This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
Author's personal copy
Available online at www.sciencedirect.com
Scripta Materialia 63 (2010) 254–257 www.elsevier.com/locate/scriptamat
Tribological behavior of carbon nanofiber–zirconia composite Pavol Hvizdosˇ,* Viktor Puchy´, Annama´ria Duszova´ and Ja´n Dusza Institute of Materials Research, Slovak Academy of Sciences, Watsonova 47, 04353 Kosˇice, Slovakia Received 3 February 2010; revised 29 March 2010; accepted 31 March 2010 Available online 3 April 2010
The friction and wear behavior of ZrO2 + 1.07 wt.% carbon nanofiber composite and monolithic ZrO2 has been investigated using the ball-on-disk technique with alumina balls as friction partners in an unlubricated condition at room temperature. The coefficient of friction for the composite was found to be significantly lower compared to that for monolithic zirconia with relatively low wear rates for both materials. The main wear mechanism in the nanocomposite was abrasion accompanied with a pull-out of the carbon nanofibers which act in the interface as a sort of lubricating media. Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Zirconia; Carbon nanofibers; Nanocomposite; Wear
Carbon-based filamentous nanomaterials as carbon nanotubes (CNTs) and carbon nanofibers (CNFs) have attracted much attention during recent years due to their outstanding mechanical properties, very good electrical characteristics and excellent thermal performance [1,2]. One of the potential applications of carbon-based filamentous nanomaterials is in the form of reinforcement in polymer and ceramic composites where they produce improved mechanical and functional properties [3]. Over the last few years new ceramic–CNT composites have been developed and a number of authors have reported improved mechanical and functional properties in the case of composites compared to monolithic materials [4–7]. Investigations to date have focused mainly on alumina-based composites, with only limited work on other systems, e.g., silicon nitride or zirconia [8,9]. Furthermore, the reinforcing element has mainly been CNTs and only a few investigations have involved CNFs. Therefore, the potential advantages of CNFs compared to CNTs (price, shape, morphology) as reinforcing elements is still largely unexplored. Only in recent years have publications demonstrated the positive effect of CNFs as reinforcing elements in alumina-, silicon carbide- and hydroxyapatite-based composites [10,11]. Maensiri et al. [10] used CNFs similar to those in the present investigation and found that the influence of CNF addition on the mechanical properties of * Corresponding author. E-mail:
[email protected]
alumina–CNF composites is very similar to that found in our investigation for zirconia–CNF composite [16]. They found an improvement in the fracture toughness of approximately 13% with 2.5 vol.% CNFs, but the hardness and bending strength decreased with increasing volume fraction of CNFs. Kobayashi and Kawai [11] reported approximately 1.6 times higher fracture toughness for CNF-reinforced hydroxyapatite composite in comparison to pure hydroxyapatite. In addition to their functional properties, investigations on ceramic–CNT or CNF composites have focused on their basic mechanical properties such as hardness, strength and fracture toughness, and only limited work on their tribological characteristics has been attempted [12–15]. In these studies it was observed that the friction coefficient was reduced by the addition of CNTs to the ceramic matrix. The tribological behavior of well-aligned alumina– CNT nanocomposites with thin and thick CNTs was investigated by Xia et al. [13]. They demonstrated that the contact and buckling of CNTs determine the frictional coefficient of the composites. They also found that the thin-walled CNT composites have nearly the same frictional properties as the alumina, while the thick-walled CNT composites show much lower coefficients of friction. Lim et al. [14] investigated the tribological behavior of alumina–CNT composites with a different volume fraction of CNTs fabricated by hot pressing and tape casting, followed by lamination and hot pressing. They found that the tape-casting process significantly improved the uniform distribution of CNT in the alumina, resulting in an enhanced wear
1359-6462/$ - see front matter Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.scriptamat.2010.03.069
Author's personal copy
P. Hvizdosˇ et al. / Scripta Materialia 63 (2010) 254–257
255
15 cm s 1) were applied. Then the load was increased to 5 N and the speed was set on 10 cm s 1. Testing was done in air, at room temperature in dry conditions. The friction coefficient was measured and the damage was studied on the worn surfaces of disks and balls using optical and scanning electron microscopy (JEOL 7000). The microstructure and basic mechanical properties of the experimental materials were studied in detail elsewhere [16,17]. The microstructure of the composite consists of a very fine matrix (with grains even smaller than those in the monolithic zirconia) with relatively well-dispersed CNFs in the matrix (Fig. 1). The smaller grain size of the zirconia in the composites compared to the monolithic material is evidence that the CNFs hinder the grain growth in the composite during sintering. In Figure 2, the results of the wear test of the ZrO2 + CNF composite and monolithic ZrO2 is illustrated at an applied load of 1 N. It was clearly evident that the friction coefficient of the composite is nearly constant during the test but varies slightly with respect to the sliding speed in the interval of 0.22–0.27. The friction coefficient of monolithic ZrO2 is higher, and exhibits values from 0.38 to 0.51 depending on the sliding speed. In Figure 2b, the influence of the applied load on the friction coefficient and wear rate of the investigated materials is illustrated at applied loads of 1 and 5 N. According to these results, the friction coefficient increases with increasing applied load from an average value of 0.25 at 1 N to a value of 0.35 at 5 N for the composite and from an average value of 0.46 at 1 N to 0.5 at 5 N for monolithic ZrO2. Figure 2b illustrates that
resistance of the composites. An et al. [15] investigated the effect of CNT addition on wear resistance of alumina–CNT composites with the CNT contents ranging from 0 to 12.5 wt.%. They found enhanced wear properties for composites with CNT content in the range 0–4 wt.% . The aim of the present contribution is to study the effect of the addition of CNFs on the tribological properties of hot-pressed zirconia/1.07 wt.% CNF composite. As starting materials, ZrO2 powder (TZ-3Y, Tosoh, Japan) and CNFs (1.07 wt.%) were used. CNFs (HTF150FF, Electrovac, Austria) with an average diameter of 80–150 nm, specific surface area in the range of 20–100 m2 g 1, Young’s modulus 500 GPa, tensile strength 7 GPa and electrical resistivity of 10–3–10– 4 X cm were used. A water-based mixture was prepared with dispersant (DODECYLE MARANIL), CNFs, ZrO2 and a polyvinylbutryl binder (MOWITAL B30T). The dispersion was primarily done using ultrasonics and a magnetic stirrer. After dispersion, the mixture was subsequently spray dried. Specimens in the form of disks with diameters of 10 and 20 mm were die-pressed and then hot-pressed at 1300 °C and 40 MPa in argon. As a reference material, monolithic ZrO2 was prepared under similar conditions (but without any binder). The wear test was carried out on a High Temperature Tribometer THT (CSM, Switzerland) using the ball-ondisk technique. As friction partners, alumina balls 6 mm in diameter were used. In the first set of experiments 1 N normal load and various sliding speeds (2.5, 5, 10,
Figure 1. Microstructure of the experimental materials: (a) characteristic microstructure of monolithic ZrO2 (SEM); (b) ZrO2/ZrO2 grain boundary in the ZrO2 + CNF composite (HREM). 0.6 ZrO2
ZrO2+CNF
0.3 0.2
2.5 cm/s 5 cm/s 10 cm/s 15 cm/s
0.1
5
10 15 distance [m]
20
1E-3
1E-4
0.3 1E-5
0.2 0.1
25
0.0
ZrO2 ZrO2+CNF
0
1
2
3 load [N]
4
5
3
0.4
wear rate [mm /m.N]
0.4
0.0 0
b
0.5
friction coefficient
friction coefficient
0.5
0.6
a
1E-6
6
Figure 2. (a) Friction coefficient during the testing of the ZrO2 + CNF composite at 1 N load compared to the results for the reference material at 10 cm s 1. (b) Friction coefficient and wear rate as functions of the applied load.
Author's personal copy
256
P. Hvizdosˇ et al. / Scripta Materialia 63 (2010) 254–257
the wear rate increased with increased applied load from 1.5 10–6 to 1.3 10–5 mm3 N 1 m 1 in the case of ZrO2 + CNF and from 2.7 10–6 to 2.2 10–5 mm3 N 1 m 1 in the case of monolithic zirconia. The fact that the wear rate of the composite and ZrO2 increased with the increased applied load to the same extent shows that there is probably no significant change in wear mechanisms of the studied materials, and only their intensity increased. The measured values of the friction coefficient are in good agreement with the results reported for similar tribocouples and experimental conditions. Suh et al. [18] investigated the sliding wear behavior of ZrO2 using the ball-on-disk method and alumina as a friction partner, similarly as in the present work, at applied loads from 19.8 to 98 N. They found slightly higher coefficient of frictions from 0.6 to 0.7 depending on the sliding speed, which were independent on the applied load. According to An et al. [15], the friction coefficient of Al2O3 + CNTs tested against silicon nitride ball decreased with increased addition of CNTs; however, the influence of CNTs is significant only at amounts higher than 4 wt.%. Moreover, the lowest value of the friction coefficient of approximately 0.3 for Al2O3 + 12 wt.% CNTs is still higher compared to the coefficient of ZrO2 + 1.07 wt.% CNF measured in the present investigation. In regards to the wear rate, our results are different to the results of An et al. and Lim et al. [14] who reported a decreased wear loss in the case of Al2O3 + CNTs with up to 4 wt.% CNT. According to the results at low normal load (1 N), both materials exhibited stable friction, which did not depend on the sliding speed. Fractographic analysis of the worn surface revealed that the main wear mechanism in the zirconia is self-polishing by low-intensity abrasion. In the previous study on the alumina composites [18], the alumina grains detached from the alumina surface caused abrasion of zirconia. However, this observation is less significant in our case because of the low applied load. Hence the increase in applied load from 1 to 5 N did not vary the wear mechanism, but slightly intensified it. The worn surfaces of the composite after the wear test at 1 N applied load revealed moderate wear with production of a low amount of debris in the form of a mixture of CNFs and zirconia particles. After the test at 5 N applied load, we found nanofiber pull-out in the case of small-diameter nanofibers oriented perpen-
a
b
Zirconia grain
Carbon nanofiber
Figure 4. Schematic illustration of wear and carbon tribofilm formation in ZrO2 + CNF composite. (a) Polished surface before the wear test, and (b) tribofilm formation during the wear test.
dicularly to the surface (Fig. 3b). The CNFs with larger diameter and nanofibers oriented not perpendicularly to the worn surface were ground at both applied loads, and the resulting graphite together with the crushed, pulledout CNFs creates the transferred film (Fig. 4). It seems that the excellent friction coefficient of the composite is probably related to the smearing of this film over the contact area, which permits easy shear and then helps to achieve a lubricating effect during sliding. A similar lubricating effect was reported previously for multi-walled (MW) CNT-containing Al2O3 composites by An et al. [12] and Yamamoto et al. [15] and for single-walled (SW) CNT solids by Yamamoto [19]. An et al. reported a 40% decrease in friction coefficient in 12 wt.% MWCNT–Al2O3 composite down to the value of 0.3 [15]. Yamamoto et al. for their composites, reached a minimum friction coefficient (with a value of 0.3) at 4 mass% of MWCNT content [12]. According to An et al., the rolling motion of CNTs at the interface between the specimen and the ball can probably lower the friction coefficient as well. This effect was not proven in the case of ZrO2 + CNF composite. Our results are, however, in a very good agreement with the results reported for SWCNT solids sliding against Si3N4 where steady-state friction coefficients of 0.22–0.24 were found [19]. The results presented here demonstrate that excellent frictional properties of ZrO2 + CNF can be achieved by incorporating a low amount (1.07 wt.%) of CNTs into the microstructure of zirconia. The coefficient of friction of ZrO2 + CNF composite was found to be significantly
Figure 3. Wear damage at normal load 5 N: (a) smooth wear tracks in monolithic ZrO2; (b) detail of the wear track in ZrO2 + CNF composite showing CNFs pull-out and transferred film on the worn surface.
Author's personal copy
P. Hvizdosˇ et al. / Scripta Materialia 63 (2010) 254–257
lower compared to that for monolithic zirconia with relatively low wear rates for both materials. It seems that this low coefficient of friction is related to the formation of carbon-based transferred film over the contact area, which permits easy shear and helps to create a lubricating effect during sliding. This work was supported by the Slovak Government through projects APVV-0034-07 and LPP0174-07 and 0203-07. Contribution from CE Nanosmart and the MNT-ERANET HANCOC Project is gratefully acknowledged. The authors thank G. Blugan and J. Kuebler for the help in processing materials and J. Morgiel in TEM analyses. [1] S. Iijima, Nature 354 (1991) 56. [2] E.V. Barrera, M.L. Shofner, E.L. Corral, in: M. Meyyappan (Ed.), Carbon Nanotubes in Science & Application, CRC Press, New York, 2004. [3] E.T. Thostenson, Z. Ren, T.-W. Chou, Compos. Sci. Technol. 61 (2001) 1899. [4] Ch. Laurent, A. Peigney, O. Dumortier, A. Rousset, J. Eur. Ceram. Soc. 18 (1998) 2005. [5] G.D. Zhan, J.D. Kuntz, J.E. Garay, A.K. Mukherjee, P. Zhu, K. Koumoto, Scripta Mater. 54 (2006) 77.
257
[6] X. Wang, N.P. Padture, H. Tanaka, Nat. Mater. 3 (2004) 539. [7] R.W. Siegel, S.K. Chang, B.J. Ash, J. Stone, P.M. Ajayan, R.W. Doremus, L.S. Schadler, Scripta Mater. 44 (2001) 2061. [8] G.D. Zhan, J.D. Kuntz, J. Wan, A.K. Mukherjee, Nat. Mater. 61 (2003) 1899. [9] C. Balazsi, Z. Ko´nya, F. We´ber, L.P. Biro´, P. Arato´, Mater. Sci. Eng. C23 (2003) 1133. [10] S. Maensiri, P. Laokul, J. Klinkaewnarong, V. Amornkitbamrung, Mater. Sci. Eng. A 447 (2007) 44. [11] S. Kobayashi, W. Kawai, Compos. A Appl. Sci. Manuf. 38 (2007) 114. [12] G. Yamamoto, M. Omori, K. Yokomizo, T. Hashida, K. Adachi, Mater. Sci. Eng. B 148 (2008) 265. [13] Z.H. Xia, J. Lou, W.A. Curtin, Scripta Mater. 58 (2008) 223. [14] D.-S. Lim, D.-H. You, H.-J. Choi, S.-H. Lim, H. Jang, Wear 259 (2005) 539. [15] J.-W. An, D.-H. You, D.-S. Lim, Wear 255 (2003) 677. [16] A. Duszova´, J. Dusza, K. Toma´sˇek, J. Morgiel, G. Blugan, J. Kuebler, Scripta Mater. 58 (2008) 520. [17] A. Duszova´, J. Dusza, K. Toma´sˇek, G. Blugan, J. Kuebler, J. Eur. Ceram. Soc. 28 (2008) 1023. [18] M.-S. Suh, Y.-H. Chae, S.-S. Kim, Wear 264 (2008) 800. [19] G. Yamamoto, T. Hashida, K. Adachi, T. Takagj, J. Nanosci. Nanotechnol. 8 (2008) 2665.
