Acta Biomaterialia 10 (2014) 2333–2340
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat
Loss of mechanical properties in vivo and bone–implant interface strength of AZ31B magnesium alloy screws with Si-containing coating Lili Tan a,⇑, Qiang Wang a,b, Xiao Lin a, Peng Wan a, Guangdao Zhang b, Qiang Zhang c, Ke Yang a,⇑ a
Institute of Metal Research, Chinese Academy of Sciences, 72 Wenhua Road, Shenyang 110016, People’s Republic of China School of Stomatology, China Medical University, Shenyang 110002, People’s Republic of China c Department of Orthopaedics, General Hospital of the People’s Liberation Army, Beijing 100853, People’s Republic of China b
a r t i c l e
i n f o
Article history: Received 30 October 2012 Received in revised form 9 December 2013 Accepted 11 December 2013 Available online 19 December 2013 Keywords: Magnesium alloy Si-containing coating Mechanical properties loss Bone–implant interface strength
a b s t r a c t In this study the loss of mechanical properties and the interface strength of coated AZ31B magnesium alloy (a magnesium–aluminum alloy) screws with surrounding host tissues were investigated and compared with non-coated AZ31B, degradable polymer and biostable titanium alloy screws in a rabbit animal model after 1, 4, 12 and 21 weeks of implantation. The interface strength was evaluated in terms of the extraction torque required to back out the screws. The loss of mechanical properties over time was indicated by one-point bending load loss of the screws after these were extracted at different times. AZ31B samples with a silicon-containing coating had a decreased degradation rate and improved biological properties. The extraction torque of Ti6Al4V, poly-L-lactide (PLLA) and coated AZ31B increased significantly from 1 week to 4 weeks post-implantation, indicating a rapid osteosynthesis process over 3 weeks. The extraction torque of coated AZ31B increased with implantation time, and was higher than that of PLLA after 4 weeks of implantation, equalling that of Ti6Al4V at 12 weeks and was higher at 21 weeks. The bending loads of non-coated AZ31B and PLLA screws degraded sharply after implantation, and that of coated AZ31B degraded more slowly. The biodegradation mechanism, the coating to control the degradation rate and the bioactivity of magnesium alloys influencing the mechanical properties loss over time and bone–implant interface strength are discussed in this study and it is concluded that a suitable degradation rate will result in an improvement in the mechanical performance of magnesium alloys, making them more suitable for clinical application. Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
1. Introduction Research into Mg-based alloys used as biodegradable materials for orthopedics applications has attracted a great deal of interest. The key drawback of bio-absorbable Mg implants is their rapid degradation, resulting in hydrogen evolution, serious local alkalization, hemolysis and loss of their original strength before bone tissues have healed [1–3]. Researchers have devoted a tremendous amount of resources to investigating the rapid degradation problem of Mg alloys; surface modification is one of the key solutions and many modification methods have also been developed for biodegradable Mg alloys [4–10]. Several authors reported good biocompatibility and lower degradation rates of Mg alloys after modification [9,11–13]. Fluoride-coated Mg–calcium alloy was reported with optimized degradation kinetics in a subcutaneous mouse model [14]. A controllable polymeric membrane fabricated by polycaprolactone and dichloromethane was adopted to control ⇑ Corresponding authors. Tel./fax: +86 24 83971676 (L. Tan), +86 24 83971628 (K. Yang). E-mail addresses: [email protected] (L. Tan), [email protected] (K. Yang).
the corrosion rate of Mg alloys, and higher volumes of new bone were observed on the polymer-coated samples through in vivo study [15]. The result of in vivo implantations of the Ca–P coated and the non-coated alloy rods showed that the Ca–P coating provided Mg with significantly good surface bioactivity and promoted early bone growth at the bone–implant interface [16]. Mechanical performance is an important consideration for biodegradable Mg alloys: the fast degradation time also induces early mechanical loss before the tissue healing occurs. It is also reported that the three-point bending strength of AZ31B in simulated body fluid (SBF) was improved by fluoride conversion coating [17] and Ca–P coating [18]. However, the mechanical performance of modified Mg alloys in vivo has not yet been addressed. The mechanical properties of biodegradable Mg alloy implants are largely dependent not only on the original mechanical properties, but also on the way in which the mechanical properties of Mg alloy degrade after implantation as well as the interface strength of Mg alloy implants with surrounding host tissues. Thus, the aim of the present study was to evaluate the loss of mechanical properties over time and the interface strength of Mg alloy implants with surrounding host tissues; the effects of the
1742-7061/$ - see front matter Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.actbio.2013.12.020
2334
L. Tan et al. / Acta Biomaterialia 10 (2014) 2333–2340
surface modification and corrosion degradation mechanism on these tissues are discussed. Biodegradable poly-L-lactide (PLLA) and biostable titanium alloy are used as the control. The widely studied Mg alloy AZ31B was used in this study, and a Si-containing coating was prepared on the surface of AZ31B samples to decrease the biodegradation rate of the alloy. Moreover, silicon is an essential trace element in skeletal development [19,20]. An appropriate silicon concentration was effective in supporting the proliferation of osteoblast-like cells and actively stimulating a biological response in MG63 cells [21]. In vitro and in vivo studies have demonstrated that the silicon released from a biodegradable silicate ceramic could induce angiogenesis during bone regeneration [22]. Therefore, in this study the Si-containing coating on the surface of AZ31B samples was expected to improve the biological properties of the implants.
2. Experimental 2.1. Materials Extruded AZ31B Mg alloy (3 wt.% Al, 1.1 wt.% Zn, 0.70 wt.% Mn, 0.01 wt.% Si, 0.002 wt.% Fe, 0.008 wt.% Cu, 0.0008 wt.% Ni and balance Mg) with the microstructure shown in Fig. 1 was used as the biodegradable Mg alloy. AZ31B screws M2.6 6.5 mm in size were machined by Trauson Holdings Company Ltd. from extruded AZ31B. In order to decrease the degradation rate of AZ31B, samples were coated with a Si-containing coating; an image of an AZ31B screw with Si-containing coating is shown in Fig. 2. For the fabrication of the Si-containing coating, the alloy samples were etched in a mixed acid solution of 1.25 g l1 oxalic acid, 50% acetic acid and 25% nitric acid for 10 s at room temperature for surface activation, and subsequently immersed in a mixing solution of NaOH (5 g l1), Na2SiO3 (10 g l1) and Na4P2O7 (50 g l1) for 120 h at 60 °C, then washed with deionized water and dried at 40 °C. The micromorphology of the coating was characterized by scanning electron microscope (SEM) with an Hitachi S-3400N, and the elements and composition of the coating were characterized by energy dispersive spectroscopy (EDS) and X-ray photoelectron spectroscopy (XPS, Thermo VG ESCALAB250). The in vitro study was conducted according to ISO10998 to evaluate the in vitro biological properties of the samples before the in vivo study. The PLLA screws were provided by Chengdu Organic Chemicals Co. Ltd., Chinese Academy of
Fig. 1. The microstructure of extruded AZ31B used in this study.
Fig. 2. The AZ31B screw with Si-containing coating.
Sciences, and the Ti6Al4V screws (the same size as the AZ31B screws), provided by Trauson Holdings Company Ltd., were used as the control. 2.2. Degradation experiments The electrochemical corrosion test was investigated by potentiodynamic polarization using a PAR Verserstat3 potentiostat controlled by VersaStudio software. A three-electrode cell featuring a Pt counter-electrode and a saturated calomel reference electrode (SCE) was employed. The experiment was conducted at a sweep rate of 0.5 mV s1 within a scan range of ±0.25 V with reference to open circuit potential (OCP) using Hanks’ solution (8 g l1 NaCl, 0.06 g l1 KH2PO4, 0.4 g l1 KCl, 0.12 g l1 Na2HPO4, 0.14 g l1 CaCl2, 0.2 g l1 MgSO4, 0.35 g l1 NaHCO3, 1.0 g l1 glucose, pH 7.4). The temperature of the solution was kept at 37 °C. Ten tablets 10 mm in diameter and 4 mm high were immersed in Hanks’ solution at 37 ± 0.5 °C with an immersion ratio of 3 cm2 ml1 (sample surface area/fluid volume). The immersion solutions were refreshed every 24 h as a simplistic model of the fluid exchange in vivo. The pH variation of the Hanks’ solution was monitored before solution refreshment to measure the pH increase of the solution in 24 h. 2.3. Surgery procedure Ethical approval was obtained for the animal tests, which were conducted according to the ISO 10993-2:1992 animal welfare requirements. 16 adult white rabbits weighing 2.5–3.0 kg, 4 months old, were used and were divided into four groups through stochastic grouping. Before the surgery, all the rabbits were subjected to clinical checks. Each rabbit was anesthetized by an intramuscular injection of Sumian Xin (0.15 ml kg1 of body weight, Animal Husbandry Research Institute, Jilin, China). The mid-diaphyseal region of the femur was exposed through a lateral approach. Two holes (2 mm diameter) were bored in both femur shafts of each rabbit using a disinfected hand-operated drill with the longitudinal axis of the drill holes perpendicular to the longitudinal axis of the femoral diaphysis, 1 cm between the two holes. When boring the holes, the operation sites were cooled with physiological saline (0.9 wt.%). Then the screws, sterilized by ethylene oxide, were driven into the predrilled holes with a torque wrench (Tohnivhi), each rabbit having four different screws implanted. An image of one of the femur shafts with inserted screws is shown in Fig. 3. The torque needed to insert the screw was read from the dial of the torque wrench, which was controlled at a similar value by the same insertion rate for all the screw implantations to ensure the same initial condition for all the implants. The muscle layers, subcutaneous tissue and skin were subsequently sewed up.
L. Tan et al. / Acta Biomaterialia 10 (2014) 2333–2340
2335
2.6. Bone–implant interface morphology observation The interface morphology of the tissue adjacent to the implant sites was observed, to provide supporting information for the outcomes from the biomechanical assessments in this study. For observation analysis, tissue blocks containing the implants were fixed in 2.5% glutaraldehyde solution for 24 h, dehydrated in ascending grades of alcohol and embedded in the epoxy resin. Then the resin blocks with bone tissues were ground perpendicular to the long axis of the implant to obtain a cross-section of the implant and surrounding bone tissue. Finally, the cross-sections were polished with diamond polishing paste to remove the scratches. The bone–implant interface was characterized by SEM and EDS. 2.7. Statistical analysis
Fig. 3. One of the femur shafts of the rabbit with inserted screws.
Post-operatively, the rabbits were allowed to move freely in their cages without external support. All the rabbits received a subcutaneous injection of gentamycin 3 days after surgery. 2.4. Extraction torque measurements Rabbits were killed by anesthesia overdose at 1, 4, 12 and 21 weeks post-implantation. The implant sites were surgically exposed; the bone and soft tissues formed on top of the implants were carefully removed to expose only the screw nut and keep the tissue around the thread untouched. Subsequently, the screws were wound out of the bone using the torque wrench with the same rate for all the screws, and the peak extraction torque required to wind out the screws was read. 2.5. Mechanical properties measurements The one-point bending test was used to evaluate the mechanical properties of the screws after these had been extracted from the femur of the rabbits at different time points. The test model is shown in Fig. 4. The screw nut was immobilized by a mechanical fixture, and a pressure bar with a crosshead and a diameter of 2 mm was engaged to move downward at a rate of 0.5 mm min1, causing samples to bend to failure under the applied load. The peak loads of the screws with different implantation times were recorded to evaluate the loss of mechanical properties over time.
Fig. 4. One-point bending test model.
Data analysis and a bivariate correlation test (Pearson) were performed by SPSS (version 13.0). Results were presented as the mean ± standard deviation (SD). Comparisons between groups were made by the t-test. Significant difference was defined at either P < 0.01 or P < 0.05. 3. Results 3.1. Degradation and biological performance of AZ31B with Sicontaining coating The surface and cross-section morphologies of Si-containing coating and its EDS are shown in Fig. 5. It can be seen that the surface of the coating is uniform and scattered with small particles; the EDS shows that Mg, O and Si elements are contained in the coating. The thickness of the coating is 1.5–2 lm. From the XPS result shown in Fig. 6, we can see that the Si-containing coating is mainly composed of Mg2SiO4, MgO and SiO2. The polarization curves of AZ31B with and without Si-containing coating are shown in Fig. 7. The corrosion potential (Ecor) and the corrosion current density (icor) derived from the polarization curves are shown in Table 2. The degradation rates of the samples are also calculated from the formula according to ASTM G102-89(2004) as follows:
CR ¼ K 1
icor
q
EW;
ð1Þ
where CR is the corrosion rate (mm year1), icor is the corrosion current density (lA cm2), K1 is 3.27 103 (mm g lA1 cm1 year1), q is the density (g cm3) and EW is the equivalent weight. Compared to non-coated AZ31B alloy, the corrosion potential of the coating shifts positively, from 1.23 V for the AZ31B to 1.16 V for the coated AZ31B; hence the coated AZ31B can inhibit anodic dissolution considerably, indicating that corrosion occurs with difficulty. Furthermore, the corrosion current density of the coating (0.51 106 A cm2) is significantly lower than that of the noncoated AZ31B (11.34 106 A cm2), confirming that the corrosion resistance of the coated AZ31B is enhanced, which is due to a diminution of the cathodic hydrogen evolution reaction after the Mg substrate had been covered with Si-containing coating [23]. From the corrosion rate result, we can see that the corrosion rate of coated AZ31B (0.011 mm year1) is more than one order less than that of AZ31B (0.248 mm year1). The polarization analysis indicates that the coating can prevent the penetration of solution and effectively protect the AZ31B substrate from corrosion. Fig. 8 shows the pH value variation of Hanks’ solution immersed with samples for 2 months. Clearly, the pH value of the solution immersed with non-coated AZ31B increases rapidly to 11.3 after the first day of immersion, then the pH decreases gradually with further immersion time owing to the formation of Mg(OH)2 on
2336
L. Tan et al. / Acta Biomaterialia 10 (2014) 2333–2340
Fig. 5. Surface (a), cross-section (b) morphologies and EDS analysis (c) of the AZ31B alloy with Si-contained coating.
Fig. 6. XPS spectra of the Si-containing coating on AZ31B alloy sample: (a) XPS specific spectrum of O1s; (b) XPS specific spectrum of Mg1s; (c) XPS specific spectrum of Si2p.
the surface. Comparatively, the pH value of the coating is 8.38 on the first day, much lower than that of untreated AZ31B and with the increase of the immersion time, the pH values remain almost stable. Moreover, biological tests in vitro for AZ31B with Si-containing coating have also been investigated, and the results are shown in Table 1. It is found that the Si-containing coating has met the requirement of the standard biocompatibility tests according to ISO10993, including cytotoxicity (rank: 0), hemolytics (rank: 0),
systemic toxicity (none), sensitization (none) and irritation (rank:
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat
Loss of mechanical properties in vivo and bone–implant interface strength of AZ31B magnesium alloy screws with Si-containing coating Lili Tan a,⇑, Qiang Wang a,b, Xiao Lin a, Peng Wan a, Guangdao Zhang b, Qiang Zhang c, Ke Yang a,⇑ a
Institute of Metal Research, Chinese Academy of Sciences, 72 Wenhua Road, Shenyang 110016, People’s Republic of China School of Stomatology, China Medical University, Shenyang 110002, People’s Republic of China c Department of Orthopaedics, General Hospital of the People’s Liberation Army, Beijing 100853, People’s Republic of China b
a r t i c l e
i n f o
Article history: Received 30 October 2012 Received in revised form 9 December 2013 Accepted 11 December 2013 Available online 19 December 2013 Keywords: Magnesium alloy Si-containing coating Mechanical properties loss Bone–implant interface strength
a b s t r a c t In this study the loss of mechanical properties and the interface strength of coated AZ31B magnesium alloy (a magnesium–aluminum alloy) screws with surrounding host tissues were investigated and compared with non-coated AZ31B, degradable polymer and biostable titanium alloy screws in a rabbit animal model after 1, 4, 12 and 21 weeks of implantation. The interface strength was evaluated in terms of the extraction torque required to back out the screws. The loss of mechanical properties over time was indicated by one-point bending load loss of the screws after these were extracted at different times. AZ31B samples with a silicon-containing coating had a decreased degradation rate and improved biological properties. The extraction torque of Ti6Al4V, poly-L-lactide (PLLA) and coated AZ31B increased significantly from 1 week to 4 weeks post-implantation, indicating a rapid osteosynthesis process over 3 weeks. The extraction torque of coated AZ31B increased with implantation time, and was higher than that of PLLA after 4 weeks of implantation, equalling that of Ti6Al4V at 12 weeks and was higher at 21 weeks. The bending loads of non-coated AZ31B and PLLA screws degraded sharply after implantation, and that of coated AZ31B degraded more slowly. The biodegradation mechanism, the coating to control the degradation rate and the bioactivity of magnesium alloys influencing the mechanical properties loss over time and bone–implant interface strength are discussed in this study and it is concluded that a suitable degradation rate will result in an improvement in the mechanical performance of magnesium alloys, making them more suitable for clinical application. Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
1. Introduction Research into Mg-based alloys used as biodegradable materials for orthopedics applications has attracted a great deal of interest. The key drawback of bio-absorbable Mg implants is their rapid degradation, resulting in hydrogen evolution, serious local alkalization, hemolysis and loss of their original strength before bone tissues have healed [1–3]. Researchers have devoted a tremendous amount of resources to investigating the rapid degradation problem of Mg alloys; surface modification is one of the key solutions and many modification methods have also been developed for biodegradable Mg alloys [4–10]. Several authors reported good biocompatibility and lower degradation rates of Mg alloys after modification [9,11–13]. Fluoride-coated Mg–calcium alloy was reported with optimized degradation kinetics in a subcutaneous mouse model [14]. A controllable polymeric membrane fabricated by polycaprolactone and dichloromethane was adopted to control ⇑ Corresponding authors. Tel./fax: +86 24 83971676 (L. Tan), +86 24 83971628 (K. Yang). E-mail addresses: [email protected] (L. Tan), [email protected] (K. Yang).
the corrosion rate of Mg alloys, and higher volumes of new bone were observed on the polymer-coated samples through in vivo study [15]. The result of in vivo implantations of the Ca–P coated and the non-coated alloy rods showed that the Ca–P coating provided Mg with significantly good surface bioactivity and promoted early bone growth at the bone–implant interface [16]. Mechanical performance is an important consideration for biodegradable Mg alloys: the fast degradation time also induces early mechanical loss before the tissue healing occurs. It is also reported that the three-point bending strength of AZ31B in simulated body fluid (SBF) was improved by fluoride conversion coating [17] and Ca–P coating [18]. However, the mechanical performance of modified Mg alloys in vivo has not yet been addressed. The mechanical properties of biodegradable Mg alloy implants are largely dependent not only on the original mechanical properties, but also on the way in which the mechanical properties of Mg alloy degrade after implantation as well as the interface strength of Mg alloy implants with surrounding host tissues. Thus, the aim of the present study was to evaluate the loss of mechanical properties over time and the interface strength of Mg alloy implants with surrounding host tissues; the effects of the
1742-7061/$ - see front matter Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.actbio.2013.12.020
2334
L. Tan et al. / Acta Biomaterialia 10 (2014) 2333–2340
surface modification and corrosion degradation mechanism on these tissues are discussed. Biodegradable poly-L-lactide (PLLA) and biostable titanium alloy are used as the control. The widely studied Mg alloy AZ31B was used in this study, and a Si-containing coating was prepared on the surface of AZ31B samples to decrease the biodegradation rate of the alloy. Moreover, silicon is an essential trace element in skeletal development [19,20]. An appropriate silicon concentration was effective in supporting the proliferation of osteoblast-like cells and actively stimulating a biological response in MG63 cells [21]. In vitro and in vivo studies have demonstrated that the silicon released from a biodegradable silicate ceramic could induce angiogenesis during bone regeneration [22]. Therefore, in this study the Si-containing coating on the surface of AZ31B samples was expected to improve the biological properties of the implants.
2. Experimental 2.1. Materials Extruded AZ31B Mg alloy (3 wt.% Al, 1.1 wt.% Zn, 0.70 wt.% Mn, 0.01 wt.% Si, 0.002 wt.% Fe, 0.008 wt.% Cu, 0.0008 wt.% Ni and balance Mg) with the microstructure shown in Fig. 1 was used as the biodegradable Mg alloy. AZ31B screws M2.6 6.5 mm in size were machined by Trauson Holdings Company Ltd. from extruded AZ31B. In order to decrease the degradation rate of AZ31B, samples were coated with a Si-containing coating; an image of an AZ31B screw with Si-containing coating is shown in Fig. 2. For the fabrication of the Si-containing coating, the alloy samples were etched in a mixed acid solution of 1.25 g l1 oxalic acid, 50% acetic acid and 25% nitric acid for 10 s at room temperature for surface activation, and subsequently immersed in a mixing solution of NaOH (5 g l1), Na2SiO3 (10 g l1) and Na4P2O7 (50 g l1) for 120 h at 60 °C, then washed with deionized water and dried at 40 °C. The micromorphology of the coating was characterized by scanning electron microscope (SEM) with an Hitachi S-3400N, and the elements and composition of the coating were characterized by energy dispersive spectroscopy (EDS) and X-ray photoelectron spectroscopy (XPS, Thermo VG ESCALAB250). The in vitro study was conducted according to ISO10998 to evaluate the in vitro biological properties of the samples before the in vivo study. The PLLA screws were provided by Chengdu Organic Chemicals Co. Ltd., Chinese Academy of
Fig. 1. The microstructure of extruded AZ31B used in this study.
Fig. 2. The AZ31B screw with Si-containing coating.
Sciences, and the Ti6Al4V screws (the same size as the AZ31B screws), provided by Trauson Holdings Company Ltd., were used as the control. 2.2. Degradation experiments The electrochemical corrosion test was investigated by potentiodynamic polarization using a PAR Verserstat3 potentiostat controlled by VersaStudio software. A three-electrode cell featuring a Pt counter-electrode and a saturated calomel reference electrode (SCE) was employed. The experiment was conducted at a sweep rate of 0.5 mV s1 within a scan range of ±0.25 V with reference to open circuit potential (OCP) using Hanks’ solution (8 g l1 NaCl, 0.06 g l1 KH2PO4, 0.4 g l1 KCl, 0.12 g l1 Na2HPO4, 0.14 g l1 CaCl2, 0.2 g l1 MgSO4, 0.35 g l1 NaHCO3, 1.0 g l1 glucose, pH 7.4). The temperature of the solution was kept at 37 °C. Ten tablets 10 mm in diameter and 4 mm high were immersed in Hanks’ solution at 37 ± 0.5 °C with an immersion ratio of 3 cm2 ml1 (sample surface area/fluid volume). The immersion solutions were refreshed every 24 h as a simplistic model of the fluid exchange in vivo. The pH variation of the Hanks’ solution was monitored before solution refreshment to measure the pH increase of the solution in 24 h. 2.3. Surgery procedure Ethical approval was obtained for the animal tests, which were conducted according to the ISO 10993-2:1992 animal welfare requirements. 16 adult white rabbits weighing 2.5–3.0 kg, 4 months old, were used and were divided into four groups through stochastic grouping. Before the surgery, all the rabbits were subjected to clinical checks. Each rabbit was anesthetized by an intramuscular injection of Sumian Xin (0.15 ml kg1 of body weight, Animal Husbandry Research Institute, Jilin, China). The mid-diaphyseal region of the femur was exposed through a lateral approach. Two holes (2 mm diameter) were bored in both femur shafts of each rabbit using a disinfected hand-operated drill with the longitudinal axis of the drill holes perpendicular to the longitudinal axis of the femoral diaphysis, 1 cm between the two holes. When boring the holes, the operation sites were cooled with physiological saline (0.9 wt.%). Then the screws, sterilized by ethylene oxide, were driven into the predrilled holes with a torque wrench (Tohnivhi), each rabbit having four different screws implanted. An image of one of the femur shafts with inserted screws is shown in Fig. 3. The torque needed to insert the screw was read from the dial of the torque wrench, which was controlled at a similar value by the same insertion rate for all the screw implantations to ensure the same initial condition for all the implants. The muscle layers, subcutaneous tissue and skin were subsequently sewed up.
L. Tan et al. / Acta Biomaterialia 10 (2014) 2333–2340
2335
2.6. Bone–implant interface morphology observation The interface morphology of the tissue adjacent to the implant sites was observed, to provide supporting information for the outcomes from the biomechanical assessments in this study. For observation analysis, tissue blocks containing the implants were fixed in 2.5% glutaraldehyde solution for 24 h, dehydrated in ascending grades of alcohol and embedded in the epoxy resin. Then the resin blocks with bone tissues were ground perpendicular to the long axis of the implant to obtain a cross-section of the implant and surrounding bone tissue. Finally, the cross-sections were polished with diamond polishing paste to remove the scratches. The bone–implant interface was characterized by SEM and EDS. 2.7. Statistical analysis
Fig. 3. One of the femur shafts of the rabbit with inserted screws.
Post-operatively, the rabbits were allowed to move freely in their cages without external support. All the rabbits received a subcutaneous injection of gentamycin 3 days after surgery. 2.4. Extraction torque measurements Rabbits were killed by anesthesia overdose at 1, 4, 12 and 21 weeks post-implantation. The implant sites were surgically exposed; the bone and soft tissues formed on top of the implants were carefully removed to expose only the screw nut and keep the tissue around the thread untouched. Subsequently, the screws were wound out of the bone using the torque wrench with the same rate for all the screws, and the peak extraction torque required to wind out the screws was read. 2.5. Mechanical properties measurements The one-point bending test was used to evaluate the mechanical properties of the screws after these had been extracted from the femur of the rabbits at different time points. The test model is shown in Fig. 4. The screw nut was immobilized by a mechanical fixture, and a pressure bar with a crosshead and a diameter of 2 mm was engaged to move downward at a rate of 0.5 mm min1, causing samples to bend to failure under the applied load. The peak loads of the screws with different implantation times were recorded to evaluate the loss of mechanical properties over time.
Fig. 4. One-point bending test model.
Data analysis and a bivariate correlation test (Pearson) were performed by SPSS (version 13.0). Results were presented as the mean ± standard deviation (SD). Comparisons between groups were made by the t-test. Significant difference was defined at either P < 0.01 or P < 0.05. 3. Results 3.1. Degradation and biological performance of AZ31B with Sicontaining coating The surface and cross-section morphologies of Si-containing coating and its EDS are shown in Fig. 5. It can be seen that the surface of the coating is uniform and scattered with small particles; the EDS shows that Mg, O and Si elements are contained in the coating. The thickness of the coating is 1.5–2 lm. From the XPS result shown in Fig. 6, we can see that the Si-containing coating is mainly composed of Mg2SiO4, MgO and SiO2. The polarization curves of AZ31B with and without Si-containing coating are shown in Fig. 7. The corrosion potential (Ecor) and the corrosion current density (icor) derived from the polarization curves are shown in Table 2. The degradation rates of the samples are also calculated from the formula according to ASTM G102-89(2004) as follows:
CR ¼ K 1
icor
q
EW;
ð1Þ
where CR is the corrosion rate (mm year1), icor is the corrosion current density (lA cm2), K1 is 3.27 103 (mm g lA1 cm1 year1), q is the density (g cm3) and EW is the equivalent weight. Compared to non-coated AZ31B alloy, the corrosion potential of the coating shifts positively, from 1.23 V for the AZ31B to 1.16 V for the coated AZ31B; hence the coated AZ31B can inhibit anodic dissolution considerably, indicating that corrosion occurs with difficulty. Furthermore, the corrosion current density of the coating (0.51 106 A cm2) is significantly lower than that of the noncoated AZ31B (11.34 106 A cm2), confirming that the corrosion resistance of the coated AZ31B is enhanced, which is due to a diminution of the cathodic hydrogen evolution reaction after the Mg substrate had been covered with Si-containing coating [23]. From the corrosion rate result, we can see that the corrosion rate of coated AZ31B (0.011 mm year1) is more than one order less than that of AZ31B (0.248 mm year1). The polarization analysis indicates that the coating can prevent the penetration of solution and effectively protect the AZ31B substrate from corrosion. Fig. 8 shows the pH value variation of Hanks’ solution immersed with samples for 2 months. Clearly, the pH value of the solution immersed with non-coated AZ31B increases rapidly to 11.3 after the first day of immersion, then the pH decreases gradually with further immersion time owing to the formation of Mg(OH)2 on
2336
L. Tan et al. / Acta Biomaterialia 10 (2014) 2333–2340
Fig. 5. Surface (a), cross-section (b) morphologies and EDS analysis (c) of the AZ31B alloy with Si-contained coating.
Fig. 6. XPS spectra of the Si-containing coating on AZ31B alloy sample: (a) XPS specific spectrum of O1s; (b) XPS specific spectrum of Mg1s; (c) XPS specific spectrum of Si2p.
the surface. Comparatively, the pH value of the coating is 8.38 on the first day, much lower than that of untreated AZ31B and with the increase of the immersion time, the pH values remain almost stable. Moreover, biological tests in vitro for AZ31B with Si-containing coating have also been investigated, and the results are shown in Table 1. It is found that the Si-containing coating has met the requirement of the standard biocompatibility tests according to ISO10993, including cytotoxicity (rank: 0), hemolytics (rank: 0),
systemic toxicity (none), sensitization (none) and irritation (rank:
