Influence of the surface nanocrystallization on the gas nitriding of Ti–6Al–4V alloy

Influence of the surface nanocrystallization on the gas nitriding of Ti–6Al–4V alloy

Applied Surface Science 286 (2013) 412–416 Contents lists available at ScienceDirect Applied Surface Science journal homepage: www.elsevier.com/loca...

2MB Sizes 0 Downloads 6 Views

Applied Surface Science 286 (2013) 412–416

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Influence of the surface nanocrystallization on the gas nitriding of Ti–6Al–4V alloy Liling Ge, Na Tian ∗ , Zhengxin Lu, Caiyin You School of Materials Science and Engineering, Xi’an University of Technology, Xi’an 710048, PR China

a r t i c l e

i n f o

Article history: Received 5 December 2012 Received in revised form 1 August 2013 Accepted 17 September 2013 Available online 25 September 2013 Keywords: Ti–6Al–4V alloy Surface nanocrystallization Gas nitriding

a b s t r a c t Supersonic Fine Particle Bombarding (SFPB) was performed on Ti–6Al–4V alloy to form surface nanostructures. The effect of the surface nanostructures on the gas nitriding was investigated by varying the nitriding temperature and time. Through SFPB process, a nanocrystalline surface layer with a mean grain size of 16 nm was obtained, presenting a high hardness of 640 HV. In comparison to the surface without SFPB treatment, the treated surface exhibited a thick nitrogen diffusion layer and high hardness under the same nitriding process. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Titanium and its alloys possess many merits of the excellent corrosion resistance, high specific strength and heat resistance, which are very favorable to the automotive and aerospace products. Ti–6Al–4V alloy is very popularly used, amount to 50% current titanium alloy production. Ti–6Al–4V exhibits not only excellent mechanical properties, but also remarkable corrosion resistance in many different corrosive environments. It can also be jointed by welding and cold working. However the low hardness and poor wear resistance have restricted their applications in engineering fields. In order to improve the surface mechanical properties of titanium alloy, surface chemical heat treatment was commonly used [1–3], such as nitriding and boronizing. The corrosion, wear resistance and surface hardness can be enhanced through gas nitriding and laser nitriding for the biomedical titanium and its alloys [2,3]. Generally, the treatment temperature for the conventional nitriding of the titanium alloy ranges from 780 ◦ C to 950 ◦ C with a processing time of several tens hours, resulting in the high energy consumption. It was reported that the surface nanocrystallization can greatly shorten nitridation duration or decrease nitridation temperature [4–7]. Through the severe plastic deformation, the nano-scale grains can be achieved and large number of defects will be induced too, such as dislocation, twins and shear band. These defects could provide the channels to the atomic diffusions besides lots of grain

∗ Corresponding author. Tel.: +86 29 82312090. E-mail address: [email protected] (N. Tian). 0169-4332/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.apsusc.2013.09.105

boundaries, promoting the dynamic process of chemical reaction. In this work, the surface nanostructures of Ti–6Al–4V alloy were obtained by Supersonic Fine Particle Bombarding (SFPB) technology [8]. Under the same nitriding process, the SFPB treated surface presented a much thicker nitriding layer, higher hardness than the un-SFPB surface.

2. Experimental procedures 2.1. Specimen preparation Cold-rolled plate of Ti–6Al–4V with a size of 60 mm × 60 mm × 3 mm was used in this study. The chemical compositions of Ti–6Al–4V are listed in Table 1. The main parameters of the SFPB process were chosen as follows: the airflow pressure was about 0.18 MPa, the diameter of steel shot was 0.3 mm, and the processing duration was 60 min. SFPB sample was cut into the size of 10 mm × 10 mm × 3 mm and then was cleaned by acetone. The samples were placed into a tube furnace for nitriding process. The nitriding process and equipment can be seen in Reference [9]. The nitriding gas was pure NH3 which was dissociated in the pre-dissociation furnace. Since nitrogen could be strongly absorbed by Ti above 600 ◦ C, the nitriding processes were done at 650 ◦ C, 700 ◦ C, 750 ◦ C, and 800 ◦ C for 3 h, 6 h, and 9 h respectively. The samples were etched by the solution of HF (3 ml) + H2 O2 (2 ml) + H2 O (10 ml) for the optical microscopy observations. The microstructure and phase constituents were examined using high resolution transmission electron microscopy (TEM, JEM-3010) and X-ray diffraction (XRD, 7000S). The hardness was measured using a Vickers microhardness tester (TUKON 2100B).

L. Ge et al. / Applied Surface Science 286 (2013) 412–416

413

Table 1 Chemical compositions of Ti–6Al–4V (mass%).

Ti–6Al–4V

Al

V

Fe

C

O

H

Ti

6.34

4.10

0.01

0.05

0.12

<0.015

Remaining

Fig. 1. XRD patterns of the surface of SFPB and un-SFPB samples.

The cross-sectional morphology was observed by a JSM-6700F scanning electron microscopy (SEM) and Olympus GX71 optical microscopy.

3. Experimental results 3.1. Microstructure and deformed layer depth after SFPB for Ti–6Al–4V Fig. 1 describes the XRD patterns of Ti–6Al–4V samples before and after SFPB. Both samples consist of ␣-Ti and ␤-Ti phases. Compared with un-SFPB sample, the diffraction peaks of the SFPB sample were broadened. The mean grain size of the surface layer of the SFPB sample is about 16 nm, calculated by scherrer-wilson function without the consideration of lattice micro-strain. Fig. 2 shows the cross-sectional SEM image and the microhardness profile of the SFPB sample along the deformed surface layer. As shown in Fig. 2a, in the severely deformed layer, the grains of ␣-Ti and ␤-Ti phases were significantly refined so as to be hardly distinguished. It can be seen in Fig. 2b that the micro-hardness of SFPB sample changes from 640 HV in the surface layer to 325 HV in the center. Based on the hardness variation, the thickness of severely deformed layer is about 30 ␮m (marked by the dashed line). The depth of the deformed layer is about 150 ␮m. There exist two stages for the hardness change with the SFPB deformed depth. The severely deformed surface layer presents very large hardness, slowly dropping from 640 HV to 542 HV. With entering the slightly deformed layer, the hardness drastically drops to 400 HV, and then gradually reaches the hardness 325 HV. Fig. 3 shows the bright field TEM images of the microstructure and corresponding selected area diffraction pattern of the deformed surface layer of the SFPB sample. It can be seen that the equiaxal nanocryatalline grains (about 25 nm) was produced by the SFPB treatment in the surface. The corresponding selected-area diffraction pattern indicates that the nanocrystalline grains exhibited random crystallographic orientations. The formation of nanocrystalline microstructure is related to the dynamic recrystallization during deformation [10–12]. The grain size of SFPB surface observed by TEM is slightly higher than the calculated value based on the XRD result, which might be attributed to the differences of two observation techniques.

Fig. 2. (a) The cross-sectional SEM image and (b) hardness profile along the deformed depth from the surface for SFPB sample.

3.2. Nitriding microstructure and hardness for SFPB and un-SFPB samples The cross-sectional optical microstructures of the SFPB and un-SFPB samples nitrided at 650 ◦ C, 700 ◦ C, 750 ◦ C and 800 ◦ C for 6 h are shown in Fig. 4 respectively. The regions with bright contrast (marked by the dashed lines) correspond to the nitride layers. Fig. 4a, c, e and g are the nitride microstructures of SFPB

Fig. 3. Bright field TEM image and selected area diffraction pattern of SFPB surface.

414

L. Ge et al. / Applied Surface Science 286 (2013) 412–416

Fig. 4. Cross-sectional observations of samples of SFPB (a, c, e and g) and un-SFPB (b, d, f and h) after nitriding at different temperature for 6 h: (a) and (b) nitriding at 650 ◦ C; (c) and (d) nitriding at 700 ◦ C; (e) and (f) nitriding at 750 ◦ C; (g) and (h) nitriding at 800 ◦ C.

samples. Fig. 4b, d, f and h are the nitride microstructures of unSFPB samples. It can be clearly seen that the thickness of the nitrided layer increased with the increase of nitriding temperature for SFPB samples, and the depths of the nitrided layer are about 10 ␮m, 20 ␮m, 30 ␮m and 45 ␮m, respectively. However, the nitrided layer could not be observed when the un-SFPB sample was nitrided at the temperature of 650 ◦ C (Fig. 4b). The nitrided layer for the unSFPB samples can be seen when the nitriding temperature is 700 ◦ C and above. The nitrided layer depth of the un-SFPB sample is much thinner than that of the SFPB samples at the same nitriding temperature below 800 ◦ C. When the nitriding temperature reaches 800 ◦ C, which lies in the range of the conventional nitriding temperature for Ti alloy: 780 ∼ 950 ◦ C, both SFPB and un-SFPB samples have a similar thickness of nitrided layer (seen in Fig. 4g and h). Fig. 5 shows the XRD patterns of the SFPB and un-SFPB samples nitrided at 750 ◦ C for 6 h. It can be seen that ␣-Ti(N) and the hard phases of Ti2 N and TiN compounds formed in the surface after

Fig. 5. X-ray diffraction patterns obtained from the surface layer after nitriding at 750 ◦ C for 6 h.

L. Ge et al. / Applied Surface Science 286 (2013) 412–416

Fig. 6. Surface hardness of SFPB and un-SFPB samples after nitriding at different temperatures for 6 h.

nitriding. Since Ti has a very high affinity with O, there exist minor TiO2 phases. The ␣-Ti(N) diffraction peak for the SFPB sample is much stronger than that for the un-SFPB one, which indicates that there are more ␣-Ti(N) in SFPB samples. This can be explained that the high density of grain boundaries and a large number of defects for SFPB sample provide the fast channel for N atom diffusion [4,5,7,9]. Fig. 6 presents the surface hardness of the SFPB and un-SFPB samples nitrided at 650 ◦ C, 700 ◦ C, 750 ◦ C and 800 ◦ C for 6 h. It can be clearly seen that the hardness of both samples holds a linear relationship with the increase of temperature from 650 ◦ C to 750 ◦ C, but the hardness increase for the SFPB sample is much quicker than that for the un-SFPB sample. At the same nitriding temperature, the surface hardness of the SFPB sample is much higher than that of the un-SFPB sample. Especially at the nitriding temperature of 750 ◦ C, the surface hardness of the SFPB sample is 1126 HV, while the hardness of the un-SFPB sample is only 613 HV. When the nitriding temperature reaches 800 ◦ C, the surface hardness is obviously enhanced up to 1100 HV for un-SFPB nitride samples, which is very close to 1150 HV for SFPB sides. The surface hardness change is consistent with the result of Fig. 4, in which the thickness of the nitriding layer is very similar for both sample while the nitriding temperature is 800 ◦ C. Fig. 7 shows the surface hardness for the SFPB and un-SFPB samples nitrided at 750 ◦ C for different time. It can be seen that the surface hardness of the SFPB sample increases obviously for 3 h and 6 h, and tends to be constant for 9 h. For the un-SFPB sample, the increase of the surface hardness is very slow with the increase of the nitriding time at lower nitriding temperature.

Fig. 7. Surface hardness of the of SFPB and un-SFPB samples varying from nitriding time.

415

Fig. 8. Hardness profiles of the surface layers of SFPB and un-SFPB nitriding samples at 750 ◦ C for 6 h.

Fig. 8 is the hardness profile of the surface layer of the SFPB and un-SFPB samples nitrided at 750 ◦ C for 6 h. It can be seen that there is higher hardness with a thicker nitrided layer on the SFPB sample. The hardness of the nitrided SFPB sample sharply decreases from the top surface to the deep matrix, whereas the hardness of the nitrided layer of un-SFPB samples gradually decreased from the surface layer to the substrate. Based on the dramatic hardness enhancement, a severe nitriding reaction happened for the SFPB surface, showing a severe nitriding layer about 50 ␮m 4. Discussion As for the gas nitriding, the most important processing parameters are temperature and time. The conventional nitriding temperature for the titanium alloy ranges from 780 ◦ C to 950 ◦ C, and nitriding time is usually several tens hours. Briefly, the higher nitriding temperature and longer nitriding time cause the thicker nitrided layer and higher surface hardness. However, too high temperature and too long time result in the high energy consumption, grain coarsening, the oxidation and distortion of the samples. The surface nanocrystallization can decrease the nitriding temperature and shorten the nitriding time [4,5,7,9]. In this study, the surface nanocrystallization for Ti–6Al–4V alloy was carried out by the SFPB. The mean size of grains in the surface is about 16 nm. The surface hardness was promoted to 640 HV from original value of 325 HV. The surface nanocrystallization also induces large number of defects due to the severe plastic deformation, such as dislocation, twins and shear band [10,11,13], which cause the high chemical activity in the surface and promote the dynamic process of the chemical reaction. The large number of the grain boundary and dislocation in the surface could provide an ideal diffusion channel for atom diffusion which can play an important role during the first interface reaction-controlled stage of the gas nitriding process [4]. As for the SFPB surface, the surface nanostructure (<50 nm at 750 ◦ C) of Ti–6Al–4V alloy is stable at the temperature from 650 to 750 ◦ C [13], so a large number of grain boundaries provide the diffusion channels for N atoms during the nitridation at below 750 ◦ C. At the same time, the increase of temperature raises the diffusion coefficient of N atom from 650 to 750 ◦ C too. N atom could be easily saturated in ␣-Ti and the new hard phases of Ti2 N and TiN formed with the increase of the N content (shown in Fig. 5). The nitrided layer depth was also increased with increasing temperature from 650 to 750 ◦ C (shown in Fig. 4a, c and e) and it can be clearly seen that the surface hardness of the SFPB samples rapidly increase with the increase of temperature from 650 ◦ C to 750 ◦ C (shown in Fig. 6). But compared with the SFPB surface, the un-SFPB surface could not supply enough diffusion channel for nitrogen atom at the nitriding temperature between 650 ◦ C to 750 ◦ C, due to the coarse grain and lack of grain boundaries. Although the diffusion coefficient

416

L. Ge et al. / Applied Surface Science 286 (2013) 412–416

increases with the increase of temperature, the nitrogen does not increase so quickly. It is difficult to make the N atom saturated in ␣-Ti and form Ti2 N and TiN phases at the low nitriding temperature. These caused the slower increase of the nitriding layer depth and surface hardness for un-SFPB sample than those for SFPB at the same nitriding temperature (shown in Fig. 4). The surface hardness of the un-SFPB sample is only 613 HV even after nitriding at 750 ◦ C, while the hardness of the SFPB sample raises to 1126 HV (shown in Fig. 6). However, when the nitriding temperature reaches 800 ◦ C (in the range of the conventional nitriding temperature for Ti alloy: 780–950 ◦ C), the grains of the surface nanostructure for the SFPB sample and the cold rolled matrix microstructure were coarsened (shown in Fig. 4g and h). This is because the nanostructure thermal stability of Ti–6Al–4V produced by SFPB was retained till 750 ◦ C [13]. The diffusion channels of the high density of grain boundaries from nanocrystalline structures disappeared at the high nitriding temperature of 800 ◦ C, causing a similar nitriding thickness for both SFPB and un-SFPB samples (shown in Fig. 4g and h). The surface hardness was obviously enhanced for the un-SFPB nitride samples, up to 1100 HV, while 1150 HV for the SFPB sides at this temperature (shown in Fig. 6). Therefore, the surface hardness of the un-SFPB side is close to that of the SFPB side. The nitridation was dominantly controlled by the diffusion at 800 ◦ C. Namely, the promotion of the nitriding temperature just increases the diffusion coefficient of N atom [4]. During the gas nitriding for the SFPB and un-SFPB side, the influence of the nitriding time on the thickness and hardness of the nitriding layer complies with the parabolic rule in the beginning of nitridation. Once the nitrides formed, it would hinder the diffusion of nitrogen atoms and the formation of succeeding nitride, resulting in the constant change of surface hardness and nitriding layer (shown in Figs. 7 and 8).The thickness and surface hardness for SFPB side are higher than those for un-SFPB side.

5. Conclusions The surface grains in Ti–6Al–4V alloy were refined by the SFPB process and the depth of the deformed layer is about 150 ␮m after the SPPB treatment. The mean grain size of the surface layer of the SFPB sample is about 16 nm. The micro-hardness of the SFPB sample increases to 640 HV from 325 HV. The nanostructures by the SFPB accelerates the nitriding speed and decreases the nitriding temperature to 650 ◦ C, much lower than the conventional nitriding temperature. SFPB process not only increases the surface hardness up to 1126 HV and but also the nitriding layer up to about 250 ␮m at a relatively low nitriding temperature of 750 ◦ C. Acknowledgement The authors gratefully acknowledge the financial support of Science and Technology Planning Project of Shaanxi Province (Grant No. 2009K06-22). References [1] A. Zhecheva, W. Sha, S. Malinovb, Surf. Coat. Technol. 200 (2005) 2192–2207. [2] S. Sathish, M. Geetha, N.D. Pandey, C. Richard, R. Asokamani, Mater. Sci. Eng. C 30 (2010) 376–382. [3] M. Nakai, M. Niinomi, T. Akahori, N. Ohtsu, H. Nishimura, H. Toda, H. Fukui, M. Ogawa, Mater. Sci. Eng. A 486 (2008) 193–201. [4] J.F. Gu, D.H. Bei, J.S. Pan, J. Lu, K. Lu, Mater. Lett. 55 (2002) 340–343. [5] W.P. Tong, N.R. Tao, Z.B. Wang, J. Lu, K. Lu, Science 299 (2003) 686–688. [6] Z.B. Wang, N.R. Tao, W.P. Tong, J. Lu, K. Lu, Acta Mater. 51 (2003) 4319–4329. [7] Y.M. Lin, J. Lu, L.P. Wang, T. Xu, Q.J. Xue, Acta Mater. 54 (2006) 5599-5605. [8] T.Y. Xiong, Z.W. Liu, Z.C. Li, J. Wu, H.Z. Jin, T.F. Li, Mater. Lett. 17 (3) (2003) 69–71. [9] L.L. G, C.H. Lu, X.T. Jing, Z.X. Lu, Trans. Mater. Heat Treat. 29 (5) (2008) 155–159. [10] M. Wen, G. Liu, J.F. Gu, W.M. Guan, J. Lu, Appl. Surf. Sci. 255 (2009) 6097–6102. [11] K.Y. Zhu, A. Vassel, F. Brisset, K. Lu, J. Lu, Acta Mater. 52 (2004) 4101–4110. [12] L.Q. Hu, J.F. Ma, B.S. Xu, Trans. Mater. Heat Treat 28 (2007) 343–347. [13] Z.Q. Guo, L.L. G, H. Yuan, C. Qing, Trans. Mater. Heat Treat. 33 (3) (2012) 115–118.