Advanced yolk–shell hydroxyapatite for bone graft materials: kilogram-scale production and structure-in vitro bioactivity relationship

Abstract

This paper introduces a facile method for synthesizing a new structured hydroxyapatite [HAp, Ca₁₀(PO₄)₆(OH)₂] material, named “yolk–shell”, by a simple spray drying process. The spray-dried precursor powders, consisting of Ca and P salts and dextrin, are transformed into powders of yolk–shell HAp by a simple combustion process under an oxygen atmosphere. For comparison, filled-structure HAp powders are prepared by spray pyrolysis in the absence of dextrin. A few small apatite crystals are found to be formed on the surfaces of the filled-structure HAp grains after 7 days of soaking in simulated body fluid (SBF). On the other hand, small acicular apatite crystals are observed on the yolk–shell grain surfaces after only 9 hours of soaking in SBF. The entire specimen surface is covered by numerous acicular and newly-formed hydroxyl carbonate apatite crystals after 1 day of soaking. These crystals are observed both at the outer and inner surfaces of the shell and the outer surface of the core. Inductively coupled plasma analysis shows that the dissolution of calcium and hydroxyl ions from yolk–shell HAp is notably increased compared with a filled-structure HAp. These results indicate that yolk–shell-structured HAp powders possess an enhanced in vitro bioactivity, which is encouraging for its potential use as a bone grafting material.

---

Introduction

Hydroxyapatite (HAp) has long been considered a representative bone grafting material owing to its similarities with the apatite in bones and teeth and its inherent biocompatibility.^1–4 However, it has been reported that longer periods of time are required for substantial bone apposition to occur on the surface of dense HAp relative to some bioactive glasses and glass ceramics, and this disadvantageously increases patient rehabilitation time.^5–11 To overcome this limitation, several methods have been proposed including introducing pores, or varying the crystallinity or powder size.^12–17 In addition, various HAp structures; including filled spheres, hollow spheres, nanorods, nanowires, nanoplates, nanoribbons, core–shell structures, flower structures, and tube structures; have been investigated to improve the osteoconductivity properties of synthetic HAp.^18–31

Among the various morphologies studied, a yolk–shell structure with a core@void@shell configuration has recently received increasing attention because of its wide variety of potential applications, including drug/gene delivery, energy conversion/storage, catalysis, nanoreactors, and sensors.^32–44 Since a yolk–shell structure contains a movable core, an interstitial hollow space, and an external shell, it is also expected to have great potential for application to the HAp used for bone grafts and drug/gene delivery. Firstly, the high surface area of a yolk–shell structure could conceivably increase the apatite solubility in body fluid, and subsequently enhance the low-crystalline hydroxyl-carbonate apatite forming capacity by increasing the degree of apatite supersaturation in body fluid. Secondly, its interstitial hollow space between the core and shell makes it possible for it to contain bioactive factors, such as bone morphogenetic proteins and cell adhesion proteins, which can encourage bone formation and cell adhesion while the hollow space is simultaneously filled with newly formed bone. Additionally, pores introduced into the surface layer of a yolk–shell structure can enable the release of inner bioactive factors from the structure, bone ingrowth from outside the shell, and nutrient delivery to the regenerated tissue formed in the yolk–shell structure's core. The design and fabrication of a HAp yolk–shell structure containing interacting pores in the surface layer has therefore drawn great attention as the scaffolds act not only as a structural support, but also provide the benefits of a fast recovery rate and high therapeutic efficacy to bone defects.

However, to the best our knowledge, a preparation process for yolk–shell-structured HAp powder suitable for bone graft substitutes has not yet been researched. The reason for this might be that HAp has a complicated crystal structure, at least in comparison to common oxides such as SiO₂. Furthermore, it is difficult to obtain singular and pure-phase HAp crystals, as HAp's sensitivity to pH during the synthesis process invariably leads to the formation of secondary phases such as Ca₃(PO₄)₂ or CaO. The production of yolk–shell structured HAp powder therefore remains both challenging and of great significance.

This paper proposes a simple and highly efficient preparation method for yolk–shell-structured HAp powder by spray drying, which is suitable for large-scale production. The characteristics and performance of the resulting powder were subsequently investigated by soaked them in a simulated body fluid (SBF). In addition, their structural effect on low-crystalline hydroxyl-carbonate apatite formation efficiency was investigated in detail.

Experimental

Yolk–shell structured HAp powders were prepared by a spray drying system, a schematic diagram of which is shown in Fig. S1.† For the synthesis of a yolk–shell structure, a calcium phosphate solution (0.2 M) with a Ca/P ratio of 1.67 was first prepared by dissolving calcium nitrate tetrahydrate [Ca(NO₃)₂·4H₂O, Aldrich], diammonium hydrogen phosphate [(NH₄)₂HPO₄, Aldrich] and dextrin [(C₆H₁₀O₅)_n, Samchun] in deionized water. The total amount of dextrin added was fixed at 100 g L⁻¹. The temperatures at the inlet and outlet of the spray dryer were 300 °C and 130 °C, respectively; while a two-fluid nozzle was used as an atomizer and the atomization pressure was 2.4 bar. The precursor powders obtained by spray drying of the precursor solution were post heat-treated at 600 °C in an oxygen gas atmosphere for 1 h. Filled-structure HAp powders were also prepared by spray pyrolysis, using the system depicted in Fig. S2.† In this instance, the starting solution of calcium phosphate (1.0 M) was prepared by dissolving the same reagents as described above in deionized water, but without the addition of dextrin. The resulting solution was then transformed into droplets using an ultrasonic spray generator (1.7 MHz, 17 vibrators). The length and diameter of the quartz reactor maintained at 1500 °C were 120 cm and 7 cm, respectively. The flow rate of air used as the carrier gas was fixed at 10 L min⁻¹.

The low-crystalline hydroxyl-carbonate apatite forming capacity of the dense and yolk–shell-structured HAp powders were assessed in SBF, which was prepared by dissolving reagent grade NaCl, NaHCO₃, KCl, K₂HPO₄·3H₂O, MgCl₂·6H₂O, CaCl₂, and Na₂SO₄ in deionized water. This solution was buffered at pH 7.25 using tris(hydroxymethyl)aminomethane with 1 M HCl at 36.5 °C, and then passed through a pre-sterilized filter unit (Millipore, 0.22 μm). The resulting HAp powders were sterilized under a UV lamp for 0.5 h, and then soaked in 30 mL of SBF at 36.5 °C for different periods of time.

The microstructures of the HAp powders were observed by scanning electron microscopy (SEM, JEOL, JSM-6060) and field-emission transmission electron microscopy (FETEM, JEOL, JEM-2100F). In addition, their crystal phases were evaluated by X-ray diffractometry (XRD, X'Pert PRO MPD) using Cu Kα radiation (λ = 1.5418 Å) at the Korea Basic Science Institute (Daegu). The functional groups of the specimens were also evaluated using Fourier transform infrared (FT-IR) spectroscopy (Spectrum 100, Perkin Elmer), in which a total of 256 scans were averaged to yield spectra at a resolution of 4 cm⁻¹. The surface areas of the crystals of the powders were measured by the Brunauer–Emmett–Teller (BET) method, using N₂ as the adsorbate gas. Finally, thermogravimetric and differential thermal analyses were performed using a Pyris 1 TGA (Perkin Elmer) and DSC8000 (Perkin Elmer) thermogravimetric analyzer, respectively, within a temperature range of 25–650 °C and at a heating rate of 10 °C min⁻¹ under a static air atmosphere.

Results and discussion

Fig. 1a shows the morphology of composite powders obtained by spray-drying, which are comprised of calcium nitrate, diammonium hydrogen phosphate, and dextrin. These spray-dried powders have a completely spherical shape, narrow size distribution, filled structure, and non-aggregation characteristics. Obtaining powders of metal salts using large-scale spray drying process is difficult under a high humidity atmosphere, because of the high hygroscopicity of the metal salts. Consequently, the addition of dextrin as a carbon source for the subsequent formation of a yolk–shell structure also plays a role as an efficient drying additive by reducing the hygroscopicity. The yolk–shell structured powders were formed by combustion of the spray-dried precursor powders at 600 °C in an O₂ gas atmosphere. In Fig. 1b and c, a yolk–shell-structured HAp powder with a distinct yolk@void@shell configuration was observed, in which the powders are completely spherical and non-aggregated. Furthermore, the TEM image in Fig. 1c distinctly shows a single shell, filled core, and contained pores in the shell layer of the HAp powders; while the selected-area electron diffraction (SAED) pattern in Fig. 1d reveals that the obtained yolk–shell powder is a polycrystalline HAp phase. The high-resolution TEM image, as seen in Fig. 1e, shows clear lattice fringes separated by 0.281 nm, which corresponds to the (211) plane of HAp (JCPDS Card no. 34-0010). In the dot-mapping images (Fig. 1f), the calcium, oxygen, and phosphorous components of the HAp are uniformly distributed all over the specimen. This means that phase separation of each component is minimized in the formation process of the yolk–shell powders. Decomposition of the calcium nitrate and diammonium hydrogen phosphate, along with carbonization of the dextrin, resulted in a CaO–P₂O₅–C intermediate product. Repeated combustion and contraction of this carbon-composite intermediate produced the final yolk–shell-structured HAp powders as shown in Fig. 2.^41,42 Specifically, the carbon decomposition and formation of HAp in the outer part of the intermediate forms a core–shell powder with a configuration of CaO–P₂O₅–C/HAp. Meanwhile, contraction of the core by further heating forms a yolk–shell powder with a configuration of CaO–P₂O₅–C@void@HAp; subsequent decomposition of this core resulting in a yolk–shell powder with a configuration of HAp@void@HAp.

Fig. 1 Morphologies and dot-mapping images of the powders: (a) SEM image of spray-dried precursor powders, (b) SEM image of yolk–shell powders, (c) TEM image of yolk–shell powders, (d) SAED pattern of yolk–shell powders, (e) HR-TEM image of yolk–shell powders, and (f) dot-mapping images of yolk–shell powders.

Fig. 2 Formation mechanism of the yolk–shell-structured hydroxyapatite powder fabricated by spray-drying using dextrin as carbon source.

The carbon components produced by dextrin are completely decomposed during the formation of yolk–shell-structured HAp powder, which agrees with the thermal analyses of the spray-dried precursor powders shown in Fig. 3a. The DSC curve shows that there was an endothermic peak at 188 °C, which may originate from the decomposition of dextrin and the initial calcium and phosphate salts. This is corroborated by the huge weight loss observed in the TGA curve at around this same temperature, with the curve becoming saturated at about 600 °C. The total weight loss of the spray-dried precursor prior to reaching 600 °C was 86%. Fig. 3b and c show the XRD and FT-IR spectra of the spray-dried precursor powders. An amorphous phase can be observed in the XRD pattern (Fig. 3b), because neither the dextrin nor the calcium and phosphate salts are decomposed during the drying process. Both CH, CH₂, and OH bands originating from the dextrin were detected. The asymmetrical bending (ν₄) modes of the PO₄³⁻ ion from the initial phosphate salt, were detected at around 609 and 574 cm⁻¹. Additionally the symmetric stretching mode (ν₃) of the NO₃⁻ ion from the initial calcium salt, was also found at around 1384 cm⁻¹ (Fig. 3c). However, the combusted powders with a clear yolk–shell structure had revealed a pure crystalline structure of a hexagonal HAp phase, as shown in Fig. 3d. The mean crystallite size of the HAp powders, as calculated from the peak width of the XRD pattern, was 33 nm. The FT-IR result in Fig. 3c also shows specific bands for HAp, the asymmetrical stretching (ν₃) and bending (ν₄) modes of the PO₄³⁻ ion were detected at around 1088, 1041, 609, and 574 cm⁻¹. The symmetric stretching modes, ν₁ and ν₂, of the PO₄³⁻ ion were also found at around 966 and 474 cm⁻¹, respectively. Additionally, the stretching mode of the OH⁻ ion was detected at around 3572 cm⁻¹.

Fig. 3 TG-DSC, XRD, and FT-IR analyses of the spray-dried precursor and post-treated yolk–shell powders: (a) TG-DSC curves of spray-dried precursor powders, (b) XRD pattern of spray-dried precursor powders, (c) FT-IR spectra, and (d) XRD pattern of yolk–shell powders.

The characteristics of the powders prepared by a one-pot spray pyrolysis process from the spray solution without dextrin are shown in Fig. 4. The XRD pattern of the powders shown in Fig. 4a reveals crystal peaks of a HAp phase without any discernible impurity peaks and a mean crystallite size of 48 nm. The FT-IR result shown in Fig. 4b also confirms the formation of a phase pure HAp powder. The SEM and TEM images shown in Fig. 4c and d show the spherical shape and filled structure of these powders, which were formed by the drying and decomposition of single droplets containing calcium nitrate and diammonium hydrogen phosphate. The mean sizes of the yolk–shell and filled-structure HAp powders were 1.6 and 0.6 μm, respectively; while their BET surface areas were 9.2 and 2.1 m² g⁻¹, respectively. The filled-structured HAp powders had dense structure without pores as shown in the N₂ adsorption–desorption isotherms in Fig. 5.

Fig. 4 Characteristics of the HAp powders prepared by a one-pot spray pyrolysis process: (a) XRD pattern, (b) FT-IR spectrum, (c) SEM image, and (d) TEM image.

Fig. 5 N₂ adsorption–desorption isotherms measured at 77 K for the filled- and yolk–shell-structured HAp powders.

The low-crystalline hydroxyl carbonate apatite-forming capacity of the filled- and yolk–shell-structured HAp powders was evaluated by soaking them in SBF for various time intervals, as shown in Fig. 6. In Fig. 6a, no microstructural changes were observable for up to 3 days of soaking (data not shown); however, a few small apatite crystals were formed on the surface of filled HAp grains after 7 days. On the other hand, small apatite crystals were newly observed on the yolk–shell grain surfaces after just 9 hours of soaking in SBF, as evidenced by the small, acicular white particles in the inset of Fig. 6b. The number and size of these small apatite crystals increased with increasing time, and could be clearly observed even under low magnification after 1 day of soaking; at which point the entire specimen surface was covered by numerous acicular hydroxyl carbonate apatite crystals. Consistent with these observations, the newly formed apatite layer gradually thickened with increasing duration of the testing period. In addition, from the fracture morphologies of yolk–shell powders soaked in SBF for 9 hours and 7 days, newly formed hydroxyl carbonate apatite crystals were observed both at the outer and inner surfaces of the shell and outer surface of the core (Fig. 6c). To explain this formation behavior of hydroxyl carbonate apatite on HAp, an ICP-AES evaluation of the SBF with soaking time was conducted (Fig. 7). In the case of the yolk–shell HAp, an increase in the calcium (Fig. 7a) and hydroxyl ion (Fig. 7c) concentration in SBF up to around 3 and 6 hours, respectively, indicates that these ions were rapidly released from the yolk–shell powders during this period. The binding energies of calcium, phosphate, and hydroxyl ion of HAp structure were −12.01, −17.06, and −6.35 eV, respectively. The smallest and highest binding energies of calcium and phosphate ions demonstrate that these elements are the easiest and the hardest to dissolve out from the HAp structure, respectively.⁴⁵ The subsequent decrease indicates that they were then consumed by newly-formed low-crystalline hydroxyl carbonate apatite crystals, as both calcium and hydroxyl ions are constituent elements of apatite. Similarly, the decrease in phosphorous concentration with time (Fig. 7b) can be attributed to this apatite formation, thus proving the high forming capacity of low crystalline hydroxyl carbonate apatite in a yolk–shell-structured HAp. In contrast, the filled-structure HAp powders produced a slight increase in the pH and calcium concentration of SBF up to 3 hours of soaking, after which there was a slight decrease. This proves the minor forming capacity of hydroxyl carbonate apatite in a filled-structure HAp. When yolk–shell HAp was soaked in tris-buffered deionized water, calcium and hydroxyl ions were released quickly in a time-dependent manner (Fig. 7d and f). In contrast, when filled structured HAp was soaked in tris-buffered deionized water, the release of constituent elements of apatite was negligible within the testing period (Fig. 7d–f). The yolk–shell structure in terms of enhancing the dissolution property is better than the filled structured powders. Taken together, these results indicate that the high specific surface area of the yolk–shell structure induces a rapid release of calcium and hydroxyl ions from the structure, subsequently increasing the degree of apatite supersaturation in SBF and thereby inducing a greater formation of apatite on the powder's surface compared with a filled structured.

Fig. 6 External surface images of (a) filled- and (b) yolk–shell-structured HAp powders after soaking in SBF for 3 h, 9 h, 1 day, and 7 days, and (c) fractured surface images of yolk–shell-structured powders after soaking in SBF for 9 h and 7 days.

Fig. 7 Changes in atomic concentrations of calcium and phosphorous, and pH of SBF and tris-buffered deionized water after soaking filled- and yolk–shell-structured HAp powders for various time intervals.

The formation behaviors of hydroxyl carbonate apatite for filled- and yolk–shell-structured specimens are described in detail in Scheme 1. In the filled-structured powder, as mentioned above, the specific surface area is relatively low compared to yolk–shell, which induces only a small release of calcium and hydroxyl ions from the structure for the nucleation of apatite. Accordingly, it is difficult to increase the degree of apatite supersaturation in body fluid, so a small amount of apatite formation occurs only at the outer surface of the powder. On the other hand, in a yolk–shell-structured powder, the apatite solubility in body fluid is enhanced by its high surface area, which subsequently enhances the low-crystalline hydroxyl-carbonate apatite forming capacity by increasing the degree of apatite supersaturation in body fluid. These conditions therefore induce the formation of hydroxyl-carbonate apatite both at the outer and inner surfaces of the shell, and the outer surface of the core. Along with its apatite forming advantages, yolk–shell-structured HAp also allows for the inclusion of bioactive factors, such as bone morphogenetic proteins and cell adhesion proteins, in the hollow space between the core and the shell. These can be released through the interacting pores in the surface layer of the powder, and subsequently improve the bone forming capacity and cell adhesion. The detailed release behavior of bioactive factors in vitro is presently being examined and will be reported in the near future. Nonetheless, yolk–shell-structured HAp does show improved in vitro bioactivity compared with a filled structure by virtue of its high specific surface area and enhanced solubility. Yolk–shell-structured HAp powders could therefore be applied to the development of novel bone-substitute materials.

Scheme 1 Schematic diagram for the formation of low-crystalline hydroxyl carbonate apatite of the filled- and yolk–shell-structured HAp powders.

Conclusions

In this study, yolk–shell-structured HAp powders are prepared for the first time and their hydroxyl carbonate apatite formation behaviors are introduced. For the large-scale production of yolk–shell-structured HAp, dextrin was selected for use as the carbon source material as well as the drying agent. Simple combustion of a spray-dried powder was then used to produce the yolk–shell-structured HAp powders. Low crystalline hydroxyl carbonate apatite was found to fully cover the surface of the yolk–shell HAp powders even after just 1 day of soaking in SBF, as its unique core@void@shell structure increased the dissolution of calcium and hydroxyl ions. In addition, the hollow space between the core and shell is expected to permit the incorporation of bioactive factors. Future studies will therefore address the potential use of yolk–shell HAp microspheres in delivery applications for bioactive factors; however, the yolk–shell HAp powders introduced in this study can already be applied to the development of novel bone-substitute materials.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (no. 2012R1A2A2A02046367). This work was supported by the Converging Research Center Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2011-50210). This work was supported by the Energy Efficiency & Resources Core Technology Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP), granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (201320200000420).

Notes and references

1. B. J. Kwon, J. Kim, Y. H. Kim, M. H. Lee, H. S. Baek, D. H. Lee, H. L. Kim, H. J. Seo, M. H. Lee and S. Y. Kwon, Artif. Organs, 2013, 37, 663.
2. M. Swetha, K. Sahithi, A. Moorthi, N. Srinivasan, K. Ramasamy and N. Selvamurugan, Int. J. Biol. Macromol., 2010, 47, 1.
3. H. Zhou and J. Lee, Acta Biomater., 2011, 7, 2769.
4. G. Zimmermann and A. Moghaddam, Injury, 2011, 42, S16.
5. X. Chatzistavrou, O. Tsigkou, H. D. Amin, K. M. Paraskevopoulos, V. Salih and A. R. Boccaccini, J. Eur. Ceram. Soc., 2012, 32, 3051.
6. A. Hoppe, N. S. Güldal and A. R. Boccaccini, Biomaterials, 2011, 32, 2757.
7. I. Izquierdo-Barba, A. J. Salinas and M. Vallet-Regí, Int. J. Appl. Glass Sci., 2013, 4, 149.
8. B. A. Oirschot, H. S. Alghamdi, T. O. Närhi, S. Anil, A. Al Farraj Aldosari, J. J. den Beucken and J. A. Jansen, Clin. Oral. Implants Res., 2014, 25, 21.
9. H. Oonishi, L. Hench, J. Wilson, F. Sugihara, E. Tsuji, M. Matsuura, S. Kin, T. Yamamoto and S. Mizokawa, J. Biomed. Mater. Res., 2000, 51, 37.
10. A. Salinas, S. Shruti, G. Malavasi, L. Menabue and M. Vallet-Regi, Acta Biomater., 2011, 7, 3452.
11. C. Tirapelli, H. Panzeri, E. Lara, R. Soares, O. Peitl and E. Zanotto, J. Oral Rehabil., 2011, 38, 253.
12. Z. Artzi, N. Givol, M. D. Rohrer, C. E. Nemcovsky, H. S. Prasad and H. Tal, J. Periodontol., 2003, 74, 1153.
13. L. Bjerre, C. Bünger, A. Baatrup, M. Kassem and T. Mygind, J. Biomed. Mater. Res., Part A, 2011, 97, 251.
14. A. L. Carvalho, P. E. Faria, M. F. Grisi, S. L. Souza, M. Taba Jr, D. B. Palioto, A. B. Novaes Jr, A. F. Fraga, L. S. Ozyegin and F. N. Oktar, J. Oral Implantol., 2007, 33, 267.
15. M. Khalid, M. Mujahid, S. Amin, R. Rawat, A. Nusair and G. Deen, Ceram. Int., 2013, 39, 39.
16. W. H. Lee, A. V. Zavgorodniy, C. Y. Loo and R. Rohanizadeh, J. Biomed. Mater. Res., Part A, 2012, 100, 1539.
17. P. Pripatnanont, T. Nuntanaranont, S. Vongvatcharanon and S. Limlertmongkol, J. Oral Maxillofac. Surg., 2007, 65, 2418.
18. F. Chen, Y. J. Zhu, K. W. Wang and K. L. Zhao, CrystEngComm, 2011, 13, 1858.
19. J. D. Chen, Y. J. Wang, K. Wei, S. H. Zhang and X. T. Shi, Biomaterials, 2007, 28, 2275.
20. J. S. Cho, D. S. Jung, J. M. Han and Y. C. Kang, J. Ceram. Process. Res., 2008, 9, 348.
21. J. S. Cho and Y. C. Kang, J. Alloys Compd., 2008, 464, 282.
22. J. S. Cho and S. H. Rhee, J. Eur. Ceram. Soc., 2013, 33, 233.
23. A. Li, K. Sun, Y. Yin, Z. Yu, Z. Lu and J. Bi, Acta Mater. Compos. Sin., 2004, 21, 98.
24. S. Li and Z. Peng, J. Wuhan Univ. Technol., Mater. Sci. Ed., 2010, 25, 252.
25. K. Lin, X. Liu, J. Chang and Y. Zhu, Nanoscale, 2011, 3, 3052.
26. M. Neumeier, L. A. Hails, S. A. Davis, S. Mann and M. Epple, J. Mater. Chem., 2011, 21, 1250.
27. K. W. Wang, Y. J. Zhu, X. Y. Chen, W. Y. Zhai, Q. Wang, F. Chen, J. Chang and Y. R. Duan, Chem.–Asian J., 2010, 5, 2477.
28. Q. Wang, W. Huang, D. Wang, B. W. Darvell, D. E. Day and M. N. Rahaman, J. Mater. Sci.: Mater. Med., 2006, 17, 641.
29. X. Wang, J. Zhuang, Q. Peng and Y. Li, Adv. Mater., 2006, 18, 2031.
30. P. Yang, Z. Quan, Z. Hou, C. Li, X. Kang, Z. Cheng and J. Lin, Biomaterials, 2009, 30, 4786.
31. H. Zhao, W. Dong, Y. Zheng, A. Liu, J. Yao, C. Li, W. Tang, B. Chen, G. Wang and Z. Shi, Biomaterials, 2011, 32, 5837.
32. M. Chen, Y. N. Kim, H. M. Lee, C. Li and S. O. Cho, J. Phys. Chem. C, 2008, 112, 8870.
33. J. Gao, G. Liang, J. S. Cheung, Y. Pan, Y. Kuang, F. Zhao, B. Zhang, X. Zhang, E. X. Wu and B. Xu, J. Am. Chem. Soc., 2008, 130, 11828.
34. J. Lee, J. C. Park and H. Song, Adv. Mater., 2008, 20, 1523.
35. X. Wang, X. L. Wu, Y. G. Guo, Y. Zhong, X. Cao, Y. Ma and J. Yao, Adv. Funct. Mater., 2010, 20, 1680.
36. H. B. Wu, J. S. Chen, H. H. Hng and X. Wen Lou, Nanoscale, 2012, 4, 2526.
37. G. Zhang, L. Yu, H. B. Wu, H. E. Hoster and X. W. Lou, Adv. Mater., 2012, 24, 4609.
38. W. Zhao, H. Chen, Y. Li, L. Li, M. Lang and J. Shi, Adv. Funct. Mater., 2008, 18, 2780.
39. Y. Zhao and L. Jiang, Adv. Mater., 2009, 21, 3621.
40. Y. Zhu, T. Ikoma, N. Hanagata and S. Kaskel, Small, 2010, 6, 471.
41. Y. J. Hong, M. Y. Son and Y. C. Kang, Adv. Mater., 2013, 25, 2279.
42. S. H. Choi and Y. C. Kang, Chem.–Eur. J., 2014, 20, 5835.
43. Y. J. Hong, J. W. Yoon, J. H. Lee and Y. C. Kang, Chem.–Eur. J., 2014, 20, 2737.
44. S. H. Choi and Y. C. Kang, Small, 2014, 10, 474.
45. J. S. Cho, S. H. Um, D. S. Yoo, Y. C. Chung, S. H. Chung, J. C. Lee and S. H. Rhee, J. Biomed. Mater. Res., Part B DOI:10.1002/jbm.b.33087.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra02925a

---