Content-dependent biomineralization activity and mechanical properties based on polydimethylsiloxane–bioactive glass–poly(caprolactone) hybrids monoliths for bone tissue regeneration

Abstract

In this study, polydimethylsiloxane–bioactive glass–poly(caprolactone) (PDMS–BG–PCL) hybrid monoliths with various PDMS–BG contents were successfully fabricated via a typical sol–gel route. As a reinforcement, the PDMS–BG was used to improve the biomineralization activity, mechanical properties and osteoblasts biocompatibility of PCL polymer. The incorporation of PCL significantly decreased the formation time and increased the toughness of crack-free PDMS–BG–PCL hybrid monoliths. The mechanical properties of PDMS–BG–PCL hybrid monoliths were significantly affected by the content of PDMS–BG and PDMS–BG–PCL (30 wt%) showed a much higher elastic modulus (328.87 ± 18.82 MPa). The hydrophilicity of PDMS–BG–PCL hybrids was also increased as the PDMS–BG increased. Additionally, the biomineralization activity of PDMS–BG–PCL hybrid monoliths could be tailored by the PDMS–BG content. All PDMS–BG–PCL hybrids could induce fast deposition of a crystalline apatite layer on their surface in SBF for 7 days. The in vitro cellular studies also showed that PDMS–BG–PCL hybrids can enhance osteoblasts attachment and cell viability compared with PCL. The crack-free monolith structure, biomimetic hybrid composition and high apatite-forming bioactivity make PDMS–BG–PCL hybrid a promising candidate as scaffolds and implants for drug delivery and bone regeneration applications.

---

1. Introduction

In recent years, there are increasing interests on designing novel bioactive biomaterials in the field of bone tissue repair and regeneration.¹ The ideal biomaterials for bone tissue regeneration are required to be bioactive, osteoconductive, biocompatible, processable and should be mechanically controlled in accordance with the native bone tissue.² Single inorganic and polymer biomaterials are difficult to meet all the requirements, due to various disadvantages. Therefore, bioactive ceramic-based polymer nanocomposites have been developed as scaffolds or implants for bone tissue engineering and regeneration.^3,4

Among nanocomposite biomaterials, bioactive glass (BG)-based polymer composites have been paid a lot attention recently.^5–7 BGs have excellent biodegradability, biocompatibility and osteoconductivity, while BG has the ability to form a chemical bonding in contact with living tissue through a bonelike apatite mineral phase.^8–10 It is very reasonable and promising by incorporation of BG nanoparticles into a polymer matrix to fabricate bioactive polymer composite biomaterials with controlled physicochemical properties. Pure BG ceramics usually show a poor mechanical properties such as brittleness and low flexibility which make them difficult to be formed into complex shapes and easy to fracture under mechanical loads.¹¹ On the other hand, polymers with biocompatibility and biodegradable properties are elastomeric and can be readily process into certain structures, but they lack high modulus to match with bone tissue.¹² In addition, biomedical polymer presents poor biomineralization activity in attaching body fluid.¹³ To solve these problems, many BG nanoparticles (BGN) reinforced polymers including poly(caprolactone) (PCL) and gelatin were fabricated and proposed as potential biomaterials for bone tissue regeneration.^14,15 This kind of synthetic polymer nanocomposites can mimic the composition of bone tissue, where the inorganic phase is equivalent to apatite and the polymer component equivalent to the collagen-rich extracellular matrix. However, BGN-based polymer systems usually possess unstable mechanical properties and non-uniform biomineralization activity, due to the poor size distribution and low interface interactions between inorganic and polymer phase.¹⁶

Recently, bioactive glass sol-based polymer hybrid biomaterials with a molecular level organic–inorganic structure were developed via a sol–gel process at room temperature.^17–19 In this process, BG sol could efficiently be hybridized with soluble polymer chains when the inorganic network was formed. For examples, molecular level-based BG-polymers such as BG–gelatin and BG–chitosan systems have been fabricated.^20–22 Generally, hydrophilic gelatin and chitosan was easily to hybridize with BG sol to form the hybrid structure. However, as hydrophobic polymer, PCL was difficult to hybridize with BG sol to form uniform hybrids in the sol–gel process, due to the poor solubility of PCL with BG sol. Therefore, few studies reported the synthesis and properties of BG–PCL hybrids monoliths by sol–gel process.²³

In previous study, we successfully fabricated the crack-free BG–PCL hybrid monoliths through PDMS modified technology in the sol–gel process.²⁴ Under the presence of polydimethylsiloxane (PDMS), BG–PCL hybrid monoliths could be formed facilely without any brittle characteristics. PDMS-derived BG–PCL hybrid monoliths (PDMS–BG–PCL) demonstrated the molecular-level microstructure and high biomineralization activity. However, the detailed effect of BG content on the surface structure, hydrophilicity, biomineralization activity and mechanical properties of PDMS–BG–PCL hybrid monoliths was still not be investigated. The osteoblasts biocompatibility of PDMS–BG–PCL was also not demonstrated. Therefore, in this study, the PDMS–BG–PCL hybrid monoliths with different BG contents were fabricated by the sol–gel process, and their properties including hydrophilicity, biomineralization activity, mechanical behavior and cellular biocompatibility were also studied in detail.

2. Experimental

2.1. Materials

Tetraethoxysilane (TEOS, Si(OC₂H₅)₄), calcium nitrite (Ca(NO₃)₂·4H₂O), iso-propyl-alcohol (IPA), tetrahydrofuran (THF), dichloromethane (DCM) and hydrochloric acid (HCl, 35%) were purchased from Guanghua Chemical Factory Co., Ltd. (Guangdong, PR China). Polydimethylsiloxane (PDMS, HO–[Si(CH₃)₂–O–]_nH, M_n = 1100) was purchased from Alfa (Alfa, USA). PCL (M_n = 80 000) was supplied by Sigma-Aldrich (Sigma-Aldrich, St. Louis, MO).

2.2. Fabrication of PDMS–BG–PCL hybrid monoliths

The PDMS–BG–PCL hybrid monoliths were produced via a PDMS modified sol–gel process. Briefly, TEOS (42 g) was first pre-hydrolyzed in aqueous solution containing iso-propyl-alcohol (IPA) of 8 mL and tetrahydrofuran (THF) of 7 mL. After 30 min, 35% HCL of 7 mL, water of 40 mL and 18 g PDMS was added to the solution until the catalysis and hydrolysis reaction for 2 h. After additional stirring for 20 hours, Ca(NO₃)₂·4H₂O, H₂O and IPA were mixed with the former solution for 1 h. Then the predetermined content of the PDMS–BG sol (0, 30, 40, 50 and 60 wt% in relation to the PCL polymer) was dispersed into 10 mL of DCM with vigorous stirring for 5 min after which the PCL pellets were added with further stirring for 24 h. The fabricated mixture was then poured into Teflon dishes and kept at 30 °C for 24 h to form PDMS–BG–PCL hybrid gel. Finally, the PDMS–BG–PCL hybrid monoliths were obtained after heating hybrid gels at 60 °C for 24 h.

2.3. Evaluation of surface morphology and structure

The surface morphologies of the hybrid monoliths were examined by the scanning electron microscopy (SEM, JEOL JSM-6390). The elemental composition was determined by an energy dispersive spectrometry (EDS). The chemical structure of the specimens was characterized by Fourier transform infrared (FTIR) absorption spectroscopy (FTIRRS: vetex70) using KBr tablets method. The crystalline composition of the samples were performed by the X-ray diffractometry with Cu Kα radiation, operated at 40 kV and 30 mA from 20 to 70° at a scanning rate of 0.02° s⁻¹ and with a step of 0.02° (XRD, D/MAX-2400).

2.4. Hydrophilicity measurement of hybrid monoliths

The hydrophilicity of samples was evaluated by dynamic water contact angle measurement (SL200KB, China). Before test, the PDMS–BG–PCL hybrid monoliths films were cut into a size of 10 mm × 40 mm and a droplet of deionized water was deposited on the surfaces of samples using 21-gauge needle. After 2 min, the water contact angles were calculated and high-resolution image was captured. At least five positions for each sample were measured.

2.5. Biomineralization activity investigation of hybrid monoliths

The biomineralization activity of the hybrid monoliths were evaluated by examining the apatite formation on their surfaces of samples after soaking for different time periods in SBF, according to our previous report.²⁵ Briefly, the specimens of 10 × 10 × 2 mm³ in size were cut and incubated in SBF (in mM: Na⁺ 142, K⁺ 5.0, Mg²⁺ 1.5, Ca²⁺ 2.5, Cl⁻ 147.8, HCO₃⁻ 4.2, HPO₄²⁻ 1.0, SO₄²⁻ 0.5), which is nearly equal to those of human blood plasma. The specimens were soaked in 25 mL SBF at 37 °C and then removed from the fluids after soaking in SBF for various periods. After washed with ion-exchanged distilled water and dried at 40 °C for 1 day, the surface apatite-forming activity of samples was evaluated through SEM, EDS, FTIR and XRD analysis.

2.6. Mechanical behavior evaluation of hybrid monoliths

The tensile mechanical properties (tensile strength and modulus) of hybrid monoliths were tested at a crosshead speed of 50 mm min⁻¹ by a universal mechanical machine with a 500 N load cell (SHT4206, MTS). The PDMS–BG–PCL hybrids and PCL with a size of 10 mm × 60 mm were used for tensile mechanical test. The stress–strain curves were recorded by the additional software of machine. The tensile modulus of samples was obtained by determining the slope of the initial linear elastic portion of stress–strain curves. At least five species per sample were measured.

2.7. Cell biocompatibility investigation of the hybrid monolith

The osteoblast cell line (MC3T3-E1) was used to study the cell biocompatibility of the hybrid materials. Cell culture was carried out under standard growth medium in a humidified atmosphere with 5% CO₂. Before cell being seeded, the hybrid monoliths were cut into a size of 10 mm × 10 mm and sterilized with ultraviolet (UV) irradiation for 30 min. Cells were seeded on the surface of the hybrid materials with a density of 5000 cells per well. Tissue culture plate (TCP) was used as a control. Cell proliferation was evaluated by using a commercial AlamarBlue™ assay kit (Life Technologies). The final fluorescent intensity of the cell solution after incubation with Alamar Blue kit was recorded by a microplate reader (Molecular Devices) according to the instruction book. At least 5 species per sample were measured to obtain mean value and standard deviation (SD). The adhesion and morphology of the cells were analyzed by using LIVE/DEAD viability kit (Molecular Probes). The staining procedure was according to the manufacture instruction. The cell morphology was observed by using a fluorescence microscope (IX53, Olympus).

2.8. Statistics

All data were expressed as mean ± standard deviation (SD). Statistical differences between different groups were evaluated by using T-test analysis. Statistical significance was represented as *p < 0.05 and **p < 0.01.

3. Results and discussion

3.1. Fabrication of PDMS–BG–PCL hybrid monoliths

The PDMS–BG–PCL hybrid monoliths could be formed through mixing BG sol and PCL solution under assistance of PDMS, as shown in Fig. 1. As a biocompatible poly(siloxane), PDMS has a typical Si–O–Si structure which possesses a strong interaction with BG sols, and the alkyl side chains can expeditiously hybridize with PCL polymer phase. Thus the crack-free PDMS–BG–PCL hybrid monoliths with molecular level inorganic–organic phase structure could be facilely fabricated. Although the PDMS–BG hybrid monoliths have been prepared in our previous report, PDMS–BG derived hybrids are inherently brittle because of the existence of the high silica phase. Therefore, incorporating PDMS–BG into the PCL matrix which is a high tough biopolymer, may fabricate molecular level PDMS–BG–PCL hybrid monoliths with highly controlled structure, mechanical properties, biomineralization activity and osteoblast biocompatibility.

Fig. 1 Chemical structure and optical images of crack-free PDMS–BG–PCL hybrid monoliths fabricated using representative sol–gel route.

3.2. Structure and morphology characterizations of PDMS–BG–PCL hybrid monoliths

The crystalline phase composition and structure of PDMS–BG–PCL hybrids with different PDMS–BG contents (0, 30, 40, 50 and 60 wt%) was analyzed by XRD, as shown in Fig. 2. The representative characteristic peaks at 2θ = 21.88° and 2θ = 23.85° were observed, which were attributed to the PCL (semi-crystalline polymer). Due to the typical amorphous structure of bioactive glass, the peaks of PCL significantly decreased as the improvement of PDMS–BG contents (0–60 wt%).

Fig. 2 XRD patterns showing the crystalline composition and structure of PDMS–BG–PCL hybrid monoliths with different PDMS–BG contents.

Fig. 3 shows the surface morphology and structures of PDMS–BG–PCL hybrid monoliths with a range of PDMS–BG contents (0, 30, 40, 50 and 60 wt%). The surface of the pure PCL indicated relatively smooth morphology, as shown in Fig. 3A. The PDMS–BG–PCL hybrids with various PDMS–BG contents (30, 40, 50 and 60 wt%) exhibited quite similar surface microstructures, suggesting the molecular-level distributions of inorganic phase in PCL matrix. The EDS spectra clearly indicated the chemical composition change of PDMS–BG–PCL hybrid monoliths as various inorganic phase contents (Fig. 4). In the EDS spectrum of PCL, no any Si and Ca element was detected. However, the Si and Ca were detected clearly as the incorporation of PDMS–BG phase (Fig. 4A). Moreover, the Si peak significantly increased with the improvement of PDMS–BG contents. These results clearly demonstrated that PDMS–BG with different content could be efficiently hybridized into PCL matrix.

Fig. 3 Surface microstructure and morphology of PDMS–BG–PCL hybrid monoliths. (A) 30 wt% PDMS–BG; (B) 40 wt% PDMS–BG; (C) 50 wt% PDMS–BG; (D) 60 wt% PDMS–BG.

Fig. 4 EDS analysis spectra of PDMS–BG–PCL hybrid monoliths showing the elemental composition. (A) 30 wt% PDMS–BG, (B) 40 wt% PDMS–BG, (C) 50 wt% PDMS–BG, (D) 60 wt% PDMS–BG.

3.3. Mechanical properties evaluation of PDMS–BG–PCL hybrid monoliths

The tensile tests are used to examine the mechanical properties of the PDMS–BG–PCL hybrid monoliths, as shown in Fig. 5. The typical tensile stress–strain behavior of PDMS–BG–PCL hybrids with a range of PDMS–BG contents (0, 30, 40, 50 and 60 wt%) are shown in Fig. 5A. All samples showed a representative elastomeric stress–strain behavior in the initial strain range of 10%. The mechanical properties (ultimate tensile strength, elastic modulus and strain at failure) of the PDMS–BG–PCL hybrids (0, 30, 40, 50 and 60 wt%) are shown in Fig. 5B–D. The ultimate tensile strength of hybrid monoliths was significantly decreased from 26.07 ± 0.29 to 8.47 ± 2.1 MPa with increasing PDMS–BG content from 0 to 60 wt% (Fig. 5B). However, the Young's modulus of PDMS–BG–PCL 60 wt% hybrids presented a significantly high value of 328.87 ± 18.82 MPa as compared to the 236.33 ± 14.8 MPa of PCL (Fig. 5C). The strain at failure exhibited a similar change tendency with ultimate tensile strength for PDMS–BG–PCL from 0 to 60 wt%. Pure PDMS–BG hybrid monoliths were very brittle biomaterials due to their inorganic phase glass gel structure.²⁶ Here the addition PCL phase significantly improved the toughness. The strain of hybrid monoliths was significantly enhanced as the increase of PCL content, indicating the improvement of toughness. On the other hand, in our hybrid systems, only PDMS–BG 30 wt% monoliths showed significantly high Young's modulus compared with PCL polymer, which may be attributed to the molecular-level distribution of inorganic phase in PCL matrix. At high PDMS–BG content, the inorganic phase may be not hybridized well with polymer phase and the uniform structure may induce the decrease of Young's modulus. PDMS–BG may be a promising reinforcement material for improving the elastic modulus of a variety of biodegradable polymers.

Fig. 5 Mechanical properties evaluation of PDMS–BG–PCL hybrid monoliths with different PDMS–BG contents. (A) Stress–strain behavior; (B) ultimate tensile strength; (C) Young's modulus; (D) elongation at break.

3.4. Hydrophilicity measurement of PDMS–BG–PCL hybrid monoliths

The hydrophilicity of biomaterials implants plays an important role in determining their biological response for bone tissue regeneration. Fig. 6 shows the water contact angle result of PCL and PDMS–BG–PCL hybrid monoliths. Pure PCL presented a hydrophobic feature with a water contact angle of 100°. Compared with PCL, PDMS–BG phase incorporation with a range of 50% to 60% decreased the water contact angle of hybrids from 100° to 64°, indicating the improvement of hydrophilicity. This result can be attributed to the high water absorption ability of PDMS–BG phase when in contacting with water.

Fig. 6 Water contact measurement of PDMS–BG–PCL hybrid monoliths with different PDMS–BG contents showing the hydrophilicity of samples.

3.5. Biomineralization activity investigations of PDMS–BG–PCL hybrid monoliths

Based on the important role of biomineralization activity for biomaterials in bone tissue regeneration, the in vitro apatite-forming bioactivity of PDMS–BG–PCL hybrid monoliths was evaluated by soaking PDMS–BG–PCL hybrid monoliths into SBF for 7 days, as shown in Fig. 7 and Fig. 8 and Fig. 9. The PDMS–BG contents (0, 30, 40, 50 and 60 wt%) significantly affected the apatite formation of hybrid monoliths. The pure PCL did not show the formation of new apatite crystals layers on their surface (Fig. 7A–B) after 7 days. As compared to the specimens before being incubated into SBF (Fig. 3), new apatite layers formed on the surface of PDMS–BG–PCL hybrids (30, 40, 50 and 60 wt%) were observed clearly (Fig. 7C–J). As the improvement of PDMS–BG content, the morphology of formed apatite nanocrystals gradually changed from rod-like to needle-like feature which is the representative biomineralization characteristics of bioactive glass materials.

Fig. 7 Surface morphology of PDMS–BG–PCL hybrid monoliths with different PDMS–BG contents after biomineralization in SBF for 7 days. (A and B) PCL; (C and D) 30 wt%; (E and F) 40 wt%; (G and H) 50 wt%; (I and J) 60 wt%.

Fig. 8 Elemental composition of PDMS–BG–PCL hybrid monoliths with different PDMS–BG contents after biomineralization in SBF for 7 days. (A) PCL; (B) 30 wt%; (C) 40 wt%; (D) 50 wt%; (E) 60 wt%.

Fig. 9 XRD patterns of PDMS–BG–PCL hybrid monoliths with different PDMS–BG contents after biomineralization in SBF for 7 days. The representative diffraction peaks of hydroxyapatite were marked in the picture.

Fig. 8A–E show the EDS spectra of the PDMS–BG–PCL hybrid monoliths with a PDMS–BG content (0, 30, 40, 50 and 60 wt%) after being soaked in SBF for 7 days. No any Ca and P peaks were observed from the EDS for PCL, indicating no apatite deposition on pure PCL surface (Fig. 8A). However, as increasing PDMS–BG contents up to 30, 40, 50 and 60 wt%, the intensity of the Ca and P increased significantly. That is to say, the as-prepared PDMS–BG–PCL hybrid monoliths can efficiently induce the formation of the apatite layer when soaking in SBF. In addition, the EDS spectra revealed the improved hydroxyapatite (HA) formation ability as increased PDMS–BG content. Our results are consistent with the literature as reported. It should be noted that the surface morphology and elemental composition of apatite formed on the surface of PDMS–BG–PCL hybrids was similar with that on PDMS–BG monoliths, after soaking in SBF for 7 days (Fig. S1 and S2†).

The crystalline phase structure of new formed apatite layer on surface of samples after in SBF for 7 days was further examined by XRD, as shown in Fig. 9. The PDMS–BG–PCL hybrid monoliths with a PDMS–BG contents (30, 40, 50 and 60 wt%) presented several characteristic peaks associated with crystalline hydroxyapatite. Compared with pure PCL, the diffraction peaks at 2θ = 21.88° and 2θ = 23.85° for PDMS–BG–PCL significantly weakened after soaking in SBF for 7 days, suggesting the mineralized apatite layer formation. The XRD diffraction peaks for 30 wt%–60 wt% PDMS–BG at 32, 39, 46 and 49° were assigned to the reflections from the (211), (310), (222) and (213) crystal planes of the HA (JCPDS No. 09-0432). The SEM, EDS and XRD analysis clearly showed that PDMS–BG incorporation could significantly improve the biomineralization activity of PDMS–BG–PCL hybrid monoliths.

3.6. Osteoblasts biocompatibility assessment of PDMS–BG–PCL hybrid monoliths

The osteoblasts biocompatibility of PDMS–BG–PCL hybrid monoliths was determined by evaluating the cells attachment and proliferation activity of the MC3T3-E1 osteoblast cell line on the surface of samples, as shown in Fig. 10. The cells showed the normal cell attachment and spreading morphology on the surface of PDMS–BG–PCL 30% and PDMS–BG–PCL 60% after culture for 3 days. No any dead cells were observed on the surface of samples, demonstrating the good cell attachment activity. In addition, the cells viability on PCL and PDMS–BG–PCL hybrid monoliths significantly increased as the extended culture periods from 1 to 5 day, which indicated that our hybrid monolith could support the osteoblasts proliferation. As compared to PCL control, the osteoblasts presented the significantly high cell viability after incubating with PDMS–BG–PCL (30 and 60%) for 1 and 3 days. At 1 day, significantly high cell viability was also observed for 60% PDMS–BG compared with 30% PDMS–BG hybrids and PCL control. These results demonstrated that our PDMS–BG–PCL hybrid monoliths possess a good osteoblasts biocompatibility and the incorporation of PDMS–BG could efficiently improve the osteoblasts activity of PCL biopolymer.

Fig. 10 Osteoblasts biocompatibility investigation of PDMS–BG–PCL hybrid monoliths with different PDMS–BG contents (PCL, 30 wt% and 60 wt% PDMS–BG). MC3T3-E1 cells attachment morphology at 3 day and proliferation activity after 1–5 days culture.

As a biocompatible hybrid polymer, PDMS has been widely used as a biomedical implant material.²⁷ PDMS possesses a typical backbone chain of Si–O–Si and side chain of alkyl, and has a high compatibility with silica-based sol phase. The side chain of alkyl in PDMS could have strong interaction with hydrophobic PCL matrix. Therefore, in this study, the PDMS–BG–PCL hybrid monoliths with crack-free structure could be fabricated successfully through sol–gel process. The representative hybrid inorganic–organic structure showed highly controlled stiffness and toughness which is very important for their bone tissue regeneration applications. The hydrophilicity and mechanical properties of hybrid monoliths could be tailored facilely by changing the PDMS–BG content. In addition, the incorporation of PDMS–BG significantly enhanced the biomineralization activity and osteoblasts biocompatibility of PCL biopolymer, which improved greatly application values of PCL biopolymer in bone tissue regeneration.

In bone tissue engineering and guiding bone tissue regeneration, the ideal biomaterials should be tough (easy processing) and bioactive (biomineralization activity for bone-bonding). In our previous study, the as-fabricated PDMS–BG hybrid monoliths possessed high bioactivity and good osteoblasts biocompatibility.^24,25 However, the PDMS–BG hybrid monoliths showed the poor toughness and was easy crumble, the processing time was also very long (one week), which are not suitable for large-scale bone tissue engineering applications. On the other hand, pure PCL polymer suffers from the poor bioactivity (apatite-forming ability), which is not benefit for the enhanced bone tissue regeneration. Compared with PDMS–BG and PCL, the fabrication of crack-free PDMS–BG–PCL hybrid monoliths is facile such as shortened processing time, and their properties are significantly improved including optimized mechanical properties and biomineralization activity. We believe that our PDMS–BG–PCL hybrid monoliths are promising in bone tissue regeneration and drug delivery applications.

4. Conclusions

Novel and crack-free bioactive PDMS–BG–PCL hybrid monoliths with various PDMS–BG contents have been successfully fabricated by a facile sol–gel process. PDMS–BG could be uniformly distributed in the PCL matrix. The PDMS–BG–PCL hybrid monoliths with a PDMS–BG content of 30 wt% showed a significantly optimized elastic modulus and toughness. The hydrophilicity of the PDMS–BG–PCL hybrid monoliths was significantly improved through the addition of the PDMS–BG to the PCL matrix. PDMS–BG–PCL hybrid can induce the fast deposition of a crystalline apatite layer on their surface in SBF, demonstrating the high biomineralization activity of our materials. The optimized PDMS–BG–PCL hybrid showed the significantly improved osteoblast biocompatibility compared with PCL. The crack-free structure, biomimetic hybrid composition, high apatite-forming bioactivity and good osteoblasts biocompatibility make PDMS–BG–PCL hybrid monolith a promising candidate as implants for drug delivery and bone regeneration applications.

Acknowledgements

This work was supported by the Research Fund for the Doctoral Program of Higher Education of China under grant 20120201130004, the Science and Technology Developing Project of Shaanxi Province (2015KW-001), the Natural Science Basic Research Plan in Shaanxi Province of China (Program No. 2015JQ5165), the Fundamental Research Funds for the Central Universities (XJJ2014090), partially the National Natural Science Foundation of China Major Research Plan on Nanomanufacturing under Grant No. 91323303, the National Natural Science Foundation of China under Grant No. 61078058, and the 111 Project of China (B14040). The SEM work was done at International Center for Dielectric Research, Xi'an Jiaotong University, Xi'an, China. The authors also thank Ms Dai for her help in using SEM and EDS.

References

1. Z. G. Chen, L. Z. Kang, Q. Y. Meng, H. Y. Liu, Z. L. Wang, Z. W. Guo and F. Z. Cui, Mater. Sci. Eng., C, 2014, 45, 94.
2. R. Mishra, S. K. Goel, K. C. Gupta and A. Kumar, Tissue Eng., Part A, 2014, 20, 751.
3. P. Fattahi, G. Yang, G. Kim and M. R. Abidian, Adv. Mater., 2014, 26, 1846.
4. H. Zhou and J. Lee, Acta Biomater., 2011, 7, 2769.
5. A. R. Boccaccini, M. Erol, W. J. Stark, D. Mohn, Z. Hong and J. F. Mano, Compos. Sci. Technol., 2010, 70, 1764.
6. D. Mohn, C. Bruhin, N. A. Luechinger, W. J. Stark, T. Imfeld and M. Zehnder, Int. Endod. J., 2010, 43, 1037.
7. N. Fong, L. A. Poole-Warren and A. Simmons, J. Biomed. Mater. Res., Part B, 2013, 101B, 310.
8. B. Lei, X. Chen, X. Han and J. Zhou, J. Mater. Chem., 2012, 22, 16906.
9. L. L. Hench and J. M. Polak, Science., 2002, 295, 1014.
10. Z. H. Zhou, B. L. Ou, T. L. Huang, W. N. Zeng, L. H. Liu, Q. Q. Liu, J. Chen, Y. M. Zhao, Z. M. Yang, D. F. Cao and S. L. He, J. Macromol. Sci., Part A: Pure Appl.Chem., 2013, 50, 1048.
11. T. Kokubo, Biomaterials, 1991, 12, 155.
12. G. Wei and P. X. Ma, Biomaterials, 2004, 25, 4749.
13. K. M. Woo, J.-H. Jun, V. J. Chen, J. Seo, J.-H. Baek, H.-M. Ryoo, G.-S. Kim, M. J. Somerman and P. X. Ma, Biomaterials, 2007, 28, 335.
14. B. Lei, K. H. Shin, D. Y. Noh, Y. H. Jo, Y. H. Koh and H. E. Kim, Mater. Sci. Eng., C, 2013, 33, 1102.
15. M. Peter, N. S. Binulal, S. V. Nair, N. Selvamurugan, H. Tamura and R. Jayakumar, Chem. Eng. J., 2010, 158, 353.
16. B. A. Allo, A. S. Rizkalla and K. Mequanint, Langmuir, 2010, 26, 18340.
17. Y. Guillemin, M. Etienne, E. Sibottier and A. Walcarius, Chem. Mater., 2011, 23, 5313.
18. E. Johan, S. U. Khan and O. F. Göbel, J. Eur. Ceram. Soc., 2010, 30, 1555.
19. B. Lei, K. H. Shin, I. H. Jo, Y. H. Koh and H. E. Kim, Mater. Chem. Phys., 2014, 145, 397.
20. Y. Xue, L. Wang, Y. Shao, J. Yan, X. Chen and B. Lei, Chem. Eng. J., 2014, 251, 158.
21. M. Yazdimamaghani, M. Razavi, D. Vashaee, V. R. Pothineni, J. Rajadas and L. Tayebi, Appl. Surf. Sci., 2015, 338, 137.
22. B. Lei, K. H. Shin, Y. H. Koh and H. E. Kim, J. Biomed. Mater. Res., Part B, 2014, 102, 1528.
23. B. A. Allo, A. S. Rizkalla and K. Mequanint, ACS Appl. Mater. Interfaces, 2012, 4, 3148.
24. J. Chen, W. Que, Y. Xing and B. Lei, Appl. Surf. Sci., 2015, 332, 542.
25. J. Chen, W. Que, Z. He and X. Zhang, Polym. Compos., 2014, 35, 1193.
26. T. Zhu and K. H. Row, J. Sep. Sci., 2012, 35, 1294.
27. G. Mehta, M. J. Kiel, J. W. Lee, N. Kotov, J. J. Linderman and S. Takayama, Adv. Funct. Mater., 2007, 17, 2701.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra09075j

---