- Email: [email protected]
Graphite Composite under Dry Sliding
Journal of Materials Science & Technology xxx (2015) 1e6
Contents lists available at ScienceDirect
Journal of Materials Science & Technology journal homepage: www.jmst.org
Friction and Wear Behavior of Resin/Graphite Composite under Dry Sliding Zhenguo Zhu, Shuo Bai*, Junfeng Wu, Li Xu, Ting Li, Yong Ren, Chang Liu Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy of Sciences, Shenyang 110016, China
a r t i c l e i n f o Article history: Received 2 April 2014 Received in revised form 27 May 2014 Available online xxx Key words: Resin/graphite composite Friction and wear “Dusting” wear Transfer film
The friction and wear behavior of resin/graphite composite has been investigated using a pin-on-disc configuration under dry sliding condition. The results showed that the resin/graphite composite exhibited much better mechanical and tribological properties compared with the unimpregnated graphite. The friction coefficient was reduced by addition of furan resin, which could also prevent the “dusting” wear at loads more than 15 MPa. The steady and lubricated transfer film was easily formed on the counterpart surface due to the interaction of furan resin and wear debris of graphite, which was useful to reduce the wear rate of the resin/graphite composite. The composite is highly promising for mechanical sealing application and can be used at high load for long time sliding. Copyright © 2014, The editorial office of Journal of Materials Science & Technology. Published by Elsevier Limited. All rights reserved.
1. Introduction Carbon‒graphite materials have been widely used in mechanical seal due to their excellent self-lubricating, chemical inertness, high thermal conductivity and outstanding temperature stability[1]. The application of traditional carbon‒graphite material is sometimes restricted by their low mechanical strength, low wear resistance and poor sealing performance, which are caused by a large portion of connected pores generated during the preparation process. To overcome these above shortcomings, some carbon‒ graphite composites have been prepared by infiltrating these materials in a second phase (metal, polymer, inorganic salt, etc.) at high pressure in the carbon industry[2‒4]. The friction and wear properties of graphite have been widely studied[1,5‒7]. For a long time, weak interplanar van der Waals interactions between planes of graphite are considered to be the origin of its low friction coefficient[8]. Nevertheless, the friction coefficient of graphite was higher than 0.5 and fluctuated greatly under vacuum or some inert atmospheres[5,9,10]. This erratic friction behavior is known as “dusting” wear owing to the rapid disintegration of graphite into a cloud of fine dust-like debris[10]. Researches show that water vapor or other condensable vapors in the environment could prevent the “dusting” wear and lead to a
* Corresponding author. Prof., Ph.D.; Tel.: þ86 24 83978648; Fax: þ86 2423891320. E-mail address: [email protected] (S. Bai).
complete lubrication even at much lower pressure. It was clearly demonstrated that the interactions between graphite and water vapor greatly influence its tribological properties[5,11]. The widely accepted friction mechanism for carbon‒graphite material inferred that the stable and complete transfer film formed on metal surface during friction against graphite could reduce the friction coefficient and wear rate, and the formation of transfer film is concerned with the structure of graphite and its interactions with the environment[1,12,13]. Compared with pure carbon‒graphite, carbon‒graphite composite exhibited different tribological properties owing to the multiphase structure and interaction among the phases during contact sliding. There were lots of researches on the friction and wear behavior of C/C composite[14‒18] and the friction coefficient of C/C composite was relatively high owing to different properties of carbon fiber and pyrolytic carbon matrix. However, if lubricated film formed between the friction surface, the C/C composite could also exhibit better tribological properties[17]. Some investigations on metal/graphite composite had been conducted[19‒21], where tribological properties of the composite were significantly enhanced by forming steady and lubricated transfer film. It was found that the transfer film usually formed when polymers slide against a metal or polymer counterfaces to affect the friction and wear behavior of the two sliding pairs[22]. Therefore, it may be a feasible way to combine resin with graphite, by which a steady and complete transfer film would form and low friction coefficient and wear rate can be achieved. Research had found that resin/graphite composite could form lubricated film at the initial
http://dx.doi.org/10.1016/j.jmst.2014.10.004 1005-0302/Copyright © 2014, The editorial office of Journal of Materials Science & Technology. Published by Elsevier Limited. All rights reserved.
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004
Z. Zhu et al. / Journal of Materials Science & Technology xxx (2015) 1e6
2
Fig. 2. Microstructure of the resin/graphite composite.
Fig. 1. Schematic of sliding friction test.
stage of the friction due to the supporting of the resin[23]. But the role of resin in resin/graphite composite is not clear during high loads friction process. In this work, a resin/graphite composite was prepared and the effect of applied loads on the tribological properties of the composite was investigated under dry sliding, which was the harshest condition for mechanical seal. Moreover, the wear mechanisms of graphite and the resin/graphite composite were discussed.
2. Experimental Work The substrate graphite material used in this study was prepared by the traditional molding method and heat-treated at 2200 C, and its degree of graphitization is 24.4%. The resin/graphite composite was prepared by impregnating furan resin at high pressure, and the specific method of preparation had been described in details in our previous publication[24]. The weight gain rate (resin content, %) of the composite was calculated using the following equation:
w¼
m m0 100% m0
and 12.8 mm in length) and the counterpart disc (45 steel) were polished with No.1200 waterproof abrasive paper and ultrasonically cleaned in acetone. A thermocouple was attached on the stationary disc approximately 2 mm from the mating face, and an infrared thermal imager (Ti32, FLUKE) was used to measure the friction temperature field. The tests were carried out at normal load from 3 MPa to 20 MPa at a constant linear velocity of 0.5 m/s for 120 min. The friction coefficient was directly recorded with a testing machine. Before and after the test, the pins were ultrasonically cleaned and dried for weighing. The mass loss of the pin was measured with an electrical balance (Mettler AE240, accuracy 0.01 mg) for the specific wear rate calculation. The specific wear rate k was calculated from the following equation:
k¼
Dm mm3 N1 m1 rNL
(2)
where Dm (mg) is the mass loss of the specimen, r (mg/mm3) is the density of the specimen, N (N) is the normal load and L (m) is the total sliding distance. The microstructure, worn surfaces of specimens, transfer film on the steel counterpart and wear debris were observed with an optical microscope (OM) and scanning electron microscope (SEM, Nova NanoSEM 430). The synchronous thermal analyses (thermogravimetry/differential scanning calorimetry/ mass spectrometer, TG/DSC/MS, Netzsch 449C Jupiter/QMS 403C) of furan resin were performed in air at a ramping rate of 5 C/min up to 300 C.
(1) 3. Results and Discussion
where w (%) is the weight gain rate, m0 (g) is the weight before impregnation, m (g) is the weight after impregnation. Compressive and flexural tests were performed on a universal testing machine. The open porosity (%) of the composite was measured by boiling method. Shore hardness (HS) was measured by using a shore hardness tester. The data reported here represent the average result of at least five tests. The friction and wear tests were conducted in a pin-on-disc contact configuration (Fig. 1) under dry sliding condition at room temperature. Before test, the pin specimen (4.8 mm in diameter
3.1. Structure and properties Some mechanical and physical properties of the graphite and resin/graphite composite are listed in Table 1. The furan resin of 6 wt% was impregnated into the graphite. The densities of the graphite and resin/graphite composite are 1.85 and 1.98 g/cm3, respectively. The compressive strength and flexural strength of the composite are 156 MPa and 58 MPa, respectively, much higher than 101 MPa and 35 MPa for the graphite. The shore hardness of the
Table 1 Mechanical and physical properties of graphite and its composite
Graphite Resin/graphite composite
Density (g/cm3)
Resin content (wt%)
Compressive strength (MPa)
Flexural strength (MPa)
Shore hardness (HS)
Open porosity (%)
1.85 ± 0.02 1.98 ± 0.02
‒ 6
101 156
35 58
72 76
10.34 0.73
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004
Z. Zhu et al. / Journal of Materials Science & Technology xxx (2015) 1e6
3
Fig. 3. Representative SEM images of the fracture surfaces of (a) the graphite, (b) the resin/graphite composite and (c) furan resin in the composite.
provides a large interface area available for achieving strong interaction between graphite and the resin. So the resin/graphite composite exhibits much better mechanical properties than graphite. 3.2. Friction and wear behavior
Fig. 4. Relationship of friction coefficient of the graphite and its composite versus applied loads.
composite reaches 76. The open porosity of the composite is only 0.73%, less than one-tenth of the graphite. The polished surface structure of the resin/graphite composite is shown in Fig. 2. It can be seen that the furan resin is impregnated well in graphite. Fig. 3 presents SEM images of the compressive fracture surfaces of the graphite and resin/graphite composite. Many connected pores are observed on the fracture surface of graphite (Fig. 3(a)). The surface is very rough, and there are many small particles. The fracture surface of the composite (Fig. 3(b)) is relatively smooth and almost no pores can be found. Fig. 3(c) shows the furan resin in the composite which undergoes a brittle fracture. As reported previously, there are many pores generated during the preparation process of graphite, which cause stress concentration and decrease the strength and hardness of graphite[25]. Obviously, high pressure impregnation is an efficient way to improve the mechanical properties of graphite because most pores in graphite can be filled by the furan resin, which heals the pore defects and
The variation of the friction coefficient of the graphite and the resin/graphite composite with applied loads against the 45 steel counterpart is shown in Fig. 4. The friction coefficient of the graphite was about 0.21 at 3 MPa, which firstly increased and then decreased when the normal load was increased to 10 MPa, and the highest friction coefficient was 0.29 at 8 MPa. The friction coefficient sharply increased to about 0.5 at 15 MPa along with the rapid disintegration of graphite into a cloud of fine dust-like debris, which is known as “dusting” wear[10]. The friction coefficient of the composite was about 0.17 at 3 MPa, which also firstly increased to 0.25 at 5 MPa and then decreased to 0.12 at 20 MPa. Compared to the graphite, the average friction coefficient of the composite was decreased by 18.3% below 10 MPa. Fig. 5 presents the friction behavior of the graphite and the resin/graphite composite at a load of 15 MPa. For the graphite, the erratic friction behavior occurred at the sliding distance about 380 m and the measured temperature is about 85 C. The accumulation of heat between the graphite pin and the steel counterpart resulted in an abrupt rising of the friction coefficient and friction temperature (Fig. 5(a)). As shown in Fig. 5(b), when the sliding distance of the composite reached about 2800 m, the temperature between the composite pin and the steel counterpart also increased to 85 C, and then the friction coefficient and temperature decreased slowly. Lancaster[26] had observed a friction and wear transition in ambient air for graphite at 150e185 C and attributed it to the desorption of physisorbed water vapor from graphite surfaces. In this study, although the thermocouple measured temperature is
Fig. 5. Friction behavior of (a) the graphite and (b) the resin/graphite composite at a load of 15 MPa.
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004
4
Z. Zhu et al. / Journal of Materials Science & Technology xxx (2015) 1e6
Fig. 6. (a) Relationship between the highest temperature and thermocouple temperature; (b) the representative friction temperature field during “dusting wear” as recorded by an infrared thermal imager.
about 85 C, the highest temperature of friction is determined to be higher than 160 C (Fig. 6(a)). So that the “dusting” wear of the graphite happened at 15 MPa is believed due to the heavy desorption of physisorbed water vapor from graphite surfaces caused by transient high temperature, where the wear of the graphite changed from mild wear into a severe wear. On the other hand, the temperature of the composite did not reach 85 C before 2800 m sliding distance, after which the friction coefficient and temperature decreased gradually. The parallel synchronous thermogravimetry/differential scanning calorimetry/mass spectroscopy (TG/DSC/MS) analyses of furan resin showed that about 10 wt % weight was lost occurred upon heating to 300 C at a ramping rate of 5 C/min, and the MS (m/e ¼ 18 H2O) signal appeared at around 150 C associated with the endothermic of the DSC profile, which meant that the decomposition of furan resin could release water and adsorb the heat (Fig. 7). The water omitted from furan resin at high temperature could supplement the water vapor desorption from the composite surface and result in the reduction of friction coefficient. On the other hand, the endothermic reaction would decrease the friction temperature and keep lubrication of the composite. So the impregnating furan resin would make the resin/ graphite composite have a better friction performance and good lubrication even at high loads. Fig. 8 shows the wear rate of the graphite and resin/graphite composite under different applied loads. Two samples present similar low wear rate (about 1.5 106 mm3 N1 m1 at loads less than 5 MPa. However, the wear rate of the resin/graphite composite obviously reduced at higher loads, such as reducing by 170% at 8 MPa and 104% at 10 MPa as compared to the graphite. The wear rate of the composite was still very low even at 15 MPa and 20 MPa while the graphite underwent “dusting” wear at 15 MPa and the
wear rate was 717 106 mm3 N1 m1, which was more than eight hundred times larger than that of the composite. The results show that the wear resistance of the resin/graphite composite under dry sliding was greatly enhanced by the addition of furan resin, especially at high loads. The reducing wear of the resin/ graphite composite was concerned with higher strength and hardness because of the enhancement of furan resin. Fig. 9 shows typical morphologies of the worn surface of the graphite and resin/graphite composite. The worn surface of the graphite shows many grooves and ridges parallel to the sliding direction (SD) under loads of both 3 MPa and 8 MPa (Figs. 9 (a, b)), which indicates that the main wear mechanism was abrasive wear. There are much shallower grooves on the worn surface of the resin/ graphite composite at 3 MPa (Fig. 9(c)), and the worn surface is smooth and grooves were considerably restricted at 8 MPa (Fig. 9(d)). These results indicate that furan resin in the composite is helpful to form a lubricated film at higher load. The representative OM images of the transfer films on the counterpart of the graphite and resin/graphite composite are given in Fig. 10. There are two regions on the transfer film of graphite (Fig. 10(a)): the black “A” region represents graphite transfer film and bright “B” is 45 steel counterpart, indicating that there was an unstable and incomplete friction film formed between the graphite and steel counterpart during sliding. However, the transfer film of the composite is very smooth and intact, which prevented the direct contact between the composite and counterpart (Fig. 10(b)). The wear debris was collected after the wear experiment and was investigated by SEM (Fig. 11). Most of the wear debris from graphite at 8 MPa was twisting and linear (Fig. 11(a)). This shape of wear debris is consistent with the grooves and ridges on the worn
Fig. 7. TG/DSC/MS profiles of the furan resin being heated to 300 C at a ramping rate of 5 C/min.
Fig. 8. Effect of normal load on the wear rate of the graphite and resin/graphite composite.
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004
Z. Zhu et al. / Journal of Materials Science & Technology xxx (2015) 1e6
5
Fig. 9. Typical morphologies of the worn surfaces of (a) 3 MPa and (b) 8 MPa of the graphite; (c) 3 MPa and (d) 8 MPa of the resin/graphite composite.
Fig. 10. OM images of the transfer films of (a) the graphite and (b) the resin/graphite composite on the 45 steel counterface.
surface due to scratched off by many hard asperities on the counterpart steel surface, and steady and complete films are difficult to be formed. The wear debris of the resin/graphite composite was layer-like flakes (Fig. 11(b)) and its morphology was similar to that of the transfer film on the counterpart. The cyclic stresses resulting from the contact area of the composite surface against the hard counterpart led to transfer film peeling off from the friction surface. It was reported that the wear rate was not the rate of fracture of the original surface but the rate of detachment of the wear debris from the friction films[27]. So this layer-like flakes debris is useful for forming steady and complete film and reducing the wear rate of the resin/graphite composite under dry sliding condition. Langlade et al. reported that some pure carbon‒graphite at some conditions is poorly adherent to a surface; the formation of a transfer film is often limited; and the film is patchy and could
not provide efficient lubrication[13]. In this study, it is difficult to form a steady and complete transfer film on the 45 steel counterpart under dry sliding for the applied graphite, and the asperities on 45 steel counterpart still could be found (Fig. 10(a)). Abrasive wear of the graphite occurred because the graphite and steel counterpart contacted directly and the friction surface was scratched by asperities on counterpart forming grooves and ridges worn surface (Figs. 9(a, b)) and twisting and linear wear debris (Fig. 11(a)). Therefore, the friction coefficient and wear rate of graphite were higher. However, the transfer film on the counterpart was steady and complete when the resin/graphite composite slid against 45 steel under dry sliding (Fig. 10(b)), so the worn surface was smooth and the friction coefficient and wear rate of the composite were relatively low. Researchers found that the additives in graphite, such as sodium thiosulfate
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004
Z. Zhu et al. / Journal of Materials Science & Technology xxx (2015) 1e6
6
Fig. 11. SEM images of the typical wear debris: (a) the graphite; (b) the resin/graphite composite.
(Na2S203), sodium molybdate (Na2MoO4), and tricresyl phosphate (TCP), would improve the formation of transfer film and the stability of friction process by interaction with graphite[28,29]. Furan resin may also interact with graphite during friction process, including physical and chemical interaction. This interaction is beneficial to the formation of a steady and complete transfer film, which provides efficient lubrication for the resin/graphite composite.
4. Conclusions The mechanical and tribological properties of the resin/graphite composite have been investigated in comparison with graphite, and the following points are concluded: (1) Furan resin is an effective reinforcement to achieve higher density, lower open porosity, and better mechanical properties for the resin/graphite composite. (2) The average friction coefficient of the resin/graphite composite was reduced by 18.3% compared with graphite below 10 MPa, and the lowest friction coefficient reached 0.12 at 20 MPa. Meanwhile, the furan resin could prevent “dusting” wear at loads more than 15 MPa because the decomposition of furan resin was endothermic reaction and released water vapor at high temperatures. (3) The wear rate of the resin/graphite composite was reduced by 170% at 8 MPa and 104% at 10 MPa, and the wear rate of “dusting” wear of graphite at 15 MPa was more than eight hundred times larger than the composite. (4) The origin of the abrasive wear for graphite is the formation of unstable and incomplete transfer film. Meanwhile, the interaction between wear debris of furan resin and graphite leads to steady and complete transfer film, especially at
higher loads, which enables relatively low friction coefficient and wear rate for the resin/graphite composite. References [1] J.A. Williams, J.H. Morris, A. Ball, Tribol. Int. 30 (1997) 663e676. [2] J.J. Ott, R.E. Matthews, G.D. Lawrence, Impregnation of Porous Body with Metal, U.S. Patent No. 3867177, 1975. [3] F.L. Abrams, J.W. Hager, R.B. Hall, Carbon or Graphite Foam Reinforced Composites, US Patent No. 5770127, 1998. [4] R.R. Paxton, G.I. Beyer, Method for Impregnating Graphite Bodies and the Article Produced Therefrom, US Patent No. 3342627, 1967. [5] R.H. Savage, J. Appl. Phys. 19 (1948) 1e10. [6] J. Spreadborough, Wear 5 (1962) 18e30. [7] J.K. Lancaster, Treatise Mater. Sci. Technol. 13 (1979) 141‒174. [8] W.L. Bragg, Introduction to Crystal Analysis, Bell, London, 1928. [9] F.P. Bowden, J.E. Young, Proc. R. Soc. Land. A 208 (1951) 444e455. [10] B.K. Yen, Wear 192 (1996) 208e215. [11] R.H. Savage, D.L. Schaefer, J. Appl. Phys. 27 (1956) 136e138. [12] W.T. Clark, A. Connolly, W. Hirst, Brit. J. Appl. Phys. 14 (1963) 20e27. [13] C. Langlade, S. Fayeulle, R. Olierb, Wear 172 (1994) 85e92. [14] J.D. Chen, J.H. Chern Lin, C.P. Ju, Wear 193 (1996) 38e47. [15] B.K. Yen, T. Ishihara, Carbon 34 (1996) 489e498. [16] V.I. Kostikov, Z.V. Eremeeva, D.A. Slyuta, D.L. Yaitskii, G.Kh Sharipzyanova, Russ. J. Non-ferrous Metals 50 (2009) 189e192. [17] X. Xiong, B.Y. Huang, Carbon 44 (2006) 463e467. [18] B.L. Lei, L.L. He, M.Z. Yi, L.P. Ran, H.J. Xu, Y.C. Ge, K. Peng, Carbon 49 (2011) 4554e4562. [19] Y.A. Wang, J.X. Li, Y. Yan, L.J. Qiao, Tribol. Int. 50 (2012) 26e34. [20] Y.B. Liu, S.C. Lim, S. Ray, P.K. Rohatgi, Wear 159 (1992) 201e205. [21] P.K. Rohatgi, S. Ray, Y. Liu, Int. Mater. Rev. 37 (1992) 129e152. [22] S. Bahadur, Wear 245 (2000) 92e95. [23] Y.F. Hu, Q.L. Wang, M. Sui, J.W. Hu, J. China Univ. Mining Technol. 37 (2008) 216e219. [24] Z.G. Zhu, J. Chen, H.J. Xu, S. Bai, T. Li, Y. Ren, H.F. Du, J.F. Wu, Carbon Tech. 32 (2013) A23eA25 (in Chinese). [25] Z.T. He, L.N. Gao, X. Wang, B.L. Zhang, W. Qi, J.L. Song, X.J. He, C. Zhang, H. Tang, H.H. Xia, X.T. Zhou, Carbon 72 (2014) 304e311. [26] J.K. Lancaster, ASLE Trans. 18 (1975) 187e201. [27] M. Godet, Approaches to Modeling of Friction and Wear, Springer-Verlag, New York, 1988, pp. 12e36. [28] C. Langlade, S. Fayeulle, R. Olierb, Thin Solid Films 237 (1994) 38e47. [29] J.K. Lancaster, Tribol. Int. 23 (1990) 371e389.
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004
Contents lists available at ScienceDirect
Journal of Materials Science & Technology journal homepage: www.jmst.org
Friction and Wear Behavior of Resin/Graphite Composite under Dry Sliding Zhenguo Zhu, Shuo Bai*, Junfeng Wu, Li Xu, Ting Li, Yong Ren, Chang Liu Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy of Sciences, Shenyang 110016, China
a r t i c l e i n f o Article history: Received 2 April 2014 Received in revised form 27 May 2014 Available online xxx Key words: Resin/graphite composite Friction and wear “Dusting” wear Transfer film
The friction and wear behavior of resin/graphite composite has been investigated using a pin-on-disc configuration under dry sliding condition. The results showed that the resin/graphite composite exhibited much better mechanical and tribological properties compared with the unimpregnated graphite. The friction coefficient was reduced by addition of furan resin, which could also prevent the “dusting” wear at loads more than 15 MPa. The steady and lubricated transfer film was easily formed on the counterpart surface due to the interaction of furan resin and wear debris of graphite, which was useful to reduce the wear rate of the resin/graphite composite. The composite is highly promising for mechanical sealing application and can be used at high load for long time sliding. Copyright © 2014, The editorial office of Journal of Materials Science & Technology. Published by Elsevier Limited. All rights reserved.
1. Introduction Carbon‒graphite materials have been widely used in mechanical seal due to their excellent self-lubricating, chemical inertness, high thermal conductivity and outstanding temperature stability[1]. The application of traditional carbon‒graphite material is sometimes restricted by their low mechanical strength, low wear resistance and poor sealing performance, which are caused by a large portion of connected pores generated during the preparation process. To overcome these above shortcomings, some carbon‒ graphite composites have been prepared by infiltrating these materials in a second phase (metal, polymer, inorganic salt, etc.) at high pressure in the carbon industry[2‒4]. The friction and wear properties of graphite have been widely studied[1,5‒7]. For a long time, weak interplanar van der Waals interactions between planes of graphite are considered to be the origin of its low friction coefficient[8]. Nevertheless, the friction coefficient of graphite was higher than 0.5 and fluctuated greatly under vacuum or some inert atmospheres[5,9,10]. This erratic friction behavior is known as “dusting” wear owing to the rapid disintegration of graphite into a cloud of fine dust-like debris[10]. Researches show that water vapor or other condensable vapors in the environment could prevent the “dusting” wear and lead to a
* Corresponding author. Prof., Ph.D.; Tel.: þ86 24 83978648; Fax: þ86 2423891320. E-mail address: [email protected] (S. Bai).
complete lubrication even at much lower pressure. It was clearly demonstrated that the interactions between graphite and water vapor greatly influence its tribological properties[5,11]. The widely accepted friction mechanism for carbon‒graphite material inferred that the stable and complete transfer film formed on metal surface during friction against graphite could reduce the friction coefficient and wear rate, and the formation of transfer film is concerned with the structure of graphite and its interactions with the environment[1,12,13]. Compared with pure carbon‒graphite, carbon‒graphite composite exhibited different tribological properties owing to the multiphase structure and interaction among the phases during contact sliding. There were lots of researches on the friction and wear behavior of C/C composite[14‒18] and the friction coefficient of C/C composite was relatively high owing to different properties of carbon fiber and pyrolytic carbon matrix. However, if lubricated film formed between the friction surface, the C/C composite could also exhibit better tribological properties[17]. Some investigations on metal/graphite composite had been conducted[19‒21], where tribological properties of the composite were significantly enhanced by forming steady and lubricated transfer film. It was found that the transfer film usually formed when polymers slide against a metal or polymer counterfaces to affect the friction and wear behavior of the two sliding pairs[22]. Therefore, it may be a feasible way to combine resin with graphite, by which a steady and complete transfer film would form and low friction coefficient and wear rate can be achieved. Research had found that resin/graphite composite could form lubricated film at the initial
http://dx.doi.org/10.1016/j.jmst.2014.10.004 1005-0302/Copyright © 2014, The editorial office of Journal of Materials Science & Technology. Published by Elsevier Limited. All rights reserved.
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004
Z. Zhu et al. / Journal of Materials Science & Technology xxx (2015) 1e6
2
Fig. 2. Microstructure of the resin/graphite composite.
Fig. 1. Schematic of sliding friction test.
stage of the friction due to the supporting of the resin[23]. But the role of resin in resin/graphite composite is not clear during high loads friction process. In this work, a resin/graphite composite was prepared and the effect of applied loads on the tribological properties of the composite was investigated under dry sliding, which was the harshest condition for mechanical seal. Moreover, the wear mechanisms of graphite and the resin/graphite composite were discussed.
2. Experimental Work The substrate graphite material used in this study was prepared by the traditional molding method and heat-treated at 2200 C, and its degree of graphitization is 24.4%. The resin/graphite composite was prepared by impregnating furan resin at high pressure, and the specific method of preparation had been described in details in our previous publication[24]. The weight gain rate (resin content, %) of the composite was calculated using the following equation:
w¼
m m0 100% m0
and 12.8 mm in length) and the counterpart disc (45 steel) were polished with No.1200 waterproof abrasive paper and ultrasonically cleaned in acetone. A thermocouple was attached on the stationary disc approximately 2 mm from the mating face, and an infrared thermal imager (Ti32, FLUKE) was used to measure the friction temperature field. The tests were carried out at normal load from 3 MPa to 20 MPa at a constant linear velocity of 0.5 m/s for 120 min. The friction coefficient was directly recorded with a testing machine. Before and after the test, the pins were ultrasonically cleaned and dried for weighing. The mass loss of the pin was measured with an electrical balance (Mettler AE240, accuracy 0.01 mg) for the specific wear rate calculation. The specific wear rate k was calculated from the following equation:
k¼
Dm mm3 N1 m1 rNL
(2)
where Dm (mg) is the mass loss of the specimen, r (mg/mm3) is the density of the specimen, N (N) is the normal load and L (m) is the total sliding distance. The microstructure, worn surfaces of specimens, transfer film on the steel counterpart and wear debris were observed with an optical microscope (OM) and scanning electron microscope (SEM, Nova NanoSEM 430). The synchronous thermal analyses (thermogravimetry/differential scanning calorimetry/ mass spectrometer, TG/DSC/MS, Netzsch 449C Jupiter/QMS 403C) of furan resin were performed in air at a ramping rate of 5 C/min up to 300 C.
(1) 3. Results and Discussion
where w (%) is the weight gain rate, m0 (g) is the weight before impregnation, m (g) is the weight after impregnation. Compressive and flexural tests were performed on a universal testing machine. The open porosity (%) of the composite was measured by boiling method. Shore hardness (HS) was measured by using a shore hardness tester. The data reported here represent the average result of at least five tests. The friction and wear tests were conducted in a pin-on-disc contact configuration (Fig. 1) under dry sliding condition at room temperature. Before test, the pin specimen (4.8 mm in diameter
3.1. Structure and properties Some mechanical and physical properties of the graphite and resin/graphite composite are listed in Table 1. The furan resin of 6 wt% was impregnated into the graphite. The densities of the graphite and resin/graphite composite are 1.85 and 1.98 g/cm3, respectively. The compressive strength and flexural strength of the composite are 156 MPa and 58 MPa, respectively, much higher than 101 MPa and 35 MPa for the graphite. The shore hardness of the
Table 1 Mechanical and physical properties of graphite and its composite
Graphite Resin/graphite composite
Density (g/cm3)
Resin content (wt%)
Compressive strength (MPa)
Flexural strength (MPa)
Shore hardness (HS)
Open porosity (%)
1.85 ± 0.02 1.98 ± 0.02
‒ 6
101 156
35 58
72 76
10.34 0.73
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004
Z. Zhu et al. / Journal of Materials Science & Technology xxx (2015) 1e6
3
Fig. 3. Representative SEM images of the fracture surfaces of (a) the graphite, (b) the resin/graphite composite and (c) furan resin in the composite.
provides a large interface area available for achieving strong interaction between graphite and the resin. So the resin/graphite composite exhibits much better mechanical properties than graphite. 3.2. Friction and wear behavior
Fig. 4. Relationship of friction coefficient of the graphite and its composite versus applied loads.
composite reaches 76. The open porosity of the composite is only 0.73%, less than one-tenth of the graphite. The polished surface structure of the resin/graphite composite is shown in Fig. 2. It can be seen that the furan resin is impregnated well in graphite. Fig. 3 presents SEM images of the compressive fracture surfaces of the graphite and resin/graphite composite. Many connected pores are observed on the fracture surface of graphite (Fig. 3(a)). The surface is very rough, and there are many small particles. The fracture surface of the composite (Fig. 3(b)) is relatively smooth and almost no pores can be found. Fig. 3(c) shows the furan resin in the composite which undergoes a brittle fracture. As reported previously, there are many pores generated during the preparation process of graphite, which cause stress concentration and decrease the strength and hardness of graphite[25]. Obviously, high pressure impregnation is an efficient way to improve the mechanical properties of graphite because most pores in graphite can be filled by the furan resin, which heals the pore defects and
The variation of the friction coefficient of the graphite and the resin/graphite composite with applied loads against the 45 steel counterpart is shown in Fig. 4. The friction coefficient of the graphite was about 0.21 at 3 MPa, which firstly increased and then decreased when the normal load was increased to 10 MPa, and the highest friction coefficient was 0.29 at 8 MPa. The friction coefficient sharply increased to about 0.5 at 15 MPa along with the rapid disintegration of graphite into a cloud of fine dust-like debris, which is known as “dusting” wear[10]. The friction coefficient of the composite was about 0.17 at 3 MPa, which also firstly increased to 0.25 at 5 MPa and then decreased to 0.12 at 20 MPa. Compared to the graphite, the average friction coefficient of the composite was decreased by 18.3% below 10 MPa. Fig. 5 presents the friction behavior of the graphite and the resin/graphite composite at a load of 15 MPa. For the graphite, the erratic friction behavior occurred at the sliding distance about 380 m and the measured temperature is about 85 C. The accumulation of heat between the graphite pin and the steel counterpart resulted in an abrupt rising of the friction coefficient and friction temperature (Fig. 5(a)). As shown in Fig. 5(b), when the sliding distance of the composite reached about 2800 m, the temperature between the composite pin and the steel counterpart also increased to 85 C, and then the friction coefficient and temperature decreased slowly. Lancaster[26] had observed a friction and wear transition in ambient air for graphite at 150e185 C and attributed it to the desorption of physisorbed water vapor from graphite surfaces. In this study, although the thermocouple measured temperature is
Fig. 5. Friction behavior of (a) the graphite and (b) the resin/graphite composite at a load of 15 MPa.
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004
4
Z. Zhu et al. / Journal of Materials Science & Technology xxx (2015) 1e6
Fig. 6. (a) Relationship between the highest temperature and thermocouple temperature; (b) the representative friction temperature field during “dusting wear” as recorded by an infrared thermal imager.
about 85 C, the highest temperature of friction is determined to be higher than 160 C (Fig. 6(a)). So that the “dusting” wear of the graphite happened at 15 MPa is believed due to the heavy desorption of physisorbed water vapor from graphite surfaces caused by transient high temperature, where the wear of the graphite changed from mild wear into a severe wear. On the other hand, the temperature of the composite did not reach 85 C before 2800 m sliding distance, after which the friction coefficient and temperature decreased gradually. The parallel synchronous thermogravimetry/differential scanning calorimetry/mass spectroscopy (TG/DSC/MS) analyses of furan resin showed that about 10 wt % weight was lost occurred upon heating to 300 C at a ramping rate of 5 C/min, and the MS (m/e ¼ 18 H2O) signal appeared at around 150 C associated with the endothermic of the DSC profile, which meant that the decomposition of furan resin could release water and adsorb the heat (Fig. 7). The water omitted from furan resin at high temperature could supplement the water vapor desorption from the composite surface and result in the reduction of friction coefficient. On the other hand, the endothermic reaction would decrease the friction temperature and keep lubrication of the composite. So the impregnating furan resin would make the resin/ graphite composite have a better friction performance and good lubrication even at high loads. Fig. 8 shows the wear rate of the graphite and resin/graphite composite under different applied loads. Two samples present similar low wear rate (about 1.5 106 mm3 N1 m1 at loads less than 5 MPa. However, the wear rate of the resin/graphite composite obviously reduced at higher loads, such as reducing by 170% at 8 MPa and 104% at 10 MPa as compared to the graphite. The wear rate of the composite was still very low even at 15 MPa and 20 MPa while the graphite underwent “dusting” wear at 15 MPa and the
wear rate was 717 106 mm3 N1 m1, which was more than eight hundred times larger than that of the composite. The results show that the wear resistance of the resin/graphite composite under dry sliding was greatly enhanced by the addition of furan resin, especially at high loads. The reducing wear of the resin/ graphite composite was concerned with higher strength and hardness because of the enhancement of furan resin. Fig. 9 shows typical morphologies of the worn surface of the graphite and resin/graphite composite. The worn surface of the graphite shows many grooves and ridges parallel to the sliding direction (SD) under loads of both 3 MPa and 8 MPa (Figs. 9 (a, b)), which indicates that the main wear mechanism was abrasive wear. There are much shallower grooves on the worn surface of the resin/ graphite composite at 3 MPa (Fig. 9(c)), and the worn surface is smooth and grooves were considerably restricted at 8 MPa (Fig. 9(d)). These results indicate that furan resin in the composite is helpful to form a lubricated film at higher load. The representative OM images of the transfer films on the counterpart of the graphite and resin/graphite composite are given in Fig. 10. There are two regions on the transfer film of graphite (Fig. 10(a)): the black “A” region represents graphite transfer film and bright “B” is 45 steel counterpart, indicating that there was an unstable and incomplete friction film formed between the graphite and steel counterpart during sliding. However, the transfer film of the composite is very smooth and intact, which prevented the direct contact between the composite and counterpart (Fig. 10(b)). The wear debris was collected after the wear experiment and was investigated by SEM (Fig. 11). Most of the wear debris from graphite at 8 MPa was twisting and linear (Fig. 11(a)). This shape of wear debris is consistent with the grooves and ridges on the worn
Fig. 7. TG/DSC/MS profiles of the furan resin being heated to 300 C at a ramping rate of 5 C/min.
Fig. 8. Effect of normal load on the wear rate of the graphite and resin/graphite composite.
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004
Z. Zhu et al. / Journal of Materials Science & Technology xxx (2015) 1e6
5
Fig. 9. Typical morphologies of the worn surfaces of (a) 3 MPa and (b) 8 MPa of the graphite; (c) 3 MPa and (d) 8 MPa of the resin/graphite composite.
Fig. 10. OM images of the transfer films of (a) the graphite and (b) the resin/graphite composite on the 45 steel counterface.
surface due to scratched off by many hard asperities on the counterpart steel surface, and steady and complete films are difficult to be formed. The wear debris of the resin/graphite composite was layer-like flakes (Fig. 11(b)) and its morphology was similar to that of the transfer film on the counterpart. The cyclic stresses resulting from the contact area of the composite surface against the hard counterpart led to transfer film peeling off from the friction surface. It was reported that the wear rate was not the rate of fracture of the original surface but the rate of detachment of the wear debris from the friction films[27]. So this layer-like flakes debris is useful for forming steady and complete film and reducing the wear rate of the resin/graphite composite under dry sliding condition. Langlade et al. reported that some pure carbon‒graphite at some conditions is poorly adherent to a surface; the formation of a transfer film is often limited; and the film is patchy and could
not provide efficient lubrication[13]. In this study, it is difficult to form a steady and complete transfer film on the 45 steel counterpart under dry sliding for the applied graphite, and the asperities on 45 steel counterpart still could be found (Fig. 10(a)). Abrasive wear of the graphite occurred because the graphite and steel counterpart contacted directly and the friction surface was scratched by asperities on counterpart forming grooves and ridges worn surface (Figs. 9(a, b)) and twisting and linear wear debris (Fig. 11(a)). Therefore, the friction coefficient and wear rate of graphite were higher. However, the transfer film on the counterpart was steady and complete when the resin/graphite composite slid against 45 steel under dry sliding (Fig. 10(b)), so the worn surface was smooth and the friction coefficient and wear rate of the composite were relatively low. Researchers found that the additives in graphite, such as sodium thiosulfate
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004
Z. Zhu et al. / Journal of Materials Science & Technology xxx (2015) 1e6
6
Fig. 11. SEM images of the typical wear debris: (a) the graphite; (b) the resin/graphite composite.
(Na2S203), sodium molybdate (Na2MoO4), and tricresyl phosphate (TCP), would improve the formation of transfer film and the stability of friction process by interaction with graphite[28,29]. Furan resin may also interact with graphite during friction process, including physical and chemical interaction. This interaction is beneficial to the formation of a steady and complete transfer film, which provides efficient lubrication for the resin/graphite composite.
4. Conclusions The mechanical and tribological properties of the resin/graphite composite have been investigated in comparison with graphite, and the following points are concluded: (1) Furan resin is an effective reinforcement to achieve higher density, lower open porosity, and better mechanical properties for the resin/graphite composite. (2) The average friction coefficient of the resin/graphite composite was reduced by 18.3% compared with graphite below 10 MPa, and the lowest friction coefficient reached 0.12 at 20 MPa. Meanwhile, the furan resin could prevent “dusting” wear at loads more than 15 MPa because the decomposition of furan resin was endothermic reaction and released water vapor at high temperatures. (3) The wear rate of the resin/graphite composite was reduced by 170% at 8 MPa and 104% at 10 MPa, and the wear rate of “dusting” wear of graphite at 15 MPa was more than eight hundred times larger than the composite. (4) The origin of the abrasive wear for graphite is the formation of unstable and incomplete transfer film. Meanwhile, the interaction between wear debris of furan resin and graphite leads to steady and complete transfer film, especially at
higher loads, which enables relatively low friction coefficient and wear rate for the resin/graphite composite. References [1] J.A. Williams, J.H. Morris, A. Ball, Tribol. Int. 30 (1997) 663e676. [2] J.J. Ott, R.E. Matthews, G.D. Lawrence, Impregnation of Porous Body with Metal, U.S. Patent No. 3867177, 1975. [3] F.L. Abrams, J.W. Hager, R.B. Hall, Carbon or Graphite Foam Reinforced Composites, US Patent No. 5770127, 1998. [4] R.R. Paxton, G.I. Beyer, Method for Impregnating Graphite Bodies and the Article Produced Therefrom, US Patent No. 3342627, 1967. [5] R.H. Savage, J. Appl. Phys. 19 (1948) 1e10. [6] J. Spreadborough, Wear 5 (1962) 18e30. [7] J.K. Lancaster, Treatise Mater. Sci. Technol. 13 (1979) 141‒174. [8] W.L. Bragg, Introduction to Crystal Analysis, Bell, London, 1928. [9] F.P. Bowden, J.E. Young, Proc. R. Soc. Land. A 208 (1951) 444e455. [10] B.K. Yen, Wear 192 (1996) 208e215. [11] R.H. Savage, D.L. Schaefer, J. Appl. Phys. 27 (1956) 136e138. [12] W.T. Clark, A. Connolly, W. Hirst, Brit. J. Appl. Phys. 14 (1963) 20e27. [13] C. Langlade, S. Fayeulle, R. Olierb, Wear 172 (1994) 85e92. [14] J.D. Chen, J.H. Chern Lin, C.P. Ju, Wear 193 (1996) 38e47. [15] B.K. Yen, T. Ishihara, Carbon 34 (1996) 489e498. [16] V.I. Kostikov, Z.V. Eremeeva, D.A. Slyuta, D.L. Yaitskii, G.Kh Sharipzyanova, Russ. J. Non-ferrous Metals 50 (2009) 189e192. [17] X. Xiong, B.Y. Huang, Carbon 44 (2006) 463e467. [18] B.L. Lei, L.L. He, M.Z. Yi, L.P. Ran, H.J. Xu, Y.C. Ge, K. Peng, Carbon 49 (2011) 4554e4562. [19] Y.A. Wang, J.X. Li, Y. Yan, L.J. Qiao, Tribol. Int. 50 (2012) 26e34. [20] Y.B. Liu, S.C. Lim, S. Ray, P.K. Rohatgi, Wear 159 (1992) 201e205. [21] P.K. Rohatgi, S. Ray, Y. Liu, Int. Mater. Rev. 37 (1992) 129e152. [22] S. Bahadur, Wear 245 (2000) 92e95. [23] Y.F. Hu, Q.L. Wang, M. Sui, J.W. Hu, J. China Univ. Mining Technol. 37 (2008) 216e219. [24] Z.G. Zhu, J. Chen, H.J. Xu, S. Bai, T. Li, Y. Ren, H.F. Du, J.F. Wu, Carbon Tech. 32 (2013) A23eA25 (in Chinese). [25] Z.T. He, L.N. Gao, X. Wang, B.L. Zhang, W. Qi, J.L. Song, X.J. He, C. Zhang, H. Tang, H.H. Xia, X.T. Zhou, Carbon 72 (2014) 304e311. [26] J.K. Lancaster, ASLE Trans. 18 (1975) 187e201. [27] M. Godet, Approaches to Modeling of Friction and Wear, Springer-Verlag, New York, 1988, pp. 12e36. [28] C. Langlade, S. Fayeulle, R. Olierb, Thin Solid Films 237 (1994) 38e47. [29] J.K. Lancaster, Tribol. Int. 23 (1990) 371e389.
Please cite this article in press as: Z. Zhu, et al., Journal of Materials Science & Technology (2015), http://dx.doi.org/10.1016/j.jmst.2014.10.004