收稿日期：23 December 2014
基金：financially supported by the National Natural Science Foundation of China (Nos. 51401088 and 51471017);the China Postdoctoral Science Foundation (No. 2014M561580);the Natural Science Foundation of Jiangsu Province (Nos. BK20140549 and BK20130519);the Natural Science Fund for Colleges and Universities in Jiangsu Province (No. 14KJB430007);the Jiangsu Planned Projects for Postdoctoral Research Funds (No. 1401107C);the Senior Intellectuals Fund of Jiangsu University (No. 13JDG098);
Wear response of metastable b-type Ti–25Nb–2Mo–4Sn alloy for biomedical applications
Shun Guo Qi Zheng Xiu-Li Hou Xiao-Nong Cheng Xin-Qing Zhao
Institute for Advanced Materials, Jiangsu University
School of Materials Science and Engineering, Beihang University
Abstract：
The wear response of a newly developed metastable b-type Ti–25Nb–2Mo–4Sn(abbreviated as Ti-2524) alloy was investigated and compared with that of(a+b)-type Ti–6Al–4V alloy. Experimental results show that solution-treated(ST) Ti-2524 specimen has the lowest wear rate due to the combined effects of excellent ductility and lubricative Nb2O5. Although similar Nb2O5 forms on the surface of the cold rolled plus annealed(CRA) Ti-2524 specimen, the beneficial effect of Nb2O5 on the wear resistance is counteracted by the increase in wear rate caused by low elongation. Thus, the wear rate of the CRA Ti-2524 alloy is higher than that of ST Ti-2524 specimen.As for the ST Ti–6Al–4V alloy, no lubricative Nb2O5 forms on its worn surface owing to the absence of Nb. In addition, the ST Ti–6Al–4V alloy exhibits an elongation roughly similar to the CRA Ti-2524 specimen. Therefore,the ST Ti–6Al–4V specimen possesses a higher wear rate than the CRA Ti-2524 specimen.
Keyword：
Titanium alloy; Wear response; Low elastic modulus; High strength;
Author： Shun Guo,e-mail: shunguo@ujs.edu.cn; Xin-Qing Zhao,e-mail: xinqing@buaa.edu.cn;
Received： 23 December 2014
1 Introduction
Titanium and its alloys become the most attractive biomaterials for implant materials due to their excellent biocompatibility, high corrosion resistance and high strength, especially relatively low Young’s modulus [1–4]. For example, the most widely used Ti–6Al–4V alloy exhibits an elastic modulus (*108 GPa) only about half of that of the conventional 316L stainless steel (*200 GPa) and Co– Cr–Mo alloy (*210 GPa). Nevertheless, the elastic modulus of (a?b)-type Ti–6Al–4V alloy is still much higher than that of human bone tissue (*30 GPa) [5–7]. This large mismatch in the elastic modulus between implant and human bone can lead to so-called stress shielding effect and eventual implant failure [1, 8]. Additionally, it is reported that many long-term health problems, such as Alzheimer disease, neuropathy and osteomalacia, are closely associated with the release of V and Al ions from the Ti–6Al–4V alloy [1, 9]. Therefore, in the past decades, a great deal of interest has been triggered in developing noncytotoxic metastable b-type Ti alloys with low elastic modulus for biomedical applications [10–12].
In addition to elastic modulus, wear resistant property is yet another essential metric of the Ti alloys used for implants [13, 14]. It has been recognized that the low wear resistance is easy to result in implant loosening and the accumulation of wear debris that can lead to adverse allergic reactions in the tissue where they are deposited [15]. Thus far, many approaches have been adopted to improve the wear resistance of implant materials, including design of alloy composition, thermo-mechanical treatment, and surface modification [13, 14, 16].
Recently, a novel metastable b-type alloy Ti-2524 (abbreviated from the chemical composition of Ti–25Nb– 2Mo–4Sn in wt%), composing entirely of non-cytotoxic elements, has been developed for biomedical applications due to its excellent biomechanical compatibility. Upon a severe cold rolling and short-time annealing process, the Ti-2524 alloy can perform a high ultimate tensile strength of 1113 MPa, while retaining a low Young’s modulus of 65 GPa at room temperature [17, 18]. However, the wear response of the Ti-2524 alloy, another key property which must take into account for implant applications, was not investigated systematically.
In this work, the wear response of the Ti-2524 alloy was investigated and compared with that of a conventional (a?b)-type alloy Ti–6Al–4V. On the basis of the combined results from microstructure observations, mechanical property tests and X-ray photoelectron spectroscopy (XPS), the influence of alloy composition and thermo-mechanical treatment on the wear response of the Ti-2524 alloy was discussed.
2 Experimental
Ingot with a composition of Ti–25Nb–2Mo–4Sn (abbreviated as Ti-2524, wt%) was fabricated by arc melting, homogenized at 1273 K for 4 h in vacuum, and forged at 1173 K into a billet. After forging, the billet was solutiontreated at 1023 K for 1 h in an evacuated quartz tube, followed by quenching into water (*298 K). The specimens that were cut from the solution-treated billet will be called the ST Ti-2524 specimens henceforth. The solutiontreated billet was cold rolled into a plate of *1 mm in thickness, with a reduction of 70 %. Following cold rolling, the plate was annealed at 748 K for 15 min and the resultant specimen will be referred to as the CRA Ti-2524 specimens thereafter. (a?b)-type Ti–6Al–4V (wt%) alloy employed in this study was supplied by Baoji Nonferrous Metals of China. The received Ti–6Al–4V alloy was solution-treated at 1323 K for 30 min, followed by quenching into water (*298 K). The corresponding specimens will be denoted as the ST Ti–6Al–4V specimens thereafter.
Room temperature phase constitutions were detected by X-ray diffraction (XRD) using Cu Ka irradiation. Microstructure observations were conducted on a JEM 2100F transmission electron microscope (TEM). Uniaxial tensile tests were performed along the rolling direction, with an initial strain rate of 1 9 10-3s-1, to obtain the tensile stress–strain curves for the ST Ti-2524, CRA Ti-2524 and ST Ti–6Al–4V specimens. Based on these curves, the elastic moduli, yield strengths, ultimate tensile strengths and elongations from different specimens were determined. Vickers hardness measurement was conducted with a load of 9.8 N for a loading time of 15 s using a hardness tester. Binding energies of surface products were investigated by XPS in a vacuum chamber. Wear properties were tested by a ball-ondisk on UMT-2 multi-functional friction and wear tester, with a sliding speed of 100 r min-1and a normal load of 3 N for a loading time of 1 h. Wear morphology was observed on a JSM-7001F scanning electron microscope (SEM).
3 Results and discussion
Figure 1 represents the XRD patterns of ST Ti-2524, CRA Ti-2524 and ST Ti–6Al–4V specimens, respectively. A combination of b parent phase and a00martensite was detected in the ST Ti-2524 specimen. Upon a severe cold rolling plus short-time annealing treatment, the a00 martensite reversibly transforms to b-phase, accompanied by the precipitation of a few a from the b-phase matrix. In contrast to the ST Ti-2524 specimen, the ST Ti–6Al–4V specimen only consists of a single a0phase after a solution treatment at 1323 K for 30 min followed by water quenching.
The bright-field TEM images of ST Ti-2524, CRA Ti2524 and ST Ti–6Al–4V specimens are shown in Fig. 2c, respectively, along with the corresponding selected area diffraction (SAD) pattern at the left bottom corner of each image. It can be seen from Fig. 2a that coarse lath-shaped phase distributes within b matrix. An analysis of the SAD pattern in Fig. 2a shows that the lath-shaped phase is a00 martensite. This suggests that the ST Ti-2524 specimen undergoes a martensitic transformation from b to a00upon quenching. Subsequent cold rolling and annealing process lead to the appearance of irregular dark areas, as shown in the bright-field image of CRA Ti-2524 specimen in Fig. 2b. This can be attributed to the dislocation tangles produced by the cold rolling process. Additionally, a few fine precipitates with a width of several nanometers, noted as a white arrow, are seen in the CRA Ti-2524 specimen. According to the corresponding SAD pattern in Fig. 2b, these fine precipitates are recognized to be a phase. A previous experimental study by the present authors demonstrates that the precipitation of nano-sized a is closely associated with high-density dislocations acting as heterogeneous nucleation for a phase [17]. From the bright-field image and SAD pattern in Fig. 2c, one can see that the ST Ti–6Al–4V specimen undergoes a martensitic transformation from b to a0upon quenching, which is different from the b ? a00martensite transformation occurred in the ST Ti-2524 specimen. The difference in martensitic transformation types between the Ti-2524 and Ti–6Al–4V alloys could be explained by the pergence of the total amount of b-stabilizers in their corresponding high-temperature b parent phases.
Fig.1 XRD patterns of ST Ti-2524, CRA Ti-2524 and ST Ti–6Al– 4V specimens
Fig.2 Bright-field TEM images of ST Ti-2524 a, CRA Ti-2524 b and ST Ti–6Al–4V c specimens, along with corresponding SAD pattern at left bottom corner of each image
The mechanical properties of ST Ti-2524, CRA Ti-2524 and ST Ti–6Al–4V alloys are presented in Table 1. It is observed that the ST Ti-2524 alloy has advantages of low Young’s modulus and high elongation, but disadvantages of low yield strength and low ultimate tensile strength. Upon a cold rolling plus annealing treatment, the Ti-2524 alloy is significantly strengthened, while keeping a low elastic modulus. Meanwhile, this remarkable strengthening effect gives rise to an increase in hardness but a decrease in elongation. Compared with the CRA Ti-2524 alloy, the ST Ti–6Al–4V alloy possesses roughly similar mechanical properties, including strength, elongation and hardness. However, the ST Ti–6Al–4V alloy performs much higher Young’s modulus than the Ti-2524 alloy, which makes it undesirable for implant applications. Based on the mechanical properties in Table 1, it can be concluded that the mechanical properties of the biomedical Ti alloys are sensitive to alloy composition and thermo-mechanical condition.
It was reported that besides mechanical properties, the surface oxides can also exert a substantial influence on the wear response of metallic materials [14]. Thus, in the present study, the influence of surface reaction products on the wear response of ST Ti-2524, CRA Ti-2524 and ST Ti–6Al–4V specimens was also taken into account. The XPS results of the worn surfaces of ST Ti-2524, CRA Ti-2524 and ST Ti–6Al–4V specimens are shown in Table 2. The primary elements on the wear surfaces of the ST and CRA Ti-2524 specimens are Ti, Nb and O. Further analysis indicates that the surface oxidation films formed on the ST and CRA Ti-2524 specimens are mainly composed of Ti O2and Nb2O5, because the binding energies of Ti at Ti32/p2 and Nb at Nb53/d2 are in good agreement with the standard data for Ti O2and Nb2O5, respectively [19]. In the case of the ST Ti–6Al–4V specimen, however, the main elements on its worn surface are Al and O and the surface oxide is Al2O3. Here, it should be noted that among the above three oxides, Nb2O5is a very good lubricant [20] and expected to be beneficial for the improvement of wear resistance.
Table 1 Mechanical properties of ST Ti-2524, CRA Ti-2524 and ST Ti–6Al–4V alloys 下载原图
Table 1 Mechanical properties of ST Ti-2524, CRA Ti-2524 and ST Ti–6Al–4V alloys
Table 2 XPS results of binding energies at worn surfaces of ST Ti2524, CRA Ti-2524 and ST Ti–6Al–4V specimens (e V) 下载原图
Table 2 XPS results of binding energies at worn surfaces of ST Ti2524, CRA Ti-2524 and ST Ti–6Al–4V specimens (e V)
Figure 3 shows the wear morphologies of ST Ti-2524, CRA Ti-2524 and ST Ti–6Al–4V specimens. It can be seen from Fig. 3a that the surface of the ST Ti-2524 specimen is deeply smeared and grooved along the direction of sliding. For the CRA Ti-2524 specimen, however, visible brittle cracks are observed across the wear grooves, as shown in Fig. 3b. Similar brittle cracks can also be found on the wear surface of the ST Ti–6Al–4V specimen (Fig. 3c). According to the mechanical property data in Table 1, it is reasonable to attribute the formation of these brittle cracks to the poor elongations of the CRA Ti-2524 and ST Ti– 6Al–4V specimens. Clearly, these brittle cracks caused by low elongation can have a negative effect on the wear resistance of the CRA Ti-2524 and ST Ti–6Al–4V specimens.
The results of the wear tests of ST Ti-2524, CRA Ti2524 and ST Ti–6Al–4V specimens sliding on stainless steel are presented in Fig. 4. It can be seen that the ST Ti2524 specimen has the lowest wear rate, and the CRA Ti2524 specimen exhibits a slightly lower wear rate than the ST Ti–6Al–4V specimen, but higher than the ST Ti-2524 specimen. The difference in wear rate can be explained from the following two important facts: (1) the formation of Nb2O5is beneficial for the improvement of wear resistance due to its good lubrication effect; (2) low elongation promotes the appearance of brittle cracks and exerts a negative effect on wear resistance. For the ST Ti-2524 specimen, the good wear response is attributed to the combined effects of excellent ductility and lubricative Nb2O5. Although similar lubricative Nb2O5is observed on the surface of the CRA Ti-2524 specimen, this beneficial effect is counteracted by the increase in wear rate caused by low elongation. As a result, the CRA Ti-2524 specimen exhibits a higher wear rate than the ST Ti-2524 specimen. In contrast to the ST and CRA Ti-2524 specimens, no visible lubricative Nb2O5is detected on the worn surface of the ST Ti–6Al–4V specimen, as shown in Table 2. In addition, the ST Ti–6Al–4V specimen possesses an elongation of 5 %, roughly similar to the CRA Ti-2524 specimen. Thus, the ST Ti–6Al–4V specimen exhibits a slightly higher wear rate than the CRA Ti-2524 specimen.
Fig.4 Results of wear tests of ST Ti-2524, CRA Ti-2524 and ST Ti– 6Al–4V specimens sliding on stainless steel
According to the above discussion, it seems that the inner microstructures of the ST Ti-2524, CRA Ti-2524 and ST Ti– 6Al–4V specimens do not exert a direct influence on the wear response. However, it should be noted that the inner microstructure can significantly affect the ductility of the specimens. For example, the CRA Ti-2524 specimen, with high-density dislocations (Fig. 2), exhibits a low ductility and thereby performs a relatively high wear rate. Therefore, inner microstructural control can also be employed to improve the wear resistance of biomedical Ti alloys.
Fig.3 SEM images of wear morphologies of ST Ti-2524 a, CRA Ti-2524 b and ST Ti–6Al–4V c specimens
4 Conclusion
In summary, the wear response of a newly developed Ti2524 alloy was investigated and compared with that of a conventional (a?b)-type alloy Ti–6Al–4V. Experimental results indicate that the ST Ti-2524 specimen has the lowest wear rate due to the combined effects of excellent ductility and lubricative Nb2O5. Similar lubricative Nb2O5also forms on the surface of the CRA Ti-2524 specimen. However, the beneficial effect of Nb2O5on the wear resistance of the CRA Ti-2524 alloy is neutralized by the increase in wear rate caused by its poor ductility. As a result, the CRA Ti-2524 alloy performs a lower wear rate than the ST Ti–6Al–4V specimen. In the case of the ST Ti– 6Al–4V alloy, no lubricative Nb2O5is detected on its worn surface owing to the absence of Nb. Meanwhile, the ST Ti– 6Al–4V alloy exhibits an elongation of 5 %, roughly similar to the CRA Ti-2524 specimen. Therefore, the ST Ti–6Al–4V specimen exhibits a slightly higher wear rate than the CRA Ti-2524 specimen.
