- Email: [email protected]
Development and degradation behavior of magnesium scaffolds coated with polycaprolactone for bone tissue engineering
Materials Letters 132 (2014) 106–110
Contents lists available at ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/matlet
Development and degradation behavior of magnesium scaffolds coated with polycaprolactone for bone tissue engineering Mostafa Yazdimamaghani a,1, Mehdi Razavi a,1, Daryoosh Vashaee b, Lobat Tayebi a,c,n a
School of Materials Science and Engineering, Helmerich Advanced Technology Research Center, Oklahoma State University, Tulsa, OK 74106, USA School of Electrical and Computer Engineering, Helmerich Advanced Technology Research Center, Oklahoma State University, Tulsa, OK 74106, USA c School of Chemical Engineering, Oklahoma State University, Stillwater, OK 74078, USA b
art ic l e i nf o
a b s t r a c t
Article history: Received 18 March 2014 Accepted 7 June 2014 Available online 18 June 2014
Rapid degradation of magnesium (Mg) alloys is the major drawback preventing these materials from being applicable as tissue engineering scaffolds. In order to resolve this issue, in this paper, porous Mg scaffolds coated by polycaprolactone (PCL) were synthesized and their material properties and in vitro biodegradation were fully examined. The results indicated that PCL coating can significantly enhance the compressive strength and degradation resistance of Mg scaffolds. We showed that while the uncoated Mg scaffold degrades completely (100% weight loss) after 72 h, the degradation (weight loss) of the Mg scaffolds coated by 3% and 6% PCL is only 36% and 23%, respectively. Thus PCL-coated Mg scaffolds, as a biodegradable metal scaffold, can potentially have a promising application in bone tissue engineering. Published by Elsevier B.V.
Keywords: Biomaterials Biodegradation Magnesium Scaffold Bone tissue engineering
1. Introduction Designing appropriate scaffolds is one of the main challenges in tissue engineering (TE) [1–4]. TE scaffolds are mostly made by ceramics, polymers and hydrogels [2,3,5]. However, lack of mechanical strength in these materials is the main concern in the applicability of the produced scaffolds especially for bone TE [6,7]. Scientists have tailored several materials as both nanoparticles and bulk materials for various biomedical applications [8]. Metals are among the most appropriate materials, in terms of mechanical strength for biomedical scaffolds and implants [9]. Since usual metallic biomaterials are not biodegradable, producing new biodegradable metallic alloys becomes an appreciated research objective [10,11]. Among biodegradable metals, Mg alloys are of particular interest as they have mechanical properties close to bone, are bioactive and encourage bone growth [12,13]. However, current synthesized Mg scaffolds have a high degradation rate, resulting in early weakening of their mechanical strength [14–16]. It is well-known that surface modification can effectively enhance the properties of metals, especially for biomedical applications [17–20]. Coating of Mg alloys seems to be necessary to improve their degradation resistance [21,22]. Polymeric materials
n Corresponding author at: School of Materials Science and Engineering, Helmerich Advanced Technology Research Center, Oklahoma State University, Tulsa, OK 74106, USA. Tel.: þ 1 9185948634. E-mail address: [email protected] (L. Tayebi). 1 These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.matlet.2014.06.036 0167-577X/Published by Elsevier B.V.
are used extensively in biomedical applications, and PCL is one of the most popular FDA approved ones [23]. Some investigations exist on the PCL coating on bulk Mg alloys and the influence of PCL on the degradation rate of non-porous bulk Mg [24–26]. However, to the best of our knowledge, there is no study on employing PCL to coat Mg porous scaffolds. In this paper, we demonstrated that PCL can be a promising candidate as a coating material for Mg scaffold as it can control the degradation rate and may maintain the temporal mechanical strength of the scaffold.
2. Materials and methods For production of Mg scaffolds, pure Mg powder was purchased from Sigma-Aldrich with the following quality: purity 499%, semispherical morphology, particle size: 150–300 mm. Carbonate hydrogen ammonium powder was employed as the space-holder with particle size in the range of 150–300 mm. The space-holder was added to the Mg powder with the volume contents of 35%. This value of volume contents was selected based on the previous report in which this amount was introduced as the most appropriate carbonate hydrogen ammonium particles amount in Mg scaffolds that resulted in the best mechanical properties of scaffolds [14]. The mixed powders were pressed at a pressure of 200 MPa. The green compacts were then heat treated to burn out the space holder particles in an oven under vacuum and to sinter the porous samples separately in a furnace under vacuum. For the heat treatment process, the samples were first heated up to 175 1C and
M. Yazdimamaghani et al. / Materials Letters 132 (2014) 106–110
stayed at this temperature for 2 h in an oven. In order to sweep away the decomposition products of ammonium carbonate, after 2 h, the door was opened slightly and the vacuum was closed. The samples were cooled in oven and then heated up to 600 1C and kept at the final temperature for 2 h in a furnace. After that, the produced scaffolds were coated with PCL. The PCL solution was prepared by mixing PCL with the average molecular weight 80,000 g/mol and dichloromethane (DCM). Two different concentrations of coating material, 3% (w/v) PCL or 6% (w/v) PCL, in solvent were applied. The coating process parameters were optimized in order to have a complete coating on the inner surfaces of the Mg scaffolds. For this purpose, the viscosity of PCL coating solution was decreased to penetrate easily into the inner surface. The viscosity of the 3% PCL and 6% PCL solutions were adjusted to about 60 and 120 cP (centipoises), respectively. The coating was performed by immersing the samples into the PCL solution for 1 h. The immersed samples were kept in desiccators connected to a vacuum pump. Vacuum facilitated pulling the PCL solution inside the pores for coating purpose. The vacuum pressure was increased gradually, but always kept below 0.7 atm. Above this pressure, some bubbles were observed in the solution which had an adverse effect on the deposition of the PCL solution on the substrate. The solution was repressurized several times to have a complete coating on the inner surfaces. The uncoated Mg scaffold, PCL-coated Mg scaffold with 3% (w/ v) PCL, and PCL-coated Mg scaffold with 6% (w/v) PCL were labeled as Mg scaffold, Mg scaffold/3% PCL and Mg scaffold/6% PCL, respectively. Microstructural studies were conducted using scanning electron microscopy (SEM: Hitachi UHR FE S-4800). Fourier transform infrared spectroscopy (FTIR, Agilent 680 IR) was utilized to identify the functional group of samples. For FTIR analysis, the outer surface of samples was examined. Total porosity (Π) of the porous scaffolds was measured according to the following equation [14]: Π ¼ ð1 ρ=ρs Þ 100%
ð1Þ
in which ρs is the density of the Mg scaffolds evaluated via the immersion method and ρ is the apparent density of the sample, which can be measured by the weight-to-volume ratio of the scaffold. The contact angle experiment was conducted using the measurement of the wetting degree between the water and the surface of massive nonporous samples. For this purpose, an optical microscope was used to take the images of 0.2 ml water droplet on the surface of different samples. The coating procedure on Mg massive samples was similar to the coating of porous sample. The samples were immersed in a physiological saline solution (PSS) with the pH value of 6.2 as a corrosive solution. Since the main aim of our work was improving the degradation properties of the Mg scaffolds specifically at the early stage of immersion, we accelerated the degradation process by employing the PSS solution with a low pH value (6.2). This approach allowed us to monitor the significant changes in degradation rates of samples at the short time. No buffer had been used during the degradation measurement. Each sample was immersed in 25 ml PSS at 37 1C for 72 h, and the changes in pH value of the solution were monitored by a pH meter (Sartorius). The degradation products were removed from the surface by immersing the degraded samples in chromic acid solution with the concentration of 180 g/L. Chromic acid reacts with the degradation products and dissolves them in the solution, with no influence on the Mg substrate. The solution was stirred manually for 20 min and renewed one time during the immersion. Removal of the degradation product was assessed by visual observation of both outside and inside (cross-section) of the scaffolds. The scaffold containing degradation products was white while the color completely turned into dark gray after removal of the degradation products. The difference in weight before and after immersion in PSS indicated the amount of weight gain, and the difference in weight before and after chromic acid immersion for removing the corrosion products indicated the amount of weight loss. The compression test was performed according to ASTM E9 standard [27] by a Shimadzu AGSX mechanical testing machine at
Mg scaffold
Mg scaffold
Mg scaffold/3% PCL
107
Mg scaffold/6% PCL
Mg scaffold/6% PCL
Fig. 1. Overall view of produced Mg scaffolds (a), SEM images of the uncoated (b) and coated (c) scaffolds and FTIR spectra from scaffolds (d).
108
M. Yazdimamaghani et al. / Materials Letters 132 (2014) 106–110
room temperature with the rate of 1 mm/min. The length to diameter ratio of samples was adjusted to 2:1. For degraded samples, after removing the degradation products from the surface, the dimension of the samples was adjusted to a length to diameter ratio of 2:1. In each time step of immersion, the degraded samples with adjusted cylindrical shape and size were evaluated by compression test. The curve of the force versus displacement during the test was plotted and it was converted to stress–strain graph. The maximum value of stress (ultimate compressive stress) was reported as the compressive strength.
3. Results and discussion Fig. 1a shows the overall view of produced Mg scaffolds. It can be observed that the samples have been composed of porosities that distributed uniformly through the sample. Fig. 1b and c presents the SEM images of the uncoated and coated scaffolds. The SEM image of cross-sectional view of the coated scaffolds indicated the presence of coating materials on the surface of Mg substrates at the center of the scaffolds (Fig. 1c). According to Fig. 1d, the FTIR spectra confirm the presence of functional groups related to PCL coating which are C–H, C–O, C–C, and C–O–C. The porosity of different uncoated and coated samples (before immersion test) measured using this method was in the range of 35–40%. According to SEM images (Figs. 1 and 4), the pore size of Mg scaffolds was 200–300 mm. In order to show the distinctive difference between 3% and 6% PCL coatings on the Mg surface, the water contact angle test was performed. The contact angle of the uncoated sample was about 491, while the contact angle of 3% PCL and 6% PCL coated samples increased to 571 and 651, respectively. The value of contact angles Mg scaffold
Mg scaffold/3%PC
Mg scaffold/6%PCL
0.6
directly depends on the surface free energy of the samples. Thus, the result of the water contact angle experiment confirmed that the hydrophobicity of surface increased by increasing the concentration of PCL. Formation of degradation products on the surface during the immersion test can increase the weight of the scaffolds. However, samples can also lose their original contents due to the actual degradation. The combination of these two procedures at this stage of the experiment is called scaffold weight gain, which is demonstrated in Fig. 2. Since the PSS media was extremely corrosive with the pH value of 6.2, a significant shrinkage in the scaffolds was occurred as can be seen in Fig. 2. However, the degradation rate of Mg alloys in vivo is known to be much lower than the in vitro degradation rate. Witte et al. [28] have reported 4 times increase of degradation rate of Mg alloy in vitro compared to in vivo. It can be seen in Fig. 3a that the weight of Mg scaffold, Mg scaffold/3% PCL and Mg scaffold/6% PCL changed from 0.3 g to 0.52 g and 0.54 g, respectively, while the uncoated Mg scaffold fully degraded after 72 h. At the next step, the degradation products were removed from the samples and weight of the scaffolds were measured and compared with their original weight before immersion test. This weight difference is called scaffold weight loss. According to Fig. 4a, after 72 h, the weight of Mg scaffold, Mg scaffold/3% PCL and Mg scaffold/6% PCL changed from 0.3 g to 0.18 g and 0.23 g, respectively, while the uncoated Mg scaffold was degraded completely after 72 h. Thus while the uncoated Mg scaffold fully degraded after 72 h, the weight loss of the Mg scaffold/3% PCL and Mg scaffold/6% PCL was 36% and 23%, respectively. The pH values of the solutions in which the scaffolds were immersed were monitored. It can be seen in Fig. 3b that the pH values of all solutions tended to increase at the early immersion stage, and then remained at a semi-stable value. The pH changes for the uncoated Mg scaffold were higher than the coated ones. The increase in pH value was the lowest for Mg scaffold/6% PCL.
Mg scaffold
0.4
Mg scaffold/3%PCL
Mg scaffold/6%PCL
0.35
0.3
0.3 0.2 0.1 0 0
24
48
72
Weight (g)
Weight (g)
0.5
0.25 0.2 0.15 0.1
Immersion time (hours)
0.05 0
24 hrs immersion
0
24
48
72
Immersion time (hours) Mg scaffold
Mg scaffold/ Mg scaffold/ 3%PCL 6%PCL
Mg scaffold
Mg scaffold/3%PCL
Mg scaffold/6%PCL
11
48 hrs immersion
Mg scaffold Mg scaffold/ Mg scaffold/ 3%PCL 6%PCL
72 hrs immersion
pH value
10 9 8 7 Mg scaffold
Mg scaffold/ 3%PCL
Mg scaffold/ 6%PCL
6 0
24
48
72
Immersion time (hours) Fig. 2. The weight change of the scaffolds versus immersion time before removing the degradation products by chromic acid indicates the weight gain of samples (a), photographs of scaffolds after 24 h (b), 48 h (c), and 72 h (d) immersion.
Fig. 3. The weight change of the scaffolds versus immersion time after removing the degradation products by chromic acid indicates the weight loss of samples (a) and pH value (b) versus immersion time.
M. Yazdimamaghani et al. / Materials Letters 132 (2014) 106–110
109
Fig. 4. Comparison of the morphology of the degraded samples after removing the degradation products: SEM micrographs of Mg scaffold (a and b), Mg scaffold/1PCL-BG (c and d) and Mg scaffold/3PCL-BG (e and f) in different magnifications which exhibit the structural details after 48 h immersion in PSS.
Different solutions include simulated body fluid (SBF) [16], physiological serum saline (PSS) [14], mixed solution of glycerin and ethanol (GE) [29], and Tris–HCl solution [30] have been used by different investigators as the immersion solutions in order to analyze the degradation behavior of different Mg scaffolds. It is known that different amount of Cl ion in chemical composition of solutions and different initial pH value leading to a different degradation behavior of Mg scaffolds. A comparison between our work and others which have used PSS solution as corrosive medium indicates the similarities in the pH values and the amount of material weight loss in the same immersion times [14]. Exploring the degradation behavior of the scaffolds, SEM images of Mg scaffold, Mg scaffold/3% PCL, and Mg scaffold/6% PCL have been compared after 72 h immersion in PSS (Fig. 4). The uncoated Mg scaffold exhibited a corroded surface with several micro-cracks on the surface of Mg grains which indicated serious local corrosion. However, the PCL coated Mg scaffolds maintained the porous structure during the 72 h of immersion. The local corrosion was observed on these samples; however, the degradation attack was less than that of the uncoated Mg scaffold. In addition, the amount of micro-cracks on the surface of Mg grains were less than those on the uncoated Mg scaffold, which may be due to the presence of PCL as a degradation barrier. The following reactions summarize the corrosion reactions of Mg [31,32]: Mg ðsÞ þ 2H2 O-MgðOHÞ2 ðsÞ þ H2 ðgÞ;
ð2Þ
ð3Þ
MgðOHÞ2 ðsÞ þ 2Cl
ðaqÞ-MgCl2 þ2OH ðaqÞ
Degradation is accompanied by an alkalization of the corrosive media due to the production of hydroxide ions (OH ). The high
proportion of hydroxide ions supports the formation of magnesium hydroxide (reaction (2)), which in turn acts as a protective layer against corrosion. Magnesium hydroxide is disrupted by chloride ions with the release of OH– (reaction (3)). The Mg(OH)2 film and the precipitation of ions on the surface cause the weight gain of the scaffolds (Fig. 2). Although Mg(OH)2 is slightly soluble in water, rigorous degradation occurs in aqueous physiological media as Mg(OH)2 reacts with Cl to form highly soluble magnesium chloride (MgCl2) and hydrogen bubbles (reaction (2)). According to Fig. 3a, the degradation rates of scaffolds were significantly influenced by different surface treatments. The PCL coating acted as a barrier layer on the surface to avoid rapid degradation of scaffolds. Hydrogen bubbles may have the role of removing the degradation products from the surface resulting in lower weight gain for the uncoated scaffolds (Fig. 2). Due to the formation of OH ion according to reaction (2), the pH of samples went up rapidly in the initial 24 h (Fig. 3b). It is worth noting that the weight change is due to both Mg scaffold substrate and PCL coating whereas the coating can degrade or take up water with time. The slower rise in pH values of the solution for the PCLcoated samples indicates a relatively gentle chemical dissolution. The compressive strength of the Mg scaffold, Mg scaffold/3% PCL, and Mg scaffold/6% PCL before and during the immersion of the samples for 24, 48, and 72 h in PSS solution are presented in Table 1. Since the PCL coating did not influence the bulk mechanical properties of the Mg scaffold, the compressive strength of all samples are similar before the immersion test (time point 0). The values of the compressive strength for the Mg scaffold/6% PCL are the highest and the values for the uncoated Mg scaffolds are the lowest compared to other samples. The uncoated Mg scaffold was degraded completely after 48 h, and as it can be observed, the
110
M. Yazdimamaghani et al. / Materials Letters 132 (2014) 106–110
Table 1 Compressive strength of the Mg scaffold, Mg scaffold/3% PCL, and Mg scaffold/6% PCL before immersion test and after the immersion of the samples for 24, 48, and 72 h in PSS solution. Compressive strength (MPa)
Mg scaffold Mg scaffold/3% PCL Mg scaffold/6% PCL
Immersion time (h) 0
24
48
72
527 3 527 3 527 3
127 2 177 2 227 2
0 107 1 147 2
0 5.8 7 1 8.8 7 2
compressive stress of the Mg scaffolds increased by the concentration of PCL. Therefore, since PCL coating improved the degradation resistance of Mg scaffolds, it enhanced the mechanical stability of samples during the immersion in the PSS solution. It can also be seen that the compressive strength of the uncoated Mg scaffolds, Mg scaffold/3% PCL and Mg scaffold/6% PCL decreased with time. Note that the uncoated Mg scaffold presented a sharp decrease of compressive strength after 24 h of immersion, which can be due to the severe local corrosion of the uncoated Mg scaffold. The compressive strengths of Mg scaffold/3% PCL and Mg scaffold/6% PCL remained higher than that of the uncoated Mg scaffold after 72 h of immersion, which is mainly due to the slower degradation rate, indicating that the PCL coating delayed the loss of mechanical strength of the Mg scaffold. 4. Conclusion Mg scaffolds were produced by powder metallurgy technique including blending–pressing–sintering method and PCL was coated on the surface of synthesized scaffolds. The PCL-coated Mg scaffolds exhibited noticeably lower degradation compared to the uncoated scaffold. According to our study, increasing the concentration of polycaprolactone from 3% (w/v) to 6% (w/v) can enhance the degradation resistance. Acknowledgments The work was partially supported by OCAST (Grant no. AR131054 8161), AFOSR (Grant no. FA9550-10-1-0010) and NSF (Grant no. 0933763).
References [1] Geng F, Tan L, Zhang B, Wu C, He Y, Yang J, et al. J Mater Sci Technol 2009;25:123–9. [2] Yazdimamaghani M, Vashaee D, Assefa S, Walker KJ, Madihally SV, Köhler GA, et al. J Biomed Nanotechnol 2014;10:911–31. [3] Shahini A, Yazdimamaghani M, Walker KJ, Eastman MA, Hatami-Marbini H, Smith BJ, et al. Int J Nanomed 2014;9:167. [4] Mozafari M, Vashaee Daryoosh, Tayebi Lobat. Electroconductive nanocomposite scaffolds: a new strategy into tissue engineering and regenerative medicine. In: Ebrahimi F, editor. Nanocomposites – new trends and developments. Croatia: INTECH; 2012. [5] Marra KG, Szem JW, Kumta PN, DiMilla PA, Weiss LE. J Biomed Mater Res 1999;47:324–35. [6] Yazdanpanah A, Kamalian R, Moztarzadeh F, Mozafari M, Ravarian R, Tayebi L. Ceram Int 2012;38:5007–14. [7] Farraro KF, Kim KE, Woo SL, Flowers JR, McCullough MB. J Biomech 2014;47:1979–86, http://dx.doi.org/10.1016/j.jbiomech.2013.12.003. [8] Rouhani P, Salahinejad E, Kaul R, Vashaee D, Tayebi L. J Alloys Compd 2013;568:102–5. [9] Salahinejad E, Hadianfard MJ, Macdonald DD, Sharifi-Asl S, Mozafari M, Walker KJ, et al. PloS One 2013;8:e61633. [10] McBride ED. J Am Med Assoc 1938;111:2464–7. [11] Witte F. Acta Biomater 2010;6:1680–92. [12] Hornberger H, Virtanen S, Boccaccini A. Acta Biomater 2012;8:2442–55. [13] Razavi M, Fathi M, Savabi O, Vashaee D, Tayebi L. Mater Sci Eng C 2014;41:168–77. [14] Zhuang H, Han Y, Feng A. Mater Sci Eng C 2008;28:1462–6. [15] Witte F, Ulrich H, Rudert M, Willbold E. J Biomed Mater Res Part A 2007;81:748–56. [16] Gu X, Zhou W, Zheng Y, Liu Y, Li Y. Mater Lett 2010;64:1871–4. [17] Niinomi M. Metall Mater Trans A 2002;33:477–86. [18] Salahinejad E, Hadianfard MJ, Macdonald DD, Sharifi S, Mozafari M, Walker KJ, et al. J Biomed Nanotechnol 2013;9:1327–35. [19] Salahinejad E, Hadianfard MJ, Macdonald DD, Mozafari M, Vashaee D, Tayebi L. Mater Lett 2013;97:162–5. [20] Razavi M, Fathi M, Savabi O, Mohammad Razavi S, Hashemi Beni B, Vashaee D, et al. Mater Lett 2013;113:174–8. [21] Razavi M, Fathi M, Savabi O, Mohammad Razavi S, Hashemi Beni B, Vashaee D, et al. Ceram Int 2014;40:3865–72. [22] Razavi M, Fathi M, Savabi O, Beni BH, Razavi SM, Vashaee D, et al. Appl Surf Sci 2014;288:130–7. [23] Kweon H, Yoo MK, Park IK, Kim TH, Lee HC, Lee H-S, et al. Biomaterials 2003;24:801–8. [24] Degner J, Singer F, Cordero L, Boccaccini AR, Virtanen S. Appl Surf Sci 2013;282:264–70. [25] Wong HM, Yeung KW, Lam KO, Tam V, Chu PK, Luk KD, et al. Biomaterials 2010;31:2084–96. [26] Chen Y, Song Y, Zhang S, Li J, Zhao C, Zhang X. Biomed Mater 2011;6:025005. [27] A. E9, Annual book of ASTM standards; 2000. [28] Witte F, Fischer J, Nellesen J, Crostack H-A, Kaese V, Pisch A, et al. Biomaterials 2006;27:1013–8. [29] Zhang X, Li X, Li J, Sun X. Prog Nat Sci: Mater Int 2013;23:183–9. [30] Wei J, Jia J, Wu F, Wei S, Zhou H, Zhang H, et al. Biomaterials 2010;31:1260–9. [31] Song G. Corros Sci 2007;49:1696–701. [32] Staiger MP, Pietak AM, Huadmai J, Dias G. Biomaterials 2006;27:1728–34.
Contents lists available at ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/matlet
Development and degradation behavior of magnesium scaffolds coated with polycaprolactone for bone tissue engineering Mostafa Yazdimamaghani a,1, Mehdi Razavi a,1, Daryoosh Vashaee b, Lobat Tayebi a,c,n a
School of Materials Science and Engineering, Helmerich Advanced Technology Research Center, Oklahoma State University, Tulsa, OK 74106, USA School of Electrical and Computer Engineering, Helmerich Advanced Technology Research Center, Oklahoma State University, Tulsa, OK 74106, USA c School of Chemical Engineering, Oklahoma State University, Stillwater, OK 74078, USA b
art ic l e i nf o
a b s t r a c t
Article history: Received 18 March 2014 Accepted 7 June 2014 Available online 18 June 2014
Rapid degradation of magnesium (Mg) alloys is the major drawback preventing these materials from being applicable as tissue engineering scaffolds. In order to resolve this issue, in this paper, porous Mg scaffolds coated by polycaprolactone (PCL) were synthesized and their material properties and in vitro biodegradation were fully examined. The results indicated that PCL coating can significantly enhance the compressive strength and degradation resistance of Mg scaffolds. We showed that while the uncoated Mg scaffold degrades completely (100% weight loss) after 72 h, the degradation (weight loss) of the Mg scaffolds coated by 3% and 6% PCL is only 36% and 23%, respectively. Thus PCL-coated Mg scaffolds, as a biodegradable metal scaffold, can potentially have a promising application in bone tissue engineering. Published by Elsevier B.V.
Keywords: Biomaterials Biodegradation Magnesium Scaffold Bone tissue engineering
1. Introduction Designing appropriate scaffolds is one of the main challenges in tissue engineering (TE) [1–4]. TE scaffolds are mostly made by ceramics, polymers and hydrogels [2,3,5]. However, lack of mechanical strength in these materials is the main concern in the applicability of the produced scaffolds especially for bone TE [6,7]. Scientists have tailored several materials as both nanoparticles and bulk materials for various biomedical applications [8]. Metals are among the most appropriate materials, in terms of mechanical strength for biomedical scaffolds and implants [9]. Since usual metallic biomaterials are not biodegradable, producing new biodegradable metallic alloys becomes an appreciated research objective [10,11]. Among biodegradable metals, Mg alloys are of particular interest as they have mechanical properties close to bone, are bioactive and encourage bone growth [12,13]. However, current synthesized Mg scaffolds have a high degradation rate, resulting in early weakening of their mechanical strength [14–16]. It is well-known that surface modification can effectively enhance the properties of metals, especially for biomedical applications [17–20]. Coating of Mg alloys seems to be necessary to improve their degradation resistance [21,22]. Polymeric materials
n Corresponding author at: School of Materials Science and Engineering, Helmerich Advanced Technology Research Center, Oklahoma State University, Tulsa, OK 74106, USA. Tel.: þ 1 9185948634. E-mail address: [email protected] (L. Tayebi). 1 These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.matlet.2014.06.036 0167-577X/Published by Elsevier B.V.
are used extensively in biomedical applications, and PCL is one of the most popular FDA approved ones [23]. Some investigations exist on the PCL coating on bulk Mg alloys and the influence of PCL on the degradation rate of non-porous bulk Mg [24–26]. However, to the best of our knowledge, there is no study on employing PCL to coat Mg porous scaffolds. In this paper, we demonstrated that PCL can be a promising candidate as a coating material for Mg scaffold as it can control the degradation rate and may maintain the temporal mechanical strength of the scaffold.
2. Materials and methods For production of Mg scaffolds, pure Mg powder was purchased from Sigma-Aldrich with the following quality: purity 499%, semispherical morphology, particle size: 150–300 mm. Carbonate hydrogen ammonium powder was employed as the space-holder with particle size in the range of 150–300 mm. The space-holder was added to the Mg powder with the volume contents of 35%. This value of volume contents was selected based on the previous report in which this amount was introduced as the most appropriate carbonate hydrogen ammonium particles amount in Mg scaffolds that resulted in the best mechanical properties of scaffolds [14]. The mixed powders were pressed at a pressure of 200 MPa. The green compacts were then heat treated to burn out the space holder particles in an oven under vacuum and to sinter the porous samples separately in a furnace under vacuum. For the heat treatment process, the samples were first heated up to 175 1C and
M. Yazdimamaghani et al. / Materials Letters 132 (2014) 106–110
stayed at this temperature for 2 h in an oven. In order to sweep away the decomposition products of ammonium carbonate, after 2 h, the door was opened slightly and the vacuum was closed. The samples were cooled in oven and then heated up to 600 1C and kept at the final temperature for 2 h in a furnace. After that, the produced scaffolds were coated with PCL. The PCL solution was prepared by mixing PCL with the average molecular weight 80,000 g/mol and dichloromethane (DCM). Two different concentrations of coating material, 3% (w/v) PCL or 6% (w/v) PCL, in solvent were applied. The coating process parameters were optimized in order to have a complete coating on the inner surfaces of the Mg scaffolds. For this purpose, the viscosity of PCL coating solution was decreased to penetrate easily into the inner surface. The viscosity of the 3% PCL and 6% PCL solutions were adjusted to about 60 and 120 cP (centipoises), respectively. The coating was performed by immersing the samples into the PCL solution for 1 h. The immersed samples were kept in desiccators connected to a vacuum pump. Vacuum facilitated pulling the PCL solution inside the pores for coating purpose. The vacuum pressure was increased gradually, but always kept below 0.7 atm. Above this pressure, some bubbles were observed in the solution which had an adverse effect on the deposition of the PCL solution on the substrate. The solution was repressurized several times to have a complete coating on the inner surfaces. The uncoated Mg scaffold, PCL-coated Mg scaffold with 3% (w/ v) PCL, and PCL-coated Mg scaffold with 6% (w/v) PCL were labeled as Mg scaffold, Mg scaffold/3% PCL and Mg scaffold/6% PCL, respectively. Microstructural studies were conducted using scanning electron microscopy (SEM: Hitachi UHR FE S-4800). Fourier transform infrared spectroscopy (FTIR, Agilent 680 IR) was utilized to identify the functional group of samples. For FTIR analysis, the outer surface of samples was examined. Total porosity (Π) of the porous scaffolds was measured according to the following equation [14]: Π ¼ ð1 ρ=ρs Þ 100%
ð1Þ
in which ρs is the density of the Mg scaffolds evaluated via the immersion method and ρ is the apparent density of the sample, which can be measured by the weight-to-volume ratio of the scaffold. The contact angle experiment was conducted using the measurement of the wetting degree between the water and the surface of massive nonporous samples. For this purpose, an optical microscope was used to take the images of 0.2 ml water droplet on the surface of different samples. The coating procedure on Mg massive samples was similar to the coating of porous sample. The samples were immersed in a physiological saline solution (PSS) with the pH value of 6.2 as a corrosive solution. Since the main aim of our work was improving the degradation properties of the Mg scaffolds specifically at the early stage of immersion, we accelerated the degradation process by employing the PSS solution with a low pH value (6.2). This approach allowed us to monitor the significant changes in degradation rates of samples at the short time. No buffer had been used during the degradation measurement. Each sample was immersed in 25 ml PSS at 37 1C for 72 h, and the changes in pH value of the solution were monitored by a pH meter (Sartorius). The degradation products were removed from the surface by immersing the degraded samples in chromic acid solution with the concentration of 180 g/L. Chromic acid reacts with the degradation products and dissolves them in the solution, with no influence on the Mg substrate. The solution was stirred manually for 20 min and renewed one time during the immersion. Removal of the degradation product was assessed by visual observation of both outside and inside (cross-section) of the scaffolds. The scaffold containing degradation products was white while the color completely turned into dark gray after removal of the degradation products. The difference in weight before and after immersion in PSS indicated the amount of weight gain, and the difference in weight before and after chromic acid immersion for removing the corrosion products indicated the amount of weight loss. The compression test was performed according to ASTM E9 standard [27] by a Shimadzu AGSX mechanical testing machine at
Mg scaffold
Mg scaffold
Mg scaffold/3% PCL
107
Mg scaffold/6% PCL
Mg scaffold/6% PCL
Fig. 1. Overall view of produced Mg scaffolds (a), SEM images of the uncoated (b) and coated (c) scaffolds and FTIR spectra from scaffolds (d).
108
M. Yazdimamaghani et al. / Materials Letters 132 (2014) 106–110
room temperature with the rate of 1 mm/min. The length to diameter ratio of samples was adjusted to 2:1. For degraded samples, after removing the degradation products from the surface, the dimension of the samples was adjusted to a length to diameter ratio of 2:1. In each time step of immersion, the degraded samples with adjusted cylindrical shape and size were evaluated by compression test. The curve of the force versus displacement during the test was plotted and it was converted to stress–strain graph. The maximum value of stress (ultimate compressive stress) was reported as the compressive strength.
3. Results and discussion Fig. 1a shows the overall view of produced Mg scaffolds. It can be observed that the samples have been composed of porosities that distributed uniformly through the sample. Fig. 1b and c presents the SEM images of the uncoated and coated scaffolds. The SEM image of cross-sectional view of the coated scaffolds indicated the presence of coating materials on the surface of Mg substrates at the center of the scaffolds (Fig. 1c). According to Fig. 1d, the FTIR spectra confirm the presence of functional groups related to PCL coating which are C–H, C–O, C–C, and C–O–C. The porosity of different uncoated and coated samples (before immersion test) measured using this method was in the range of 35–40%. According to SEM images (Figs. 1 and 4), the pore size of Mg scaffolds was 200–300 mm. In order to show the distinctive difference between 3% and 6% PCL coatings on the Mg surface, the water contact angle test was performed. The contact angle of the uncoated sample was about 491, while the contact angle of 3% PCL and 6% PCL coated samples increased to 571 and 651, respectively. The value of contact angles Mg scaffold
Mg scaffold/3%PC
Mg scaffold/6%PCL
0.6
directly depends on the surface free energy of the samples. Thus, the result of the water contact angle experiment confirmed that the hydrophobicity of surface increased by increasing the concentration of PCL. Formation of degradation products on the surface during the immersion test can increase the weight of the scaffolds. However, samples can also lose their original contents due to the actual degradation. The combination of these two procedures at this stage of the experiment is called scaffold weight gain, which is demonstrated in Fig. 2. Since the PSS media was extremely corrosive with the pH value of 6.2, a significant shrinkage in the scaffolds was occurred as can be seen in Fig. 2. However, the degradation rate of Mg alloys in vivo is known to be much lower than the in vitro degradation rate. Witte et al. [28] have reported 4 times increase of degradation rate of Mg alloy in vitro compared to in vivo. It can be seen in Fig. 3a that the weight of Mg scaffold, Mg scaffold/3% PCL and Mg scaffold/6% PCL changed from 0.3 g to 0.52 g and 0.54 g, respectively, while the uncoated Mg scaffold fully degraded after 72 h. At the next step, the degradation products were removed from the samples and weight of the scaffolds were measured and compared with their original weight before immersion test. This weight difference is called scaffold weight loss. According to Fig. 4a, after 72 h, the weight of Mg scaffold, Mg scaffold/3% PCL and Mg scaffold/6% PCL changed from 0.3 g to 0.18 g and 0.23 g, respectively, while the uncoated Mg scaffold was degraded completely after 72 h. Thus while the uncoated Mg scaffold fully degraded after 72 h, the weight loss of the Mg scaffold/3% PCL and Mg scaffold/6% PCL was 36% and 23%, respectively. The pH values of the solutions in which the scaffolds were immersed were monitored. It can be seen in Fig. 3b that the pH values of all solutions tended to increase at the early immersion stage, and then remained at a semi-stable value. The pH changes for the uncoated Mg scaffold were higher than the coated ones. The increase in pH value was the lowest for Mg scaffold/6% PCL.
Mg scaffold
0.4
Mg scaffold/3%PCL
Mg scaffold/6%PCL
0.35
0.3
0.3 0.2 0.1 0 0
24
48
72
Weight (g)
Weight (g)
0.5
0.25 0.2 0.15 0.1
Immersion time (hours)
0.05 0
24 hrs immersion
0
24
48
72
Immersion time (hours) Mg scaffold
Mg scaffold/ Mg scaffold/ 3%PCL 6%PCL
Mg scaffold
Mg scaffold/3%PCL
Mg scaffold/6%PCL
11
48 hrs immersion
Mg scaffold Mg scaffold/ Mg scaffold/ 3%PCL 6%PCL
72 hrs immersion
pH value
10 9 8 7 Mg scaffold
Mg scaffold/ 3%PCL
Mg scaffold/ 6%PCL
6 0
24
48
72
Immersion time (hours) Fig. 2. The weight change of the scaffolds versus immersion time before removing the degradation products by chromic acid indicates the weight gain of samples (a), photographs of scaffolds after 24 h (b), 48 h (c), and 72 h (d) immersion.
Fig. 3. The weight change of the scaffolds versus immersion time after removing the degradation products by chromic acid indicates the weight loss of samples (a) and pH value (b) versus immersion time.
M. Yazdimamaghani et al. / Materials Letters 132 (2014) 106–110
109
Fig. 4. Comparison of the morphology of the degraded samples after removing the degradation products: SEM micrographs of Mg scaffold (a and b), Mg scaffold/1PCL-BG (c and d) and Mg scaffold/3PCL-BG (e and f) in different magnifications which exhibit the structural details after 48 h immersion in PSS.
Different solutions include simulated body fluid (SBF) [16], physiological serum saline (PSS) [14], mixed solution of glycerin and ethanol (GE) [29], and Tris–HCl solution [30] have been used by different investigators as the immersion solutions in order to analyze the degradation behavior of different Mg scaffolds. It is known that different amount of Cl ion in chemical composition of solutions and different initial pH value leading to a different degradation behavior of Mg scaffolds. A comparison between our work and others which have used PSS solution as corrosive medium indicates the similarities in the pH values and the amount of material weight loss in the same immersion times [14]. Exploring the degradation behavior of the scaffolds, SEM images of Mg scaffold, Mg scaffold/3% PCL, and Mg scaffold/6% PCL have been compared after 72 h immersion in PSS (Fig. 4). The uncoated Mg scaffold exhibited a corroded surface with several micro-cracks on the surface of Mg grains which indicated serious local corrosion. However, the PCL coated Mg scaffolds maintained the porous structure during the 72 h of immersion. The local corrosion was observed on these samples; however, the degradation attack was less than that of the uncoated Mg scaffold. In addition, the amount of micro-cracks on the surface of Mg grains were less than those on the uncoated Mg scaffold, which may be due to the presence of PCL as a degradation barrier. The following reactions summarize the corrosion reactions of Mg [31,32]: Mg ðsÞ þ 2H2 O-MgðOHÞ2 ðsÞ þ H2 ðgÞ;
ð2Þ
ð3Þ
MgðOHÞ2 ðsÞ þ 2Cl
ðaqÞ-MgCl2 þ2OH ðaqÞ
Degradation is accompanied by an alkalization of the corrosive media due to the production of hydroxide ions (OH ). The high
proportion of hydroxide ions supports the formation of magnesium hydroxide (reaction (2)), which in turn acts as a protective layer against corrosion. Magnesium hydroxide is disrupted by chloride ions with the release of OH– (reaction (3)). The Mg(OH)2 film and the precipitation of ions on the surface cause the weight gain of the scaffolds (Fig. 2). Although Mg(OH)2 is slightly soluble in water, rigorous degradation occurs in aqueous physiological media as Mg(OH)2 reacts with Cl to form highly soluble magnesium chloride (MgCl2) and hydrogen bubbles (reaction (2)). According to Fig. 3a, the degradation rates of scaffolds were significantly influenced by different surface treatments. The PCL coating acted as a barrier layer on the surface to avoid rapid degradation of scaffolds. Hydrogen bubbles may have the role of removing the degradation products from the surface resulting in lower weight gain for the uncoated scaffolds (Fig. 2). Due to the formation of OH ion according to reaction (2), the pH of samples went up rapidly in the initial 24 h (Fig. 3b). It is worth noting that the weight change is due to both Mg scaffold substrate and PCL coating whereas the coating can degrade or take up water with time. The slower rise in pH values of the solution for the PCLcoated samples indicates a relatively gentle chemical dissolution. The compressive strength of the Mg scaffold, Mg scaffold/3% PCL, and Mg scaffold/6% PCL before and during the immersion of the samples for 24, 48, and 72 h in PSS solution are presented in Table 1. Since the PCL coating did not influence the bulk mechanical properties of the Mg scaffold, the compressive strength of all samples are similar before the immersion test (time point 0). The values of the compressive strength for the Mg scaffold/6% PCL are the highest and the values for the uncoated Mg scaffolds are the lowest compared to other samples. The uncoated Mg scaffold was degraded completely after 48 h, and as it can be observed, the
110
M. Yazdimamaghani et al. / Materials Letters 132 (2014) 106–110
Table 1 Compressive strength of the Mg scaffold, Mg scaffold/3% PCL, and Mg scaffold/6% PCL before immersion test and after the immersion of the samples for 24, 48, and 72 h in PSS solution. Compressive strength (MPa)
Mg scaffold Mg scaffold/3% PCL Mg scaffold/6% PCL
Immersion time (h) 0
24
48
72
527 3 527 3 527 3
127 2 177 2 227 2
0 107 1 147 2
0 5.8 7 1 8.8 7 2
compressive stress of the Mg scaffolds increased by the concentration of PCL. Therefore, since PCL coating improved the degradation resistance of Mg scaffolds, it enhanced the mechanical stability of samples during the immersion in the PSS solution. It can also be seen that the compressive strength of the uncoated Mg scaffolds, Mg scaffold/3% PCL and Mg scaffold/6% PCL decreased with time. Note that the uncoated Mg scaffold presented a sharp decrease of compressive strength after 24 h of immersion, which can be due to the severe local corrosion of the uncoated Mg scaffold. The compressive strengths of Mg scaffold/3% PCL and Mg scaffold/6% PCL remained higher than that of the uncoated Mg scaffold after 72 h of immersion, which is mainly due to the slower degradation rate, indicating that the PCL coating delayed the loss of mechanical strength of the Mg scaffold. 4. Conclusion Mg scaffolds were produced by powder metallurgy technique including blending–pressing–sintering method and PCL was coated on the surface of synthesized scaffolds. The PCL-coated Mg scaffolds exhibited noticeably lower degradation compared to the uncoated scaffold. According to our study, increasing the concentration of polycaprolactone from 3% (w/v) to 6% (w/v) can enhance the degradation resistance. Acknowledgments The work was partially supported by OCAST (Grant no. AR131054 8161), AFOSR (Grant no. FA9550-10-1-0010) and NSF (Grant no. 0933763).
References [1] Geng F, Tan L, Zhang B, Wu C, He Y, Yang J, et al. J Mater Sci Technol 2009;25:123–9. [2] Yazdimamaghani M, Vashaee D, Assefa S, Walker KJ, Madihally SV, Köhler GA, et al. J Biomed Nanotechnol 2014;10:911–31. [3] Shahini A, Yazdimamaghani M, Walker KJ, Eastman MA, Hatami-Marbini H, Smith BJ, et al. Int J Nanomed 2014;9:167. [4] Mozafari M, Vashaee Daryoosh, Tayebi Lobat. Electroconductive nanocomposite scaffolds: a new strategy into tissue engineering and regenerative medicine. In: Ebrahimi F, editor. Nanocomposites – new trends and developments. Croatia: INTECH; 2012. [5] Marra KG, Szem JW, Kumta PN, DiMilla PA, Weiss LE. J Biomed Mater Res 1999;47:324–35. [6] Yazdanpanah A, Kamalian R, Moztarzadeh F, Mozafari M, Ravarian R, Tayebi L. Ceram Int 2012;38:5007–14. [7] Farraro KF, Kim KE, Woo SL, Flowers JR, McCullough MB. J Biomech 2014;47:1979–86, http://dx.doi.org/10.1016/j.jbiomech.2013.12.003. [8] Rouhani P, Salahinejad E, Kaul R, Vashaee D, Tayebi L. J Alloys Compd 2013;568:102–5. [9] Salahinejad E, Hadianfard MJ, Macdonald DD, Sharifi-Asl S, Mozafari M, Walker KJ, et al. PloS One 2013;8:e61633. [10] McBride ED. J Am Med Assoc 1938;111:2464–7. [11] Witte F. Acta Biomater 2010;6:1680–92. [12] Hornberger H, Virtanen S, Boccaccini A. Acta Biomater 2012;8:2442–55. [13] Razavi M, Fathi M, Savabi O, Vashaee D, Tayebi L. Mater Sci Eng C 2014;41:168–77. [14] Zhuang H, Han Y, Feng A. Mater Sci Eng C 2008;28:1462–6. [15] Witte F, Ulrich H, Rudert M, Willbold E. J Biomed Mater Res Part A 2007;81:748–56. [16] Gu X, Zhou W, Zheng Y, Liu Y, Li Y. Mater Lett 2010;64:1871–4. [17] Niinomi M. Metall Mater Trans A 2002;33:477–86. [18] Salahinejad E, Hadianfard MJ, Macdonald DD, Sharifi S, Mozafari M, Walker KJ, et al. J Biomed Nanotechnol 2013;9:1327–35. [19] Salahinejad E, Hadianfard MJ, Macdonald DD, Mozafari M, Vashaee D, Tayebi L. Mater Lett 2013;97:162–5. [20] Razavi M, Fathi M, Savabi O, Mohammad Razavi S, Hashemi Beni B, Vashaee D, et al. Mater Lett 2013;113:174–8. [21] Razavi M, Fathi M, Savabi O, Mohammad Razavi S, Hashemi Beni B, Vashaee D, et al. Ceram Int 2014;40:3865–72. [22] Razavi M, Fathi M, Savabi O, Beni BH, Razavi SM, Vashaee D, et al. Appl Surf Sci 2014;288:130–7. [23] Kweon H, Yoo MK, Park IK, Kim TH, Lee HC, Lee H-S, et al. Biomaterials 2003;24:801–8. [24] Degner J, Singer F, Cordero L, Boccaccini AR, Virtanen S. Appl Surf Sci 2013;282:264–70. [25] Wong HM, Yeung KW, Lam KO, Tam V, Chu PK, Luk KD, et al. Biomaterials 2010;31:2084–96. [26] Chen Y, Song Y, Zhang S, Li J, Zhao C, Zhang X. Biomed Mater 2011;6:025005. [27] A. E9, Annual book of ASTM standards; 2000. [28] Witte F, Fischer J, Nellesen J, Crostack H-A, Kaese V, Pisch A, et al. Biomaterials 2006;27:1013–8. [29] Zhang X, Li X, Li J, Sun X. Prog Nat Sci: Mater Int 2013;23:183–9. [30] Wei J, Jia J, Wu F, Wei S, Zhou H, Zhang H, et al. Biomaterials 2010;31:1260–9. [31] Song G. Corros Sci 2007;49:1696–701. [32] Staiger MP, Pietak AM, Huadmai J, Dias G. Biomaterials 2006;27:1728–34.