Template-free synthesis of uniform magnetic mesoporous TiO₂ nanospindles for highly selective enrichment of phosphopeptides

Abstract

A novel, simple and template-free strategy was developed for the synthesis of uniform magnetic mesoporous TiO₂ nanospindles. Unlike previous synthetic approaches which mainly involve the strict control of sol–gel processes, templates, or capping agents, our strategy is based on an additional post-hydrolysis route, giving rise to mesoporous TiO₂ with a high surface area (366 m² g⁻¹). After crystallization, the mesoporous TiO₂ nanospindles obtained a high surface area (∼117 m² g⁻¹) and a large pore size (∼5.4 nm), and exhibited a high magnetic susceptibility (∼18.0 emu g⁻¹), as well as showed a remarkable performance for highly selective and rapid enrichment of phosphopeptides with an unprecedented enrichment capacity of ∼300 mg g⁻¹. These understandings open up a new way for the preparation of functional mesoporous TiO₂-based materials with high surface areas in a template-free manner, which is of significant importance for applications from both environmental and industrial points of view.

---

Titanium dioxide (TiO₂) has received increasing attention over the past decades because of its potential applications in diverse fields, including photocatalysis, dye-sensitized solar cells, lithium-ion batteries, catalyst supports, as well as selective separation and adsorption of biomolecules.^1–6 Most of these applications are strongly dependent on the crystal structure, particle size, crystallinity, and specific surface area of TiO₂ materials.^7–11 It is known that mesoporous materials possess high surface areas, large pore volumes and uniform mesopore channels that can not only extraordinarily increase the density of accessible active sites, but also greatly facilitate mass diffusion within frameworks.^12,13 As a result, much work has been done to synthesize mesoporous TiO₂ materials by using various strategies such as the soft-templating route, which relies on the synergistic self-assembly of surfactant templates and inorganic precursors.^14–16 Since it has been well established to produce mesoporous silica, there is a great tendency to extend this methodology to mesoporous TiO₂ analogues.¹⁷ However, it is difficult to control the reactivity of titanium precursors compared to silicates, because they are highly sensitive to water or even moisture and tend to hydrolyze fast and form dense precipitates prior to assembly with the surfactant templates.¹⁸ To overcome this problem, a nanocasting method based on mesoporous silica or carbon as a hard template has been applied to synthesize various mesoporous titanias.^19,20 However, the nanocasting method is costly, time-consuming, and laborious. Moreover, it is noted that most of the present strategies for the synthesis of mesoporous TiO₂ involve the use of templates, which should be removed after the synthesis by calcination or etching to make the pores accessible.²¹ In addition, the resultant mesoporous TiO₂ powders are difficult to separate from the catalytic or adsorption system because of their small sizes.^22,23 These challenges greatly prevent their practical and widespread applications. Therefore, an efficient, facile, general and environmentally friendly method for the synthesis of mesoporous TiO₂ with a high surface area and magnetic recyclable properties is greatly desirable.

Protein phosphorylation, one of the most important post-translational modifications, plays a vital role in regulating many cellular processes such as growth, division, signaling transduction, and so on.^24,25 It is therefore of great importance to identify the phosphorylation sites and quantify their dynamic changes for the fundamental understanding of biological processes. Recently, mass spectrometry (MS) has been presented as a reliable and powerful tool for the analysis of protein phosphorylation owing to its ultrahigh sensitivity, wide dynamic range, and rapid sequence mapping. However, the identification and characterization of phosphopeptides remain challenging due to their low dynamic stoichiometry and the signal suppression of nonphosphorylated peptides. Consequently, robust and highly selective phosphorylated protein/peptide enrichment methods are in great demand for MS-based phosphoproteome analysis.

Unlike previous approaches mentioned above which mainly involve the strict control of sol–gel processes, templates, or capping agents, we report herein a novel strategy for the synthesis of mesoporous TiO₂ nanomaterials. We term this new concept as “post-hydrolysis”. In a practical design, the amorphous TiO₂ derived from the Stöber sol–gel method before calcination is subjected to an additional ultrasonic treatment in water, which greatly promotes further hydrolysis of unhydrolyzed alkoxy moieties and the condensation of the amorphous TiO₂ framework, giving rise to mesoporous TiO₂ with a high surface area (366 m² g⁻¹). It is featured with simplicity, reproducibility and template-free. After crystallization and reduction, the resultant mesoporous TiO₂ nanospindles obtained a high surface area (∼117 m² g⁻¹) and large pore size (∼5.4 nm), and showed a remarkable performance for highly selective and rapid enrichment of phosphopeptides, as well as exhibited a fast response to an external magnetic field. We found that without the additional post-hydrolysis process, the end products could be conglutinated bulky materials with a low surface area (∼21 m² g⁻¹), exhibiting a poor performance for selective enrichment of phosphopeptides. In light of these findings, functional mesoporous TiO₂-based materials with high surface areas can be facilely and widely synthesized in a template-free manner, which is of significant importance for applications from both environmental and industrial points of view.

The synthetic process for the uniform magnetic mesoporous TiO₂ nanospindles (denoted as Fe₃O₄@mTiO₂) is illustrated in Fig. 1. In the first step, a compact amorphous TiO₂ layer was deposited on the surface of α-Fe₂O₃ nanospindles via a Stöber coating method,^26,27 resulting in the α-Fe₂O₃@TiO₂ core–shell nanospindles. In the second step, the resultant products were immersed into water and then subjected to the additional post-hydrolysis process (for details, please see ESI†), which led to the formation of mesoporous nanospindles (denoted as α-Fe₂O₃@mTiO₂). After a thermal treatment and a reduction with H₂, uniform magnetic mesoporous Fe₃O₄@mTiO₂ nanospindles were obtained with a highly crystalline anatase shell. In a control experiment, the α-Fe₂O₃@TiO₂ nanospindles were directly annealed without the additional post-hydrolysis process, which resulted in bulky materials (denoted as Fe₃O₄@TiO₂).

Fig. 1 The template-free strategy towards the synthesis of uniform core–shell structured magnetic mesoporous TiO₂ nanospindles (Fe₃O₄@mTiO₂) based on a new post-hydrolysis process.

Uniform α-Fe₂O₃ nanospindles with a size of ∼100 nm in diameter and ∼480 nm in length were synthesized as cores via a hydrothermal method (Fig. S1†). In the facile Stöber system, titanium species from the hydrolysis and condensation of tetrabutyl titanate (TBOT) can deposit on the α-Fe₂O₃ nanospindles, leading to the formation of α-Fe₂O₃@TiO₂ core–shell nanospindles with a diameter of ∼220 nm and a length of ∼600 nm (Fig. 2a and S2a†). Field-emission scanning electron microscopy (FESEM) and transmission electron microcopy (TEM) images of the α-Fe₂O₃@TiO₂ nanospindles (Fig. S2b and S3a†) clearly reveal the well-defined core–shell structure and the presence of cross-linking between adjacent nanospindles (marked with arrows). The TEM image of a single nanospindle (Fig. 2b) shows the obvious tearing junctions (marked with a circle) on the surface of the TiO₂ shells, which may be induced by external forces. It suggests that the linkage is not a physical aggregation but a continuous TiO₂ network. In addition, the TiO₂ shells appear to be a compact matrix with a thickness of ∼60 nm (Fig. 2c). After the post-hydrolysis process, uniform mesoporous α-Fe₂O₃@mTiO₂ core–shell nanospindles are obtained (Fig. 2d and S2c†). FESEM and TEM images clearly demonstrate that the α-Fe₂O₃@mTiO₂ nanospindles are individually separated without linkages, and the previously smooth surface of the TiO₂ shells is converted into a rough one (Fig. S2d and S3b†). More importantly, the TiO₂ shells exhibit a highly mesoporous structure, which is derived from the voids between the aggregated TiO₂ oligomers (Fig. 2e and f). N₂ sorption isotherms (Fig. 3A) of the α-Fe₂O₃@mTiO₂ nanospindles show typical type IV curves and a distinct hysteresis loop close to the H1 type, with an increase in the adsorption branch at a relative pressure (P/P₀) of 0.2–0.5, indicating that the TiO₂ shells contain uniform mesopores. The Brunauer–Emmett–Teller (BET) surface area and pore volume of the α-Fe₂O₃@mTiO₂ nanospindles are calculated to be ∼366 m² g⁻¹ and 0.3 cm³ g⁻¹, respectively, which are much higher than those of the α-Fe₂O₃@TiO₂ nanospindles (∼119 m² g⁻¹, 0.1 cm³ g⁻¹). Moreover, the corresponding pore size distribution curve (Fig. 3B) of the α-Fe₂O₃@mTiO₂ spindles derived from the adsorption branch of the isotherms by using the Barrett–Joyner–Halenda (BJH) method clearly shows a uniform pore size of ∼2.6 nm, larger than that of the α-Fe₂O₃@TiO₂ spindles (<1.7 nm). The physical data are summarized in Table S1.† These imply that the post-hydrolysis process can greatly improve the monodispersity and mesoporosity of the TiO₂ shells.

Fig. 2 TEM images of (a–c) the α-Fe₂O₃@TiO₂ core–shell nanospindles prepared via a Stöber coating method and (d–f) the uniform mesoporous α-Fe₂O₃@mTiO₂ core–shell nanospindles obtained after the post-hydrolysis process.

Fig. 3 N₂ sorption isotherms (A), the corresponding pore size distribution curves (B) and the O 1s high-resolution XPS spectra (C) of (a) the α-Fe₂O₃@TiO₂ core–shell nanospindles and (b) the uniform mesoporous α-Fe₂O₃@mTiO₂ core–shell nanospindles.

X-ray photoelectron spectroscopy (XPS) was used to investigate the surface molecular states of the TiO₂ shells before and after the post-hydrolysis process. The survey spectra (Fig. S4†) of both the α-Fe₂O₃@TiO₂ and α-Fe₂O₃@mTiO₂ nanospindles show three well-resolved peaks of C 1s, Ti 2p, and O 1s without the typical one of Fe element, further indicating that the α-Fe₂O₃ core is uniformly coated by a thick TiO₂ layer. The high content of carbon in the α-Fe₂O₃@TiO₂ nanospindles clearly demonstrates the presence of plentiful alkoxy moieties (R) in the matrix of the TiO₂ shells, suggesting the incompletely hydrolyzed nature of TBOT under the low content of ammonia in this system. After the post-hydrolysis process, the carbon content decreases from 52.9 to 39.0%, but the ratio of Ti/O is almost unchanged (Table S2†), indicating that the alkoxy moieties can be partly removed in water under ultrasound. The O 1s high-resolution XPS spectra clearly demonstrate the variation in the chemical states of the oxygen atom (Fig. 3C). The spectra can be deconvoluted into three single peaks corresponding to Ti–O (530.0 eV), O–H (531.6 eV), and C–O (533.0 eV) functional groups,²⁸ respectively; the relative ratios of these peaks are listed in Table S1.† It can be clearly observed that the content of the C–O group decreases from 22.9 to 11.5%; in contrast, that of the OH group increases from 26.9 to 39.0%. In addition, the concentration of Ti–O species remains almost unchanged. All the results suggest that the Stöber-derived TiO₂ shells contain a large number of alkoxy moieties; and the corresponding Ti–OR moieties can be hydrolyzed into Ti–OH groups in water under ultrasound. FTIR spectroscopy (Fig. S5†) was also used to characterize the structural change of the TiO₂ shells upon the reaction with H₂O. The peaks in the range of 3100–3600 cm⁻¹ and at 1626 cm⁻¹ can be assigned to the –OH stretching and deformation vibrations.²⁹ Bands at 2956 and 2922 cm⁻¹ can be attributed to the –CH₂– and/or –CH₃ stretching vibration, whereas the peaks in the region of 1400–1600 cm⁻¹ correspond to the deformation vibrations, further revealing the presence of alkoxy moieties in the TiO₂ matrix. The variations of the relative intensities of the hydroxyl and alkyl groups further confirm that post-hydrolysis occurs during the ultrasonic treatment.

Notably, when directly annealed at 500 °C in air without the post-hydrolysis process, the nanospindles appear to be fused together (denoted as α-Fe₂O₃@TiO₂-500) and the TiO₂ shells are composed of large merged nanoparticles with a size of ∼15 nm, as revealed by FESEM images (Fig. 4a and S6a†). The TEM image further demonstrates that the α-Fe₂O₃@TiO₂-500 samples are conglutinated bulky materials (Fig. 4b). In contrast, discrete and uniform nanospindles are well retained when crystallizing the α-Fe₂O₃@mTiO₂ ones (Fig. S6b†). The XPS spectrum and TG analysis of the α-Fe₂O₃@mTiO₂-500 nanospindles indicate that the left alkoxy moieties can be effectively removed during the annealing process at 500 °C in air (Fig. S7†). FESEM and TEM images of the α-Fe₂O₃@mTiO₂-500 nanospindles (Fig. 4c and d and S8a†) clearly show that the TiO₂ shells are constructed of numerous aggregated nanoparticles. Moreover, the HRTEM image (Fig. S8b†) shows that the TiO₂ nanoparticles are well crystallized with a size of ∼6 nm and a d-spacing of 0.35 nm, well-matched to the d₁₀₁ of anatase,³⁰ and packed into the highly mesoporous structures. XRD patterns (Fig. S9†) of the resultant core–shell spindles reveal that the amorphous TiO₂ shells are well crystallized into the anatase phase.³⁰ Interestingly, it can be identified that the anatase characteristic peaks of the α-Fe₂O₃@mTiO₂-500 nanospindles are much broader than those of the α-Fe₂O₃@TiO₂-500 samples; calculated from the Scherrer formula, the average crystal sizes of TiO₂ are estimated to be ∼5 and 15 nm, respectively, which are in good agreement with the FESEM results. N₂ sorption isotherms of the resultant materials show type IV curves with a distinct hysteresis loop close to the H1 type (Fig. S10A†). The BET surface area and pore volume of the α-Fe₂O₃@mTiO₂-500 nanospindles are calculated to be ∼117 m² g⁻¹ and 0.2 cm³ g⁻¹, respectively, which are much higher than those of the α-Fe₂O₃@TiO₂-500 samples (∼21 m² g⁻¹, 0.1 cm³ g⁻¹). In contrast, the corresponding pore size of the α-Fe₂O₃@mTiO₂-500 spindles (∼5.4 nm) is smaller than that of the α-Fe₂O₃@TiO₂-500 samples (∼7.3 nm) (Fig. S10B†). This further confirms that the TiO₂ shells of the α-Fe₂O₃@mTiO₂-500 spindles are made of small nanocrystals, resulting in higher porosity and smaller voids, whereas those of the α-Fe₂O₃@TiO₂-500 samples are different, as illustrated in the inset of Fig. S10B†.

Fig. 4 FESEM and TEM images of the bulk α-Fe₂O₃@TiO₂-500 samples (a and b), and the uniform mesoporous α-Fe₂O₃@mTiO₂-500 core–shell nanospindles (c and d).

In this case, since the low content of ammonia is used as a catalyst, the hydrolysis and condensation of TBOT are directed toward the middle structures rather than the ends of completely cross-linked chains based on the positively charged values (δ) of the Ti atom in the partially hydrolyzed species (Fig. S11a†),³¹ which leads to a continuous and highly branched polymer matrix possessing a large number of unhydrolyzed alkoxy groups (Fig. 1) in the as-made TiO₂ shell. When being subjected to an additional ultrasonic water treatment process, the Ti–OR moieties easily undergo further hydrolysis to form Ti–OH groups with the elimination of ROH molecules (Fig. 1 and S11b†) due to the high sensitivity to moisture. Meanwhile, the overall continuous and low-polymerized matrix would be further condensed and transformed into an aggregate of small TiO₂ grains (Fig. 1 and S12†). Such synergistic effects leave mesopores in the amorphous TiO₂ frameworks. In addition, the little linkages of the polymer networks between the α-Fe₂O₃@TiO₂ nanospindles can be easily broken down owing to the high reactivity of water molecules under ultrasound, leading to the formation of discrete nanospindles. When annealed at high temperatures, it is known that the grain growth of TiO₂ experiences a localized crystallization process into the well-defined nanocrystals.³² Since the disconnected boundary nature of the parent α-Fe₂O₃@mTiO₂ nanospindles require an additional high activation barrier for reconstruction and mass transport, the resultant α-Fe₂O₃@mTiO₂-500 nanospindles can retain the discrete and uniform core–shell structure during the thermal treatment at 500 °C. Meanwhile, the small internal packing TiO₂ grains would grow large into anatase nanocrystals, which leads to relatively large packing pore sizes (Fig. S12†). In contrast, conglutinated bulky materials are obtained with large merged nanocrystals as the initial α-Fe₂O₃@TiO₂ nanospindles approach one another and continuous boundaries exist in the adjacent TiO₂ shell matrix (Fig. 1 and S12†). With these understandings, uniform functional mesoporous TiO₂ shelled core–shell nanoparticles with high surface areas can be facilely synthesized via a new route without using templates, for example, SiO₂@mTiO₂ and Fe₃O₄@mTiO₂ nanoparticles (Fig. S13†).

It has been well demonstrated that hematite can be easily converted to magnetite by the reduction with H₂.³³ Here, we explore this conversion reaction to introduce magnetic functionality into the resultant mesoporous α-Fe₂O₃@mTiO₂-500 nanospindles under a continuous flow of H₂–air at 400 °C. XRD patterns (Fig. S14A†) clearly reveal that α-Fe₂O₃ cores are well reduced to Fe₃O₄, and the TiO₂ shells show no apparent change in the crystal phase and size. However, the average crystal size of Fe₃O₄ nanoparticles is estimated by using the Scherrer formula to be ∼13 nm, much smaller than that of α-Fe₂O₃ ones (∼50 nm) owing to the crystal reconstruction during the conversion reaction.³³ Notably, the resultant uniform mesoporous Fe₃O₄@mTiO₂ core–shell nanospindles show a superparamagnetic behavior with little detectable remanence or coercivity (Fig. S14B†). The saturation magnetization value is measured to be ∼18.0 emu g⁻¹. As expected, the mesoporous Fe₃O₄@mTiO₂ core–shell nanospindles can be conveniently separated upon application of an external magnetic field (Fig. S14B,† inset). This feature is greatly desirable as it can be employed to separate TiO₂ nanoparticles in a reaction medium. In addition, TEM images of mesoporous Fe₃O₄@mTiO₂ nanospindles (Fig. S14C†) further demonstrate that the uniform core–shell nanostructures are well maintained during the H₂ reduction process.

Recently, TiO₂ has been widely applied for the enrichment of phosphopeptides, where the phosphate functional group can bind to the surface of TiO₂ particles through bridging bidentate bonds.^34,35 In this respect, the resultant mesoporous Fe₃O₄@mTiO₂ core–shell nanospindles are used to enrich phosphopeptides from a complex mixture and separated by an external magnetic field, as illustrated in Fig. 5a. As a basic test, the mesoporous Fe₃O₄@mTiO₂ nanospindles are first used to isolate phosphopeptides from tryptic digests of the standard protein β-casein. A direct analysis of the β-casein digest by MALDI-TOF MS indicates that only three phosphopeptides are detected with weak intensities (Fig. S15a†). In contrast, after being enriched by the mesoporous Fe₃O₄@mTiO₂ nanospindles (Fig. S15b†), almost all the peaks in the mass spectrum are related to phosphopeptides, and no non-phosphorylated peptides are observed. Additionally, the signal-to-noise (S/N) ratio for phosphopeptides is significantly improved, indicating that the mesoporous Fe₃O₄@mTiO₂ core–shell nanospindles are promising for the selective enrichment of phosphopeptides.

Fig. 5 (a) Schematic illustration of the selective enrichment process for phosphorylated peptides by using the uniform mesoporous magnetic Fe₃O₄@mTiO₂ core–shell nanospindles for MALDI MS analysis. Mass spectra of a peptide mixture from β-casein and BSA with a molar ratio of 1 : 500 before enrichment (b), and after enrichment with the mesoporous Fe₃O₄@mTiO₂ (c) and bulk Fe₃O₄@TiO₂ (d) samples, respectively. (e) The enrichment capacity analysis of the mesoporous Fe₃O₄@mTiO₂ and bulk Fe₃O₄@TiO₂ samples. The asterisks indicate the phosphopeptides.

The selectivity of enrichment is further evaluated by capturing phosphopeptides from a more complex mixture of tryptic β-casein and BSA digests with a molar ratio of 1 : 500. It can be clearly identified that no phosphopeptides are detected from the initial mixture due to the presence of a high-abundance of non-phosphopeptides from BSA (Fig. 5b). By contrast, after the enrichment by the mesoporous Fe₃O₄@mTiO₂ nanospindles, the signals related to phosphopeptides are highly enhanced and dominate the mass spectrum with a very clean background (Fig. 5c), which is much better than that for the TiO₂ nanoparticles reported previously.^36–39 In a control experiment, the bulk Fe₃O₄@TiO₂ samples are also employed to enrich phosphopeptides from the complex mixture. Fig. 5d shows that the mass spectrum is dominated by nonphosphorylated peptides' signals and three phosphopeptides with weak intensities can be detected. This is because large merged TiO₂ nanoparticles with a low surface area are produced during thermal annealing without the post-hydrolysis process, which leads to lower active sites for the adsorption of phosphopeptides. To study the enrichment capacity of phosphopeptides with the resultant products, various concentrations of tryptic β-casein digests are prepared and then incubated with an equivalent amount of products. After magnetically separating the products, the supernatants are collected and subjected to MALDI-MS analysis. The products would reach the maximum enrichment capacity once the signal of phosphopeptides could be detected in the supernatants. As a result, the enrichment capacity (Fig. 5e) of the mesoporous Fe₃O₄@mTiO₂ nanospindles is estimated to be ∼300 mg g⁻¹, which is much higher than that of the bulk Fe₃O₄@TiO₂ samples (∼75 mg g⁻¹). This can be attributed to the small particle size and high surface area of the TiO₂ shells in the mesoporous Fe₃O₄@mTiO₂ nanospindles, which possess more active sites for phosphopeptide loading.

In summary, we developed a novel, simple and template-free strategy, which is based on a new post-hydrolysis concept, for the synthesis of uniform mesoporous TiO₂ nanospindles in a facile Stöber system. We have demonstrated that the TiO₂ shells derived from the Stöber system are incompletely hydrolyzed networks, and the unhydrolyzed alkoxy groups can be easily removed by water, yielding mesoporous TiO₂ with a high surface area. Benefitting from the well-defined structural features of the parent α-Fe₂O₃@mTiO₂ nanospindles including the disconnected boundary nature and the internal packing of TiO₂ grains, uniform mesoporous magnetic TiO₂ nanospindles are obtained with a high surface area (∼117 m² g⁻¹) and large pore size (∼5.4 nm) without the use of templates or capping agents. Most importantly, the resultant Fe₃O₄@mTiO₂ nanospindles exhibit a superparamagnetic behavior with a high magnetic susceptibility of ∼18.0 emu g⁻¹, and a remarkable performance for highly selective and rapid enrichment of phosphopeptides with an unprecedented enrichment capacity of ∼300 mg g⁻¹. We believe that these new understandings can be exploited to develop versatile and template-free approaches for fabricating functional mesoporous TiO₂-based materials with exceptionally high surface areas.

Acknowledgements

This work was supported by the State Key Basic Research Program of the PRC (2012CB224805, 2012CB910602, 2013CB934104), NSFC of China (Grant no. 21210004, 21025519, 21335002) and King Saud University research project (RGP-VPP-036).

References

1. X. Chen and S. S. Mao, Chem. Rev., 2007, 107, 2891.
2. G. Liu, L. Wang, H. G. Yang, H. M. Cheng and G. Q. Lu, J. Mater. Chem., 2010, 20, 831.
3. L. Zhao, X. F. Chen, X. C. Wang, Y. J. Zhang, W. Wei, Y. H. Sun, M. Antonietti and M. M. Titirici, Adv. Mater., 2010, 22, 3317.
4. W. F. Ma, Y. Zhang, L. L. Li, L. J. You, P. Zhang, Y. T. Zhang, J. M. Li, M. Yu, J. Guo, H. J. Lu and C. C. Wang, ACS Nano, 2012, 6, 3179.
5. S. H. Liu, Z. Y. Wang, C. Yu, H. B. Wu, G. Wang, Q. Dong, J. S. Qiu, A. Eychmuller and X. W. Lou, Adv. Mater., 2013, 25, 3462.
6. D. H. Chen and R. A. Caruso, Adv. Funct. Mater., 2013, 23, 1356.
7. J. S. Chen, Y. L. Tan, C. M. Li, Y. L. Cheah, D. Luan, S. Madhavi, F. Y. C. Boey, L. A. Archer and X. W. Lou, J. Am. Chem. Soc., 2010, 132, 6124.
8. J. B. Joo, Q. Zhang, I. Lee, M. Dahl, F. Zaera and Y. D. Yin, Adv. Funct. Mater., 2012, 22, 166.
9. W. Li, Y. Deng, Z. Wu, X. Qian, J. Yang, Y. Wang, D. Gu, F. Zhang, B. Tu and D. Y. Zhao, J. Am. Chem. Soc., 2011, 133, 15830.
10. R. Zhang, B. Tu and D. Zhao, Chem. – Eur. J., 2010, 16, 9977.
11. D. H. Chen, F. Z. Huang, Y. B. Cheng and R. A. Caruso, Adv. Mater., 2009, 21, 2206.
12. W. Li, Q. Yue, Y. Deng and D. Zhao, Adv. Mater., 2013, 25, 5129.
13. W. Li and D. Y. Zhao, Chem. Commun., 2013, 49, 943.
14. D. M. Antonelli and J. Y. Ying, Angew. Chem., Int. Ed., 1995, 34, 2014.
15. P. Yang, D. Zhao, D. I. Margolese, B. F. Chmelka and G. D. Stucky, Nature, 1998, 396, 152.
16. D. Chen, L. Cao, F. Huang, P. Imperia, Y.-B. Cheng and R. A. Caruso, J. Am. Chem. Soc., 2010, 132, 4438.
17. D. Gu and F. Schüth, Chem. Soc. Rev., 2014, 43, 313.
18. W. Li, Z. Wu, J. Wang, A. A. Elzatahry and D. Zhao, Chem. Mater., 2014, 26, 287.
19. W. Yue, C. Randorn, P. S. Attidekou, Z. Su, J. T. S. Irvine and W. Zhou, Adv. Funct. Mater., 2009, 19, 2826.
20. Z. Zhang, F. Zuo and P. Feng, J. Mater. Chem., 2010, 20, 2206.
21. R. Y. Zhang, A. A. Elzatahry, S. S. Al-Deyab and D. Y. Zhao, Nano Today, 2012, 7, 344.
22. J. S. Chen, C. Chen, J. Liu, R. Xu, S. Z. Qiao and X. W. Lou, Chem. Commun., 2011, 47, 2631.
23. M. Ye, Q. Zhang, Y. Hu, J. Ge, Z. Lu, L. He, Z. Chen and Y. D. Yin, Chem. – Eur. J., 2010, 16, 6243.
24. J. V. Olsen, B. Blagoev, F. Gnad, B. Macek, C. Kumar, P. Mortensen and M. Mann, Cell, 2006, 127, 635.
25. E. L. Huttlin, M. P. Jedrychowski, J. E. Elias, T. Goswami, R. Rad, S. A. Beausoleil, J. Villen, W. Haas, M. E. Sowa and S. P. Gygi, Cell, 2010, 143, 1174.
26. W. Li, J. Yang, Z. Wu, J. Wang, B. Li, S. Feng, Y. Deng, F. Zhang and D. Zhao, J. Am. Chem. Soc., 2012, 134, 11864.
27. W. Li and D. Zhao, Adv. Mater., 2013, 25, 142.
28. D. Li, X. Lu, H. Lin, F. Ren and Y. Leng, J. Mater. Sci.: Mater. Med., 2013, 24, 489.
29. Y. D. Wang, C. L. Ma, X. D. Sun and H. D. Li, J. Non-Cryst. Solids, 2003, 319, 109.
30. H. B. Wu, H. H. Hng and X. W. Lou, Adv. Mater., 2012, 24, 2567.
31. J. Livage, M. Henry and C. Sanchez, Prog. Solid State Chem., 1988, 18, 259.
32. W. Li, F. Wang, S. S. Feng, J. X. Wang, Z. K. Sun, B. Li, Y. H. Li, J. P. Yang, A. A. Elzatahry, Y. Y. Xia and D. Y. Zhao, J. Am. Chem. Soc., 2013, 135, 18300.
33. X. W. Lou and L. A. Archer, Adv. Mater., 2008, 20, 1853.
34. Z. D. Lu, M. M. Ye, N. Li, W. W. Zhong and Y. D. Yin, Angew. Chem., Int. Ed., 2010, 49, 1862.
35. Z. Lu, J. Duan, L. He, Y. Hu and Y. D. Yin, Anal. Chem., 2010, 82, 7249.
36. Y. Li, J. S. Wu, D. W. Qi, X. Q. Xu, C. H. Deng, P. Y. Yang and X. M. Zhang, Chem. Commun., 2008, 564.
37. Y. Li, X. Q. Xu, D. W. Qi, C. H. Deng, P. Y. Yang and X. M. Zhang, J. Proteome Res., 2008, 7, 2526.
38. C. T. Chen and Y. C. Chen, J. Biomed. Nanotechnol., 2008, 4, 73.
39. J. H. Wu, X. S. Li, Y. Zhao, Q. Gao, L. Guo and Y. Q. Feng, Chem. Commun., 2010, 46, 9031.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4mh00030g

---