature can also increase the density of the final product due to the high diffusion rate between the contact particles. So the increase of compact pressure and sintering temperature can provide better densification with low porosity level during the manufacturing process. The low porosity level is important in improving the wear behaviour because pores can act as a crack propagation points due to stress concentration during wear test [20]. In addition, the higher density increases the electrical conductivity of the final product.
Fig. 2 Steps for determining the optimal level of controlled multiple factors. 
S/N ratio and normalized S/N ratio of the performance characteristics.
Deviation sequences, GRC, GRG and order for each experiment.
Fig. 3 GRG for each experiment. 
Mean GRG for the input parameters.
Confirmation tests.
Fig. 4 XRD of sintered coppergraphite composite. 
ANOVA table.
4 Conclusion
In this work, coppergraphite composite was fabricated via powder metallurgy route with different values of parameters (compact pressure, sintering temperature and graphite content) based on L9 orthogonal array. Opimization of the manufacturing process to improve the multiple performance characteristics (viz. electrical conductivity, densification parameter and wear rate) were performed using GRA. According to the results of this paper, the following can be concluded:

the optimal process parameters level for optimum multiperformance characteristics was obtained as A3 B3 C1: compaction pressure 750 MPa, sintering temperature 950 °C and graphite content 10 Vol.%;

based on experimental results and analysis, the compaction pressure, sintering temperature and graphite content are significant factors that affect the quality of manufacturing coppergraphite composite;

graphite content was found to have the strongest effect on the multiperformance characteristics among the input process parameters followed by compaction pressure and sintering temperature;

GRA is a very useful technique to optimize the manufacturing process of metal matrix composites. It does not involve any complicated mathematical theory, computation or simulation. Therefore, GRA concept can be utilized without any statistical background in the industry.
References
 Y. Zhan, G. Zhang, Graphite and SiC hybrid particles reinforced copper composite and its tribological characteristic, J. Mater. Sci. Lett. 22, 1087 (2003) [CrossRef] [Google Scholar]
 K. Rajkumar, S. Aravindan, Tribological behavior of microwave processed copper–nanographite composites, Tribol. Int. 57, 282 (2013) [CrossRef] [Google Scholar]
 K. Rajkumar, S. Aravindan, M.S. Kulkarni, Wear and life characteristics of microwavesintered coppergraphite composite, J. Mater. Eng. Perform. 21, 2389 (2012) [CrossRef] [Google Scholar]
 I. Rodrigues, M. Guedes, A.C. Ferro, Microstructural changes in copper–graphite–alumina nanocomposites produced by mechanical alloying, Microsc. Microanal. 21, 120 (2015) [CrossRef] [Google Scholar]
 H. Cao, Z. Qian, L. Zhang, J. Xiao, K. Zhou, Tribological behavior of Cu matrix composites containing graphite and tungsten disulfide, Tribol. Trans. 57, 1037 (2014) [CrossRef] [Google Scholar]
 Z. Tao, Q. Guoa, X. Gaoa, L. Liua, The wettability and interface thermal resistance of copper/graphite system with an addition of chromium, Mater. Chem. Phys. 128, 228 (2011) [CrossRef] [Google Scholar]
 M.D. Montasser, J. Kyoo Lim, Manufacturing processes and properties of copper–graphite composites produced by high frequency induction heating sintering, J. Compos. Mater. 41, 2183 (2007) [CrossRef] [Google Scholar]
 J. Kovacik, S. Emmer, J. Bielek, L. Kelesi, Effect of composition on friction coefficient of Cu–graphite composites, Wear 265, 417 (2008) [CrossRef] [Google Scholar]
 A. Mazloum, J. Kovacik, S. Emmer, I. Sevostianov, Copper–graphite composites: thermal expansion, thermal and electrical conductivities, and crossproperty connections, J. Mater. Sci., 51(17), 7977 (2016) [CrossRef] [Google Scholar]
 M. Hussain, S. Khan, P. Sarmah, Optimization of powder metallurgy processing parameters of Al_{2}O_{3}/Cu composite through Taguchi method with Grey relational analysis, J. King Saud Univ. Eng. Sci. (2019) [Google Scholar]
 H. Jailani, A. Rajadurai, B. Mohan, A. Kumar, T. Sornakumar, Multiresponse optimisation of sintering parameters of Al–Si alloy/fly ash composite using Taguchi method and grey relational analysis, Int. J. Adv. Manuf. Technol. 45(3–4), 362 (2009) [CrossRef] [Google Scholar]
 A. Sadeghi, A. Babakhani, S. Zebarjad, H. Mostajabodaveh, Use of grey relational analysis for multiobjective optimisation of NiTiCu shape memory alloy produced by powder metallurgy process, J. Intell. Mater. Syst. Struct. 25(16), 2093 (2014) [CrossRef] [Google Scholar]
 F. Puh, T. Segota, Z. Jurkovic, Optimisation of hard turning process parameters with PCBN tool based on the Taguchi method, Tech. Gaz. 19, 415 (2012) [Google Scholar]
 N. Senthilkumar, J. Sudha, V. Muthukumar, A greyfuzzy approach for optimising machining parameters and the approach angle in turning AISI 1045 steel, Adv. Prod. Eng. Manag. 10, 195 (2015) [Google Scholar]
 J. Deng, Introduction to grey system theory, J. Grey Syst. 1, 1 (1989) [Google Scholar]
 C.C. Tsao, Grey–Taguchi method to optimise the milling parameters of aluminum alloy, Int. J. Adv. Manuf. Technol. 40, 41 (2009) [CrossRef] [Google Scholar]
 Y.F. Hsiao, Y.S. Tarng, W.J. Huang, Optimisation of plasma arc welding parameters by using the Taguchi method with the grey relational analysis, Mater. Manuf. Process. 23, 51 (2010) [CrossRef] [Google Scholar]
 Y. Zhan, G. Zhang, Y. Wu, Effect of surface metallization of graphite on the tribological properties of copper hybrid composites, Scand. J. Metall. 33, 80 (2004) [CrossRef] [Google Scholar]
 K. Rajkumar, S. Aravindan, Microwave sintering of copper graphite composites, J. Mater. Process. Technol. 209, 5601 (2009) [CrossRef] [Google Scholar]
 Glenn Kwabena Gyimah, Ping Huang, Dong Chen, Dry sliding studies of porosity on sintered Cubased brake materials, Trans. Control Mech. Syst. 2, 219 (2013) [Google Scholar]
Cite this article as: Alaa H. Jaafar, Haydar AlEthari, Optimization of manufacturing coppergraphite composite for electrical contact applications using grey relational analysis, Matériaux & Techniques 108, 101 (2020)
All Tables
All Figures
Fig. 1 Interrelationship between the factors and responses. 

In the text 
Fig. 2 Steps for determining the optimal level of controlled multiple factors. 

In the text 
Fig. 3 GRG for each experiment. 

In the text 
Fig. 4 XRD of sintered coppergraphite composite. 

In the text 
Les statistiques affichées correspondent au cumul d'une part des vues des résumés de l'article et d'autre part des vues et téléchargements de l'article pleintexte (PDF, FullHTML, ePub... selon les formats disponibles) sur la platefome Vision4Press.
Les statistiques sont disponibles avec un délai de 48 à 96 heures et sont mises à jour quotidiennement en semaine.
Le chargement des statistiques peut être long.