Nanotechnology 17, (2006) 5759–5764 The effects of interfacial bonding on mechanical properties of single-walled carbon nanotube reinforced copper matrix nanocomposites Byengsoo Lim, Chul-ju Kim, Bumjoon Kim, Untae Shim, Seyoung Oh, Byung-ho Sung, Jee-hoon Choi and Seunghyun Baik Abstract The effects of interfacial bonding on mechanical properties of single- walled carbon nanotube reinforced copper matrix nanocomposites were investigated. The nanocomposites were fabricated by means of a powder metallurgy process, which consists of mixing carbon nanotubes with matrix powder followed by hot-pressing. The mixing process was carried out by ultrasonicating the nanotubes and copper powder in ethanol. The interfacial strength between the nanotubes and the copper matrix was improved by coating the nanotubes with nickel. The displacement rate of the nanotube reinforced nanocomposites was found to increase at 200 ◦C, whereas that of the nickel-coated nanotube reinforced nanocomposites significantly decreased. The incorporation of carbon nanotubes and nickel-coated carbon nanotubes in the copper matrix composites improved tribological properties compared with those of pure copper specimens. (Some figures in this article are in colour only in the electronic version) Introduction Nanosized reinforcements have received considerable attention because of the enhanced mechanical property that can be imparted to the metal matrix as compared with conventional micron-sized reinforcements The incorporation of carbon nanotubes in the development of advanced engineering composites with improved mechanical and tribological properties has become an interesting research area A weak mechanical bonding state was observed for the interface between the carbon fibre and the copper matrixdue to the poor wettability between carbon and copper Experiments (copper powder mixture preparation) SWNT were dispersed in ethanol by an intensive sonication for 5 min at 540 W. Copper powder was added to the solution and mixed with nanotubes suspended in (a) (b) ethanol by ultrasonicating at Figure 1. SEM images of copper powder. (a) A copper particle before the mixing process (b) nanotubes a power level of 5 W. attached to the surface of a copper particle after the The solution was heated to a mixing process. temperature of 50 ◦C and ultrasonicated until most of the ethanol was evaporated. Experiments (Nickel electroless plating) Preactivation of carbon nanotubes was accomplished by ultrasonically dispersing carbon nanotubes in a solution of 0.1 M SnCl2/0.1 M HCl for 30 min. The Sn+2 sensitized carbon nanotubes were further activated in an aqueous solution of 0.0014 M PdCl2/0.25 M HCl for another 30 min. The activated nanotubes were then Figure 2. Nickel-coated carbon nanotubes. introduced into an electroless plating bath. The composition of the plating solution includes NiCl2·6H2O, NiSO4·6H2O, Na2HC6H5O7·1.5H2O, NaH2PO2·2H2O, NH4Cl and Pb(NO3)2. After the coating process, nickel coated nanotubes were mixed with copper powder by the mechanical mixing process. Figure 3. EDX analysis of nickel-coated nanotubes. Experiments (Hot-pressing of the nanotube-copper powder mixtures) Hot-pressing was performed in a graphite mould by uniaxial pressurization in a vacuum of 10−3 Torr to prevent oxidization of the powder mixtures. The diameter of the specimens was Figure 4. A schematic diagram of the 15 mm and the thickness was 0.8 hot-pressing apparatus. mm. The temperature was ramped to 600 ◦C in 40 min and kept constant for 30 min. Finally, the temperature was cooled down to room temperature in 20 min. Figure 5. Temperature and pressure conditions for the hot- pressing process of specimens with a diameter of 15 mm. Experiments (High temperature displacement rate tests and dry friction tests) The hot-pressed specimen was ground into the shape of a thin square (10 × 10 × 0.5 mm3), and a constant load of 245 N was applied to the specimen by a ceramic (Si3N4) ball with a diameter of 2.4 mm. Displacement of the specimen was measured by LVDT with an accuracy of 10−3 mm. The sliding tests were carried out without lubricant for 400 s at room temperature. All specimens were Figure 6. A schematic diagram of the small punch creep tester ultrasonically cleaned with acetone before the tests. A sliding speed of 0.02 m s−1 (60 rpm) was used and a normal contact load of 5.2 N was applied in all tests. Results and discussion A pure copper specimen, hot-pressed using copper powder of about 10–15 μm, was used as a reference material. For the nanotube reinforced copper matrix composite (0.5 vol%), the displacement rate rapidly increased and the rupture time was shorter than that of the pure copper specimen. This might be due to the poor interfacial strength between carbon and copper. The nickel-coated nanotube reinforced copper matrix composite showed the longest rupture time which is six times longer than that of the pure copper Figure 7. The effects of interfacial strength specimen. In addition, it showed the on displacement rates. slowest displacement rate. Such remarkable strengthening by nickel- coated nanotube reinforcements was due to the high load transfer efficiency of nickel-coated nanotubes in the copper matrix. Results and discussion (a) shows the ruptured section of the pure copper specimen. (b) shows the ruptured section of the nanotube reinforced nanocomposites (0.5 vol%). Nanotubes came into contact with each other rather than with the copper matrix. These aggregates prevented effective bonding between the copper particles and the nanotubes. The aggregates were detached from the copper matrix, and this type of interface was not strong enough to transfer the stress of the matrix to the reinforcements. (c) shows the ruptured section of the nickel-coated nanotube reinforced nanocomposites (0.5 vol%). Nickelcoated nanotubes were pulled out during the fracture process probably due to the increased interfacial bonding. Good interfacial strength was obtained due to dissolution of copper– nickel solid solution at the interface between nickel-coated nanotubes and copper particles. (b) (c) (a) Figure 8. SEM images of ruptured sections of specimens (a) pure copper specimen (b) nanotube reinforced nanocomposites (0.5 vol%) (c) nickel-coated nanotube reinforced nanocomposites (0.5 vol%). Results and discussion The friction coefficients of the pure copper specimen started from about 0.2 and increased continuously with time. The incorporation of carbon nanotubes and nickel-coated carbon nanotubes in the copper matrix decreased the friction coefficients and increased intervals up to the onset of scuffing, compared with those of the pure copper specimen. The scuffing failure time of the nickel- coated nanotube reinforced nanocomposites was shorter than that of Figure 9. The effects of carbon nanotubes and the nanotube reinforced nanocomposites. nickel-coated carbon nanotubes on the This could be due to the increased tribological properties of nanocomposites. adhesion wear between nickel and the steel ball. The excellent adhesion between nickel and steel is a well known phenomenon. Conclusion We have investigated the effects of interfacial bonding on mechanical properties of single-walled carbon nanotube reinforced copper matrix nanocomposites. Carbon nanotube – copper powder mixtures were prepared by a mechanical mixing process, and nanocomposites were fabricated by hotpressing. The interfacial bonding between the nanotubes and the copper matrix was significantly improved by coating the nanotubes with nickel. The displacement characteristics of nanocomposites at a high temperature were investigated to evaluate the interfacial strength between the copper powder and the carbon nanotubes. The displacement rate of the nanotube reinforced nanocomposites was found to increase, whereas that of the nickel-coated nanotube reinforced nanocomposites significantly decreased. The incorporation of carbon nanotubes and nickel-coated carbon nanotubes in the copper matrix composites improved tribological properties compared with those of pure copper specimens.