- Email: [email protected]
Microstructure, corrosion behavior and cytotoxicity of biodegradable Mg–Sn implant alloys prepared by sub-rapid solidification
Materials Science and Engineering C 54 (2015) 245–251
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Microstructure, corrosion behavior and cytotoxicity of biodegradable Mg–Sn implant alloys prepared by sub-rapid solidification Chaoyong Zhao a, Fusheng Pan a,b,c,⁎, Shuang Zhao a, Hucheng Pan a, Kai Song a, Aitao Tang a,b a b c
College of Materials Science and Engineering, Chongqing University, Chongqing 400044, China National Engineering Research Center for Magnesium Alloys, Chongqing University, Chongqing 400044, China Chongqing Academy of Science and Technology, Chongqing 401123, China
a r t i c l e
i n f o
Article history: Received 2 August 2014 Received in revised form 22 April 2015 Accepted 11 May 2015 Available online 13 May 2015 Keywords: Magnesium alloy Corrosion Biodegradable Tin Orthopedic implant
a b s t r a c t In this study, biodegradable Mg–Sn alloys were fabricated by sub-rapid solidification, and their microstructure, corrosion behavior and cytotoxicity were investigated by using optical microscopy, scanning electron microscopy equipped with an energy dispersive X-ray spectroscopy, X-ray diffraction, immersion test, potentiodynamic polarization test and cytotoxicity test. The results showed that the microstructure of Mg–1Sn alloy was almost equiaxed grain, while the Mg–Sn alloys with higher Sn content (Sn ≥ 3 wt.%) displayed α-Mg dendrites, and the secondary dendrite arm spacing of the primary α-Mg decreased significantly with increasing Sn content. The Mg–Sn alloys consisted of primary α-Mg matrix, Sn-rich segregation and Mg2Sn phase, and the amount of Mg2Sn phases increased with increasing Sn content. Potentiodynamic polarization and immersion tests revealed that the corrosion rates of Mg–Sn alloys increased with increasing Sn content. Cytotoxicity test showed that Mg– 1Sn and Mg–3Sn alloys were harmless to MG63 cells. These results of the present study indicated that Mg–1Sn and Mg–3Sn alloys were promising to be used as biodegradable implants. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Magnesium and its alloys have obtained increasing attention as biodegradable orthopedic implants due to their good biocompatibility, similar mechanical properties to natural bone and biodegradability in human physiological environment [1–4]. However, the rapid corrosion rate of magnesium alloys is a major obstacle to their clinical application [2]. The undesirable corrosion resistance of magnesium originates from the loose and porous oxide films on its surface, which only provide limited protection to the magnesium substrate [5]. Element alloying was usually used to improve the corrosion resistance of magnesium alloys [2]. The addition of aluminum (Al) and/or rare earth (RE) elements to magnesium were examples [6,7]. Unfortunately, Al and/or RE might induce latent toxic and harmful effects on the human body [8,9]. Therefore, alloying elements must be chosen with careful consideration of possible toxic effects [10,11]. At present, one or more elements with non-toxicity or low toxicity were chosen as alloying elements to improve the corrosion resistance of magnesium, such as Mg–Ca, Mg–Zn, Mg–Sr, Mg–Zr–Sr, Mg–Ca–Sr, Mg–Zn–Sr, Mg– Zn–Mn, Mg–Zn–Ca, Mg–Zn–Mn–Ca [12–24]. However, the corrosion resistance of these alloys needed to be further improved for clinic applications [7,14,19]. ⁎ Corresponding author at: College of Materials Science and Engineering, Chongqing University, Chongqing 400044, China. E-mail address: [email protected] (F. Pan).
http://dx.doi.org/10.1016/j.msec.2015.05.042 0928-4931/© 2015 Elsevier B.V. All rights reserved.
Tin (Sn) was one of the most essential elements in the human body. Gu et al. studied the in vitro corrosion and biocompatibility of Mg–1Sn alloy and conjectured that Sn might be a good alloying element for implant magnesium alloys [10]. Liu et al. fabricated the Mg–Sn alloys using the conventional casting method, and investigated its microstructure and mechanical properties. They reported that tin addition showed a refinement effect on the secondary dendrite arm spacing of the α-Mg phase, and Mg–5Sn alloy had the best mechanical properties [25]. However, there was lack of research about the effect of volume fraction and existence format of secondary phases on the corrosion behavior of Mg– Sn alloys. Furthermore, the cytotoxicity evaluation of the Mg–Sn alloys was also important for biomedical applications, which was scarce at present. In this study, Mg–Sn alloys were prepared by a sub-rapid solidification process [26], and their microstructure, corrosion behavior and cytotoxicity were investigated for orthopedic applications. 2. Materials and methods 2.1. Materials preparation The Mg–Sn alloys used for this investigation were the Mg–1Sn, Mg– 3Sn, Mg–5Sn and Mg–7Sn alloys with nominal composition of 1 wt.% Sn, 3 wt.% Sn, 5 wt.% Sn and 7 wt.% Sn, respectively. They were prepared using pure magnesium (99.98 wt.%) and pure tin (99.9 wt.%) ingots. The melting process was carried out in a middle-frequency induction
246
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
Fig. 1. Optical micrographs of the Mg–Sn alloys with different Sn content. (a) Mg–1Sn, (b) Mg–3Sn, (c) Mg–5Sn and (d) Mg–7Sn.
furnace under a mixed gas atmosphere of SF6 and CO2, and the stainless steel crucible with the melt was quenched in salt water [26]. Disk (10 × 10 × 2 mm3) samples were machined from the ingots and ground using silicon carbide (SiC) grinding paper up to 1000 grit for the microstructure characterization and immersion test. 2.2. Microstructure characterization Microstructure observation of samples was carried out on an optical microscopy and a scanning electron microscopy (SEM, TESCAN VEGA II LMU) equipped with an energy dispersive X-ray spectroscopy (EDX). The phase of samples was examined using an X-ray diffraction (XRD, Rigaku D/MAX-2500PC). 2.3. Immersion and potentiodynamic polarization tests The weight of the Mg–Sn alloy samples before immersion test was measured. The samples were exposed to Hank's solution for 500 h at 37 °C [21], and the ratio of sample surface area to the volume of Hank's
solution was 1 cm2: 20 ml according to ASTMG31-72 standard [27]. After immersion, the immersed samples were removed, rinsed and dried. The corrosion morphologies and chemical composition of the immersed samples were characterized by SEM and XRD, respectively. Some immersed samples were cleaned with chromate acid (200 g/l CrO3 + 10 g/l AgNO3) to remove their corrosion products, and the weight of the cleaned samples was also measured. The corrosion rate (CR) was calculated in mg cm−2 d−1 according to the literature's equation [18,28]. The weight loss rate could be converted to the average corrosion rate (Rw, mm/year) according to the following reaction [18,28]: Rw ¼ 2:1 CR : Potentiodynamic polarization test was performed in Hank's solution using a three-electrode cell with a saturated calomel electrode as the reference electrode, a platinum plate as the counter electrode and the as-prepared sample with an exposed area of 1.13 cm2 as the working electrode. In addition, the commercial AZ31 magnesium alloy was used as a control. All samples were ground using silicon carbide (SiC) grinding paper up to 1000 grit, cleaned and dried. The sample was immersed in Hank's solution for 900 s or 3600 s before undertaking the test. The potentiodynamic curves were obtained using an electrochemical workstation (CHI660E, China). The measurements started from − 300 mV vs. open circuit potential (OCP) at a constant scan rate of 1 mV s−1 and terminated until +300 mV vs. OCP. An average of three measurements was taken for each group. The corrosion current density (icorr, mA/cm2) was related to the average corrosion rate (Ri, mm/year) according to the following reaction [18,28]: Ri ¼ 22:85 icorr :
2.4. Cytotoxicity test
Fig. 2. XRD patterns of the Mg–Sn alloys with different Sn content.
MG63 cells and Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum were used in this study. The cytotoxicity of the Mg–Sn alloys was evaluated by indirect cell viability assay. The ion extracts were obtained from 10 × 10 × 2 mm3 of the Mg–Sn alloy samples by incubating in 2.8 ml medium for 24 h in a humidified atmosphere of 5% CO2 at 37 °C [29]. The DMEM medium was used as the negative control. The MG63 cells were seeded onto 96-
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
247
Fig. 3. SEM micrographs of the Mg–Sn alloys with different Sn content. (a) Mg–1Sn, (b) Mg–3Sn, (c) Mg–5Sn and (d) Mg–7Sn.
well cell culture plates at 5 × 103 cells per 100 μl medium in each well and incubated 24 h in a humidified atmosphere with 5% CO2 at 37 °C for cell attachment. Then, the medium in each well was replaced with 100 μl of extract or DMEM medium, and incubated for 2, 4 and 6 days. At the end of each time point, cell morphology was observed by optical microscopy. After that, 20 μl MTT solution (5 mg/ml) was added to each well and further incubated for 4 h. The medium was then replaced by 150 μl dimethyl sulfoxide (DMSO). The optical density (OD) measurements of 100 μl supernatant were conducted at 490 nm.
3. Results and discussion 3.1. Microstructure of the Mg–Sn alloys Fig. 1 showed the optical metallographic images of the Mg–Sn alloys. It could be seen that the Mg–1Sn alloy was almost equiaxed grain structure (Fig. 1a), while the Mg–Sn alloys with higher Sn content (Sn ≥ 3 wt.%) displayed α-Mg dendrites, and the secondary dendrite arm spacing of the primary α-Mg decreased with increasing Sn content. This trend of the refinement of the secondary dendrite arm spacing by the addition of tin was similar to the previous report [25]. The formation of dendrites was that tin was partitioned into the liquid ahead of the solidification front, resulting in a constitutional supercooling zone of liquid ahead of the interface [25]. The constitutional supercooling resulted from the tin enrichment promoted nucleation and hindered the fast growth of grain, thus showing a refinement effect in Mg–Sn alloys [25]. More tin addition caused a stronger effect of constitutional supercooling, which resulted in the decrease of the secondary dendrite arm spacing [25].
The XRD patterns of the Mg–Sn alloys were shown in Fig. 2. Only peaks corresponding to α-Mg phase were found in the XRD patterns for Mg–1Sn and Mg–3Sn alloys probably due to the low volume fractions of the second phase. On the other hand, Mg2Sn peaks could be clearly identified in the Mg–5Sn and Mg–7Sn alloys. It could also be observed that the diffraction intensities of Mg2Sn phases increased with increasing Sn content. Figs. 3 and 4 showed the SEM images and EDS analysis of the Mg–Sn alloys with different Sn content. It could be observed that the microstructure of the Mg–1Sn alloy consisted of primary α-Mg and Sn-rich segregation, as was shown in Figs. 3a and 4a. The chemical composition of Sn-rich segregation identified by EDS (Fig. 4d) showed that the atomic ratio of Mg to Sn was about 74:1, which was not likely attributed to the Mg2Sn phase. With increasing Sn content up to 3 wt.%, the alloy was mainly composed of primary α-Mg, Sn-rich net-segregation and Mg2Sn second phase. The volume fraction of the second phase (white area) increased considerably with further increasing Sn content. The high magnification SEM images of Mg–7Sn alloy in Fig. 4b and c showed that two types of second phase presented in this alloy, and their atomic ratio of Mg to Sn were about 3.5/1 and 4.8/1, respectively, which deviated from that of Mg2Sn. It was due to the small size of the second phase and the α-Mg matrix. According to Mg–Sn binary diagram, the saturation solid solubility of tin in magnesium was 14.85 wt.% at 561.2 °C, which decreased sharply to 0.45 wt.% at 200 °C and almost zero at room temperature [30]. The saturation solid solubility decreased fast with decreasing temperature, which caused the tin atoms to precipitate and formed the secondary phase Mg2Sn during the solidification process [30]. However, the microstructure of the Mg–Sn alloys in the present study differed from that of the previous study due to the different solidification method [25]. In the
248
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
Fig. 4. SEM micrographs and EDS analysis of the Mg–1Sn and Mg–7Sn alloys. (a) Mg–1Sn, (b) and (c) Mg–7Sn, (d) EDS analysis corresponding to assigned A area, (e) EDS analysis corresponding to assigned B area, (f) EDS analysis corresponding to assigned C area.
present study, much Sn-rich segregation was presented besides Mg2Sn phases in the Mg–Sn alloys. It was probably that the solidification rate of the Mg–Sn alloys was faster in the salt water than in the iron mold, so the Sn-rich segregation could not transform into Mg2Sn phase.
[17]. The Mg–Sn alloys dissolved according to the following reaction [13]:
3.2. Immersion test
As chloride ions presented in the Hank's solution, Mg(OH)2 would react with Cl− to form highly soluble magnesium chloride and OH−, which resulted in the increase of pH value of Hank's solution [2,13,18]. XRD patterns of Mg–Sn alloys after immersion in Hank's solution for 500 h in Fig. 6 revealed that Mg(OH)2 existed in the corrosion products and the diffraction intensities of Mg(OH)2 increased with increasing Sn content. The undissolved Mg(OH)2 in the corrosion products on the surfaces of Mg alloy might result from their rapid corrosion rate, and different corrosion rate of Mg–Sn alloys led to different diffraction intensities. 2+ , etc. in Hank's solution and high pH led to the Furthermore, PO3− 4 , Ca formation of the corrosion products of phosphates containing magnesium/calcium [19]. Fig. 7 showed the weight loss rate of the Mg–Sn alloys after immersion in Hank's solution for 500 h. The weight loss of magnesium alloy was an indication of corrosion [18]. The correlated average corrosion rate (Rw, mm/year) was listed in Table 1. It could be seen in Fig. 7 and Table 1 that the weight loss rate of the Mg–Sn alloys increased in the order of Mg–1Sn b Mg–3Sn b Mg–5Sn b Mg–7Sn, indicating that Mg– 1Sn alloy showed the best corrosion resistance and Mg–7Sn alloy showed the worst one. The results of immersion test clearly indicated that the corrosion resistance of Mg–Sn alloys decreased with increasing Sn content.
Fig. 5 showed the surface morphologies of the Mg–Sn alloys after immersion in Hank's solution for 500 h. It could be observed that the surfaces of the Mg–1Sn, Mg–3Sn and Mg–5Sn alloys were covered by a black film and white corrosion products, and the amount of the white corrosion products increased with increasing Sn content, indicating that the corrosion resistance of the Mg–Sn alloys decreased with increasing Sn content. A large amount of deep corrosion pits presented on the surface of Mg–7Sn alloy, which indicated that it showed a higher corrosion rate than the other Mg–Sn alloys. The SEM images of Mg–1Sn alloy at a higher magnification revealed that cracks could be seen in the black film as a result of dehydration of the surface layer in the air [13]. It was well-known that the corrosion rate of magnesium and its alloys depended on their microstructure [31–35]. As shown in the microstructure characterization, the volume fraction and existence format of secondary phases of Mg–Sn alloys differed with increasing Sn content (Figs. 2–4). The microstructure of the Mg–1Sn alloy consisted of primary α-Mg and Sn-rich segregation (Figs. 3a and 4a). Besides primary αMg and Sn-rich net-segregation, Mg2Sn phase appeared when the Sn content increased to 3 wt.%, and the volume fraction of the second phases increased with increasing Sn content. In general, the second phase had more positive corrosion potential than that of the α-phase [36]. Therefore, the second phases might act as the cathode and the αMg as the anode, which resulted in galvanic corrosion after immersion in Hank's solution. Since an increase of the second phases would lead to an increase in micro-galvanic couples, this could explain why the corrosion rate of the Mg–Sn alloys increased with increasing Sn content
Mg þ 2H2 O ¼ MgðOHÞ2 þ H2 ↑:
3.3. Potentiodynamic polarization test Fig. 8 showed the potentiodynamic polarization curves of the AZ31 and Mg–Sn alloys after immersion in Hank's solution with different time. Table 2 listed the electrochemical parameters of the AZ31 and
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
249
Fig. 5. Surface morphologies of the Mg–Sn alloys after immersion in Hank's solution for 500 h. (a) Mg–1Sn, (b) Mg–3Sn, (c) Mg–5Sn and (d) Mg–7Sn. The inserted photos were enlarged views of local area.
Mg–Sn alloys in Hank's solution obtained from the polarization curves by Tafel extrapolation and the correlated average corrosion rate (Ri, mm/year). Some information could be obtained from Fig. 8 and Table 2. Firstly, the increase of the icorr of the Mg–Sn alloys was observed with increasing Sn content during the whole immersion time, which indicated that the increase of corrosion rates of the Mg–Sn alloys in the following order: Mg–1Sn b Mg–3Sn b Mg–5Sn b Mg–7Sn [13]. The
Fig. 7. Weight loss rate of the Mg–Sn alloys after immersion in Hank's solution for 500 h.
Table 1 Values measured from weight loss rate for Mg–Sn alloys in Hank's solution.
Fig. 6. XRD patterns of the Mg–Sn alloys after immersion in Hank's solution for 500 h.
Alloy
Rw (mm/year)
Mg–1Sn Mg–3Sn Mg–5Sn Mg–7Sn
0.052 ± 0.005 0.094 ± 0.011 0.190 ± 0.013 0.295 ± 0.004
250
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
Fig. 8. Potentiodynamic polarization curves of the AZ31 and Mg–Sn alloys after immersion in Hank's solution with different time. (a) Immersion for 900 s and (b) immersion for 3600 s.
Table 2 Values measured from the polarization curves for the AZ31 and Mg–Sn alloys after immersion in Hank's solution for 900 s and 3600 s. Alloy
Ecorr (VSCE)
icorr (μA/cm2)
Ri (mm/year)
900 s AZ31 Mg–1Sn Mg–3Sn Mg–5Sn Mg–7Sn
−1.512 ± 0.003 −1.641 ± 0.018 −1.563 ± 0.003 −1.531 ± 0.011 −1.522 ± 0.009
Ecorr (VSCE)
icorr (μA/cm2)
Ri (mm/year)
15.25 ± 2.046 5.313 ± 0.569 7.343 ± 1.436 14.726 ± 3.514 22.183 ± 10.372
0.348 ± 0.047 0.121 ± 0.013 0.168 ± 0.033 0.337 ± 0.080 0.507 ± 0.237
3600 s 34.293 ± 0.27 8.393 ± 0.826 12.36 ± 2.649 21.75 ± 4.973 28.797 ± 10.937
0.784 ± 0.006 0.191 ± 0.019 0.282 ± 0.061 0.497 ± 0.114 0.658 ± 0.250
results of potentiodynamic polarization test were in consistent with that of immersion test. Secondly, the icorr of the Mg–Sn alloys with the same Sn content decreased over time, indicating increased corrosion resistance. This phenomenon might be caused by the corrosion products as the partial protective film on the surface of the samples after longer immersion time. Thirdly, icorr of the Mg–Sn alloys with low Sn content (≤5 wt.%) was smaller that that of AZ31 sample during the whole immersion time, indicating better corrosion resistance of the Mg–Sn alloys with low Sn content.
3.4. Cytotoxicity test Mg–1Sn and Mg–3Sn alloys were selected to evaluate their cytotoxicity by examining both the viability and morphologies of MG63 cells. Fig. 9 showed the cell viability of MG63 cells cultured in Mg–1Sn and
−1.496 ± 0.008 −1.605 ± 0.025 −1.544 ± 0.013 −1.532 ± 0.005 −1.521 ± 0.002
Mg–3Sn alloy extracts for 2, 4 and 6 days compared with the negative control. It could be seen that cells cultured in both Mg–1Sn and Mg– 3Sn alloy extracts had a comparable and even higher absorbance than the control during 2, 4 and 6 days. Although there was no significant difference between the MG63 cells in the extracts and those in the negative control, the results demonstrated that MG63 cells exhibited good growth in both Mg–1Sn and Mg–3Sn alloy extracts and these alloys met the requirement of cell toxicity according to ISO 10993-5:1999 [29]. Fig. 10 showed the morphologies of the MG63 cells cultured in negative control, Mg–1Sn and Mg–3Sn alloy extracts for 2, 4 and 6 days. Normal and healthy morphologies of MG63 cells were observed in both Mg–1Sn and Mg–3Sn alloy extracts, which was similar to that of the negative control. The results of cytotoxicity test showed that both Mg–1Sn and Mg– 3Sn alloys had good cytocompatibility and were safe for biomedical applications. 4. Conclusion The addition of Sn element significantly refined the secondary dendrite arm spacing of the Mg–Sn alloys prepared by sub-rapid solidification and increased the amount of the second phase. The corrosion resistance of the Mg–Sn alloys, which was related to the volume fraction and existence format of secondary phases, decreased with increasing Sn content. The in vitro cytotoxicity evaluation using MG63 cells revealed that the Mg– 1Sn and Mg–3Sn alloys were safe as biodegradable implant alloys. Acknowledgments
Fig. 9. Cell viability of MG63 expressed as a percentage of the viability of cells in the negative control after 2, 4 and 6 days incubation in Mg–1Sn and Mg–3Sn alloy extracts.
The authors acknowledged the financial support from the Ministry of Science & Technology of China (973 Project 2013CB632200 and Key Technologies Project 2012BAF09B04), and the Chongqing Municipal Government (CSTC2013JCYJC60001, Two River Scholar Project and The Chief Scientist Studio Project, and Chongqing Postdoctoral Science Foundation Special funded Project XM20120041).
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
251
Fig. 10. Optical morphologies of MG63 cells after 2, 4 and 6 day incubation. (a) Negative control, (b) Mg–1Sn alloy extracts and (c) Mg–3Sn alloy extracts.
References [1] N. Li, Y. Zheng, J. Mater. Sci. Technol. 29 (2013) 489–502. [2] M.P. Staiger, A.M. Pietak, J. Huadmai, G. Dias, Biomaterials 27 (2006) 1728–1734. [3] F. Witte, N. Hort, C. Vogt, S. Cohen, K.U. Kainer, R. Willumeit, F. Feyerabend, Curr. Opin. Solid State Mater. Sci. 12 (2008) 63–72. [4] Y. Zheng, X. Gu, F. Witte, Mater. Sci. Eng. R 77 (2014) 1–34. [5] Y. Song, E.H. Han, K. Dong, D. Shan, C.D. Yim, B.S. You, Corros. Sci. 72 (2013) 133–143. [6] F. Witte, J. Fischer, J. Nellesen, H.A. Crostack, V. Kaese, A. Pisch, F. Beckmann, H. Windhagen, Biomaterials 27 (2006) 1013–1018. [7] F. Witte, V. Kaese, H. Haferkamp, E. Switzer, A. Meyer-Lindenberg, C. Wirth, H. Windhagen, Biomaterials 26 (2005) 3557–3563. [8] S.S.A. El-Rahman, Pharmacol. Res. 47 (2003) 189–194. [9] Y. Nakamura, Y. Tsumura, Y. Tonogai, T. Shibata, Y. Ito, Toxicol. Sci. 37 (1997) 106–116. [10] X. Gu, Y. Zheng, Y. Cheng, S. Zhong, T. Xi, Biomaterials 30 (2009) 484–498. [11] G. Song, Corros. Sci. 49 (2007) 1696–1701. [12] Y. Wan, G. Xiong, H. Luo, F. He, Y. Huang, X. Zhou, Mater. Des. 29 (2008) 2034–2037. [13] H.R.B. Rad, M.H. Idris, M.R.A. Kadir, S. Farahany, Mater. Des. 33 (2012) 88–97. [14] Z. Li, X. Gu, S. Lou, Y. Zheng, Biomaterials 29 (2008) 1329–1344. [15] H.S. Brar, J. Wong, M.V. Manuel, J. Mech. Behav. Biomed. Mater. 7 (2012) 87–95. [16] Y. Li, C. Wen, D. Mushahary, R. Sravanthi, N. Harishankar, G. Pande, P. Hodgson, Acta Biomater. 8 (2012) 3177–3188. [17] X. Gu, X. Xie, N. Li, Y. Zheng, L. Qin, Acta Biomater. 8 (2012) 2360–2374. [18] S. Cai, T. Lei, N. Li, F. Feng, Mater. Sci. Eng. C 32 (2012) 2570–2577.
[19] S. Zhang, X. Zhang, C. Zhao, J. Li, Y. Song, C. Xie, H. Tao, Y. Zhang, Y. He, Y. Jiang, Acta Biomater. 6 (2010) 626–640. [20] E. Zhang, D. Yin, L. Xu, L. Yang, K. Yang, Mater. Sci. Eng. C 29 (2009) 987–993. [21] E. Zhang, L. Yang, Mater. Sci. Eng. A 497 (2008) 111–118. [22] I.S. Berglund, H.S. Brar, N. Dolgova, A.P. Acharya, B.G. Keselowsky, M. Sarntinoranont, M.V. Manuel, J. Biomed. Mater. Res. B 100 (2012) 1524–1534. [23] H. Du, Z. Wei, X. Liu, E. Zhang, Mater. Chem. Phys. 125 (2011) 568–575. [24] F. Rosalbino, S. De Negri, A. Saccone, E. Angelini, S. Delfino, J. Mater. Sci. Mater. Med. 21 (2010) 1091–1098. [25] H. Liu, Y. Chen, Y. Tang, S. Wei, G. Niu, J. Alloys Compd. 440 (2007) 122–126. [26] H. Zhang, D. Zhang, C. Ma, S. Guo, Mater. Lett. 92 (2013) 45–48. [27] ASTM-G31-72, Standard practice for laboratory immersion corrosion testing of metals, Annual Book of ASTM Standards, American Society for Testing and Materials, Philadelphia, Pennsylvania, USA, 2004. [28] Z. Shi, M. Liu, A. Andrej, Corros. Sci. 22 (2007) 1806–1814. [29] ISO-10993-5, Biological Evaluation of Medical Devices — Part 5: Tests for Cytotoxicity: In Vitro Methods, ANSI/AAMI, Arlington, VA, 1999. [30] T.B. Massalski, H. Okamoto, P. Subramanian, L. Kacprzak, Binary Alloy Phase Diagrams, ASM international, 1990. [31] G.L. Song, Z. Xu, Electrochim. Acta 55 (2010) 4148–4161. [32] G. Ben-Hamu, D. Eliezer, K. Shin, S. Cohen, J. Alloys Compd. 431 (2007) 269–276. [33] A. Pardo, M. Merino, A. Coy, F. Viejo, R. Arrabal, S. Feliú Jr., Electrochim. Acta 53 (2008) 7890–7902. [34] T. Zhang, Y. Shao, G. Meng, Z. Cui, F. Wang, Corros. Sci. 53 (2011) 1960–1968. [35] Y. Song, E.H. Han, D. Shan, C.D. Yim, B.S. You, Corros. Sci. 60 (2012) 238–245. [36] G. Song, A. Atrens, Adv. Eng. Mater. 5 (2003) 837–858.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Microstructure, corrosion behavior and cytotoxicity of biodegradable Mg–Sn implant alloys prepared by sub-rapid solidification Chaoyong Zhao a, Fusheng Pan a,b,c,⁎, Shuang Zhao a, Hucheng Pan a, Kai Song a, Aitao Tang a,b a b c
College of Materials Science and Engineering, Chongqing University, Chongqing 400044, China National Engineering Research Center for Magnesium Alloys, Chongqing University, Chongqing 400044, China Chongqing Academy of Science and Technology, Chongqing 401123, China
a r t i c l e
i n f o
Article history: Received 2 August 2014 Received in revised form 22 April 2015 Accepted 11 May 2015 Available online 13 May 2015 Keywords: Magnesium alloy Corrosion Biodegradable Tin Orthopedic implant
a b s t r a c t In this study, biodegradable Mg–Sn alloys were fabricated by sub-rapid solidification, and their microstructure, corrosion behavior and cytotoxicity were investigated by using optical microscopy, scanning electron microscopy equipped with an energy dispersive X-ray spectroscopy, X-ray diffraction, immersion test, potentiodynamic polarization test and cytotoxicity test. The results showed that the microstructure of Mg–1Sn alloy was almost equiaxed grain, while the Mg–Sn alloys with higher Sn content (Sn ≥ 3 wt.%) displayed α-Mg dendrites, and the secondary dendrite arm spacing of the primary α-Mg decreased significantly with increasing Sn content. The Mg–Sn alloys consisted of primary α-Mg matrix, Sn-rich segregation and Mg2Sn phase, and the amount of Mg2Sn phases increased with increasing Sn content. Potentiodynamic polarization and immersion tests revealed that the corrosion rates of Mg–Sn alloys increased with increasing Sn content. Cytotoxicity test showed that Mg– 1Sn and Mg–3Sn alloys were harmless to MG63 cells. These results of the present study indicated that Mg–1Sn and Mg–3Sn alloys were promising to be used as biodegradable implants. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Magnesium and its alloys have obtained increasing attention as biodegradable orthopedic implants due to their good biocompatibility, similar mechanical properties to natural bone and biodegradability in human physiological environment [1–4]. However, the rapid corrosion rate of magnesium alloys is a major obstacle to their clinical application [2]. The undesirable corrosion resistance of magnesium originates from the loose and porous oxide films on its surface, which only provide limited protection to the magnesium substrate [5]. Element alloying was usually used to improve the corrosion resistance of magnesium alloys [2]. The addition of aluminum (Al) and/or rare earth (RE) elements to magnesium were examples [6,7]. Unfortunately, Al and/or RE might induce latent toxic and harmful effects on the human body [8,9]. Therefore, alloying elements must be chosen with careful consideration of possible toxic effects [10,11]. At present, one or more elements with non-toxicity or low toxicity were chosen as alloying elements to improve the corrosion resistance of magnesium, such as Mg–Ca, Mg–Zn, Mg–Sr, Mg–Zr–Sr, Mg–Ca–Sr, Mg–Zn–Sr, Mg– Zn–Mn, Mg–Zn–Ca, Mg–Zn–Mn–Ca [12–24]. However, the corrosion resistance of these alloys needed to be further improved for clinic applications [7,14,19]. ⁎ Corresponding author at: College of Materials Science and Engineering, Chongqing University, Chongqing 400044, China. E-mail address: [email protected] (F. Pan).
http://dx.doi.org/10.1016/j.msec.2015.05.042 0928-4931/© 2015 Elsevier B.V. All rights reserved.
Tin (Sn) was one of the most essential elements in the human body. Gu et al. studied the in vitro corrosion and biocompatibility of Mg–1Sn alloy and conjectured that Sn might be a good alloying element for implant magnesium alloys [10]. Liu et al. fabricated the Mg–Sn alloys using the conventional casting method, and investigated its microstructure and mechanical properties. They reported that tin addition showed a refinement effect on the secondary dendrite arm spacing of the α-Mg phase, and Mg–5Sn alloy had the best mechanical properties [25]. However, there was lack of research about the effect of volume fraction and existence format of secondary phases on the corrosion behavior of Mg– Sn alloys. Furthermore, the cytotoxicity evaluation of the Mg–Sn alloys was also important for biomedical applications, which was scarce at present. In this study, Mg–Sn alloys were prepared by a sub-rapid solidification process [26], and their microstructure, corrosion behavior and cytotoxicity were investigated for orthopedic applications. 2. Materials and methods 2.1. Materials preparation The Mg–Sn alloys used for this investigation were the Mg–1Sn, Mg– 3Sn, Mg–5Sn and Mg–7Sn alloys with nominal composition of 1 wt.% Sn, 3 wt.% Sn, 5 wt.% Sn and 7 wt.% Sn, respectively. They were prepared using pure magnesium (99.98 wt.%) and pure tin (99.9 wt.%) ingots. The melting process was carried out in a middle-frequency induction
246
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
Fig. 1. Optical micrographs of the Mg–Sn alloys with different Sn content. (a) Mg–1Sn, (b) Mg–3Sn, (c) Mg–5Sn and (d) Mg–7Sn.
furnace under a mixed gas atmosphere of SF6 and CO2, and the stainless steel crucible with the melt was quenched in salt water [26]. Disk (10 × 10 × 2 mm3) samples were machined from the ingots and ground using silicon carbide (SiC) grinding paper up to 1000 grit for the microstructure characterization and immersion test. 2.2. Microstructure characterization Microstructure observation of samples was carried out on an optical microscopy and a scanning electron microscopy (SEM, TESCAN VEGA II LMU) equipped with an energy dispersive X-ray spectroscopy (EDX). The phase of samples was examined using an X-ray diffraction (XRD, Rigaku D/MAX-2500PC). 2.3. Immersion and potentiodynamic polarization tests The weight of the Mg–Sn alloy samples before immersion test was measured. The samples were exposed to Hank's solution for 500 h at 37 °C [21], and the ratio of sample surface area to the volume of Hank's
solution was 1 cm2: 20 ml according to ASTMG31-72 standard [27]. After immersion, the immersed samples were removed, rinsed and dried. The corrosion morphologies and chemical composition of the immersed samples were characterized by SEM and XRD, respectively. Some immersed samples were cleaned with chromate acid (200 g/l CrO3 + 10 g/l AgNO3) to remove their corrosion products, and the weight of the cleaned samples was also measured. The corrosion rate (CR) was calculated in mg cm−2 d−1 according to the literature's equation [18,28]. The weight loss rate could be converted to the average corrosion rate (Rw, mm/year) according to the following reaction [18,28]: Rw ¼ 2:1 CR : Potentiodynamic polarization test was performed in Hank's solution using a three-electrode cell with a saturated calomel electrode as the reference electrode, a platinum plate as the counter electrode and the as-prepared sample with an exposed area of 1.13 cm2 as the working electrode. In addition, the commercial AZ31 magnesium alloy was used as a control. All samples were ground using silicon carbide (SiC) grinding paper up to 1000 grit, cleaned and dried. The sample was immersed in Hank's solution for 900 s or 3600 s before undertaking the test. The potentiodynamic curves were obtained using an electrochemical workstation (CHI660E, China). The measurements started from − 300 mV vs. open circuit potential (OCP) at a constant scan rate of 1 mV s−1 and terminated until +300 mV vs. OCP. An average of three measurements was taken for each group. The corrosion current density (icorr, mA/cm2) was related to the average corrosion rate (Ri, mm/year) according to the following reaction [18,28]: Ri ¼ 22:85 icorr :
2.4. Cytotoxicity test
Fig. 2. XRD patterns of the Mg–Sn alloys with different Sn content.
MG63 cells and Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum were used in this study. The cytotoxicity of the Mg–Sn alloys was evaluated by indirect cell viability assay. The ion extracts were obtained from 10 × 10 × 2 mm3 of the Mg–Sn alloy samples by incubating in 2.8 ml medium for 24 h in a humidified atmosphere of 5% CO2 at 37 °C [29]. The DMEM medium was used as the negative control. The MG63 cells were seeded onto 96-
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
247
Fig. 3. SEM micrographs of the Mg–Sn alloys with different Sn content. (a) Mg–1Sn, (b) Mg–3Sn, (c) Mg–5Sn and (d) Mg–7Sn.
well cell culture plates at 5 × 103 cells per 100 μl medium in each well and incubated 24 h in a humidified atmosphere with 5% CO2 at 37 °C for cell attachment. Then, the medium in each well was replaced with 100 μl of extract or DMEM medium, and incubated for 2, 4 and 6 days. At the end of each time point, cell morphology was observed by optical microscopy. After that, 20 μl MTT solution (5 mg/ml) was added to each well and further incubated for 4 h. The medium was then replaced by 150 μl dimethyl sulfoxide (DMSO). The optical density (OD) measurements of 100 μl supernatant were conducted at 490 nm.
3. Results and discussion 3.1. Microstructure of the Mg–Sn alloys Fig. 1 showed the optical metallographic images of the Mg–Sn alloys. It could be seen that the Mg–1Sn alloy was almost equiaxed grain structure (Fig. 1a), while the Mg–Sn alloys with higher Sn content (Sn ≥ 3 wt.%) displayed α-Mg dendrites, and the secondary dendrite arm spacing of the primary α-Mg decreased with increasing Sn content. This trend of the refinement of the secondary dendrite arm spacing by the addition of tin was similar to the previous report [25]. The formation of dendrites was that tin was partitioned into the liquid ahead of the solidification front, resulting in a constitutional supercooling zone of liquid ahead of the interface [25]. The constitutional supercooling resulted from the tin enrichment promoted nucleation and hindered the fast growth of grain, thus showing a refinement effect in Mg–Sn alloys [25]. More tin addition caused a stronger effect of constitutional supercooling, which resulted in the decrease of the secondary dendrite arm spacing [25].
The XRD patterns of the Mg–Sn alloys were shown in Fig. 2. Only peaks corresponding to α-Mg phase were found in the XRD patterns for Mg–1Sn and Mg–3Sn alloys probably due to the low volume fractions of the second phase. On the other hand, Mg2Sn peaks could be clearly identified in the Mg–5Sn and Mg–7Sn alloys. It could also be observed that the diffraction intensities of Mg2Sn phases increased with increasing Sn content. Figs. 3 and 4 showed the SEM images and EDS analysis of the Mg–Sn alloys with different Sn content. It could be observed that the microstructure of the Mg–1Sn alloy consisted of primary α-Mg and Sn-rich segregation, as was shown in Figs. 3a and 4a. The chemical composition of Sn-rich segregation identified by EDS (Fig. 4d) showed that the atomic ratio of Mg to Sn was about 74:1, which was not likely attributed to the Mg2Sn phase. With increasing Sn content up to 3 wt.%, the alloy was mainly composed of primary α-Mg, Sn-rich net-segregation and Mg2Sn second phase. The volume fraction of the second phase (white area) increased considerably with further increasing Sn content. The high magnification SEM images of Mg–7Sn alloy in Fig. 4b and c showed that two types of second phase presented in this alloy, and their atomic ratio of Mg to Sn were about 3.5/1 and 4.8/1, respectively, which deviated from that of Mg2Sn. It was due to the small size of the second phase and the α-Mg matrix. According to Mg–Sn binary diagram, the saturation solid solubility of tin in magnesium was 14.85 wt.% at 561.2 °C, which decreased sharply to 0.45 wt.% at 200 °C and almost zero at room temperature [30]. The saturation solid solubility decreased fast with decreasing temperature, which caused the tin atoms to precipitate and formed the secondary phase Mg2Sn during the solidification process [30]. However, the microstructure of the Mg–Sn alloys in the present study differed from that of the previous study due to the different solidification method [25]. In the
248
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
Fig. 4. SEM micrographs and EDS analysis of the Mg–1Sn and Mg–7Sn alloys. (a) Mg–1Sn, (b) and (c) Mg–7Sn, (d) EDS analysis corresponding to assigned A area, (e) EDS analysis corresponding to assigned B area, (f) EDS analysis corresponding to assigned C area.
present study, much Sn-rich segregation was presented besides Mg2Sn phases in the Mg–Sn alloys. It was probably that the solidification rate of the Mg–Sn alloys was faster in the salt water than in the iron mold, so the Sn-rich segregation could not transform into Mg2Sn phase.
[17]. The Mg–Sn alloys dissolved according to the following reaction [13]:
3.2. Immersion test
As chloride ions presented in the Hank's solution, Mg(OH)2 would react with Cl− to form highly soluble magnesium chloride and OH−, which resulted in the increase of pH value of Hank's solution [2,13,18]. XRD patterns of Mg–Sn alloys after immersion in Hank's solution for 500 h in Fig. 6 revealed that Mg(OH)2 existed in the corrosion products and the diffraction intensities of Mg(OH)2 increased with increasing Sn content. The undissolved Mg(OH)2 in the corrosion products on the surfaces of Mg alloy might result from their rapid corrosion rate, and different corrosion rate of Mg–Sn alloys led to different diffraction intensities. 2+ , etc. in Hank's solution and high pH led to the Furthermore, PO3− 4 , Ca formation of the corrosion products of phosphates containing magnesium/calcium [19]. Fig. 7 showed the weight loss rate of the Mg–Sn alloys after immersion in Hank's solution for 500 h. The weight loss of magnesium alloy was an indication of corrosion [18]. The correlated average corrosion rate (Rw, mm/year) was listed in Table 1. It could be seen in Fig. 7 and Table 1 that the weight loss rate of the Mg–Sn alloys increased in the order of Mg–1Sn b Mg–3Sn b Mg–5Sn b Mg–7Sn, indicating that Mg– 1Sn alloy showed the best corrosion resistance and Mg–7Sn alloy showed the worst one. The results of immersion test clearly indicated that the corrosion resistance of Mg–Sn alloys decreased with increasing Sn content.
Fig. 5 showed the surface morphologies of the Mg–Sn alloys after immersion in Hank's solution for 500 h. It could be observed that the surfaces of the Mg–1Sn, Mg–3Sn and Mg–5Sn alloys were covered by a black film and white corrosion products, and the amount of the white corrosion products increased with increasing Sn content, indicating that the corrosion resistance of the Mg–Sn alloys decreased with increasing Sn content. A large amount of deep corrosion pits presented on the surface of Mg–7Sn alloy, which indicated that it showed a higher corrosion rate than the other Mg–Sn alloys. The SEM images of Mg–1Sn alloy at a higher magnification revealed that cracks could be seen in the black film as a result of dehydration of the surface layer in the air [13]. It was well-known that the corrosion rate of magnesium and its alloys depended on their microstructure [31–35]. As shown in the microstructure characterization, the volume fraction and existence format of secondary phases of Mg–Sn alloys differed with increasing Sn content (Figs. 2–4). The microstructure of the Mg–1Sn alloy consisted of primary α-Mg and Sn-rich segregation (Figs. 3a and 4a). Besides primary αMg and Sn-rich net-segregation, Mg2Sn phase appeared when the Sn content increased to 3 wt.%, and the volume fraction of the second phases increased with increasing Sn content. In general, the second phase had more positive corrosion potential than that of the α-phase [36]. Therefore, the second phases might act as the cathode and the αMg as the anode, which resulted in galvanic corrosion after immersion in Hank's solution. Since an increase of the second phases would lead to an increase in micro-galvanic couples, this could explain why the corrosion rate of the Mg–Sn alloys increased with increasing Sn content
Mg þ 2H2 O ¼ MgðOHÞ2 þ H2 ↑:
3.3. Potentiodynamic polarization test Fig. 8 showed the potentiodynamic polarization curves of the AZ31 and Mg–Sn alloys after immersion in Hank's solution with different time. Table 2 listed the electrochemical parameters of the AZ31 and
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
249
Fig. 5. Surface morphologies of the Mg–Sn alloys after immersion in Hank's solution for 500 h. (a) Mg–1Sn, (b) Mg–3Sn, (c) Mg–5Sn and (d) Mg–7Sn. The inserted photos were enlarged views of local area.
Mg–Sn alloys in Hank's solution obtained from the polarization curves by Tafel extrapolation and the correlated average corrosion rate (Ri, mm/year). Some information could be obtained from Fig. 8 and Table 2. Firstly, the increase of the icorr of the Mg–Sn alloys was observed with increasing Sn content during the whole immersion time, which indicated that the increase of corrosion rates of the Mg–Sn alloys in the following order: Mg–1Sn b Mg–3Sn b Mg–5Sn b Mg–7Sn [13]. The
Fig. 7. Weight loss rate of the Mg–Sn alloys after immersion in Hank's solution for 500 h.
Table 1 Values measured from weight loss rate for Mg–Sn alloys in Hank's solution.
Fig. 6. XRD patterns of the Mg–Sn alloys after immersion in Hank's solution for 500 h.
Alloy
Rw (mm/year)
Mg–1Sn Mg–3Sn Mg–5Sn Mg–7Sn
0.052 ± 0.005 0.094 ± 0.011 0.190 ± 0.013 0.295 ± 0.004
250
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
Fig. 8. Potentiodynamic polarization curves of the AZ31 and Mg–Sn alloys after immersion in Hank's solution with different time. (a) Immersion for 900 s and (b) immersion for 3600 s.
Table 2 Values measured from the polarization curves for the AZ31 and Mg–Sn alloys after immersion in Hank's solution for 900 s and 3600 s. Alloy
Ecorr (VSCE)
icorr (μA/cm2)
Ri (mm/year)
900 s AZ31 Mg–1Sn Mg–3Sn Mg–5Sn Mg–7Sn
−1.512 ± 0.003 −1.641 ± 0.018 −1.563 ± 0.003 −1.531 ± 0.011 −1.522 ± 0.009
Ecorr (VSCE)
icorr (μA/cm2)
Ri (mm/year)
15.25 ± 2.046 5.313 ± 0.569 7.343 ± 1.436 14.726 ± 3.514 22.183 ± 10.372
0.348 ± 0.047 0.121 ± 0.013 0.168 ± 0.033 0.337 ± 0.080 0.507 ± 0.237
3600 s 34.293 ± 0.27 8.393 ± 0.826 12.36 ± 2.649 21.75 ± 4.973 28.797 ± 10.937
0.784 ± 0.006 0.191 ± 0.019 0.282 ± 0.061 0.497 ± 0.114 0.658 ± 0.250
results of potentiodynamic polarization test were in consistent with that of immersion test. Secondly, the icorr of the Mg–Sn alloys with the same Sn content decreased over time, indicating increased corrosion resistance. This phenomenon might be caused by the corrosion products as the partial protective film on the surface of the samples after longer immersion time. Thirdly, icorr of the Mg–Sn alloys with low Sn content (≤5 wt.%) was smaller that that of AZ31 sample during the whole immersion time, indicating better corrosion resistance of the Mg–Sn alloys with low Sn content.
3.4. Cytotoxicity test Mg–1Sn and Mg–3Sn alloys were selected to evaluate their cytotoxicity by examining both the viability and morphologies of MG63 cells. Fig. 9 showed the cell viability of MG63 cells cultured in Mg–1Sn and
−1.496 ± 0.008 −1.605 ± 0.025 −1.544 ± 0.013 −1.532 ± 0.005 −1.521 ± 0.002
Mg–3Sn alloy extracts for 2, 4 and 6 days compared with the negative control. It could be seen that cells cultured in both Mg–1Sn and Mg– 3Sn alloy extracts had a comparable and even higher absorbance than the control during 2, 4 and 6 days. Although there was no significant difference between the MG63 cells in the extracts and those in the negative control, the results demonstrated that MG63 cells exhibited good growth in both Mg–1Sn and Mg–3Sn alloy extracts and these alloys met the requirement of cell toxicity according to ISO 10993-5:1999 [29]. Fig. 10 showed the morphologies of the MG63 cells cultured in negative control, Mg–1Sn and Mg–3Sn alloy extracts for 2, 4 and 6 days. Normal and healthy morphologies of MG63 cells were observed in both Mg–1Sn and Mg–3Sn alloy extracts, which was similar to that of the negative control. The results of cytotoxicity test showed that both Mg–1Sn and Mg– 3Sn alloys had good cytocompatibility and were safe for biomedical applications. 4. Conclusion The addition of Sn element significantly refined the secondary dendrite arm spacing of the Mg–Sn alloys prepared by sub-rapid solidification and increased the amount of the second phase. The corrosion resistance of the Mg–Sn alloys, which was related to the volume fraction and existence format of secondary phases, decreased with increasing Sn content. The in vitro cytotoxicity evaluation using MG63 cells revealed that the Mg– 1Sn and Mg–3Sn alloys were safe as biodegradable implant alloys. Acknowledgments
Fig. 9. Cell viability of MG63 expressed as a percentage of the viability of cells in the negative control after 2, 4 and 6 days incubation in Mg–1Sn and Mg–3Sn alloy extracts.
The authors acknowledged the financial support from the Ministry of Science & Technology of China (973 Project 2013CB632200 and Key Technologies Project 2012BAF09B04), and the Chongqing Municipal Government (CSTC2013JCYJC60001, Two River Scholar Project and The Chief Scientist Studio Project, and Chongqing Postdoctoral Science Foundation Special funded Project XM20120041).
C. Zhao et al. / Materials Science and Engineering C 54 (2015) 245–251
251
Fig. 10. Optical morphologies of MG63 cells after 2, 4 and 6 day incubation. (a) Negative control, (b) Mg–1Sn alloy extracts and (c) Mg–3Sn alloy extracts.
References [1] N. Li, Y. Zheng, J. Mater. Sci. Technol. 29 (2013) 489–502. [2] M.P. Staiger, A.M. Pietak, J. Huadmai, G. Dias, Biomaterials 27 (2006) 1728–1734. [3] F. Witte, N. Hort, C. Vogt, S. Cohen, K.U. Kainer, R. Willumeit, F. Feyerabend, Curr. Opin. Solid State Mater. Sci. 12 (2008) 63–72. [4] Y. Zheng, X. Gu, F. Witte, Mater. Sci. Eng. R 77 (2014) 1–34. [5] Y. Song, E.H. Han, K. Dong, D. Shan, C.D. Yim, B.S. You, Corros. Sci. 72 (2013) 133–143. [6] F. Witte, J. Fischer, J. Nellesen, H.A. Crostack, V. Kaese, A. Pisch, F. Beckmann, H. Windhagen, Biomaterials 27 (2006) 1013–1018. [7] F. Witte, V. Kaese, H. Haferkamp, E. Switzer, A. Meyer-Lindenberg, C. Wirth, H. Windhagen, Biomaterials 26 (2005) 3557–3563. [8] S.S.A. El-Rahman, Pharmacol. Res. 47 (2003) 189–194. [9] Y. Nakamura, Y. Tsumura, Y. Tonogai, T. Shibata, Y. Ito, Toxicol. Sci. 37 (1997) 106–116. [10] X. Gu, Y. Zheng, Y. Cheng, S. Zhong, T. Xi, Biomaterials 30 (2009) 484–498. [11] G. Song, Corros. Sci. 49 (2007) 1696–1701. [12] Y. Wan, G. Xiong, H. Luo, F. He, Y. Huang, X. Zhou, Mater. Des. 29 (2008) 2034–2037. [13] H.R.B. Rad, M.H. Idris, M.R.A. Kadir, S. Farahany, Mater. Des. 33 (2012) 88–97. [14] Z. Li, X. Gu, S. Lou, Y. Zheng, Biomaterials 29 (2008) 1329–1344. [15] H.S. Brar, J. Wong, M.V. Manuel, J. Mech. Behav. Biomed. Mater. 7 (2012) 87–95. [16] Y. Li, C. Wen, D. Mushahary, R. Sravanthi, N. Harishankar, G. Pande, P. Hodgson, Acta Biomater. 8 (2012) 3177–3188. [17] X. Gu, X. Xie, N. Li, Y. Zheng, L. Qin, Acta Biomater. 8 (2012) 2360–2374. [18] S. Cai, T. Lei, N. Li, F. Feng, Mater. Sci. Eng. C 32 (2012) 2570–2577.
[19] S. Zhang, X. Zhang, C. Zhao, J. Li, Y. Song, C. Xie, H. Tao, Y. Zhang, Y. He, Y. Jiang, Acta Biomater. 6 (2010) 626–640. [20] E. Zhang, D. Yin, L. Xu, L. Yang, K. Yang, Mater. Sci. Eng. C 29 (2009) 987–993. [21] E. Zhang, L. Yang, Mater. Sci. Eng. A 497 (2008) 111–118. [22] I.S. Berglund, H.S. Brar, N. Dolgova, A.P. Acharya, B.G. Keselowsky, M. Sarntinoranont, M.V. Manuel, J. Biomed. Mater. Res. B 100 (2012) 1524–1534. [23] H. Du, Z. Wei, X. Liu, E. Zhang, Mater. Chem. Phys. 125 (2011) 568–575. [24] F. Rosalbino, S. De Negri, A. Saccone, E. Angelini, S. Delfino, J. Mater. Sci. Mater. Med. 21 (2010) 1091–1098. [25] H. Liu, Y. Chen, Y. Tang, S. Wei, G. Niu, J. Alloys Compd. 440 (2007) 122–126. [26] H. Zhang, D. Zhang, C. Ma, S. Guo, Mater. Lett. 92 (2013) 45–48. [27] ASTM-G31-72, Standard practice for laboratory immersion corrosion testing of metals, Annual Book of ASTM Standards, American Society for Testing and Materials, Philadelphia, Pennsylvania, USA, 2004. [28] Z. Shi, M. Liu, A. Andrej, Corros. Sci. 22 (2007) 1806–1814. [29] ISO-10993-5, Biological Evaluation of Medical Devices — Part 5: Tests for Cytotoxicity: In Vitro Methods, ANSI/AAMI, Arlington, VA, 1999. [30] T.B. Massalski, H. Okamoto, P. Subramanian, L. Kacprzak, Binary Alloy Phase Diagrams, ASM international, 1990. [31] G.L. Song, Z. Xu, Electrochim. Acta 55 (2010) 4148–4161. [32] G. Ben-Hamu, D. Eliezer, K. Shin, S. Cohen, J. Alloys Compd. 431 (2007) 269–276. [33] A. Pardo, M. Merino, A. Coy, F. Viejo, R. Arrabal, S. Feliú Jr., Electrochim. Acta 53 (2008) 7890–7902. [34] T. Zhang, Y. Shao, G. Meng, Z. Cui, F. Wang, Corros. Sci. 53 (2011) 1960–1968. [35] Y. Song, E.H. Han, D. Shan, C.D. Yim, B.S. You, Corros. Sci. 60 (2012) 238–245. [36] G. Song, A. Atrens, Adv. Eng. Mater. 5 (2003) 837–858.