Surface & Coatings Technology 200 (2006) 4870 4875 www.elsevier.

com/locate/surfcoat

Electrodeposition and characterization of NiCocarbon nanotubes composite coatings

L. Shia,b, C.F. Suna,b, P. Gaoa, F. Zhoua, W.M. Liua,*
a

State Key Laboratory of Solid Lubrication, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, PR China b Graduate School of Chinese Academy of Science, Beijing 100039, PR China Received 17 December 2004; accepted in revised form 27 April 2005 Available online 20 June 2005

Abstract Ni Co carbon nanotubes composite coatings were prepared by electrodeposition in a Ni Co plating bath containing carbon nanotubes (CNTs) to be co-deposited. The polarization behavior of the composite plating bath was examined on a PAR-273A potentiostat/galvanostat device. The friction and wear behaviors of the Ni Co CNTs composite coatings were evaluated on a UMT-2MT test rig in a ball-on-disk contact mode. The morphologies of the original and worn surfaces of the composite coatings were observed on an atom force microscope (AFM) and scanning electron microscope. It was found that the introduction of the carbon nanotubes in the electrolyte caused the shift towards larger negatives of the reduction potential of the Ni Co alloy coating, and the co-deposited CNTs had no significant effect on the electrodeposition process of the Ni Co alloy coating. However, the co-deposited CNTs led to changes in the composition and structure of the composite coatings as well. Namely, the peak width of the Ni Co solid solution for the composite coating is broader than that of the Ni Co alloy coating and the composite coating possess higher microhardness and elastic modulus than Ni Co alloy coating. The co-deposited CNTs were uniformly distributed in the Ni Co matrix and contributed to greatly increase the microhardness and tribological properties of the Ni Co alloy coating. D 2005 Elsevier B.V. All rights reserved.
Keywords: Electrodeposition; Ni Co alloy coating; CNTs; Composite coating; Friction and wear behavior

1. Introduction The thickness of the magnetic layer used in microelectrical mechanical system (MEMS) can vary from a few nanometers to a few millimeters, depending on the applications, and the magnetic thin films must have good adhesion and corrosion resistance and low-stress, and be thermally stable with excellent magnetic properties. As a kind of typical magnetic layers for MEMS, electrodeposited Ni Co alloy coatings have been widely used as recording head materials in computer hard drive industries. Since metal-matrix composite coatings usually have significantly improved mechanical strength, wear resistance and corrosion resistance, and desired chemical and biological compatibilities than the alloy coatings [1 5], it is essential
* Corresponding author. Tel.: +86 931 4968166; fax: +86 931 8277088. E-mail address: wmliu@ns.lzb.ac.cn (W.M. Liu). 0257-8972/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2005.04.037

and feasible to improve the comprehensive properties of Ni Co alloy coatings by introducing third phase reinforcements to generate Ni Co based composite coatings. This is especially so when carbon nanotubes (CNTs) are selected as the third phase, because carbon nanotubes possess many remarkable properties such as high strength and elastic modulus, good flexibility, and unique conductivity. For example, the Youngs modulus of a single-wall nanotube was theoretically estimated to be as high as 5 TPa, and the averaged Youngs modulus and bending strength of an isolated multiwalled CNT were measured to be 1.8 TPa and 14.2 GPa. This is why CNTs have been finding tremendous application in the microdevices such as nano diodes, nano transistors, and needle-like tips of atomic imaging machines. The greatly increased wear resistance of hot-pressed Al2O3 composite and electroless Ni P coating by the incorporation of CNTs provides the recent evidences to the successful use of CNTs as a kind of reinforcing agent [6,7].

L. Shi et al. / Surface & Coatings Technology 200 (2006) 4870 4875

4871

With those perspectives in mind and with a view to the advantages including precisely controlled near room temperature operation, low energy requirements, rapid deposition rates, capability to handle complex geometries, low cost, and simple scale-up with easily maintained equipment, of electrodeposition, we anticipate that it would be readily feasible to significantly improve the mechanical and tribological properties of electrodeposited Ni Co alloy coatings by the inclusion of CNTs as the reinforcing agent. Although CNTs and some kinds of ceramic nano-particulates have been extensively focused on in terms of their effect on the mechanical and tribological properties of electrodeposited and electroless metal-matrix coatings [2,3,7,11], no work has been so far reported on the preparation and properties of electrodeposited Ni Co CNTs composite coatings. Thus in the present work, Ni Co CNTs nanocomposite coatings were prepared by the electrodeposition in a nickel cobalt plating bath containing carbon nanotubes. This article deals with the surface morphologies, microstructures, and mechanical and tribological properties of the target composite coatings.

Fig. 1. Contact configuration of the friction pair.

2. Experimental 2.1. Synthesis and purification of CNTs The CNTs were synthesized by the catalytic decomposition of CH4 gas in the presence of iron catalyst supported on MgO powder [8,9]. A CH4/H2 mixture (CH4 : H2 = 50 : 1, V: V) of a pressure 600 Pa was introduced into a quartz chamber heated to 1173 K to allow the generation of the CNTs. The as-prepared CNTs were successively immersed in a concentrated hydrochloric acid and nitric acid, each for 12 h, and cleaned by distilled water to remove the catalyst impurities. After being filtered, the cleaned CNTs were dried at 393 K in an oven. The resultant black powders were heated at 693 K for 1 h in air and refluxed in a dilute nitric acid for 24 h, which gave birth to the target pure CNTs. 2.2. Preparation and characterization of Ni Co CNTs composite coatings Analytical reagents and distilled water were used to prepare the plating solution. The plating baths used to prepare the Ni Co CNTs composite coatings were composed of 220 g / l NiSO4, 40 g / l NiCl2, 20 g / l CoSO4, 35 g / l H3BO3, 1 g / l carbon tube, and a proper amount of surfactant. The electroplating tests were performed on a 273A potentiostat/galvanostat device (EG and G Princeton Applied Research). The bath was stirred using a magnetic stirrer. The experiments were conducted at 45 -C and current density of 50 mA cm 2. A platinum plate of 40 40 mm2 was used as the anode, while a saturated calomel electrode (SCE) was used as the reference electrode. A

stainless steel plate (20 30 3 mm) of a surface roughness below 0.3 Am (R a) was used as the cathode substrate to be plated. The polarization curves for the electrolyte at a sweep rate of 0.1 mV/s were recorded also on the 273A potentiogalvanostat device controlled by a PC. The surface morphologies of the target composite coatings were observed using an atom force microscope (AFM) which had a sharp Si3N4 tip radius approximately 20 nm and a surface scanning area of 10 10 Am2. The wear tracks of the composite coatings were observed using a scanning electron microscope (JEOLJSM-5600LV). The phase structures of the composite coatings were analyzed on an X-ray diffractometer (Philips X Pert-MRD). The nanoindentation tests involving loading and unloading courses were conducted on a Nano Test-600 device (Micro Materials, Wrexham, UK), with a trigonal (Berkovich) diamond tip, to determine the hardness of the composite coatings. The approach of Oliver and Pharr [10] was used to realize the tip correction and data reduction, and the computer provides accordingly the microhardness and elastic modulus by Indentatin software and load unload curve. The load and spatial resolutions of the equipment were 1 mN and 0.1 nm, respectively. Five indentations on different parts of each sample were performed and the averaged hardness values were given in this article. The deviation of the average hardness is less than 5%. The tribological behaviors of the electrodeposited coatings reciprocally sliding against SAE52100 steel ball (f3 mm) were examined on a UMT-2MT tribometer in a ball-on-disk configuration (see Fig. 1). The sliding was performed at an amplitude of 5 mm, a normal load of 0.54.0 N, and a frequency of 3.05.0 Hz. All the friction and wear tests were performed under unlubricated condition at room temperature and in ambient air (relative humidity 52%56%). The friction coefficient was recorded continuously during the tests.

3. Results and discussion 3.1. Polarization behavior of Ni Co CNTs electrolyte Fig. 2 shows the cathodic polarization behavior of Ni Co and Ni Co CNTs electrolytes. It is seen that the introduction of the carbon nanotubes in the electrolyte

4872
-0.01 0.00 0.01 0.02

L. Shi et al. / Surface & Coatings Technology 200 (2006) 4870 4875

Ni-Co

b
Ni-Co Ni-Co-CNT

0.03 0.04 0.05

Intensity (cps)

j (Acm )

-2

a
0.06 0.07 -1.0

-0.9

-0.8

-0.7 E (V vs. SCE)

-0.6

-0.5

20

30

40

50

60

70

80

90

2Theta (degrees)

Fig. 2. Cathodic polarization curves for the deposition of Ni Co and Ni Co CNTs in the plating baths.

Fig. 4. XRD patterns of electrodeposited (a) Ni Co and (b) Ni Co CNTs composite coatings.

causes the shift towards larger negatives of the reduction potential of Ni Co, but the slope of the reduction curve keeps unchanged. Such a shift to a lower value in the

reduction potential was attributed to the decrease in the active surface area of the cathode owing to the adsorption of the carbon nanotubes and might also relate to the decrease in the ionic transportation by the carbon nanotubes, which did not significantly affect the electrochemical reaction mechanism. This observation conforms to what was reported by Wu et al [11]. 3.2. Surface morphology of and structure the composite coating Fig. 3 shows the AFM images of the Ni Co and Ni Co CNTs composite coating. It is seen that the Ni Co CNTs composite coating has much smaller particle size, more uniform and compact coating in appearance than Ni Co coating, which indicates that the co-deposited CNTs were uniformly distributed in the Ni Co matrix of the composite coating and the presence of CNTs results in decreasing the size of particles. It could be rational to suppose that the carbon nanotubes uniformly distributed in
12

10

Load (mN)

b a

0 0 50 100 Depth (nm) 150 200

Fig. 3. AFM images of (a) Ni Co and (b) Ni Co CNTs composite coating.

Fig. 5. Typical load displacement curves of (a) Ni Co and (b) Ni Co CNTs composite coatings.

L. Shi et al. / Surface & Coatings Technology 200 (2006) 4870 4875
0.65 0.60 0.55

4873

Friction coefficient

0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0 500 1000 1500 2000 2500 3000 Sliding cycles 0.5N 3Hz 2.0N 4Hz 4.0N 5Hz

0.45 0.40

intensity of the diffraction peaks of the Ni Co solid solution in the composite coating is lower, while the peak width of the Ni Co solid solution for the composite coating is broader than that of the Ni Co alloy coating. This is attributed to the decrease in the grain size of the Ni Co CNTs composite coating by the addition of carbon nanotubes into the plating bath, which is proved by the average grain sizes of the coatings calculated from the diffraction peak widths (with the instrumental width eliminated) using Scherrer equation [12]. The Ni Co matrix in the composite coating had an average grain size of 17 nm, much smaller than 42 nm, that of the Ni Co alloy coating electrodeposited under the same condition. It is has been confirmed by AFM images. Namely, the growth of the electrodeposited layer is a competition between the nucleation and crystal growth. The carbon nanotubes provide more nucleation sites and hence retard the crystal growth, subsequently the corresponding Ni Co matrix in the composite coating has a smaller crystal size. 3.3. Mechanical and tribological properties Fig. 5 shows the typical loa unload curves for the Ni Co and Ni Co CNTs composite coating subject to the nanoindentation measurement at a depth of 200 nm. It was calculated based on the load unload curves given in Fig. 5 that the Ni Co CNTs composite coating had a nanoindentation hardness of 5.87 GPa and an elastic modulus of 236 GPa, which were higher than 4.41 GPa and 202 GPa, the hardness and elastic modulus of the Ni Co alloy coating. This implies that the Ni Co CNTs composite coatings could have better tribological properties than the Ni Co alloy coatings. Fig. 6 shows the variation of the friction coefficients for the Ni Co coating and Ni Co CNTs composite coating with sliding cycles. It is seen that the Ni Co CNTs composite coating has much smaller friction coefficient than the Ni Co alloy coating under the same test conditions, and the friction coefficient of two kinds of coatings gradually increases with increasing sliding cycles.

Friction coefficient

0.35 0.30 0.25 0.20 0.15 0 500 1000 1500 2000 2500

0.5N 3Hz 2.0N 4Hz 4.0N 5Hz 3000

Sliding cycles

Fig. 6. Friction coefficient of (a) Ni Co and (b) Ni Co CNTs composite coatings/GCr15 steel ball as a function of sliding cycles.

the Ni Co matrix contributed to significantly increase the mechanical properties and wear resistance of the Ni Co alloy coating. Fig. 4 shows the XRD patterns of the electrodeposited Ni Co coating and Ni Co CNTs composite coating. It can be concluded from the XRD patterns that the electrodeposited Ni Co alloy is composed of a solid solution. The

Fig. 7. SEM morphologies of (a) wear scar of the steel ball and (b) C distribution thereon (against Ni Co CNTs composite coating at 2.0 N and 4 Hz for 3000 cycles).

4874

L. Shi et al. / Surface & Coatings Technology 200 (2006) 4870 4875

Besides, the friction coefficient of the Ni Co CNTs composite coating decreases with increasing normal load and sliding speed. Namely, the averaged friction coefficient of the composite coating decreased from 0.37 at 0.5 N (sliding frequency 3 Hz) to 0.22 at 4.0 N (sliding frequency 5 Hz), which could be attributed to the formation of a lubricious transfer layer on the counterpart surface during sliding. This observation conforms to the typical evolution of the friction coefficient as a function of sliding cycle for the carbon coatings [13]. In other words, the contact temperature between the counterpart ball and the composite coating increases with increasing the normal load and sliding speed, which contributes to speed the transfer and accumulation of the lubricious wear debris on the counterpart rubbing surface. Subsequently, the friction is reduced owing to the formation of the lubricious transfer layer. Fig. 7 shows the SEM morphologies of the wear scar of the steel ball and C distribution thereon. It is seen that a great amount of wear debris were embedded in the worn steel ball surface (see Fig. 7a), while a small amount of C was transferred from the composite coating to the rubbing surface of the steel ball (see Fig. 7b). The SEM morphologies of the worn surfaces of Ni Co coating and Ni Co CNTs composite coating are shown in Fig. 8. It is seen that the wear track of the Ni Co alloy coating shows signs of severe adhesion, scuffing, and plastic deformation, and some large wear debris is generated thereon (see Fig. 8a). Different from the Ni Co alloy coating, the adhesion and scuffing on the worn surface of the Ni Co CNTs composite coating is significantly abated, though the scuffing and plastic deformation signs still remain in the latter case (see Fig. 8b). The significantly decreased adhesion could correspond to the better wearresistance of the composite coating than the Ni Co alloy coating. This is rational, since the CNTs transferred onto the rubbing surface of the counterpart steel ball could serve as the spacers to prevent the rough contact between the coating and the steel ball, which thereby contributes to reduce the wear of the frictional pair. In addition, CNTs would more easily slide or roll between the mating metal and coating surfaces, resulting in a great decrease in the friction

coefficient. In general, the Ni Co coating and Ni Co CNTs composite coating show obvious differences in the worn surface morphologies, which corresponds well to their different friction and wear behaviors.

4. Conclusions It was feasible to prepare Ni Co CNTs composite coating by properly incorporating carbon nanotubes to be co-deposited in the Ni Co plating bath. The addition of the carbon nanotubes to the electrolyte resulted in the shift towards larger negatives of the reduction potential of Ni Co but had no significant effect on the electrodepostion process. Moreover, the incorporation of CNTs in the plating bath led to the changes in the structures and enhances mechanical and tribological behaviors of the composite coatings as compared to the Ni Co alloy coatings. Namely, the Ni Co CNTs composite coatings had better mechanical and tribological properties than the Ni Co alloy coatings, which was attributed to the grain-fining and dispersive strengthening effects of the co-deposited carbon nanotubes. At the same time, a transfer layer of carbon was formed on the rubbing surface of counterpart steel ball, which corresponded to the significantly retarded adhesion, scuffing and lower friction and wear of the composite coating than the Ni Co alloy coating as well. Besides, the worn surface morphologies of the Ni Co CNTs composite coatings are much different from that of the Ni Co coating, which corresponded to their different friction and wear behaviors as well.

Acknowledgements The work was supported by the National Natural Science Foundation of China (Grant No.50405040), the Ministry of Science and Technology of China (Grant No.2002AA302609), and the Innovative Group Foundation from NSFC (Grant No.50421502). The authors thank Prof. J. Z. Zhao and Mr. D. K. Song for their help in the SEM, EDS, and XRD measurements.

Fig. 8. SEM morphology of the worn surfaces of (a) Ni Co coating at 0.5 N and 3 Hz for 3000 cycles and (b) Ni Co CNTs composite coating at 4.0 N and 5 Hz for 3000 cycles.

L. Shi et al. / Surface & Coatings Technology 200 (2006) 4870 4875

4875

References
[1] L. Benea, P.L. Bonora, A. Borello, S. Martelli, Wear 249 (2002) 995. [2] X.C. Li, Z.W. Li, Mater. Sci. Eng., A Struct. Mater.: Prop. Microstruct. Process. 00 (2003) 1. [3] A.F. Zimmerman, G. Palumbo, K.T. Aust, U. Erb, Mater. Sci. Eng., A Struct. Mater.: Prop. Microstruct. Process. 328 (2002) 137. [4] L. Rapoport, M. Lvovsky, I. Lapsker, V. Leshchinsky, Y. Volovik, Y. Feldman, A. Margolin, R. Rosentsveig, R. Tenne, Nano Lett. 1 (2001) 137. [5] C.B. Wang, D.L. Wang, W.X. Chen, Y.Y. Wang, Wear 253 (2002) 563.

[6] J.W. An, D.H. You, D.S. Lim, Wear 255 (2003) 677. [7] W.X. Chen, J.P. Tu, H.Y. Gan, Z.D. Xu, Q.G. Wang, J.Y. Lee, Z.L. Liu, X.B. Zhang, Surf. Coat. Technol. 160 (2002) 68. [8] J. Kong, A.M. Cassell, H. Dai, Chem. Phys. Lett. 292 (1998) 567. [9] M. Su, B. Zheng, J. Liu, Chem. Phys. Lett. 322 (2000) 321. [10] W.C. Oliver, G.M. Pharr, J. Mater. Res. 7 (1992) 1564. [11] G. Wu, N. Li, D.R. Zhou, K.C. Mitsuo, Surf. Coat. Technol. 176 (2003) 157. [12] B.D. Cullity, Elements of X-ray Diffraction, 2nd edition, Addison Wesley Publishing, London, 1978. [13] S.K. Field, M. Jarratt, D.G. Teer, Tribol. Int. 37 (2004) 949.