- Email: [email protected]
Mechanical properties of Ti(C0.7N0.3) film produced by plasma electrolytic carbonitriding of Ti6Al4V alloy
Applied Surface Science 254 (2008) 6350–6357
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Mechanical properties of Ti(C0.7N0.3) film produced by plasma electrolytic carbonitriding of Ti6Al4V alloy Xin-Mei Li, Yong Han * State-key Laboratory for Mechanical Behavior of Materials, Xi’an Jiaotong University, Xi’an 710049, PR China
A R T I C L E I N F O
A B S T R A C T
Article history: Received 1 October 2007 Received in revised form 16 March 2008 Accepted 28 March 2008 Available online 4 April 2008
Porous nanocrystalline Ti(C0.7N0.3) film on Ti6Al4V substrate was prepared by plasma electrolytic carbonitriding (PECN). The film was characterized and analyzed by using a variety of analytical techniques, such as XRD, SEM, EDX, TEM, FESEM, Rockwell C indenter, scratch tester, Vickers microhardness tester and ring-on-block tribometer. The results showed that the film was about 15 mm thick and its hardness was Hv 2369 at a load of 0.2 N. The adhesion of the film was characterized by Lc and Pc value, and was found to be about 42 N and more than 800 N, respectively. The friction coefficients and wear volume loss of the PECN-treated samples sliding against a steel counterpart were much less than those of the untreated Ti6Al4V. The film possessed a good wear-resistance and antifriction under oillubricated condition due to its high hardness, adhesion and fracture toughness. Also, the porous surface morphology of the Ti(C0.7N0.3) film contributed to the enhanced tribological resistance by promoting the formation of lubricant film and entrapping wear debris. ß 2008 Elsevier B.V. All rights reserved.
PACS: 61.82.Rx 68.35.Gy 81.65. Lp Keywords: Plasma electrolytic carbonitriding Ti(C0.7N0.3) thick film Ti6Al4V Adhesion Wear-resistance
1. Introduction Titanium alloys are attractive materials due to its low density, very good resistance to corrosion and high relative strength. However, their resistance to frictional wear is poor. Surface engineering techniques are widely used to improve the surface hardness and wear resistance of titanium alloys, and they have been realized as promising methods to deposit TiC, TiN and Ti(C,N) hard films on titanium alloys. Up to now, various techniques have been developed to deposit or form titanium carbide/nitride films on titanium alloys, such as physical vapor deposition (PVD) [1,2], chemical vapor deposition (CVD) [3,4], ion beam assisted deposition (IBAD) [5–7], and plasma carbiding/nitriding [8,9]. However, the titanium carbide/nitride films produced by IBAD on Ti6Al4V alloy are too thin (generally less than 3 mm) and the loadbearing capacity is poor [7]. The adhesion of the titanium carbide/ nitride films obtained by PVD and CVD are insufficient owing to the increased internal stresses in the films [10], and thereby delamination of the films has been observed under wear condition [11]. Although plasma carbiding/nitridng treatment of titanium at
* Corresponding author. Fax: +86 29 82663453. E-mail address: [email protected] (Y. Han). 0169-4332/$ – see front matter ß 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2008.03.172
700–900 8C leads to firmly adhered titanium carbide/nitride layer, the formation of the compound layer with the thickness more than 10 mm requires much higher temperature and longer treatment time [8,12], which results in worse fatigue behavior and earlier fracture due to the aging of the substrate material [7]. It has been proved that the improvement efficacy of the wear resistance and load-bearing capacity strongly depend on the thickness and adhesion of the hard films [13,14], thus alternative methods to produce thick and firmly adhered titanium carbide/ nitride films on titanium alloy at low temperature are still worth exploring. It is found that plasma electrolytic oxidation (PEO) can achieve a relatively fast conversion of titanium surface, at near-to-ambient bulk temperature, into a titanium oxide ceramic layer [15–17]. These titanium oxide layers amount to 10 mm in thickness and firmly bond to the titanium substrate. Based on the principle involved in PEO, a novel surface modification technology named as plasma electrolytic carbonitriding (PECN), has been developed for surface modification of pure titanium and thereby formation of thick Ti(CxN1x) films in our previous work [18,19], however, still unclear how about the mechanical properties of the modified layer. In the present work, the mechanical properties of the PECN-formed Ti(CxN1x) film on Ti6Al4V alloy are investigated, especially the adhesion, hardness, and friction and wear behavior.
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
2. Experimental Ti6Al4V plates with sizes of 30 mm 10 mm 1 mm were used as substrates. The plates were polished to a mirror finish with an aqueous silica suspension, and ultrasonically cleaned with acetone and distilled water. For PECN treatment, a pulse power supply was employed, and a Ti6Al4V plate was used as a cathode while a graphite plate was used as an anode in an electrolytic cell. A mixed organic solution of triethanolamine [N(CH2CH2OH)3] and formamide [HCONH2], and other proprietary ingredients (added primarily for adjustment of electrical conductivity) were chosen as an electrolyte. The applied voltage, pulse frequency and duty cycle were fixed at 600 V, 100 Hz and 40%, respectively, and the Ti6Al4V plates were discharge-treated for 2.5 h. The electrolyte bath was water-cooled and its temperature was maintained lower than 30 8C. After the PECN treatment, the obtained samples were washed with distilled water and dried at room temperature. The phase components of the PECN-treated samples were analyzed with X-ray diffraction (XRD) using a Cu Ka radiation. Scanning electron microscopy (SEM) and field emission scanning electron microscopy (FESEM) were employed to observe the morphologies, thickness and grain sizes of the formed film. Rockwell C indenter (conical diamond with 1208 included angle and 0.2 mm tip radius) was used to evaluate the adhesion of the PECN-formed film. In the indentation method, the force which separates a film from a substrate, expressed by Pc value, can characterize the adhesion of a film [13]. A series of loads were applied to the surface of the PECN-treated sample, and the resulting damages of the film around the indentations were examined using SEM. As a comparison, the adhesion of the film was also investigated using a WS-92 automatic scratch tester. A diamond stylus of 200 mm radius with an apex angle of 1208 was drawn over the film at a speed of 6 mm min1 under a continuous progressive normal force. The smallest load at which the film is damaged is designated the critical load, Lc, which gives an indirect indication of the adhesion [20,21]. The scratch tester was equipped with acoustic emission monitoring equipment, which was used as an on-line failure monitor. Critical load thresholds and detailed morphology of the scratch pattern were correlated by SEM and backscattered electron microscopy (BSE) observations. Five scratch tests were performed to obtain mean and standard deviation values for each film. Hardness measurements were performed on the surface of the PECN-treated sample using a Vickers microhardness tester at a load of 0.2 N. The hardness value is an average of 10 measurements. Fracture toughness of the film was roughly estimated from Vickers indentations at a load of 10 N. Friction and wear tests were carried out on a ring-on-block reciprocating sliding tribometer under lubricated condition as follows. The blocks (8 mm diameter) were made of the PECN-treated and untreated Ti6Al4V, respectively, and the rings (40 mm external diameter) were made of middle carbon steel with a hardness of HRC 48–52. The ring was rotated against the block at a speed of 0.42 m s1. Commercial 30# engine oil and 0.9% saline solution were used as lubricant, respectively, and drop rates of them were about 16–20 drops/ min. The friction coefficients were calculated from the moment automatically recorded by the friction and wear tester according to the following relationship:
m¼
M RF
volume loss was determined by measuring the cross-sectional area of the wear scar with a profilometer. Worn surfaces were investigated using SEM coupled with energy-dispersive X-ray spectrometer (EDX). At least three repetitions of each test were carried out. 3. Results and discussion The XRD pattern of the sample PECN discharge-treated for 2.5 h is shown in Fig. 1. The formed film is composed of Ti(C0.7N0.3). As shown in Fig. 2, the Ti(C0.7N0.3) film exhibits a porous surface morphology with a pore size of 5 mm. These pores are well separated and homogeneously distributed over the surface of the film. When observed at high magnification (Fig. 3), the film matrix consists of much fine particles with the size of 40–60 nm, suggesting that the PECN-formed Ti(C0.7N0.3) film is nanocrystallized. From the cross-sectional view (Fig. 4), the thickness of the Ti(C0.7N0.3) film is 15 mm. The surface and cross-sectional morphology and crystallization characteristics of the PECN-formed Ti(C0.7N0.3) film on Ti6Al4V are similar to those on commercial titanium [18,19]. The adhesion of the PECN-formed Ti(C0.7N0.3) film can be roughly evaluated using indentation test according to the damage type of the film [20]. Fig. 5 shows the indentation morphologies of the film subjected to different loads. No cracks and spallations can be observed around the indentation at a load of 800 N; however, the radial and annular cracks appear with increasing the applied
Fig. 1. XRD patterns of the PECN-treated samples for the discharge time of (a) 0 and (b) 2.5 h.
(1)
where m is the friction coefficient, M is the moment, R is the external diameter of the ring, and F is the applied load. The wear
6351
Fig. 2. SEM photograph of the sample PECN discharge-treated for 2.5 h.
6352
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
Fig. 4. Cross-sectional morphology of the sample PECN discharge-treated for 2.5 h. Fig. 3. FESEM photograph of the sample PECN discharge-treated for 2.5 h.
load up to 950 N, indicating that the Pc value of the Ti(C0.7N0.3) film is more than 800 N. For thin films under such a high load, a considerable substrate deformation is inevitable, thus resulting in the failure for most of the hard films. For example, when subjected to indentation test, for a sputtered TiN film (2 mm thick) on hardened M2 steel substrate, flaking off in a large annular area occurred at 600 N load; for an ion plated TiN film (2 mm thick) on hardened M2 steel substrate, crack and flaking in a small area were detected at 600 N, however, annular flaking occurred at 600 N with increasing the TiN film up to 4 mm in thickness [22]. The Pc value generally increases with the augment of the substrate hardness, it can be concluded, therefore, the PECN-formed Ti(C0.7N0.3) film possesses high adhesion.
Scratch test is generally accepted as one of the simple means in assessing adhesion of a film on its substrate [23–25]. In the scratch test, the load at which failure is first observed to occur regularly along the scratch track, correlating with a sudden increase in acoustic emission, is termed as the critical load Lc. The typical acoustic emission signal of the PECN-treated sample in the scratch test is shown in Fig. 6. The Lc value of the Ti(C0.7N0.3) film is about 42 N in average. Fig. 7 shows the typical scratched morphologies for PECN-treated sample on one scratch track made at continuous progressive load. Buckling failure is observed within the scratch track and the Ti(C0.7N0.3) film tends towards spallation in the track only, which is a ductile failure mode according to [20]. As reported by Takadoum and Houmid
Fig. 5. Indentation morphologies of the PECN-treated sample at the loads of (a) 300 N, (b) 500 N, (c) 800 N and (d) 950 N.
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
Fig. 6. Acoustic emission signal of the PECN-treated sample in scratch test.
[26], when an ion plated TiN film on 35CD4 steel substrate subjected to the scratch test, for the film thickness of 1.5 mm, only superficial semicircular crack traces were detected at 2 N load while spalling observed at 5 N; for the film thickness of 3 mm, spalling took place at 3 N; for the film thickness of 5 mm, spalling occurred at 2 N. Arai et al. [22] also reported, when an ion plated TiN film on hardened M2 steel substrate subjected to the scratch test, for the film thickness of 1 mm, only cracks were observed at 1 N while chipping detected at 2 N; for the film thickness of 4 mm, small amounts of flaking appeared at 2 N. However, for a sputtered TiN film (2 mm thick) on hardened M2 steel substrate, flaking in a large area occurred at 2 N. Besides, some researchers directly utilized the critical load to indicate cracking resistance [27,28], or some even termed it ‘‘scratch toughness’’ [29,30]. For example, a magnetron sputtered
6353
Ti1xAlxN nanocomposite thin film (hardness 28.5 GPa, thickness 0.9 mm) on silicon substrate which reached 70 N load without brittle failure was considered having good scratch toughness. In comparison, films of TiN (0.9 mm, Lc = 30–40 N) and AlN (0.6 mm, Lc = 20 N) were not so good in scratch toughness [31,32]. Like the Pc value, the Lc value tends to increase with the augment of the substrate hardness. Consequently, it also can be confirmed that the PECN-formed Ti(C0.7N0.3) film possesses high adhesion, toughness and load-bearing capacity. The microhardness of the sample PECN discharge-treated for 2.5 h is Hv2369, over 5 times of that of the Ti6Al4V substrate (Hv 420). As shown in Fig. 8(a), the Vickers microhardness indentation is quite small in size, due to the rather low load. However, the diagonal of the indentation is distinct. In the case of brittle materials and coatings, the technique of indentation fracture permits to estimate the fracture toughness from the length of the cracks induced around the indentation [33–35]. To roughly estimate the fracture toughness of the PECN-formed Ti(C0.7N0.3) film, a load as high as 10 N is applied to the film, and no cracks are observed around the indentation impression (Fig. 8(b)). As pointed out by Feng et al. [36] and Veprek et al. [37], TiN coatings and TiN/a-Si3N4 coatings were thought to be of high fracture toughness on the basis of the fact that no radial cracks appeared around the indentation impression at loads of 0.25 and 1 N, respectively. Compared with the aforementioned results, it can be expected that the PECN-formed Ti(C0.7N0.3) film possesses a considerable fracture toughness. In addition, the impression region of the Ti(C0.7N0.3) film becomes dense (Fig. 8(b)), exhibiting a significant plastic deformation. The considerable plasticity and ductility of the PECN-formed film is thought to be due to the fine grain size. Moreover, the porous surface morphology is advantageous for reducing brittle failures and relaxing stress [38]. The friction and wear behavior of the PECN-treated and untreated Ti6Al4V samples sliding against a steel counterpart are shown in Fig. 9. The friction coefficients of PECN-treated sample under oil-lubricated condition are significantly reduced
Fig. 7. SEM (a and c) and BSE (b and d) photographs of scratched morphologies for the PECN-treated sample. The point where the failure is first observed is denoted as ‘‘F’’, and (c and d) are its part magnifications, respectively. Direction of scratch is shown by arrows.
6354
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
Fig. 8. Vickers indentation views of the PECN-formed film at the loads of (a) 0.2 N and (b) 10 N.
to 0.10–0.11, about four times of magnitude less than in the case of the untreated Ti6Al4V. Furthermore, it tends to decrease with increasing the sliding distance. Compared with the untreated Ti6Al4V sample, the PECN-treated sample shows a pronounced improvement in wear-resistance. For example, the wear volume loss of the PECN-treated sample subjected to a sliding distance of 1507 m under oil-lubricated condition is 0.588 102 mm3, about 1/3000 of that of the untreated Ti6Al4V sample subjected to 452 m. For the untreated Ti6Al4V sample under 0.9% saline solution lubricated condition, a sliding distance above 50 m causes severe wear and wide-range oscillation of the friction coefficients. For the PECN-treated sample under 0.9% saline solution lubricated condition, the friction coefficients increase gradually, however, its increment rates are far slower than those of the untreated Ti6Al4V sample whether under oil-lubricated or 0.9% saline solution lubricated condition. Accordingly, the resulting wear volume loss of the PECN-treated sample under 0.9% saline solution lubricated condition is also much less than that of the untreated Ti6Al4V sample. In general, PECN-formed Ti(C0.7N0.3) film possesses good wear-resistance and antifriction, especially under oil-lubricated condition. Fig. 10 presents the wear morphologies of the PECN-treated and untreated Ti6Al4V samples. The wear track of the untreated Ti6Al4V sample is wide and deep, full of microcutting and deep grooves (Fig. 10(a) and (b)). In addition, Fe is not detected on the worn surface (Fig. 11(a)), indicating no metal transfer from the steel counterpart onto the untreated Ti6Al4V during the wear test. It is evident that the untreated Ti6Al4V sample experienced severe abrasive wear. Since the untreated Ti6Al4V is softer than the steel counterpart, metal on the contact tip can act as a cutting/machining edge on the untreated Ti6Al4V sample, thus resulting in the deep grooves, gouging and rapid failure of the sample surface. For the PECN-treated sample, however, the wear behavior is quite different. Almost no measurable wear can be detected for the PECN-treated sample under oil-lubricated condition and the only observable feature is a polishing of the sample surface. The wear track is narrow and shallow, and the worn surface is smooth, without any plough and detachment (Fig. 10(c) and (d)), suggesting that the PECNtreated sample only suffers mild abrasive wear under oillubricated condition. Compared with the initial porous surface morphology (Fig. 2), the porosity of the worn surface is decreased and some debris are embedded in pores (Fig. 10(d)). EDX reveals that these debris are mainly composed of iron chips (Fig. 11(b)), exhibiting a metal transfer from the steel counterpart onto the Ti(C0.7N0.3) film during the wear test.
It also can be confirmed by the EDX spectrum shown in Fig. 11(c). Despite appearance of fine and shallow scratches parallel to relative movement direction, the wear track of the PECN-treated sample under 0.9% saline solution lubricated
Fig. 9. Variations of (a) friction coefficient m and (b) wear volume loss of the samples under condition I, II, III and IV at a load of 200 N. I, untreated Ti6Al4V sample subjected to a sliding distance of 452 m under oil-lubricated condition; II, PECNtreated sample subjected to a sliding distance of 1507 m under oil-lubricated condition; III, untreated Ti6Al4V sample subjected to a sliding distance of 50 m under 0.9% saline solution lubricated condition; IV, PECN-treated sample subjected to a sliding distance of 1507 m under 0.9% saline solution lubricated condition.
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
6355
Fig. 10. SEM photographs of wear tracks for the samples under different conditions of wear test at a load of 200 N. (a and b) Untreated Ti6Al4V sample subjected to a sliding distance of 452 m under oil-lubricated condition; (c and d) PECN-treated sample subjected to a sliding distance of 1507 m under oil-lubricated condition; (e and f) PECNtreated sample subjected to a sliding distance of 1507m under 0.9% saline solution lubricated condition. (b), (d) and (f) are the part magnifications of (a), (c) and (e), respectively.
condition is still narrow and shallow, and the wear surface is smooth (Fig. 10(e) and (f)). Besides, the metal transfer from the steel counterpart onto the Ti(C0.7N0.3) film and the oxidation of the film occur (Fig. 11(d)). Abrasive wear is still predominant for the PECN-treated sample under 0.9% saline solution lubricated condition. Although there is no change in the wear mechanism for the PECN-treated and untreated Ti6Al4V samples under lubricated condition, the wear loss of the former is far less than that of the latter. The high hardness, adhesion and fracture toughness of the PECN-formed Ti(C0.7N0.3) film should be responsible for the improved wear behavior. The porous surface morphology of the Ti(C0.7N0.3) film is also beneficial to
reduce the friction and wear for the following reasons. On the one hand, it can entrap the grease oil, promote the formation of an even and continuous lubricant film and block the penetration of the contact tip, thus reducing the friction force, friction coefficient and wear rate. On the other hand, it can lodge the wear particles and debris so as to alleviate the abrasive wear. It has been reported that the porous Al2O3 ceramic layer was beneficial to reduce the friction coefficient and improve wear resistance under lubricated condition [39,40]. Consequently, the firmly adhesive, porous and nanocrystalline Ti(C0.7N0.3) film produced in this study is expected to have significant applications under lubricated condition.
6356
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
Fig. 11. EDX analyses of wear tracks for the samples under different conditions of wear test at a load of 200 N. (a) Untreated Ti6Al4V sample subjected to a sliding distance of 452 m under oil-lubricated condition; (b and c) PECN-treated sample subjected to a sliding distance of 1507 m under oil-lubricated condition, and (b) is the EDX spectrum of the gray packing in Fig. 10(d); (d) PECN-treated sample subjected to a sliding distance of 1507 m under 0.9% saline solution lubricated condition.
4. Conclusions Plasma electrolytic carbonitriding can form Ti(C0.7N0.3) thick film on Ti6Al4V substrate. The film is about 15 mm thick, exhibiting nanocrystalline characterization and porous morphology. The friction coefficients and wear volume loss of the Ti(C0.7N0.3) coated samples sliding against a steel counterpart are much less than those of the untreated Ti6Al4V. The Ti(C0.7N0.3) film possesses good wear-resistance and antifriction under oillubricated condition due to its high hardness, adhesion, fracture toughness, and porous surface morphology. Such Ti(C0.7N0.3) film is expected to be a promising candidate as a protective coating of Ti6Al4V alloy in order to extend wear life. Acknowledgments We appreciate the National Natural Science Foundation of China (Grant number 50671078), and National High Technology Research and Development Program of China (Grant number 2006AA03Z0447) for financially supporting this work.
References [1] M. Tamaki, H. Kuwahara, Y. Tomli, N. Yamamoto, J. Mater. Synth. Process. 6 (3) (1998) 215–219. [2] H.J. Klkman, Surf. Coat. Tech. 72 (1995) 30–36. [3] K.T. Rie, T. Stucky, R.A. Silva, E. Leitao, Surf. Coat. Tech. 74–75 (1995) 973–980. [4] Donghau Kuo, Kwonwen Huang, Surf. Coat. Tech. 135 (2001) 150–157. [5] Branko Skoric, Damir Kakas, Natasa Bibic, Milan Rakita, Surf. Sci. 566–568 (2004) 40–44. [6] F. Torregrosa, L. Barralier, L. Roux, Thin Solid Films 266 (1995) 245. [7] V. Fouquet, L. Pichon, A. Straboni, M. Drouet, Surf. Coat. Tech. 186 (2004) 34–39. [8] K. Matsuura, M. Kudoh, Acta Mater. 50 (2002) 2693–2700. [9] Taek-Soo. Kim, Yong-Gwon. Park, Myeong-Yong. Wey, Mater. Sci. Eng. A 361 (2003) 275–280. [10] V. Teixeira, Thin Solid Films 392 (2001) 276–281. [11] P. Kaestner, J. Olfe, J.-W. He, K.-T. Rie, Surf. Coat. Tech. 142–144 (2001) 928– 933. [12] M. Tamaki, Y. Tomii, N. Yamamoto, Plasmas Ions 3 (2000) 33–39. [13] Y.Y. Guu, J.F. Lin, Surf. Coat. Tech. 85 (1996) 146–155. [14] E. Posti, I. Nieminen, Mater. Manuf. Process 4 (2) (1989) 239–252. [15] A.L. Yerokhin, X. Nie, A. Leyland, A. Matthews, Surf. Coat. Tech. 122 (1999) 73–93. [16] L. Rama Krishna, K.R.C. Somaraju, G. Sundararajan, Surf. Coat. Tech. 163–164 (2003) 484–490. [17] A.L. Yerokhin, X. Nie, A. Leyland, A. Matthews, Surf. Coat. Tech. 130 (2000) 195– 206. [18] X.M. Li, Y. Han, Electrochem. Commun. 8 (2006) 267–272.
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357 [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]
X.M. Li, Y. Han, Y.S. Li, Surf. Coat. Tech. 201 (2007) 5326–5329. W. Heinke, A. Leyland, A. Matthews, Thin Solid Films 270 (1995) 431–438. S.J. Bull, Tribol. Int. 30 (7) (1997) 491–498. T. Arai, H. Fujita, M. Watanabe, Thin Solid Films 154 (1987) 387–401. V. Bellido-Gonzalez, N. Stefanopoulos, F. Deguilhen, Surf. Coat. Tech. 74–75 (1995) 884–889. X.D. Li, B. Bhushan, J. Mater. Res. 14 (6) (1999) 2328–2337. S. Sundararajan, B. Bhushan, J. Mater. Res. 16 (2) (2001) 437–445. J. Takadoum, B.H. Houmid, Surf. Coat. Tech. 96 (1997) 272–282. A.A. Voevodin, C. Rebholz, J.M. Schneider, P. Stevenson, A. Matthews, Surf. Coat. Tech. 73 (1995) 185–197. E. Harry, A. Rouzaud, P. Juliet, Y. Pauleau, M. Ignat, Surf. Coat. Tech. 116–119 (1999) 172–175. A.A. Voevodin, J.S. Zabinski, Thin Solid Films 370 (2000) 223–231. A.A. Voevodin, J.S. Zabinski, J. Mater. Sci. 33 (2) (1998) 319–327.
6357
[31] P.W. Shum, K.Y. Li, Z.F. Zhou, Y.G. Shen, Surf. Coat. Tech. 185 (2004) 245–253. [32] Sam Zhang, Deen Sun, Yongqing Fu, Hejun Du, Surf. Coat. Tech. 198 (2005) 74–84. [33] G.R. Antis, P. Chanticul, B.R. Lawn, D.B. Marshall, J. Am. Ceram. Soc. 64 (9) (1981) 533–538. [34] T.M. Payne, S.A. Dillich, Surf. Coat. Tech. 35 (1988) 299–308. [35] C.B. Ponton, R.D. Rawlings, Mater. Sci. Tech. 5 (1989) 865–872. [36] Wenran Feng, Dianran Yan, Jining He, Guling Zhang, Guangliang Chen, Weichao Gu, Size Yang, Appl. Surf. Sci. 243 (2005) 204–213. [37] S. Veprek, A. Niederhofer, K. Moto, Surf. Coat. Tech. 133–134 (2000) 152–159. [38] Xuetong Sun, Zhaohua Jiang, Shigang Xin, Zhongping Yao, Thin Solid Films 471 (2005) 194–199. [39] P.A. Dearnley, J. Gummersbach, H. Weiss, A.A. Ogwu, T.J. Davies, Wear 225–229 (1999) 127–134. [40] A.A. Voevodin, A.L. Yerokhin, V.V. Lyubimov, M.S. Donley, J.S. Zabinske, Surf. Coat. Tech. 86–87 (1996) 516–521.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Mechanical properties of Ti(C0.7N0.3) film produced by plasma electrolytic carbonitriding of Ti6Al4V alloy Xin-Mei Li, Yong Han * State-key Laboratory for Mechanical Behavior of Materials, Xi’an Jiaotong University, Xi’an 710049, PR China
A R T I C L E I N F O
A B S T R A C T
Article history: Received 1 October 2007 Received in revised form 16 March 2008 Accepted 28 March 2008 Available online 4 April 2008
Porous nanocrystalline Ti(C0.7N0.3) film on Ti6Al4V substrate was prepared by plasma electrolytic carbonitriding (PECN). The film was characterized and analyzed by using a variety of analytical techniques, such as XRD, SEM, EDX, TEM, FESEM, Rockwell C indenter, scratch tester, Vickers microhardness tester and ring-on-block tribometer. The results showed that the film was about 15 mm thick and its hardness was Hv 2369 at a load of 0.2 N. The adhesion of the film was characterized by Lc and Pc value, and was found to be about 42 N and more than 800 N, respectively. The friction coefficients and wear volume loss of the PECN-treated samples sliding against a steel counterpart were much less than those of the untreated Ti6Al4V. The film possessed a good wear-resistance and antifriction under oillubricated condition due to its high hardness, adhesion and fracture toughness. Also, the porous surface morphology of the Ti(C0.7N0.3) film contributed to the enhanced tribological resistance by promoting the formation of lubricant film and entrapping wear debris. ß 2008 Elsevier B.V. All rights reserved.
PACS: 61.82.Rx 68.35.Gy 81.65. Lp Keywords: Plasma electrolytic carbonitriding Ti(C0.7N0.3) thick film Ti6Al4V Adhesion Wear-resistance
1. Introduction Titanium alloys are attractive materials due to its low density, very good resistance to corrosion and high relative strength. However, their resistance to frictional wear is poor. Surface engineering techniques are widely used to improve the surface hardness and wear resistance of titanium alloys, and they have been realized as promising methods to deposit TiC, TiN and Ti(C,N) hard films on titanium alloys. Up to now, various techniques have been developed to deposit or form titanium carbide/nitride films on titanium alloys, such as physical vapor deposition (PVD) [1,2], chemical vapor deposition (CVD) [3,4], ion beam assisted deposition (IBAD) [5–7], and plasma carbiding/nitriding [8,9]. However, the titanium carbide/nitride films produced by IBAD on Ti6Al4V alloy are too thin (generally less than 3 mm) and the loadbearing capacity is poor [7]. The adhesion of the titanium carbide/ nitride films obtained by PVD and CVD are insufficient owing to the increased internal stresses in the films [10], and thereby delamination of the films has been observed under wear condition [11]. Although plasma carbiding/nitridng treatment of titanium at
* Corresponding author. Fax: +86 29 82663453. E-mail address: [email protected] (Y. Han). 0169-4332/$ – see front matter ß 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2008.03.172
700–900 8C leads to firmly adhered titanium carbide/nitride layer, the formation of the compound layer with the thickness more than 10 mm requires much higher temperature and longer treatment time [8,12], which results in worse fatigue behavior and earlier fracture due to the aging of the substrate material [7]. It has been proved that the improvement efficacy of the wear resistance and load-bearing capacity strongly depend on the thickness and adhesion of the hard films [13,14], thus alternative methods to produce thick and firmly adhered titanium carbide/ nitride films on titanium alloy at low temperature are still worth exploring. It is found that plasma electrolytic oxidation (PEO) can achieve a relatively fast conversion of titanium surface, at near-to-ambient bulk temperature, into a titanium oxide ceramic layer [15–17]. These titanium oxide layers amount to 10 mm in thickness and firmly bond to the titanium substrate. Based on the principle involved in PEO, a novel surface modification technology named as plasma electrolytic carbonitriding (PECN), has been developed for surface modification of pure titanium and thereby formation of thick Ti(CxN1x) films in our previous work [18,19], however, still unclear how about the mechanical properties of the modified layer. In the present work, the mechanical properties of the PECN-formed Ti(CxN1x) film on Ti6Al4V alloy are investigated, especially the adhesion, hardness, and friction and wear behavior.
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
2. Experimental Ti6Al4V plates with sizes of 30 mm 10 mm 1 mm were used as substrates. The plates were polished to a mirror finish with an aqueous silica suspension, and ultrasonically cleaned with acetone and distilled water. For PECN treatment, a pulse power supply was employed, and a Ti6Al4V plate was used as a cathode while a graphite plate was used as an anode in an electrolytic cell. A mixed organic solution of triethanolamine [N(CH2CH2OH)3] and formamide [HCONH2], and other proprietary ingredients (added primarily for adjustment of electrical conductivity) were chosen as an electrolyte. The applied voltage, pulse frequency and duty cycle were fixed at 600 V, 100 Hz and 40%, respectively, and the Ti6Al4V plates were discharge-treated for 2.5 h. The electrolyte bath was water-cooled and its temperature was maintained lower than 30 8C. After the PECN treatment, the obtained samples were washed with distilled water and dried at room temperature. The phase components of the PECN-treated samples were analyzed with X-ray diffraction (XRD) using a Cu Ka radiation. Scanning electron microscopy (SEM) and field emission scanning electron microscopy (FESEM) were employed to observe the morphologies, thickness and grain sizes of the formed film. Rockwell C indenter (conical diamond with 1208 included angle and 0.2 mm tip radius) was used to evaluate the adhesion of the PECN-formed film. In the indentation method, the force which separates a film from a substrate, expressed by Pc value, can characterize the adhesion of a film [13]. A series of loads were applied to the surface of the PECN-treated sample, and the resulting damages of the film around the indentations were examined using SEM. As a comparison, the adhesion of the film was also investigated using a WS-92 automatic scratch tester. A diamond stylus of 200 mm radius with an apex angle of 1208 was drawn over the film at a speed of 6 mm min1 under a continuous progressive normal force. The smallest load at which the film is damaged is designated the critical load, Lc, which gives an indirect indication of the adhesion [20,21]. The scratch tester was equipped with acoustic emission monitoring equipment, which was used as an on-line failure monitor. Critical load thresholds and detailed morphology of the scratch pattern were correlated by SEM and backscattered electron microscopy (BSE) observations. Five scratch tests were performed to obtain mean and standard deviation values for each film. Hardness measurements were performed on the surface of the PECN-treated sample using a Vickers microhardness tester at a load of 0.2 N. The hardness value is an average of 10 measurements. Fracture toughness of the film was roughly estimated from Vickers indentations at a load of 10 N. Friction and wear tests were carried out on a ring-on-block reciprocating sliding tribometer under lubricated condition as follows. The blocks (8 mm diameter) were made of the PECN-treated and untreated Ti6Al4V, respectively, and the rings (40 mm external diameter) were made of middle carbon steel with a hardness of HRC 48–52. The ring was rotated against the block at a speed of 0.42 m s1. Commercial 30# engine oil and 0.9% saline solution were used as lubricant, respectively, and drop rates of them were about 16–20 drops/ min. The friction coefficients were calculated from the moment automatically recorded by the friction and wear tester according to the following relationship:
m¼
M RF
volume loss was determined by measuring the cross-sectional area of the wear scar with a profilometer. Worn surfaces were investigated using SEM coupled with energy-dispersive X-ray spectrometer (EDX). At least three repetitions of each test were carried out. 3. Results and discussion The XRD pattern of the sample PECN discharge-treated for 2.5 h is shown in Fig. 1. The formed film is composed of Ti(C0.7N0.3). As shown in Fig. 2, the Ti(C0.7N0.3) film exhibits a porous surface morphology with a pore size of 5 mm. These pores are well separated and homogeneously distributed over the surface of the film. When observed at high magnification (Fig. 3), the film matrix consists of much fine particles with the size of 40–60 nm, suggesting that the PECN-formed Ti(C0.7N0.3) film is nanocrystallized. From the cross-sectional view (Fig. 4), the thickness of the Ti(C0.7N0.3) film is 15 mm. The surface and cross-sectional morphology and crystallization characteristics of the PECN-formed Ti(C0.7N0.3) film on Ti6Al4V are similar to those on commercial titanium [18,19]. The adhesion of the PECN-formed Ti(C0.7N0.3) film can be roughly evaluated using indentation test according to the damage type of the film [20]. Fig. 5 shows the indentation morphologies of the film subjected to different loads. No cracks and spallations can be observed around the indentation at a load of 800 N; however, the radial and annular cracks appear with increasing the applied
Fig. 1. XRD patterns of the PECN-treated samples for the discharge time of (a) 0 and (b) 2.5 h.
(1)
where m is the friction coefficient, M is the moment, R is the external diameter of the ring, and F is the applied load. The wear
6351
Fig. 2. SEM photograph of the sample PECN discharge-treated for 2.5 h.
6352
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
Fig. 4. Cross-sectional morphology of the sample PECN discharge-treated for 2.5 h. Fig. 3. FESEM photograph of the sample PECN discharge-treated for 2.5 h.
load up to 950 N, indicating that the Pc value of the Ti(C0.7N0.3) film is more than 800 N. For thin films under such a high load, a considerable substrate deformation is inevitable, thus resulting in the failure for most of the hard films. For example, when subjected to indentation test, for a sputtered TiN film (2 mm thick) on hardened M2 steel substrate, flaking off in a large annular area occurred at 600 N load; for an ion plated TiN film (2 mm thick) on hardened M2 steel substrate, crack and flaking in a small area were detected at 600 N, however, annular flaking occurred at 600 N with increasing the TiN film up to 4 mm in thickness [22]. The Pc value generally increases with the augment of the substrate hardness, it can be concluded, therefore, the PECN-formed Ti(C0.7N0.3) film possesses high adhesion.
Scratch test is generally accepted as one of the simple means in assessing adhesion of a film on its substrate [23–25]. In the scratch test, the load at which failure is first observed to occur regularly along the scratch track, correlating with a sudden increase in acoustic emission, is termed as the critical load Lc. The typical acoustic emission signal of the PECN-treated sample in the scratch test is shown in Fig. 6. The Lc value of the Ti(C0.7N0.3) film is about 42 N in average. Fig. 7 shows the typical scratched morphologies for PECN-treated sample on one scratch track made at continuous progressive load. Buckling failure is observed within the scratch track and the Ti(C0.7N0.3) film tends towards spallation in the track only, which is a ductile failure mode according to [20]. As reported by Takadoum and Houmid
Fig. 5. Indentation morphologies of the PECN-treated sample at the loads of (a) 300 N, (b) 500 N, (c) 800 N and (d) 950 N.
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
Fig. 6. Acoustic emission signal of the PECN-treated sample in scratch test.
[26], when an ion plated TiN film on 35CD4 steel substrate subjected to the scratch test, for the film thickness of 1.5 mm, only superficial semicircular crack traces were detected at 2 N load while spalling observed at 5 N; for the film thickness of 3 mm, spalling took place at 3 N; for the film thickness of 5 mm, spalling occurred at 2 N. Arai et al. [22] also reported, when an ion plated TiN film on hardened M2 steel substrate subjected to the scratch test, for the film thickness of 1 mm, only cracks were observed at 1 N while chipping detected at 2 N; for the film thickness of 4 mm, small amounts of flaking appeared at 2 N. However, for a sputtered TiN film (2 mm thick) on hardened M2 steel substrate, flaking in a large area occurred at 2 N. Besides, some researchers directly utilized the critical load to indicate cracking resistance [27,28], or some even termed it ‘‘scratch toughness’’ [29,30]. For example, a magnetron sputtered
6353
Ti1xAlxN nanocomposite thin film (hardness 28.5 GPa, thickness 0.9 mm) on silicon substrate which reached 70 N load without brittle failure was considered having good scratch toughness. In comparison, films of TiN (0.9 mm, Lc = 30–40 N) and AlN (0.6 mm, Lc = 20 N) were not so good in scratch toughness [31,32]. Like the Pc value, the Lc value tends to increase with the augment of the substrate hardness. Consequently, it also can be confirmed that the PECN-formed Ti(C0.7N0.3) film possesses high adhesion, toughness and load-bearing capacity. The microhardness of the sample PECN discharge-treated for 2.5 h is Hv2369, over 5 times of that of the Ti6Al4V substrate (Hv 420). As shown in Fig. 8(a), the Vickers microhardness indentation is quite small in size, due to the rather low load. However, the diagonal of the indentation is distinct. In the case of brittle materials and coatings, the technique of indentation fracture permits to estimate the fracture toughness from the length of the cracks induced around the indentation [33–35]. To roughly estimate the fracture toughness of the PECN-formed Ti(C0.7N0.3) film, a load as high as 10 N is applied to the film, and no cracks are observed around the indentation impression (Fig. 8(b)). As pointed out by Feng et al. [36] and Veprek et al. [37], TiN coatings and TiN/a-Si3N4 coatings were thought to be of high fracture toughness on the basis of the fact that no radial cracks appeared around the indentation impression at loads of 0.25 and 1 N, respectively. Compared with the aforementioned results, it can be expected that the PECN-formed Ti(C0.7N0.3) film possesses a considerable fracture toughness. In addition, the impression region of the Ti(C0.7N0.3) film becomes dense (Fig. 8(b)), exhibiting a significant plastic deformation. The considerable plasticity and ductility of the PECN-formed film is thought to be due to the fine grain size. Moreover, the porous surface morphology is advantageous for reducing brittle failures and relaxing stress [38]. The friction and wear behavior of the PECN-treated and untreated Ti6Al4V samples sliding against a steel counterpart are shown in Fig. 9. The friction coefficients of PECN-treated sample under oil-lubricated condition are significantly reduced
Fig. 7. SEM (a and c) and BSE (b and d) photographs of scratched morphologies for the PECN-treated sample. The point where the failure is first observed is denoted as ‘‘F’’, and (c and d) are its part magnifications, respectively. Direction of scratch is shown by arrows.
6354
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
Fig. 8. Vickers indentation views of the PECN-formed film at the loads of (a) 0.2 N and (b) 10 N.
to 0.10–0.11, about four times of magnitude less than in the case of the untreated Ti6Al4V. Furthermore, it tends to decrease with increasing the sliding distance. Compared with the untreated Ti6Al4V sample, the PECN-treated sample shows a pronounced improvement in wear-resistance. For example, the wear volume loss of the PECN-treated sample subjected to a sliding distance of 1507 m under oil-lubricated condition is 0.588 102 mm3, about 1/3000 of that of the untreated Ti6Al4V sample subjected to 452 m. For the untreated Ti6Al4V sample under 0.9% saline solution lubricated condition, a sliding distance above 50 m causes severe wear and wide-range oscillation of the friction coefficients. For the PECN-treated sample under 0.9% saline solution lubricated condition, the friction coefficients increase gradually, however, its increment rates are far slower than those of the untreated Ti6Al4V sample whether under oil-lubricated or 0.9% saline solution lubricated condition. Accordingly, the resulting wear volume loss of the PECN-treated sample under 0.9% saline solution lubricated condition is also much less than that of the untreated Ti6Al4V sample. In general, PECN-formed Ti(C0.7N0.3) film possesses good wear-resistance and antifriction, especially under oil-lubricated condition. Fig. 10 presents the wear morphologies of the PECN-treated and untreated Ti6Al4V samples. The wear track of the untreated Ti6Al4V sample is wide and deep, full of microcutting and deep grooves (Fig. 10(a) and (b)). In addition, Fe is not detected on the worn surface (Fig. 11(a)), indicating no metal transfer from the steel counterpart onto the untreated Ti6Al4V during the wear test. It is evident that the untreated Ti6Al4V sample experienced severe abrasive wear. Since the untreated Ti6Al4V is softer than the steel counterpart, metal on the contact tip can act as a cutting/machining edge on the untreated Ti6Al4V sample, thus resulting in the deep grooves, gouging and rapid failure of the sample surface. For the PECN-treated sample, however, the wear behavior is quite different. Almost no measurable wear can be detected for the PECN-treated sample under oil-lubricated condition and the only observable feature is a polishing of the sample surface. The wear track is narrow and shallow, and the worn surface is smooth, without any plough and detachment (Fig. 10(c) and (d)), suggesting that the PECNtreated sample only suffers mild abrasive wear under oillubricated condition. Compared with the initial porous surface morphology (Fig. 2), the porosity of the worn surface is decreased and some debris are embedded in pores (Fig. 10(d)). EDX reveals that these debris are mainly composed of iron chips (Fig. 11(b)), exhibiting a metal transfer from the steel counterpart onto the Ti(C0.7N0.3) film during the wear test.
It also can be confirmed by the EDX spectrum shown in Fig. 11(c). Despite appearance of fine and shallow scratches parallel to relative movement direction, the wear track of the PECN-treated sample under 0.9% saline solution lubricated
Fig. 9. Variations of (a) friction coefficient m and (b) wear volume loss of the samples under condition I, II, III and IV at a load of 200 N. I, untreated Ti6Al4V sample subjected to a sliding distance of 452 m under oil-lubricated condition; II, PECNtreated sample subjected to a sliding distance of 1507 m under oil-lubricated condition; III, untreated Ti6Al4V sample subjected to a sliding distance of 50 m under 0.9% saline solution lubricated condition; IV, PECN-treated sample subjected to a sliding distance of 1507 m under 0.9% saline solution lubricated condition.
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
6355
Fig. 10. SEM photographs of wear tracks for the samples under different conditions of wear test at a load of 200 N. (a and b) Untreated Ti6Al4V sample subjected to a sliding distance of 452 m under oil-lubricated condition; (c and d) PECN-treated sample subjected to a sliding distance of 1507 m under oil-lubricated condition; (e and f) PECNtreated sample subjected to a sliding distance of 1507m under 0.9% saline solution lubricated condition. (b), (d) and (f) are the part magnifications of (a), (c) and (e), respectively.
condition is still narrow and shallow, and the wear surface is smooth (Fig. 10(e) and (f)). Besides, the metal transfer from the steel counterpart onto the Ti(C0.7N0.3) film and the oxidation of the film occur (Fig. 11(d)). Abrasive wear is still predominant for the PECN-treated sample under 0.9% saline solution lubricated condition. Although there is no change in the wear mechanism for the PECN-treated and untreated Ti6Al4V samples under lubricated condition, the wear loss of the former is far less than that of the latter. The high hardness, adhesion and fracture toughness of the PECN-formed Ti(C0.7N0.3) film should be responsible for the improved wear behavior. The porous surface morphology of the Ti(C0.7N0.3) film is also beneficial to
reduce the friction and wear for the following reasons. On the one hand, it can entrap the grease oil, promote the formation of an even and continuous lubricant film and block the penetration of the contact tip, thus reducing the friction force, friction coefficient and wear rate. On the other hand, it can lodge the wear particles and debris so as to alleviate the abrasive wear. It has been reported that the porous Al2O3 ceramic layer was beneficial to reduce the friction coefficient and improve wear resistance under lubricated condition [39,40]. Consequently, the firmly adhesive, porous and nanocrystalline Ti(C0.7N0.3) film produced in this study is expected to have significant applications under lubricated condition.
6356
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357
Fig. 11. EDX analyses of wear tracks for the samples under different conditions of wear test at a load of 200 N. (a) Untreated Ti6Al4V sample subjected to a sliding distance of 452 m under oil-lubricated condition; (b and c) PECN-treated sample subjected to a sliding distance of 1507 m under oil-lubricated condition, and (b) is the EDX spectrum of the gray packing in Fig. 10(d); (d) PECN-treated sample subjected to a sliding distance of 1507 m under 0.9% saline solution lubricated condition.
4. Conclusions Plasma electrolytic carbonitriding can form Ti(C0.7N0.3) thick film on Ti6Al4V substrate. The film is about 15 mm thick, exhibiting nanocrystalline characterization and porous morphology. The friction coefficients and wear volume loss of the Ti(C0.7N0.3) coated samples sliding against a steel counterpart are much less than those of the untreated Ti6Al4V. The Ti(C0.7N0.3) film possesses good wear-resistance and antifriction under oillubricated condition due to its high hardness, adhesion, fracture toughness, and porous surface morphology. Such Ti(C0.7N0.3) film is expected to be a promising candidate as a protective coating of Ti6Al4V alloy in order to extend wear life. Acknowledgments We appreciate the National Natural Science Foundation of China (Grant number 50671078), and National High Technology Research and Development Program of China (Grant number 2006AA03Z0447) for financially supporting this work.
References [1] M. Tamaki, H. Kuwahara, Y. Tomli, N. Yamamoto, J. Mater. Synth. Process. 6 (3) (1998) 215–219. [2] H.J. Klkman, Surf. Coat. Tech. 72 (1995) 30–36. [3] K.T. Rie, T. Stucky, R.A. Silva, E. Leitao, Surf. Coat. Tech. 74–75 (1995) 973–980. [4] Donghau Kuo, Kwonwen Huang, Surf. Coat. Tech. 135 (2001) 150–157. [5] Branko Skoric, Damir Kakas, Natasa Bibic, Milan Rakita, Surf. Sci. 566–568 (2004) 40–44. [6] F. Torregrosa, L. Barralier, L. Roux, Thin Solid Films 266 (1995) 245. [7] V. Fouquet, L. Pichon, A. Straboni, M. Drouet, Surf. Coat. Tech. 186 (2004) 34–39. [8] K. Matsuura, M. Kudoh, Acta Mater. 50 (2002) 2693–2700. [9] Taek-Soo. Kim, Yong-Gwon. Park, Myeong-Yong. Wey, Mater. Sci. Eng. A 361 (2003) 275–280. [10] V. Teixeira, Thin Solid Films 392 (2001) 276–281. [11] P. Kaestner, J. Olfe, J.-W. He, K.-T. Rie, Surf. Coat. Tech. 142–144 (2001) 928– 933. [12] M. Tamaki, Y. Tomii, N. Yamamoto, Plasmas Ions 3 (2000) 33–39. [13] Y.Y. Guu, J.F. Lin, Surf. Coat. Tech. 85 (1996) 146–155. [14] E. Posti, I. Nieminen, Mater. Manuf. Process 4 (2) (1989) 239–252. [15] A.L. Yerokhin, X. Nie, A. Leyland, A. Matthews, Surf. Coat. Tech. 122 (1999) 73–93. [16] L. Rama Krishna, K.R.C. Somaraju, G. Sundararajan, Surf. Coat. Tech. 163–164 (2003) 484–490. [17] A.L. Yerokhin, X. Nie, A. Leyland, A. Matthews, Surf. Coat. Tech. 130 (2000) 195– 206. [18] X.M. Li, Y. Han, Electrochem. Commun. 8 (2006) 267–272.
X.-M. Li, Y. Han / Applied Surface Science 254 (2008) 6350–6357 [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]
X.M. Li, Y. Han, Y.S. Li, Surf. Coat. Tech. 201 (2007) 5326–5329. W. Heinke, A. Leyland, A. Matthews, Thin Solid Films 270 (1995) 431–438. S.J. Bull, Tribol. Int. 30 (7) (1997) 491–498. T. Arai, H. Fujita, M. Watanabe, Thin Solid Films 154 (1987) 387–401. V. Bellido-Gonzalez, N. Stefanopoulos, F. Deguilhen, Surf. Coat. Tech. 74–75 (1995) 884–889. X.D. Li, B. Bhushan, J. Mater. Res. 14 (6) (1999) 2328–2337. S. Sundararajan, B. Bhushan, J. Mater. Res. 16 (2) (2001) 437–445. J. Takadoum, B.H. Houmid, Surf. Coat. Tech. 96 (1997) 272–282. A.A. Voevodin, C. Rebholz, J.M. Schneider, P. Stevenson, A. Matthews, Surf. Coat. Tech. 73 (1995) 185–197. E. Harry, A. Rouzaud, P. Juliet, Y. Pauleau, M. Ignat, Surf. Coat. Tech. 116–119 (1999) 172–175. A.A. Voevodin, J.S. Zabinski, Thin Solid Films 370 (2000) 223–231. A.A. Voevodin, J.S. Zabinski, J. Mater. Sci. 33 (2) (1998) 319–327.
6357
[31] P.W. Shum, K.Y. Li, Z.F. Zhou, Y.G. Shen, Surf. Coat. Tech. 185 (2004) 245–253. [32] Sam Zhang, Deen Sun, Yongqing Fu, Hejun Du, Surf. Coat. Tech. 198 (2005) 74–84. [33] G.R. Antis, P. Chanticul, B.R. Lawn, D.B. Marshall, J. Am. Ceram. Soc. 64 (9) (1981) 533–538. [34] T.M. Payne, S.A. Dillich, Surf. Coat. Tech. 35 (1988) 299–308. [35] C.B. Ponton, R.D. Rawlings, Mater. Sci. Tech. 5 (1989) 865–872. [36] Wenran Feng, Dianran Yan, Jining He, Guling Zhang, Guangliang Chen, Weichao Gu, Size Yang, Appl. Surf. Sci. 243 (2005) 204–213. [37] S. Veprek, A. Niederhofer, K. Moto, Surf. Coat. Tech. 133–134 (2000) 152–159. [38] Xuetong Sun, Zhaohua Jiang, Shigang Xin, Zhongping Yao, Thin Solid Films 471 (2005) 194–199. [39] P.A. Dearnley, J. Gummersbach, H. Weiss, A.A. Ogwu, T.J. Davies, Wear 225–229 (1999) 127–134. [40] A.A. Voevodin, A.L. Yerokhin, V.V. Lyubimov, M.S. Donley, J.S. Zabinske, Surf. Coat. Tech. 86–87 (1996) 516–521.