- Email: [email protected]
Microstructures and properties of titanium nitride films prepared by pulsed laser deposition at different substrate temperatures
Applied Surface Science 357 (2015) 473–478
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Microstructures and properties of titanium nitride films prepared by pulsed laser deposition at different substrate temperatures Hongjian Guo a,b,c , Wenyuan Chen a,b , Yu Shan a , Wenzhen Wang a , Zhenyu Zhang c , Junhong Jia a,∗ a
State Key Laboratory of Solid Lubrication, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, China Graduate University of Chinese Academy of Sciences, Beijing 100049, China c Lanzhou Institute of Technology, Lanzhou 730000, China b
a r t i c l e
i n f o
Article history: Received 16 April 2015 Received in revised form 1 September 2015 Accepted 6 September 2015 Available online 8 September 2015 Keywords: Titanium nitride film Pulsed laser deposition (PLD) Substrate temperature Microstructure Tribology property
a b s t r a c t The nanostructured titanium nitride (TiN) films were fabricated by pulsed laser deposition (PLD) technique at different substrate temperatures under residual vacuum, and the influence of substrate temperatures on the microstructure, mechanical and tribological properties of TiN films was investigated and discussed. The results shown that the consistent stoichiometric TiN films were obtained and the grain size increased from 10.5 to 38.7 nm with the increasing of substrate temperature. The hardness of films decreased with the substrate temperatures increasing, the highest hardness reached to 30.6 GPa at the substrate temperature of 25 ◦ C, and the critical load increased first and decreased at 500 ◦ C, the highest critical load was 23.8 N at the substrate temperature of 300 ◦ C. The film deposited at the substrate temperature of 25 ◦ C registered the lowest friction coefficient of 0.088 and wear rate of 7.8 × 10−7 mm3 /(N m). The excellent tribological performance of the films was attributed to the small grain size, high hardness and smooth surface of the film. © 2015 Elsevier B.V. All rights reserved.
1. Introduction As a transitional metal nitride, titanium nitride (TiN) has been exploited systematically in applications for hard and protective films due to its excellent properties such as outstanding hardness, high strength and rigidity, good stability at high temperatures and excellent wear resistance [1,2]. In the past few decades, many researchers have studied the TiN films by a variety of film deposition techniques, including the physical vapor deposition (PVD) [3], chemical vapor deposition (CVD) [4,5], magnetron sputtering (MS) [6–10], ion implantation [11,12], thermal spraying [13,14] and so on. Compared with above methods, the pulsed laser deposition (PLD) technique [15–17] has proven to be rapid, efficient, easy operation and cost-effective in the fabrication of high-quality thin films. It is a kind of physical vapor deposition technique which has controllable deposition parameters, including laser energy, repetition rate, substrate temperature, deposition time, gas flux as well as target–substrate distance. Growth conditions such as deposition atmosphere [18,19], repetition rate [20] and substrate temperature [21–23] can hugely modify the crystalline quality, microstructure,
∗ Corresponding author. E-mail address: [email protected] (J. Jia). http://dx.doi.org/10.1016/j.apsusc.2015.09.061 0169-4332/© 2015 Elsevier B.V. All rights reserved.
thickness, surface morphology and mechanical properties of films, which have been investigated in great detail. In previous study, Lippert [21] used the PLD technique to fabricate the yttria-stabilized zirconia (YSZ) films at room and high temperatures, which exhibited a uniform isotropic structure in the case of room temperature deposition and an oriented columnar growth at substantially higher substrate temperatures of 400–700 ◦ C. Matei [23] fabricated the vanadium nitride films at room temperature and 500 ◦ C by the PLD technique. They found that the substrate temperatures greatly influenced the thickness of the films which was higher when the films were obtained at 500 ◦ C (30 nm) compared to room temperature (17 nm), and the surface roughness value increased from 0.37 to 0.46 nm with the raising of substrate temperature. The substrate temperature can effectively and directly enhance the ad-atom mobility through the temperature-dependent thermal vibration [24,25], hugely modify the crystalline quality, microstructure and surface morphology evolution of the thin films, fundamentally influence the mechanical properties of films. However, to the best of our knowledge, the investigations on the growth conditions, such as substrate temperatures, of TiN films deposited by PLD technique are really rare. Moreover, most of the TiN films were deposited under N2 or Ar + N2 atmosphere using Ti target to obtain the chemical-reacted TiN phases in literatures, while in present work, the standard stoichiometric TiN films were
474
H. Guo et al. / Applied Surface Science 357 (2015) 473–478
deposited by PLD at different substrate temperatures under residual vacuum using TiN target directly, and we reported the first time for the effect of substrate temperatures on the microstructures and properties of TiN films scientifically. It is expected that the consistent stoichiometric nanostructured TiN films with satisfied mechanical and tribological properties could be fabricated by controlling the depositing substrate temperatures and the potential application of TiN films as anti-wear and protective coating. 2. Experimental details 2.1. Films preparation The films were deposited by PLD technique using a KrF excimer laser ( = 248 nm, pulse duration = 25 ns, COMPexPro 205) to ablate the TiN target with the dimension of 60 mm × 5 mm and 99.99% purity. The deposition chamber was evacuated by a mechanical pump and the molecular pumping to 6.0 × 10−5 Pa prior to ablating the target. The deposition parameters were as follows: laser energy = 400 mJ, repetition rate = 10 Hz, 36,000 pulses for each sample. The films were deposited on silicon wafer at nominal substrate temperatures of 25 ◦ C, 100 ◦ C, 300 ◦ C, 500 ◦ C, and then the samples were denoted as TiN1, TiN2, TiN3, TiN4, respectively. The silicon wafer was ultrasonically cleaned with acetone and alcohol and deionized water for 20 min, dried with pure nitrogen before placed to the substrate holder. The target was placed parallel to substrate at a distance of approximately 50 mm. Both the target and the substrate were cleaned with argon plasma for 30 min. The target rotated at a speed of 15 rpm during the laser ablating in order to avoid the formation of deep craters, and the substrate rotated at a speed of 10 rpm to obtain uniform films. 2.2. Films characterization Crystalline quality, mass density and grain size of the films were identified by means of X-ray diffraction (X’Pert-MRD, Philips, Cu K␣ radiation, = 0.154056 nm) technique at a potential of 40 kV and current of 60 mA, the scanning range of 2 was from 25◦ to 80◦ , at the grazing incidence angle of 1–5◦ differently. The data were analyzed with Jade 6.0 software and peaks were identified by comparing with standard ICSD patterns (89/54378) data files. The crystallite size of the films was calculated by the Scherrer equation and Williamson–Hall plot method [18]. The surface morphologies of the films were characterized by atom force microscope (AFM, AIST-NT Smart SPM, USA) with a conventional rectangular cantilever (tip curvature ≤10 nm). The thickness of the films was determined by ellipsometer (L116E, Gaertner, USA), which was equipped with a He–Ne laser (632.8 nm) set at an incident angle of 50◦ . The refraction coefficient of silicon was set as 3.85, extinction coefficient was 0.02i. Ten points along the diameter of the substrates were taken, and the average value of film thickness was reported. Micro-hardness and elastic modulus of the films were measured using in situ nanomechanical testing system (TI950, Hysitron TriboIndenter, USA) with a cube-corner diamond tip and set to run five indents on each sample. In order to exclude the influence of substrates, the nanoindentation experiments were performed in displacement control with a contact depth up to 100 nm. Hardness and elastic modulus were determined from the load displacement data following the model of Oliver and Pharr [26]. The adhesion strength was tested by a conventional scratch method using a Rockwell penetrator (diameter = 200 m). Scratch test was driven across the films at a linear increase of the load from 0.03 N up to 50 N in 1 min and the scratch length was 4 mm. At least five scratches were done for each sample and adhesion strengths
Fig. 1. GIXRD patterns of the TiN films deposited at different substrate temperatures.
were obtained by averaging the five different scratch tests. The force correspond to the delamination of the film and was referred to as critical load (Lc) and was determined by correlation of three methods of observations: changes in the friction coefficient as a function of scratch length, changes in the penetration curves as a function of scratch length and microscopic observations [22]. The friction and wear performances of the films were conducted by a ball-on-disk tribometer (UMT-3, Bruker). The disk was the TiN film, and the pair part was Al2 O3 ceramic ball (3 mm in diameter). The tests were run at the normal load of 2 N, the sliding velocity of 0.02 m/s, the sliding distance of 50 m at room temperature in atmosphere with a relative humidity of 40 ± 5%. The wear volume loss was evaluated by a NanoMap 500LS three-dimensional (3D) surface profiler with a stylus tip in tapping mode. The wear rates (KW ) were calculated using the equation of KW = VW /(P × L), where VW is the wear volume loss in mm3 , P is the normal load applied in N, and L is the sliding distance in meter (m). 3. Results and discussion 3.1. GIXRD analysis Fig. 1 presents the grazing incidence X-ray diffraction (GIXRD) patterns of TiN films deposited at different substrate temperatures, indicating that the consistent stoichiometric titanium nitride films were obtained. It can be seen that the TiN films are highly oriented in [2 0 0] orientation only when deposited at a substrate temperature of 25 ◦ C. As the substrate temperature increasing, the additional randomly oriented grains could be observed, the [1 1 1], [2 2 0] peaks appears at the substrate temperature of 100 ◦ C, and the [3 1 1], [2 2 2] appears at the substrate temperature above 300 ◦ C. However, the preferential [2 0 0] peak remains the strongest line, indicating it retains the largest volume fraction. The preferred orientation observed in the films as a function of the deposition temperatures can be explained by considering the competition between surface energy and epitaxy in accordance with Krzanowski’s work [16], where the TiN films shown a strong [2 0 0] orientation at all temperatures with additional grain orientations present at 400 and 600 ◦ C. This result is also similar to the TiN film deposited at the temperature of 700 ◦ C, where the strongest TiN [2 0 0] peak is clearly observed [17]. Owing to TiN has a B1 structure with low energy [1 0 0] planes, surface energy effects dominate leading to [2 0 0] oriented films at low temperatures. At higher temperatures, with higher atomic mobility, there is more of a tendency
H. Guo et al. / Applied Surface Science 357 (2015) 473–478
475
Table 1 Structural parameters of the TiN films deposited at different substrate temperatures. Samples
Lattice parameter (nm)
Crystallite size (nm)
Micro-strain (%)
Density (g/cm3 )
Surface roughness (RMS) (nm)
TiN1 TiN2 TiN3 TiN4
0.422 0.422 0.422 0.424
10.5 18.6 22.8 38.7
0.586 0.305 0.332 0.413
5.395 5.378 5.275 4.814
0.4 0.6 0.7 2.9
for epitaxy leading to the nucleation of [1 1 1] grains. Other orientations observed, such as [2 2 0] and [3 1 1] may be due to a transition occurring from [1 1 1] to [2 0 0] during growth. However, surface energy effects ultimately dominate, leading to a primary [2 0 0] orientation at all temperatures [13,16]. These changes in texture of the TiN thin films are due to one or combination of factors such as strain energy, surface free energy, surface diffusion and ad-atom mobility, the influence of each factor depends on the processing conditions [8]. Concerning these changes in peaks intensity and width, a more careful analysis were conducted based on Fig. 1, the calculated Lattice parameter, crystallite size, micro-strain and density are tabulated in Table 1. There was no significant change in the lattice parameters of the deposited films at different substrate temperatures. While the crystallite sizes of the films was increased from 10.5 to 38.7 nm with the increasing of substrate temperatures, the density was slightly changed from around 4.8 to 5.3 g/cm3 . Furthermore, in order to investigate crystal orientation variation with depth of the films, the TiN4 film was identified by GIXRD at angles of 1◦ , 3◦ , 5◦ , respectively. The results shown that the diffraction patterns had no obvious changes for the TiN4 film, but the diffraction peak of substrate silicon changed apparently from 1◦ to 5◦ , dominating a uniform crystalline structures in the TiN films. 3.2. Microstructure and surface morphology analysis Fig. 2 shows the thickness and deposition rate of the TiN films deposited at different substrate temperatures. It can be found that the film thickness and the deposition rate increased with the substrate temperatures. The growth rate and the crystallization were enhanced at higher deposition temperatures compared to room temperature have been confirmed by several experiments and theoretical studies [23–25] due to that the PLD films growth depended on the mobility of ad-atoms and their diffusion. At the low substrate temperatures, the existing islands on the substrate surface could hardly capture the long-range ad-atoms, and a number of small-size islands appeared at the early stage of deposition. When the temperature was elevated, which could promote the ad-atom’s
mobility leading to its accelerated motion along the random directions, the diffusion rate was enhanced and the migrating ad-atoms had a probability to meet and aggregate together, and results in enhancing the crystallization at higher substrate temperatures. In our case, the mobility of ad-atoms on the pre-deposited surface was generally enhanced at high substrate temperatures, crystallite size or grain growth was increased, and resulted in film thickness increasing, which was in agreement with the literature [23] that the films obtained at 500 ◦ C was thicker than the films deposited at room temperature. The surface morphologies of the deposited films were examined by AFM. As shown in Fig. 3, the deposited films were very flat and smooth, especially the RMS value of TiN1 films was merely 0.4 nm (calculated in area of 2 m × 2 m), the RMS value of TiN4 films was 2.9 nm, much rougher than other films (Table 1). These films were similar with the TiN films deposited under residual vacuum and an atmosphere of 10−2 –10 Pa N2 by PLD technique, where the surface roughness was 0.6–0.9 nm [27] and 0.8–1.0 nm [18], respectively. There are nearly no large particles on the surface of the films even in area of 20 m × 20 m, but some giant-sized particles of up to 3 m in diameter were also found clearly on the pulsed laser deposited TiN films [15]. The roughness varied with the increasing of substrate temperatures might be due to the enhanced crystallization at higher substrate temperatures [25]. The substrate temperature can directly and effectively enhance the ad-atoms’ mobility through the temperature-dependent thermal vibration [24], which was contributed to nucleation, crystal, diffusion, and growth of films. At low temperatures, thin film grows dispersedly due to the worse ad-atom mobility, the existing islands could hardly capture the long-range ad-atoms, and a number of small-size islands formed on the deposited films. It is obvious that the island size in this case is much smaller than others. When the substrate temperature increased, the motion of ad-atoms would be enhanced. That is to say, the higher temperature can promote the ad-atom’s mobility, which leads to their accelerated motion along the random directions on the surface, the aggregation of neighboring ad-atoms appears so that thin film grows in dendritic mode and the average island size becomes bigger than the lower temperatures, resulting in the decreased density and fluctuation of the island (Fig. 3, TiN4). The result was also similar to Kınacı’s work [28] that the roughness root mean square (RMS) value of the RF magnetron sputtered SrTiO3 film increased from 0.295 nm at the temperature of 200 ◦ C to 1.195 nm at 500 ◦ C. It can be noted that the surface of films becomes rougher with the elevation of the substrate temperatures
3.3. Mechanical and tribological properties of films
Fig. 2. Thickness and deposition rate of the TiN films deposited at different substrate temperatures.
Fig. 4 gives the critical load, nanoindentation hardness and elastic modulus of TiN films deposited at different substrate temperatures. It can be clearly seen that the critical load of the TiN films increased first and then decreased at the substrate temperature of 500 ◦ C, in there, the TiN3 film holds the highest critical load of 23.8 N implied the strongest bond to the substrate. The increased critical load of films could be explained by the formation of pseudo-diffusion interfaces between the substrate and film due to the higher substrate temperature [24]. While, at even higher
476
H. Guo et al. / Applied Surface Science 357 (2015) 473–478
Fig. 3. AFM images of the TiN films deposited at different substrate temperatures.
substrate temperature to 500 ◦ C, the film surface becomes coarse and not compact due to that the active ad-atoms escape from the substrate surface and randomly migration would be enhanced. Thus, the TiN4 exhibited the lowest critical load, While it might also be due to the larger grain size and lower density. It must be noted that there was small difference of critical load among the TiN films as the TiN3 film with maximum of 23.8 N and the TiN4 film with minimum of 20.1 N. The hardness of films decreased with the increasing of substrate temperatures, while the hardness (22.5–30.6 GPa) was comparable to that of TiN films (17.3–26.7 GPa) [9], (12.7–27.2 GPa) [10] deposited by DC magnetron sputtering. However, which was lower than that of the TiN film deposited by PLD technique in available literatures (35–40 GPa) [18,27], where the TiN film deposited in some N2 and CH4 atmosphere. Moreover, this might be due to the controlled loads were applied in the previous studies [18,27], and the experiments were performed in a controlled contact depth in our case. Since the films were very smooth uniformly, the surface roughness should not affect the results, but the small grain size coupled with high micro-strain and mass densities might be the most important reasons. There was a wealth of information available that the deformation mechanisms are different at the
Fig. 4. Variation on critical load, hardness and elastic modulus of the TiN films deposited at different substrate temperatures.
different grain sizes of the materials. When the grain size was in the range of 1–20 nm, grain-boundary sliding become important in this regime and had been identified that grain-boundary effects dominated the deformation process [29]. The TiN1 film deposited at low temperature exhibited the smallest grain size of 10.5 nm and highest density of 5.395 g/cm3 , which performed the largest hardness for that when exerted an external load on the TiN1 film, where the sliding occurred at every grain boundary and dislocation density increased due to the most grain boundaries, and could resist the external load validly. Moreover, the dislocation occurred in case which recovered rarely and exhibited the deformation at last, so the elastic modulus of TiN1 film was lowest. In contrast to the above statement, the TiN4 film held the largest grain size and lowest density, so exhibited the lowest hardness of 22.5 GPa and highest elastic modulus value of 297 GPa. Our findings are agreed well with Chaim’s results [30], where the measured elastic moduli decreased with increase in the relative density, and the films also exhibited a sharp decrease of the elastic modulus with decrease in grain size, the measured hardness was found to decrease with increase in grain size. The average friction coefficients of different TiN films paired with alumina ball were shown in Fig. 5. The lowest friction coefficient reached to 0.088 for the TiN1 film, even if the friction coefficient of TiN4 film was less than 0.14 during the test. It was found that the friction coefficients were much lower compared to that of TiNx films ( ∼ 0.3) prepared by PLD in previous work [31], where the wear tests were performed with a Si3 N4 ceramic ball (3 mm in diameter) at a load of 0.98 N under the humid air (relative humidity 50%), and also lower than the DC magnetron sputtered TiN films with friction coefficients of 0.2–0.5, where [9] the wear tests were performed with a WC ball of 4 mm diameter at a load of 2 N under the same environment condition to our course. The friction coefficients of the deposited TiN films were stable during the wear tests, as shown in Fig. 6. The steady friction coefficient of the films can be reasonably attributed to uniform and smooth films obtained by PLD technique. It might also be concluded that every film deposited at different substrate temperatures was greatly uniform. For the wear rate of the TiN films deposited at different substrate temperatures, as shown in Fig. 7, the wear rate increased with
H. Guo et al. / Applied Surface Science 357 (2015) 473–478
477
grain-boundary sliding. When the external load exerted on the film and moved at a constant speed, the film started to deform. The friction force along the film surface could be partially canceled out by the grain-boundary sliding, especially for the TiN1 film owned the smallest grain size and most gain boundaries. Additionally, the surface roughness of the film coupled with high hardness were critical for tribological performances [10], so the lowest roughness and the highest hardness might be another reasons for the TiN1 film which exhibited the lowest wear rate. The results were also in agreement with the theory [9,32] that the ratio of hardness and elastic modulus (H/E) was possible to evaluate the wear behavior of the films, in which the wear resistance of materials as a function of the H/E ratio, the maximum H/E ratio value correspond to the high anti-wear resistance, as shown in Fig. 7. 4. Conclusions Fig. 5. The average friction coefficient of the TiN films deposited at different substrate temperatures.
Fig. 6. The friction vs sliding distance curves of the TiN films deposited at different substrate temperatures.
The homogeneous TiN films were fabricated by PLD technique at different substrate temperatures, and the influence of substrate temperatures on the microstructure, mechanical and tribological properties of TiN films was investigated and discussed. Results from these studies revealed that: (1) The consistent stoichiometric nanostructured TiN films were fabricated at different substrate temperatures and the grain size increased from 10.5 to 38.7 nm with the raising of substrate temperature. (2) The hardness of the films decreased with grain size increasing, but the elastic modulus of films had the inverse relationship. The maximum hardness of the films reached to 30.6 GPa for that deposited at the substrate temperature of 25 ◦ C. (3) The TiN film deposited at the substrate temperature of 25 ◦ C registered the lowest friction coefficient about 0.088 and wear rate of 7.8 × 10−7 mm3 /(N m). The best synthesized mechanical and tribological properties were reliant on the smallest grain size, smoothest surface, and the highest hardness. Acknowledgments The authors gratefully acknowledge the National Natural Science Foundation of China (Grant No. 51175490), National Defense Science and Technology Innovation Foundation of Chinese Academy of Sciences (Grant No. CXJJ-14-M39), and the Project of Gansu Province Longyuan Young Creative Talents (Grant No. 2015GS06462) in China for providing the financial support. References
Fig. 7. Wear rate and H/E ratio of the TiN films deposited at different substrate temperatures.
the raising of substrate temperature. From the wear test results, it can be stated that the TiN1 film exhibited the best wear resistance among the films, and registered the lowest wear rate of 7.8 × 10−7 mm3 /(N m) which was about one-fifth of that of the TiN4 film to 35.5 × 10−7 mm3 /(Nm). The fine grain may be contributed to the low friction coefficient and wear rate of the deposited films. The plastic deformation mechanism was attributed to the dominance of
[1] J.M. Lackner, Industrially-scaled large-area and high-rate tribological coating by pulsed laser deposition, Surf. Coat. Technol. 200 (2005) 1439–1444. [2] C.K. Akkan, A. May, M. Hammadeh, H. Abdul-Khaliq, O.C. Aktas, Matrix shaped pulsed laser deposition: new approach to large area and homogeneous deposition, Appl. Surf. Sci. 302 (2014) 149–152. [3] U.P. LeClair, G.P. Berera, J.S. Moodera, Titanium nitride thin films obtained by a modified physical vapor deposition process, Thin Solid Films 376 (2000) 9–15. [4] H.-E. Cheng, Y.-W. Wen, Correlation between process parameters, microstructure and hardness of titanium nitride films by chemical vapor deposition, Surf. Coat. Technol. 179 (2004) 103–109. [5] B.H. Park, Y. Kim, K.H. Kim, Effect of silicon addition on microstructure and mechanical property of titanium nitride film prepared by plasma-assisted chemical vapor deposition, Thin Solid Films 384 (1999) 210–214. [6] S. Shayestehaminzadeh, T.K. Tryggvason, L. Karlsson, S. Olafsson, J.T. Gudmundsson, The properties of TiN ultra-thin films grown on SiO2 substrate by reactive high power impulse magnetron sputtering under various growth angles, Thin Solid Films 548 (2013) 354–357. [7] D. Martínez-Martínez, C. López-Cartes, A. Fernández, J.C. Sánchez-López, Exploring the benefits of depositing hard TiN thin films by non-reactive magnetron sputtering, Appl. Surf. Sci. 275 (2013) 121–126. [8] V. Chawla, R. Jayaganthan, R. Chandra, Structural characterizations of magnetron sputtered nanocrystalline TiN thin films, Mater. Charact. 59 (2008) 1015–1020.
478
H. Guo et al. / Applied Surface Science 357 (2015) 473–478
[9] S. Kataria, S.K. Srivastava, P. Kumar, G. Srinivas, J. Siju, J. Khan, D.V. Sridhara Rao, H.C. Barshilia, Nanocrystalline TiN coatings with improved toughness deposited by pulsing the nitrogen flow rate, Surf. Coat. Technol. 206 (2012) 4279–4286. [10] D.D. Kumar, N. Kumar, S. Kalaiselvam, S. Dash, R. Jayavel, Micro-tribo-mechanical properties of nanocrystalline TiN thin films for small scale device applications, Tribol. Int. 88 (2015) 25–30. [11] G.J.W.N. Huang, Y. Leng, Y.X. Leng, H. Sun, P. Yang, J.Y. Chen, J. Wang, P.K. Chu, Deformation behavior of titanium nitride film prepared by plasma immersion ion implantation and deposition, Surf. Coat. Technol. 156 (2002) 170–175. ´ M. Novakovic, ´ A. Traverse, K. Zhang, N. Bibic, ´ H. Hofsäss, K.P. Lieb, [12] M. Popovic, Modifications of reactively sputtered titanium nitride films by argon and vanadium ion implantation: microstructural and opto-electric properties, Thin Solid Films 531 (2013) 189–196. [13] V.-A. S¸erban, R.A. Ros¸u, A.I. Bucur, D.R. Pascu, Deposition of titanium nitride layers by electric arc – reactive plasma spraying method, Appl. Surf. Sci. 265 (2013) 245–249. [14] R. Ros¸u, V.-A. S¸erban, A. Bucur, M. Popescu, D. Ut¸u, Caracterisation of titanium nitride layers deposited by reactive plasma spraying, J. Technol. Plast. 36 (2011). [15] P.P.O. Chu Chen, H. Wang, Fabrication of TiN thin film by shadow-masked pulsed laser deposition, Thin Solid Films 382 (2001) 275–279. [16] J.E.K.A.R. Phani, Preferential growth of Ti and TiN films on Si(1 1 1) deposited by pulsed laser deposition, Appl. Surf. Sci. 174 (2001) 132–137. [17] K.S.K. Obata, H. Takai, K. Midorikaw, TiN growth on Si(1 0 0) by pulsed laser deposition using homogenized KrF excimer laser beam, Appl. Surf. Sci. 138–139 (1999) 335–339. [18] D. Craciun, G. Socol, N. Stefan, G. Dorcioman, M. Hanna, C.R. Taylor, E. Lambers, V. Craciun, The effect of deposition atmosphere on the chemical composition of TiN and ZrN thin films grown by pulsed laser deposition, Appl. Surf. Sci. 302 (2014) 124–128. [19] S.N. Grigoriev, V.Y. Fominski, R.I. Romanov, M.A. Volosova, Structural modification and tribological behavior improvement of solid lubricating WSex coatings during pulsed laser deposition in buffer He-Gas, J. Frict. Wear 34 (2013) 262–269. [20] L. Guan, D. Zhang, X. Li, Z. Li, Role of pulse repetition rate in film growth of pulsed laser deposition, Nucl. Instrum. Methods Phys. Res. B: Beam Interact. Mater. Atoms 266 (2008) 57–62.
[21] S. Heiroth, R. Ghisleni, T. Lippert, J. Michler, A. Wokaun, Optical and mechanical properties of amorphous and crystalline yttria-stabilized zirconia thin films prepared by pulsed laser deposition, Acta Mater. 59 (2011) 2330–2340. ´ [22] Ł. Kaczmarek, A. Kopia, K. Kyzioł, W. Szymanski, Ł. Kołodziejczyk, J. ´ J. Kleczewska, Wear resistant carbon coatings deposited at room Gawronski, temperature by pulsed laser deposition method on 7075 aluminum alloy, Vacuum 97 (2013) 20–25. [23] C.M. Ghimbeu, F. Sima, R.V. Ostaci, G. Socol, I.N. Mihailescu, C. Vix-Guterl, Crystalline vanadium nitride ultra-thin films obtained at room temperature by pulsed laser deposition, Surf. Coat. Technol. 211 (2012) 158–162. [24] D. Zhang, L. Guan, Z. Li, G. Pan, X. Tan, L. Li, Simulation of island aggregation influenced by substrate temperature, incidence kinetic energy and intensity in pulsed laser deposition, Appl. Surf. Sci. 253 (2006) 874–880. [25] S.K. Pandey, O.P. Thakur, R. Raman, A. Goyal, A. Gupta, Structural and optical properties of YSZ thin films grown by PLD technique, Appl. Surf. Sci. 257 (2011) 6833–6836. [26] W.C. Oliver, An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments, J. Mater. Res. 7 (1992) 1564–1583. [27] D. Craciun, N. Stefan, G. Socol, G. Dorcioman, E. McCumiskey, M. Hanna, C.R. Taylor, G. Bourne, E. Lambers, K. Siebein, V. Craciun, Very hard TiN thin films grown by pulsed laser deposition, Appl. Surf. Sci. 260 (2012) 2–6. [28] B. Kınacı, N. Akın, I˙ . Kars Durukan, T. Memmedli, S. Özc¸elik, The study on characterizations of SrTiO3 thin films with different growth temperatures, Superlattices Microstruct. 76 (2014) 234–243. [29] M.A. Meyers, A. Mishra, D.J. Benson, Mechanical properties of nanocrystalline materials, Prog. Mater. Sci. 51 (2006) 427–556. [30] R. Chaim, Effect of grain size on elastic modulus and hardness of nanocrystalline ZrO2 -3 wt% Y2 O3 ceramic, J. Mater. Sci. 39 (2004) 3057– 3061. [31] X.H. Zheng, J.P. Tu, R.G. Song, Microstructure and tribological performance of CNx –TiNx composite films prepared by pulsed laser deposition, Mater. Des. 31 (2010) 1716–1719. [32] D. Craciun, G. Socol, D.V. Cristea, M. Stoicanescu, N. Olah, K. Balazs, N. Stefan, E. Lambers, V. Craciun, Mechanical properties of pulsed laser deposited nanocrystalline SiC films, Appl. Surf. Sci. 336 (2015) 391–395.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Microstructures and properties of titanium nitride films prepared by pulsed laser deposition at different substrate temperatures Hongjian Guo a,b,c , Wenyuan Chen a,b , Yu Shan a , Wenzhen Wang a , Zhenyu Zhang c , Junhong Jia a,∗ a
State Key Laboratory of Solid Lubrication, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, China Graduate University of Chinese Academy of Sciences, Beijing 100049, China c Lanzhou Institute of Technology, Lanzhou 730000, China b
a r t i c l e
i n f o
Article history: Received 16 April 2015 Received in revised form 1 September 2015 Accepted 6 September 2015 Available online 8 September 2015 Keywords: Titanium nitride film Pulsed laser deposition (PLD) Substrate temperature Microstructure Tribology property
a b s t r a c t The nanostructured titanium nitride (TiN) films were fabricated by pulsed laser deposition (PLD) technique at different substrate temperatures under residual vacuum, and the influence of substrate temperatures on the microstructure, mechanical and tribological properties of TiN films was investigated and discussed. The results shown that the consistent stoichiometric TiN films were obtained and the grain size increased from 10.5 to 38.7 nm with the increasing of substrate temperature. The hardness of films decreased with the substrate temperatures increasing, the highest hardness reached to 30.6 GPa at the substrate temperature of 25 ◦ C, and the critical load increased first and decreased at 500 ◦ C, the highest critical load was 23.8 N at the substrate temperature of 300 ◦ C. The film deposited at the substrate temperature of 25 ◦ C registered the lowest friction coefficient of 0.088 and wear rate of 7.8 × 10−7 mm3 /(N m). The excellent tribological performance of the films was attributed to the small grain size, high hardness and smooth surface of the film. © 2015 Elsevier B.V. All rights reserved.
1. Introduction As a transitional metal nitride, titanium nitride (TiN) has been exploited systematically in applications for hard and protective films due to its excellent properties such as outstanding hardness, high strength and rigidity, good stability at high temperatures and excellent wear resistance [1,2]. In the past few decades, many researchers have studied the TiN films by a variety of film deposition techniques, including the physical vapor deposition (PVD) [3], chemical vapor deposition (CVD) [4,5], magnetron sputtering (MS) [6–10], ion implantation [11,12], thermal spraying [13,14] and so on. Compared with above methods, the pulsed laser deposition (PLD) technique [15–17] has proven to be rapid, efficient, easy operation and cost-effective in the fabrication of high-quality thin films. It is a kind of physical vapor deposition technique which has controllable deposition parameters, including laser energy, repetition rate, substrate temperature, deposition time, gas flux as well as target–substrate distance. Growth conditions such as deposition atmosphere [18,19], repetition rate [20] and substrate temperature [21–23] can hugely modify the crystalline quality, microstructure,
∗ Corresponding author. E-mail address: [email protected] (J. Jia). http://dx.doi.org/10.1016/j.apsusc.2015.09.061 0169-4332/© 2015 Elsevier B.V. All rights reserved.
thickness, surface morphology and mechanical properties of films, which have been investigated in great detail. In previous study, Lippert [21] used the PLD technique to fabricate the yttria-stabilized zirconia (YSZ) films at room and high temperatures, which exhibited a uniform isotropic structure in the case of room temperature deposition and an oriented columnar growth at substantially higher substrate temperatures of 400–700 ◦ C. Matei [23] fabricated the vanadium nitride films at room temperature and 500 ◦ C by the PLD technique. They found that the substrate temperatures greatly influenced the thickness of the films which was higher when the films were obtained at 500 ◦ C (30 nm) compared to room temperature (17 nm), and the surface roughness value increased from 0.37 to 0.46 nm with the raising of substrate temperature. The substrate temperature can effectively and directly enhance the ad-atom mobility through the temperature-dependent thermal vibration [24,25], hugely modify the crystalline quality, microstructure and surface morphology evolution of the thin films, fundamentally influence the mechanical properties of films. However, to the best of our knowledge, the investigations on the growth conditions, such as substrate temperatures, of TiN films deposited by PLD technique are really rare. Moreover, most of the TiN films were deposited under N2 or Ar + N2 atmosphere using Ti target to obtain the chemical-reacted TiN phases in literatures, while in present work, the standard stoichiometric TiN films were
474
H. Guo et al. / Applied Surface Science 357 (2015) 473–478
deposited by PLD at different substrate temperatures under residual vacuum using TiN target directly, and we reported the first time for the effect of substrate temperatures on the microstructures and properties of TiN films scientifically. It is expected that the consistent stoichiometric nanostructured TiN films with satisfied mechanical and tribological properties could be fabricated by controlling the depositing substrate temperatures and the potential application of TiN films as anti-wear and protective coating. 2. Experimental details 2.1. Films preparation The films were deposited by PLD technique using a KrF excimer laser ( = 248 nm, pulse duration = 25 ns, COMPexPro 205) to ablate the TiN target with the dimension of 60 mm × 5 mm and 99.99% purity. The deposition chamber was evacuated by a mechanical pump and the molecular pumping to 6.0 × 10−5 Pa prior to ablating the target. The deposition parameters were as follows: laser energy = 400 mJ, repetition rate = 10 Hz, 36,000 pulses for each sample. The films were deposited on silicon wafer at nominal substrate temperatures of 25 ◦ C, 100 ◦ C, 300 ◦ C, 500 ◦ C, and then the samples were denoted as TiN1, TiN2, TiN3, TiN4, respectively. The silicon wafer was ultrasonically cleaned with acetone and alcohol and deionized water for 20 min, dried with pure nitrogen before placed to the substrate holder. The target was placed parallel to substrate at a distance of approximately 50 mm. Both the target and the substrate were cleaned with argon plasma for 30 min. The target rotated at a speed of 15 rpm during the laser ablating in order to avoid the formation of deep craters, and the substrate rotated at a speed of 10 rpm to obtain uniform films. 2.2. Films characterization Crystalline quality, mass density and grain size of the films were identified by means of X-ray diffraction (X’Pert-MRD, Philips, Cu K␣ radiation, = 0.154056 nm) technique at a potential of 40 kV and current of 60 mA, the scanning range of 2 was from 25◦ to 80◦ , at the grazing incidence angle of 1–5◦ differently. The data were analyzed with Jade 6.0 software and peaks were identified by comparing with standard ICSD patterns (89/54378) data files. The crystallite size of the films was calculated by the Scherrer equation and Williamson–Hall plot method [18]. The surface morphologies of the films were characterized by atom force microscope (AFM, AIST-NT Smart SPM, USA) with a conventional rectangular cantilever (tip curvature ≤10 nm). The thickness of the films was determined by ellipsometer (L116E, Gaertner, USA), which was equipped with a He–Ne laser (632.8 nm) set at an incident angle of 50◦ . The refraction coefficient of silicon was set as 3.85, extinction coefficient was 0.02i. Ten points along the diameter of the substrates were taken, and the average value of film thickness was reported. Micro-hardness and elastic modulus of the films were measured using in situ nanomechanical testing system (TI950, Hysitron TriboIndenter, USA) with a cube-corner diamond tip and set to run five indents on each sample. In order to exclude the influence of substrates, the nanoindentation experiments were performed in displacement control with a contact depth up to 100 nm. Hardness and elastic modulus were determined from the load displacement data following the model of Oliver and Pharr [26]. The adhesion strength was tested by a conventional scratch method using a Rockwell penetrator (diameter = 200 m). Scratch test was driven across the films at a linear increase of the load from 0.03 N up to 50 N in 1 min and the scratch length was 4 mm. At least five scratches were done for each sample and adhesion strengths
Fig. 1. GIXRD patterns of the TiN films deposited at different substrate temperatures.
were obtained by averaging the five different scratch tests. The force correspond to the delamination of the film and was referred to as critical load (Lc) and was determined by correlation of three methods of observations: changes in the friction coefficient as a function of scratch length, changes in the penetration curves as a function of scratch length and microscopic observations [22]. The friction and wear performances of the films were conducted by a ball-on-disk tribometer (UMT-3, Bruker). The disk was the TiN film, and the pair part was Al2 O3 ceramic ball (3 mm in diameter). The tests were run at the normal load of 2 N, the sliding velocity of 0.02 m/s, the sliding distance of 50 m at room temperature in atmosphere with a relative humidity of 40 ± 5%. The wear volume loss was evaluated by a NanoMap 500LS three-dimensional (3D) surface profiler with a stylus tip in tapping mode. The wear rates (KW ) were calculated using the equation of KW = VW /(P × L), where VW is the wear volume loss in mm3 , P is the normal load applied in N, and L is the sliding distance in meter (m). 3. Results and discussion 3.1. GIXRD analysis Fig. 1 presents the grazing incidence X-ray diffraction (GIXRD) patterns of TiN films deposited at different substrate temperatures, indicating that the consistent stoichiometric titanium nitride films were obtained. It can be seen that the TiN films are highly oriented in [2 0 0] orientation only when deposited at a substrate temperature of 25 ◦ C. As the substrate temperature increasing, the additional randomly oriented grains could be observed, the [1 1 1], [2 2 0] peaks appears at the substrate temperature of 100 ◦ C, and the [3 1 1], [2 2 2] appears at the substrate temperature above 300 ◦ C. However, the preferential [2 0 0] peak remains the strongest line, indicating it retains the largest volume fraction. The preferred orientation observed in the films as a function of the deposition temperatures can be explained by considering the competition between surface energy and epitaxy in accordance with Krzanowski’s work [16], where the TiN films shown a strong [2 0 0] orientation at all temperatures with additional grain orientations present at 400 and 600 ◦ C. This result is also similar to the TiN film deposited at the temperature of 700 ◦ C, where the strongest TiN [2 0 0] peak is clearly observed [17]. Owing to TiN has a B1 structure with low energy [1 0 0] planes, surface energy effects dominate leading to [2 0 0] oriented films at low temperatures. At higher temperatures, with higher atomic mobility, there is more of a tendency
H. Guo et al. / Applied Surface Science 357 (2015) 473–478
475
Table 1 Structural parameters of the TiN films deposited at different substrate temperatures. Samples
Lattice parameter (nm)
Crystallite size (nm)
Micro-strain (%)
Density (g/cm3 )
Surface roughness (RMS) (nm)
TiN1 TiN2 TiN3 TiN4
0.422 0.422 0.422 0.424
10.5 18.6 22.8 38.7
0.586 0.305 0.332 0.413
5.395 5.378 5.275 4.814
0.4 0.6 0.7 2.9
for epitaxy leading to the nucleation of [1 1 1] grains. Other orientations observed, such as [2 2 0] and [3 1 1] may be due to a transition occurring from [1 1 1] to [2 0 0] during growth. However, surface energy effects ultimately dominate, leading to a primary [2 0 0] orientation at all temperatures [13,16]. These changes in texture of the TiN thin films are due to one or combination of factors such as strain energy, surface free energy, surface diffusion and ad-atom mobility, the influence of each factor depends on the processing conditions [8]. Concerning these changes in peaks intensity and width, a more careful analysis were conducted based on Fig. 1, the calculated Lattice parameter, crystallite size, micro-strain and density are tabulated in Table 1. There was no significant change in the lattice parameters of the deposited films at different substrate temperatures. While the crystallite sizes of the films was increased from 10.5 to 38.7 nm with the increasing of substrate temperatures, the density was slightly changed from around 4.8 to 5.3 g/cm3 . Furthermore, in order to investigate crystal orientation variation with depth of the films, the TiN4 film was identified by GIXRD at angles of 1◦ , 3◦ , 5◦ , respectively. The results shown that the diffraction patterns had no obvious changes for the TiN4 film, but the diffraction peak of substrate silicon changed apparently from 1◦ to 5◦ , dominating a uniform crystalline structures in the TiN films. 3.2. Microstructure and surface morphology analysis Fig. 2 shows the thickness and deposition rate of the TiN films deposited at different substrate temperatures. It can be found that the film thickness and the deposition rate increased with the substrate temperatures. The growth rate and the crystallization were enhanced at higher deposition temperatures compared to room temperature have been confirmed by several experiments and theoretical studies [23–25] due to that the PLD films growth depended on the mobility of ad-atoms and their diffusion. At the low substrate temperatures, the existing islands on the substrate surface could hardly capture the long-range ad-atoms, and a number of small-size islands appeared at the early stage of deposition. When the temperature was elevated, which could promote the ad-atom’s
mobility leading to its accelerated motion along the random directions, the diffusion rate was enhanced and the migrating ad-atoms had a probability to meet and aggregate together, and results in enhancing the crystallization at higher substrate temperatures. In our case, the mobility of ad-atoms on the pre-deposited surface was generally enhanced at high substrate temperatures, crystallite size or grain growth was increased, and resulted in film thickness increasing, which was in agreement with the literature [23] that the films obtained at 500 ◦ C was thicker than the films deposited at room temperature. The surface morphologies of the deposited films were examined by AFM. As shown in Fig. 3, the deposited films were very flat and smooth, especially the RMS value of TiN1 films was merely 0.4 nm (calculated in area of 2 m × 2 m), the RMS value of TiN4 films was 2.9 nm, much rougher than other films (Table 1). These films were similar with the TiN films deposited under residual vacuum and an atmosphere of 10−2 –10 Pa N2 by PLD technique, where the surface roughness was 0.6–0.9 nm [27] and 0.8–1.0 nm [18], respectively. There are nearly no large particles on the surface of the films even in area of 20 m × 20 m, but some giant-sized particles of up to 3 m in diameter were also found clearly on the pulsed laser deposited TiN films [15]. The roughness varied with the increasing of substrate temperatures might be due to the enhanced crystallization at higher substrate temperatures [25]. The substrate temperature can directly and effectively enhance the ad-atoms’ mobility through the temperature-dependent thermal vibration [24], which was contributed to nucleation, crystal, diffusion, and growth of films. At low temperatures, thin film grows dispersedly due to the worse ad-atom mobility, the existing islands could hardly capture the long-range ad-atoms, and a number of small-size islands formed on the deposited films. It is obvious that the island size in this case is much smaller than others. When the substrate temperature increased, the motion of ad-atoms would be enhanced. That is to say, the higher temperature can promote the ad-atom’s mobility, which leads to their accelerated motion along the random directions on the surface, the aggregation of neighboring ad-atoms appears so that thin film grows in dendritic mode and the average island size becomes bigger than the lower temperatures, resulting in the decreased density and fluctuation of the island (Fig. 3, TiN4). The result was also similar to Kınacı’s work [28] that the roughness root mean square (RMS) value of the RF magnetron sputtered SrTiO3 film increased from 0.295 nm at the temperature of 200 ◦ C to 1.195 nm at 500 ◦ C. It can be noted that the surface of films becomes rougher with the elevation of the substrate temperatures
3.3. Mechanical and tribological properties of films
Fig. 2. Thickness and deposition rate of the TiN films deposited at different substrate temperatures.
Fig. 4 gives the critical load, nanoindentation hardness and elastic modulus of TiN films deposited at different substrate temperatures. It can be clearly seen that the critical load of the TiN films increased first and then decreased at the substrate temperature of 500 ◦ C, in there, the TiN3 film holds the highest critical load of 23.8 N implied the strongest bond to the substrate. The increased critical load of films could be explained by the formation of pseudo-diffusion interfaces between the substrate and film due to the higher substrate temperature [24]. While, at even higher
476
H. Guo et al. / Applied Surface Science 357 (2015) 473–478
Fig. 3. AFM images of the TiN films deposited at different substrate temperatures.
substrate temperature to 500 ◦ C, the film surface becomes coarse and not compact due to that the active ad-atoms escape from the substrate surface and randomly migration would be enhanced. Thus, the TiN4 exhibited the lowest critical load, While it might also be due to the larger grain size and lower density. It must be noted that there was small difference of critical load among the TiN films as the TiN3 film with maximum of 23.8 N and the TiN4 film with minimum of 20.1 N. The hardness of films decreased with the increasing of substrate temperatures, while the hardness (22.5–30.6 GPa) was comparable to that of TiN films (17.3–26.7 GPa) [9], (12.7–27.2 GPa) [10] deposited by DC magnetron sputtering. However, which was lower than that of the TiN film deposited by PLD technique in available literatures (35–40 GPa) [18,27], where the TiN film deposited in some N2 and CH4 atmosphere. Moreover, this might be due to the controlled loads were applied in the previous studies [18,27], and the experiments were performed in a controlled contact depth in our case. Since the films were very smooth uniformly, the surface roughness should not affect the results, but the small grain size coupled with high micro-strain and mass densities might be the most important reasons. There was a wealth of information available that the deformation mechanisms are different at the
Fig. 4. Variation on critical load, hardness and elastic modulus of the TiN films deposited at different substrate temperatures.
different grain sizes of the materials. When the grain size was in the range of 1–20 nm, grain-boundary sliding become important in this regime and had been identified that grain-boundary effects dominated the deformation process [29]. The TiN1 film deposited at low temperature exhibited the smallest grain size of 10.5 nm and highest density of 5.395 g/cm3 , which performed the largest hardness for that when exerted an external load on the TiN1 film, where the sliding occurred at every grain boundary and dislocation density increased due to the most grain boundaries, and could resist the external load validly. Moreover, the dislocation occurred in case which recovered rarely and exhibited the deformation at last, so the elastic modulus of TiN1 film was lowest. In contrast to the above statement, the TiN4 film held the largest grain size and lowest density, so exhibited the lowest hardness of 22.5 GPa and highest elastic modulus value of 297 GPa. Our findings are agreed well with Chaim’s results [30], where the measured elastic moduli decreased with increase in the relative density, and the films also exhibited a sharp decrease of the elastic modulus with decrease in grain size, the measured hardness was found to decrease with increase in grain size. The average friction coefficients of different TiN films paired with alumina ball were shown in Fig. 5. The lowest friction coefficient reached to 0.088 for the TiN1 film, even if the friction coefficient of TiN4 film was less than 0.14 during the test. It was found that the friction coefficients were much lower compared to that of TiNx films ( ∼ 0.3) prepared by PLD in previous work [31], where the wear tests were performed with a Si3 N4 ceramic ball (3 mm in diameter) at a load of 0.98 N under the humid air (relative humidity 50%), and also lower than the DC magnetron sputtered TiN films with friction coefficients of 0.2–0.5, where [9] the wear tests were performed with a WC ball of 4 mm diameter at a load of 2 N under the same environment condition to our course. The friction coefficients of the deposited TiN films were stable during the wear tests, as shown in Fig. 6. The steady friction coefficient of the films can be reasonably attributed to uniform and smooth films obtained by PLD technique. It might also be concluded that every film deposited at different substrate temperatures was greatly uniform. For the wear rate of the TiN films deposited at different substrate temperatures, as shown in Fig. 7, the wear rate increased with
H. Guo et al. / Applied Surface Science 357 (2015) 473–478
477
grain-boundary sliding. When the external load exerted on the film and moved at a constant speed, the film started to deform. The friction force along the film surface could be partially canceled out by the grain-boundary sliding, especially for the TiN1 film owned the smallest grain size and most gain boundaries. Additionally, the surface roughness of the film coupled with high hardness were critical for tribological performances [10], so the lowest roughness and the highest hardness might be another reasons for the TiN1 film which exhibited the lowest wear rate. The results were also in agreement with the theory [9,32] that the ratio of hardness and elastic modulus (H/E) was possible to evaluate the wear behavior of the films, in which the wear resistance of materials as a function of the H/E ratio, the maximum H/E ratio value correspond to the high anti-wear resistance, as shown in Fig. 7. 4. Conclusions Fig. 5. The average friction coefficient of the TiN films deposited at different substrate temperatures.
Fig. 6. The friction vs sliding distance curves of the TiN films deposited at different substrate temperatures.
The homogeneous TiN films were fabricated by PLD technique at different substrate temperatures, and the influence of substrate temperatures on the microstructure, mechanical and tribological properties of TiN films was investigated and discussed. Results from these studies revealed that: (1) The consistent stoichiometric nanostructured TiN films were fabricated at different substrate temperatures and the grain size increased from 10.5 to 38.7 nm with the raising of substrate temperature. (2) The hardness of the films decreased with grain size increasing, but the elastic modulus of films had the inverse relationship. The maximum hardness of the films reached to 30.6 GPa for that deposited at the substrate temperature of 25 ◦ C. (3) The TiN film deposited at the substrate temperature of 25 ◦ C registered the lowest friction coefficient about 0.088 and wear rate of 7.8 × 10−7 mm3 /(N m). The best synthesized mechanical and tribological properties were reliant on the smallest grain size, smoothest surface, and the highest hardness. Acknowledgments The authors gratefully acknowledge the National Natural Science Foundation of China (Grant No. 51175490), National Defense Science and Technology Innovation Foundation of Chinese Academy of Sciences (Grant No. CXJJ-14-M39), and the Project of Gansu Province Longyuan Young Creative Talents (Grant No. 2015GS06462) in China for providing the financial support. References
Fig. 7. Wear rate and H/E ratio of the TiN films deposited at different substrate temperatures.
the raising of substrate temperature. From the wear test results, it can be stated that the TiN1 film exhibited the best wear resistance among the films, and registered the lowest wear rate of 7.8 × 10−7 mm3 /(N m) which was about one-fifth of that of the TiN4 film to 35.5 × 10−7 mm3 /(Nm). The fine grain may be contributed to the low friction coefficient and wear rate of the deposited films. The plastic deformation mechanism was attributed to the dominance of
[1] J.M. Lackner, Industrially-scaled large-area and high-rate tribological coating by pulsed laser deposition, Surf. Coat. Technol. 200 (2005) 1439–1444. [2] C.K. Akkan, A. May, M. Hammadeh, H. Abdul-Khaliq, O.C. Aktas, Matrix shaped pulsed laser deposition: new approach to large area and homogeneous deposition, Appl. Surf. Sci. 302 (2014) 149–152. [3] U.P. LeClair, G.P. Berera, J.S. Moodera, Titanium nitride thin films obtained by a modified physical vapor deposition process, Thin Solid Films 376 (2000) 9–15. [4] H.-E. Cheng, Y.-W. Wen, Correlation between process parameters, microstructure and hardness of titanium nitride films by chemical vapor deposition, Surf. Coat. Technol. 179 (2004) 103–109. [5] B.H. Park, Y. Kim, K.H. Kim, Effect of silicon addition on microstructure and mechanical property of titanium nitride film prepared by plasma-assisted chemical vapor deposition, Thin Solid Films 384 (1999) 210–214. [6] S. Shayestehaminzadeh, T.K. Tryggvason, L. Karlsson, S. Olafsson, J.T. Gudmundsson, The properties of TiN ultra-thin films grown on SiO2 substrate by reactive high power impulse magnetron sputtering under various growth angles, Thin Solid Films 548 (2013) 354–357. [7] D. Martínez-Martínez, C. López-Cartes, A. Fernández, J.C. Sánchez-López, Exploring the benefits of depositing hard TiN thin films by non-reactive magnetron sputtering, Appl. Surf. Sci. 275 (2013) 121–126. [8] V. Chawla, R. Jayaganthan, R. Chandra, Structural characterizations of magnetron sputtered nanocrystalline TiN thin films, Mater. Charact. 59 (2008) 1015–1020.
478
H. Guo et al. / Applied Surface Science 357 (2015) 473–478
[9] S. Kataria, S.K. Srivastava, P. Kumar, G. Srinivas, J. Siju, J. Khan, D.V. Sridhara Rao, H.C. Barshilia, Nanocrystalline TiN coatings with improved toughness deposited by pulsing the nitrogen flow rate, Surf. Coat. Technol. 206 (2012) 4279–4286. [10] D.D. Kumar, N. Kumar, S. Kalaiselvam, S. Dash, R. Jayavel, Micro-tribo-mechanical properties of nanocrystalline TiN thin films for small scale device applications, Tribol. Int. 88 (2015) 25–30. [11] G.J.W.N. Huang, Y. Leng, Y.X. Leng, H. Sun, P. Yang, J.Y. Chen, J. Wang, P.K. Chu, Deformation behavior of titanium nitride film prepared by plasma immersion ion implantation and deposition, Surf. Coat. Technol. 156 (2002) 170–175. ´ M. Novakovic, ´ A. Traverse, K. Zhang, N. Bibic, ´ H. Hofsäss, K.P. Lieb, [12] M. Popovic, Modifications of reactively sputtered titanium nitride films by argon and vanadium ion implantation: microstructural and opto-electric properties, Thin Solid Films 531 (2013) 189–196. [13] V.-A. S¸erban, R.A. Ros¸u, A.I. Bucur, D.R. Pascu, Deposition of titanium nitride layers by electric arc – reactive plasma spraying method, Appl. Surf. Sci. 265 (2013) 245–249. [14] R. Ros¸u, V.-A. S¸erban, A. Bucur, M. Popescu, D. Ut¸u, Caracterisation of titanium nitride layers deposited by reactive plasma spraying, J. Technol. Plast. 36 (2011). [15] P.P.O. Chu Chen, H. Wang, Fabrication of TiN thin film by shadow-masked pulsed laser deposition, Thin Solid Films 382 (2001) 275–279. [16] J.E.K.A.R. Phani, Preferential growth of Ti and TiN films on Si(1 1 1) deposited by pulsed laser deposition, Appl. Surf. Sci. 174 (2001) 132–137. [17] K.S.K. Obata, H. Takai, K. Midorikaw, TiN growth on Si(1 0 0) by pulsed laser deposition using homogenized KrF excimer laser beam, Appl. Surf. Sci. 138–139 (1999) 335–339. [18] D. Craciun, G. Socol, N. Stefan, G. Dorcioman, M. Hanna, C.R. Taylor, E. Lambers, V. Craciun, The effect of deposition atmosphere on the chemical composition of TiN and ZrN thin films grown by pulsed laser deposition, Appl. Surf. Sci. 302 (2014) 124–128. [19] S.N. Grigoriev, V.Y. Fominski, R.I. Romanov, M.A. Volosova, Structural modification and tribological behavior improvement of solid lubricating WSex coatings during pulsed laser deposition in buffer He-Gas, J. Frict. Wear 34 (2013) 262–269. [20] L. Guan, D. Zhang, X. Li, Z. Li, Role of pulse repetition rate in film growth of pulsed laser deposition, Nucl. Instrum. Methods Phys. Res. B: Beam Interact. Mater. Atoms 266 (2008) 57–62.
[21] S. Heiroth, R. Ghisleni, T. Lippert, J. Michler, A. Wokaun, Optical and mechanical properties of amorphous and crystalline yttria-stabilized zirconia thin films prepared by pulsed laser deposition, Acta Mater. 59 (2011) 2330–2340. ´ [22] Ł. Kaczmarek, A. Kopia, K. Kyzioł, W. Szymanski, Ł. Kołodziejczyk, J. ´ J. Kleczewska, Wear resistant carbon coatings deposited at room Gawronski, temperature by pulsed laser deposition method on 7075 aluminum alloy, Vacuum 97 (2013) 20–25. [23] C.M. Ghimbeu, F. Sima, R.V. Ostaci, G. Socol, I.N. Mihailescu, C. Vix-Guterl, Crystalline vanadium nitride ultra-thin films obtained at room temperature by pulsed laser deposition, Surf. Coat. Technol. 211 (2012) 158–162. [24] D. Zhang, L. Guan, Z. Li, G. Pan, X. Tan, L. Li, Simulation of island aggregation influenced by substrate temperature, incidence kinetic energy and intensity in pulsed laser deposition, Appl. Surf. Sci. 253 (2006) 874–880. [25] S.K. Pandey, O.P. Thakur, R. Raman, A. Goyal, A. Gupta, Structural and optical properties of YSZ thin films grown by PLD technique, Appl. Surf. Sci. 257 (2011) 6833–6836. [26] W.C. Oliver, An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments, J. Mater. Res. 7 (1992) 1564–1583. [27] D. Craciun, N. Stefan, G. Socol, G. Dorcioman, E. McCumiskey, M. Hanna, C.R. Taylor, G. Bourne, E. Lambers, K. Siebein, V. Craciun, Very hard TiN thin films grown by pulsed laser deposition, Appl. Surf. Sci. 260 (2012) 2–6. [28] B. Kınacı, N. Akın, I˙ . Kars Durukan, T. Memmedli, S. Özc¸elik, The study on characterizations of SrTiO3 thin films with different growth temperatures, Superlattices Microstruct. 76 (2014) 234–243. [29] M.A. Meyers, A. Mishra, D.J. Benson, Mechanical properties of nanocrystalline materials, Prog. Mater. Sci. 51 (2006) 427–556. [30] R. Chaim, Effect of grain size on elastic modulus and hardness of nanocrystalline ZrO2 -3 wt% Y2 O3 ceramic, J. Mater. Sci. 39 (2004) 3057– 3061. [31] X.H. Zheng, J.P. Tu, R.G. Song, Microstructure and tribological performance of CNx –TiNx composite films prepared by pulsed laser deposition, Mater. Des. 31 (2010) 1716–1719. [32] D. Craciun, G. Socol, D.V. Cristea, M. Stoicanescu, N. Olah, K. Balazs, N. Stefan, E. Lambers, V. Craciun, Mechanical properties of pulsed laser deposited nanocrystalline SiC films, Appl. Surf. Sci. 336 (2015) 391–395.