Nanoscience and NanotechnologyVol. 14, 8750–8755, 2014
www.aspbs.com/jnn
Preparation and Characterization of Copper-GraphiteComposites by Electrical Explosion of Wire in Liquid
T. N. Bien1, W. H. Gu1, L. H. Bac1�2, and J. C. Kim1�∗1School of Materials Science Engineering, University of Ulsan Daehak-Ro 93, Nam-Gu,
Ulsan, 680-749, South Korea2School of Engineering Physics, Hanoi University of Science and Technology, Hanoi, Vietnam
Copper-graphite nanocomposites containing 5 vol.% graphite were prepared by a powder metallurgyroute using an electrical wire explosion (EEW) in liquid method and spark plasma sintering (SPS)process. Graphite rods with a 0.3 mm diameter and copper wire with a 0.2 mm diameter were usedas raw materials for EEWin liquid. To compare, a pure copper and copper-graphite mixture wasalso prepared. The fabricated graphite was in the form of a nanosheet, onto which copper particleswere coated. Sintering was performed at 900 �C at a heating rate of 30 �C/min for 10 min andunder a pressure of 70 MPa. The density of the sintered composite samples was measured by theArchimedes method. A wear test was performed by a ball-on-disc tribometer under dry conditionsat room temperature in air. The presence of graphite effectively reduced the wear of composites.The copper-graphite nanocomposites prepared by EEW had lower wear rates than pure coppermaterial and simple mixed copper-graphite.
Keywords: Copper-Graphite Nanocomposites, Electrical Explosion of Wire in Liquid, SparkPlasma Sintering, Wear Resistant.
1. INTRODUCTIONMetal-graphite composites are attractive materials formany applications such as engine brushes and genera-tors or sliding contacts,1 self-lubrication parts for automo-tive pistons,2 and heat sink elements in multi-functionalelectronic packaging systems.3 Copper-graphite compos-ites are widely used in tribological engineering parts.4
Copper-graphite composites combine the positive charac-teristics of both components, i.e., high thermal and elec-trical conductivity of the copper with the low thermalexpansion coefficient and good lubricating properties ofthe graphite.5
Copper-graphite composites are typically prepared by apowder metallurgy (PM) process as the PM process offersthe possibility of obtaining uniform parts and reducingproduction costs. However, PM has certain limitations pri-marily related to the poor affinity between copper andgraphite, which gives rise to weak interfaces with a neg-ative effect on the structural, mechanical, and electricalproperties of the material.5�6 The lack of wetting between
∗Author to whom correspondence should be addressed.
copper and graphite during composite processing can beovercome by coating the graphite particles with copperbefore consolidation. Moustafa et al.7�8 fabricated copper-graphite composites by a PM route using Cu-coatedgraphite powders and a mixture of copper and graphitepowders. The copper-coated graphite powders possessedlower wear rates and friction coefficients than those madefrom pure copper and non-copper-coated graphite. More-over, the copper-coated composites show a higher densityand yield strength.In this work, the electrical explosion of wire (EEW)
in liquid was introduced to fabricate copper-graphitenanocomposites. The EEW process has attracted attentionfor fabrication of various nanosized powders due to thesimple and low-cost production.9–11 Homogenous copper-coated graphite nanocomposite powders were first pro-duced by EEW, and then the sintered composites wereproduced by SPS.
2. EXPERIMENTAL DETAILSThe EEW in liquid was used to prepare graphitenanosheets and copper powder. The experimental setup
Delivered by Publishing Technology to: INST. FOR APPLIED SCIENCE & ENGINEERING RESEARCH/ACADEMIA SINICAIP: 140.109.57.11 On: Sat, 08 Nov 2014 10:13:56
Copyright: American Scientific Publishers
Bien et al. Preparation and Characterization of Copper-Graphite Composites by Electrical Explosion of Wire in Liquid
Table I. Conditions of the wire electrical explosion process.
Capacitance 30 �FCharging voltage 3 kVCu wire diameter 0.2 mmGraphite rod diameter 0.3 mmWire length of one explosion 25 mmAmbient liquid Ethanol
of the EEW process was described previously.12 Typi-cal experimental conditions are summarized in Table I.Graphite rods with a diameter of 0.3 mm were used toprepare the graphite nanosheets. Before explosion, thegraphite rods were treated with a mixture of concentratedsulfuric acid and nitric acid (4:1, v/v). The acid-treatedgraphite rods were washed several times with deionizedwater and then dried at 60 �C to remove any remainingwater after they were immersed in a mixed acid solutionfor 120 hrs. The dried graphite rods were exploded inethanol to produce the graphite suspension.The graphite suspension was used as a working liquid
for the preparation of copper powder for EEW. Copperwire (Nicola, 99.9%) with a diameter of 0.2 mm was usedfor the explosion. Copper explosions were performed at anumber sufficient to result in 95 vol.% copper. The cop-per powder-coated graphite nanocomposite powders werecollected from the suspension by drying at 80 �C undervacuum. The reduction of the composite powders was per-formed at 700 �C for 2 hrs in order to remove the oxidephase.The bulk samples were prepared by a spark plasma
sintering (SPS) machine (Sumitomo Coal Mining Co.,Model 515S). The composite powder (4 g) was placed in agraphite die with an inner diameter of 15 mm. A pressureof 70 MPa was applied, and the sample was heated from
Figure 1. Schematic illustration of the experimental procedure.
Table II. Parameters of ball-on-disc tribometer testing.
room temperature to a sintering temperature at a heatingrate of 30 �C/min and then held for 10 min. The electricpower was turned off after the holding time, and the sam-ple was cooled without control of the cooling rate. Thedensity of the sintered composite samples was measuredby the Archimedes method. A schematic illustration of theexperimental procedure is shown in Figure 1.The wear properties of the sintered composites were
measured using a ball-on-disc tribometer. The tested sam-ples were in the shape of a disc shape 15-mm in diam-eter and 2-mm thick. The ball was tungsten carbide witha 7-mm diameter. The wear tests were performed at var-ious loads, distances, and speeds, as shown in Table II.Each sample was weighed before and after each wear test.For comparison, a pure copper and copper-graphite mix-ture was also prepared. The copper-graphite mixture wasprepared by a ball milling method. Milling was performedat room temperature for 30 min at 700 rpm using 5-mmdiameter chrome steel balls, and the ball-to-powder wt.ratio was maintained at approximately 20:1. The sinteringof pure copper and copper-graphite mixture was the sameas that of the sample prepared by EEW.
3. RESULTS AND DISCUSSIONFigure 2 shows the morphology of graphite prepared byEEW in ethanol. The graphite after explosion was brokeninto a sheet with a thickness of ∼ 20 nm. The graphitesheets are of relatively uniform size with a range from 1to 3 �m.After explosion of copper in the graphite suspension,
the copper-graphite suspension was dried and observed bySEM. As shown in Figure 3, the copper particles had aspherical shape. Copper particles size can be classified
Figure 2. FESEM image of graphite nanosheets prepared by EEW.
J. Nanosci. Nanotechnol. 14, 8750–8755, 2014 8751
Delivered by Publishing Technology to: INST. FOR APPLIED SCIENCE & ENGINEERING RESEARCH/ACADEMIA SINICAIP: 140.109.57.11 On: Sat, 08 Nov 2014 10:13:56
Copyright: American Scientific Publishers
Preparation and Characterization of Copper-Graphite Composites by Electrical Explosion of Wire in Liquid Bien et al.
Figure 3. FESEM micrographs of copper-graphite composite at (a) low and (b) high magnification.
into two groups: nanoparticles with a diameter smallerthan 100 nm and fine particles with a diameter larger than100 nm. The prepared graphite was in the form of a sheet,as described above. The copper particles were coated onthe surface of the graphite sheets. The coating of coppernanoparticles on graphite can overcome the limitations ofpoor affinity between copper and graphite and producedbetter properties of graphite-copper composites, which areshown below.Figure 4 shows the EDS results of copper-graphite
nanocomposite powders before and after reduction inhydrogen at 700 �C for 2 hrs. The nanocomposite powdersprepared by EEW in ethanol contained oxygen. The oxy-gen in the sample may be due to the oxidation of copperduring explosion by oxygen dilution in the liquid. Cop-per was in a vapor state when exploding and could havereacted with the oxygen present in ethanol. Oxidation canalso occur during handling. The nanopowders are easilyoxidized when they contact oxygen in air. The compositepowder with reduced hydrogen was pure, containing nooxygen.Figure 5 presents XRD patterns of the copper-graphite
composite before and after reduction. The compositepowder before reduction showed three main characteris-tic diffraction peaks for copper at 2� = 43�3, 50.4 and74.0 degrees, corresponding to (111), (200) and (220)crystal planes, respectively. This pattern confirms the for-mation of the FCC copper phase. The small peak at2� = 26�5� was attributed to graphite. A very low peakat 2� = 36�4 belonged to cuprous oxide (Cu2O). After
Figure 4. EDS profiles of a copper-graphite composite (a) before and(b) after reduction.
reduction of the composite powder, the peak disappearedbecause the copper oxide phase was reduced to purecopper.Figures 6(a)–(c) show optical microscopic images of
bulk samples sintered at different temperatures for a10 min holding time. The sintering temperature waschanged from 800 to 900 �C. The microstructure of thesintered sample depended on the sintering temperature.The densification increased with increase in the sinter-ing temperature. The white area on the images illustratesa good sintering area with few pores and good connec-tivity between the particles. The sample from sinteringat 800 �C had the largest black area (pores). Therefore,the porosity was dominant at low sintering temperature.When the sintering temperature increased, the porositygradually decreased. The density of these samples verifiedthe sintering results. The density of the sample sinteredat 800 �C was only 6.9 g/cm3 and the relative densitywas approximately 80%. The value reached 7.5 g/cm3 forthe sample sintered at 870 �C with a relative density ofapproximately 87%. The greatest density was 7.9 g/cm3,for the sample sintered at 900 �C, and the relative densitywas approximately 91%. The microstructure of the copper-graphite composite sample fabricated by milling containedmany small black areas, and the black areas dominated
Figure 5. XRD patterns of copper-graphite composite powders(a) before and (b) after reduction.
8752 J. Nanosci. Nanotechnol. 14, 8750–8755, 2014
Delivered by Publishing Technology to: INST. FOR APPLIED SCIENCE & ENGINEERING RESEARCH/ACADEMIA SINICAIP: 140.109.57.11 On: Sat, 08 Nov 2014 10:13:56
Copyright: American Scientific Publishers
Bien et al. Preparation and Characterization of Copper-Graphite Composites by Electrical Explosion of Wire in Liquid
Figure 6. Microstructures of copper-graphite composites prepared byEEW after sintering by SPS at (a) 800 �C, (b) 870 �C, (c) 900 �C,(d) composites prepared by milling and sintering at 900 �C and (e) purecopper.
the total surface area (Fig. 6(d)). In comparison with sam-ples fabricated by EEW, the microstructure exhibited weakcontact between the copper, graphite, and copper matrix,which resulted in a smaller density (7.6 g/cm3) than thatof the copper-graphite prepared by EEW (7.9 g/cm3). Themicrostructure of the copper samples was dense with nopores, as observed in the optical image (Fig. 6(e)). Thedensity of these samples was very high, with a value of8.8 g/cm3.The effects of load, speed, and distance on the specific
wear rate are shown in Figures 7–9. These variables were
Figure 7. The effect of distance on weight loss with a load of 10 Nand speed of 50 rpm.
Figure 8. The effect of speed on weight loss with a distance of 32 mand load of 15 N.
manipulated and the weight loss was measured after eachvariation. The specific wear rate was obtained by divid-ing the weight loss by the density, the total sliding dis-tance, and the normal load of the test. Figure 7 illustratesthat the specific wear rate depended on the distance of theball travel over the surface. When ball slid on the surface,wear occurred and the weight of the samples decreased.For some initial slides of the ball on the surface, the wearsurface was relatively rough. Wear occurred easily, andthe wear rate increased quickly, showing a steep slope ofthe curve. The surface was gradually glazed; therefore, thewear rate declined. The pure copper had a much higherwear rate than copper-graphite composites. The low spe-cific wear rate of the copper-graphite composite was dueto a smeared graphite layer at the sliding surface, whichacted as a solid lubricant. Alternatively, in the case of thecopper-graphite composites, the wear rate of the copper-graphite composite prepared by EEW was lower than thatof the mixed composite. The low wear of this sample wasmost likely due to its hardness. The hardness of the EEWcomposite was 97 HV, whereas that obtained by mixing
Figure 9. The effect of load on weight loss with a distance of 32 mand speed of 50 rpm.
J. Nanosci. Nanotechnol. 14, 8750–8755, 2014 8753
Delivered by Publishing Technology to: INST. FOR APPLIED SCIENCE & ENGINEERING RESEARCH/ACADEMIA SINICAIP: 140.109.57.11 On: Sat, 08 Nov 2014 10:13:56
Copyright: American Scientific Publishers
Preparation and Characterization of Copper-Graphite Composites by Electrical Explosion of Wire in Liquid Bien et al.
Figure 10. Photographs of the wear tracks and corresponding worn surfaces of (a), (d) copper and copper-graphite composites prepared by (b), (e)milling and (c), (f) EEW.
was 85 HV. The relationship between wear and hardnesshas been widely investigated,7�13�14 showing that samplewear decreases with increasing hardness. A harder sampleis able to better resist wear than a softer sample. This studyshowed that the hardness of the copper-graphite obtainedby EEW was greater than that of copper-graphite obtainedby milling. Therefore, the lower wear rate of samples fab-ricated by EEW may be affected by their hardness.Figure 8 illustrates that the wear rate did not depend on
the speed as none of the curves demonstrate a change inspeed. The EEW sample showed the optimal wear, similarto the wear distance case in Figure 7. Figure 9 presentedthe effects of load on the wear properties. The load dra-matically affected the specific wear rate. The specific wearrate in all samples was greatest at 10 N, decreased at 15 N,and then increased at a load of 20 N. The copper-coatedgraphite composites prepared by EEW had the smallestspecific wear rate for all ranges of the applied normalload.Figure 10 showed the photographs of the worn sur-
face of the copper, mixed copper-graphite, and copper-graphite prepared by EEW. The wear track illustrates thatthe copper sample had the largest track. The mixed copper-graphite had a narrower track than the pure copper sam-ple but a larger track than the copper-graphite preparedby EEW. Furthermore, the surface damage of the copper-graphite prepared by EEW was much less than that ofpure copper and mixed copper-graphite. The worn sur-face of the copper-graphite sample obtained by EEW was
much smoother than that of pure copper and copper-graphite mixed samples. Major deformation appeared onthe pure copper milled composite. The copper surfacewithout graphite experienced high friction; therefore, thesliding surface was highly deformed. However, the copper-graphite milling sample also exhibited high deformationas it is weaker bonding between graphite and the cop-per matrix. The weak bonding leads to rapid removal ofgraphite and copper from the surface, which could notwithstand the friction force and high deformation formedon the sliding surface.
4. CONCLUSIONSCopper-graphite nanocomposites were prepared by EEWin liquid. Graphite was in the form of a nanosheet, and asmall amount of copper oxide was detected in the com-posite powder. Pure copper-graphite nanocomposites wereobtained after reduction of the composite in hydrogen at700 �C for 2 hrs. The composite powder sintered by SPS at900 �C for 10 min had good sintering ability and high den-sity. The copper-graphite nanocomposites exhibited a sig-nificant decrease of wear. Additionally, the specific wearrate of copper-graphite prepared by EEW was less thanthat of pure copper and mixed copper-graphite compos-ites. This work demonstrated a novel method to fabricatecopper-graphite composites.
Acknowledgment: This work was supported bythe 2013 Research Fund of University of Ulsan.
8754 J. Nanosci. Nanotechnol. 14, 8750–8755, 2014
Delivered by Publishing Technology to: INST. FOR APPLIED SCIENCE & ENGINEERING RESEARCH/ACADEMIA SINICAIP: 140.109.57.11 On: Sat, 08 Nov 2014 10:13:56
Copyright: American Scientific Publishers
Bien et al. Preparation and Characterization of Copper-Graphite Composites by Electrical Explosion of Wire in Liquid
References and Notes1. J. W. Kaczmar, K. Pietrzak, and W. Włosinski, J. Mater. Process
Tech. 106, 58 (2000).2. A. Rodríguez-Guerrero, S. A. Sánchez, J. Narciso, E. Louis, and
F. Rodríguez-Reinoso, Acta Mater. 54, 1821 (2006).3. T. Oku, A. Kurumada, T. Sogabe, T. Oku, T. Hiraoka, and K. Kuroda,
J. Nucl. Mater. 257, 59 (1998).4. P. K. Rohatgi, S. Ray, and Y. Liu, Int. Mater. Rev. 37, 129 (1992).5. J. M. Casstevens, H. G. Rylander, and Z. Eliezer, Wear 48, 121
(1978).6. K. C. Owen, M. J. Wang, C. Persad, and Z. Eliezer, Wear 120, 117
(1987).7. S. F. Moustafa, S. A. El-Badry, A. M. Sanad, and B. Kieback, Wear
253, 699 (2002).
8. S. F. Moustafa, S. A. El-Badry, and A. M. Sanad, Powder Metall.40, 201 (1997).
9. Y. A. Kotov and O. M. Samatov, Nanostruct. Mater. 12, 119(1999).
10. G. S. Yun, L. H. Bac, J. S. Kim, Y. S. Kwon, H. S. Choi, and J. C.Kim, J. Alloys and Compounds 509S, S348 (2011).
11. L. H. Bac, Y. S. Kwon, J. S. Kim, Y. I. Lee, D. W. Lee, and J. C.Kim, Mater. Res. Bull. 45, 352 (2010).
12. L. H. Bac, J. S. Kim, and J. C. Kim, Res. Chem. Intermediat. 36, 795(2010).
13. M. Ghorbani, M. Mazaheri, and A. Afshar, Surf. Coat. Tech. 190, 32(2005).
14. I. Ahmad, A. Kennedy, and Y. Q. Zhu, Wear 269, 71 (2010).
Received: 17 April 2013. Accepted: 10 December 2013.
