Controlled synthesis of mesoporous nanostructured anatase TiO₂ on a genetically modified Escherichia coli surface for high reversible capacity and long-life lithium-ion batteries

Abstract

TiO₂ is a promising anode material for lithium-ion batteries. The electrochemical performance of TiO₂ can be improved by optimization of nanostructures. The present study was proposed to control the synthesis of mesoporous nanostructured anatase TiO₂ on a genetically modified Escherichia coli surface. A recombinant protein INP-SiliSila containing functional domains of silicatein-α and silaffin was constructed and expressed on the E. coli surface. Deposition of the TiO₂ precursor was facilitated by INP-SiliSila on the E. coli surface. Upon calcination, TiO₂ coating on the E. coli surface transformed to anatase and formed well-defined rod-shaped particles. The electrochemical performance of the as-prepared anatase TiO₂ as anode electrodes was improved and better than that of most reported ones. The present study not only provides an organism-based approach for fabricating nanostructured anatase TiO₂ with enhanced electrochemical performance, but also opens a new avenue to take advantage of genetically modified bacterial surfaces in the synthesis and structure control of materials.

---

Introduction

Lithium-ion batteries (LIBs) are widely used in consumer electronics, transportation and large-scale renewable energy storage.¹ The fast development of electric vehicles challenges scientists to design and synthesise materials with superior energy density, excellent capacity and long cycle life. Over the past two decades, TiO₂ has been extensively investigated for its chemical and physical properties, mainly excellent photocatalytic and electrochemical properties. Applications of TiO₂ include photocatalysis, photovoltaics, photocatalytic water splitting and electrochemical energy storage.^2,3 TiO₂ is a promising substitute for commercial graphite anode material for LIBs since it shows high electrochemical activity, low toxicity, chemical stability, low price, and safety.^3,4 As an anode material, anatase TiO₂ operates at a relatively high lithium insertion/extraction voltage. It may efficiently avoid the formation of solid electrolyte interface (SEI) layers on the anode and enhance safety of batteries. The volume change of TiO₂ is negligible (less than 4%) during lithium ion intercalation/deintercalation processes. This affords TiO₂ outstanding structure stability and ensuring extended cycle life.

However, the poor ionic and electronic conductivity of TiO₂ limit the specific capacity of TiO₂ and affect the use of TiO₂ in high-power/energy-density LIBs.^5,6 To address these issues, TiO₂ with different morphologies and hierarchical structures have been investigated.⁷ It was found that tailoring the particle size of TiO₂ and constructing porous channels can improve the practical capacity and rate capability by providing good access of electrolyte to the electrode surface, shortening the Li⁺ intercalation/deintercalation pathway, and facilitating charge across the electrode/electrolyte interface.⁸ For example, Lupo et al. prepared mesoporous TiO₂ anatase nanocrystals with good rate capability and excellent stability upon very prolonged cycling.⁹ Wang et al. synthesized a well-designed mesoporous nanostructured TiO₂ with fine anatase crystalline and good electrochemical performance.¹⁰ Ren et al. synthesized uniform multishelled TiO₂ hollow microspheres and achieved a superior capacity with minimal irreversible capacity.¹¹ Yang et al. prepared sandwich-like porous TiO₂reduced graphene oxide (rGO) composites and showed high capacity.¹² Liu et al. synthesized TiO₂/graphitic carbon hollow spheres with a high specific capacity of 137 mA h g⁻¹ (after 1000 cycles) at a current density of 1 A g⁻¹.¹³ Li et al. used a general strategy to synthesize uniform mesoporous TiO₂/graphene/mesoporous TiO₂ sandwich-like nanosheets and exhibit extra high capacity of 247 mA h g⁻¹ at a current density of 20 mA g⁻¹.¹⁴ These studies suggest that mesoporous anatase TiO₂ nanostructure improves the surface area and provides more sites for Li⁺ insertion and high capability for Li⁺ storage. Therefore, controlling and optimizing the nanostructure of TiO₂ is essential for achieving high performance of LIBs.

There have been many approaches for synthesizing nanostructured materials.^2,15,16 Inspired by natural materials synthesis processes such as bones and shells, scientists took advantages of biological scaffolds and biocatalytic methods for synthesizing and controlling nanostructured materials.^17,18 For example, Lee et al. equipped multiple viruses with peptides having affinity to single-walled carbon nanotubes and amorphous iron phosphate, and produced materials with power performances and excellent capacity retention.¹⁹ Chen et al. used proteins as the inducer and achieved the protein-mediated synthesis of nanostructured titania with different polymorphs.²⁰ Xue et al. investigated the bacterial surface in controlling and directing silica phase transformation.²¹ Ping et al. and Wang et al. controlled mineralization under the functions of recombinant proteins.^22,23

This study aims to control the synthesis of nanostructured anatase TiO₂ on genetically modified Escherichia coli surface. Functional domains of silaffin and silicatein-α were selected to modify the E. coli surface by means of bacterial surface display technique under the function of ice nucleation protein (INP). Silicatein is a member of the cathepsin L family.²⁴ It is often used in biocatalytic nanomaterial synthesis including silica, titanium dioxide, gallium oxide, barium oxofluorotitanate.^25–28 The silaffin polypeptides are derived from Cylindrotheca fusiformis. These peptides are major components of the intricate silica cell wall. Recent studies suggest that these peptides are capable of precipitating silica or other materials in vitro and at ambient conditions.^29,30 Bacterial surface display is a technique that modifies the cell surface at the genetic level.^31,32 Ice nucleation protein (INP) is often used for anchoring proteins of interest on cell surface.^33–35 It facilitates the presentation of recombinant proteins or peptides on cell surface and provided the cell surface with new functions or properties. In the present study, we examined the morphology and physical properties of nanostructured TiO₂ synthesized on genetically modified E. coli surface. Electrochemical performance of as-prepared TiO₂ was investigated to find whether the material could exhibit high reversible capacity, excellent cycling performance and superior rate capacity.

Experimental

Gene selection and construction of expression vectors

The DNA segment containing segments of INP, silicatein-α and silaffin was synthesized by Sangon Biotech Co. and cloned into plasmid pET28a (Novagen, Germany) via conventional molecular manipulation. The plasmid was named as pXSH1 (pET28a-INP-SiliSila). A plasmid pXSH2 (pET28a-INP) harbouring INP was constructed as the control group to pXSH1.

Protein expression on bacterial surface

Protein expression on bacterial surface was based on previously reported procedures.^31,35 A single bacterial colony of E. coli BL21 (λDE3) harboring pXSH1 or pXSH2 was inoculated in Luria–Bertani (LB) medium containing 30 μg mL⁻¹ kanamycin and shaking at 37 °C overnight. The cell suspension was inoculated into LB medium and followed by shaking at 37 °C until an optical density at 600 nm (OD600) of 0.5–0.6 was reached. Protein expression was initiated by supplying with 1 mM isoprophyl-β-D-thiogactopyranoside (IPTG) and continuing shaking at 37 °C for 3 hours. Cells were then harvested by centrifugation at 6000g, 4 °C for 10 minutes. For detecting protein expression on bacterial surface, mixed inner and outer membrane fragments of E. coli cell membranes were prepared by cell lysis using a French press³⁶ and subjected to SDS-PAGE analysis.

Preparation of TiO₂ materials on E. coli surface

Preparation of TiO₂ materials was based on Li et al. and other approaches.^16,18,21,26 Briefly, E. coli cells with proteins of interest on cell surface were washed three times with TBS (150 mM NaCl, 50 mM Tris, pH 7.0). Cells were resuspended in TBS containing 10 wt% titanium(IV) bis (ammonium lactato) dihydroxide solution (Ti-BALDH, Sigma-Aldrich, USA). The mixture was shaken (220 rpm) for 4 days at 37 °C, followed by incubating for another 2 days at 37 °C to precipitate titanium. The solution was then transferred into water bath at 80 °C for 4 days to allow mineralization and incubated at 37 °C for another 3 days to complete mineralization. The product was harvested by centrifugation at 8000g, 4 °C for 5 minutes and washed three times with TBS, followed by overnight treatment of vacuum freeze drying. Phase transition of bio-deposited titanium was completed by calcination at 600 °C in a muffle furnace for 240 minutes, followed by controlled cooling.

Characterization of nanostructured TiO₂ materials

The morphology of TiO₂ materials was imaged using a S4800 (Hitachi Company) scanning electron microscope (SEM), fitted with a field-emission source operating at 5 kV. The TiO₂ samples were subjected to SEM imaging without additional coating treatment. TEM and HRTEM images of the samples were recorded on carbon-coated copper grids by using a JEM-2100F microscope at an acceleration voltage of 200 kV (JEOL, Japan). Analyses of crystalline and amorphous phases were performed on an D8 Advance (Bruker AXS Company) X-ray diffractometer with Cu Kα radiation (40 kV, 40 mA). XRD patterns were collected at a 2θ range of 10–80°.

Electrochemical measurements

Electrochemical experiments were performed with Swagelok-type cells with pure lithium metal as both the counter electrode and the reference electrode at room temperature. In order to make working electrodes, slurry containing 70 wt% active TiO₂ materials, 20 wt% conductive agent and 10 wt% polymer binder (polyvinylidenedifluoride, PVDF) were dispersed in N-methyle-2-pyrrolidone (NMP). Then the slurry was coated on copper foil. Highly pure lithium foil was used as the counter electrode while celgard 2325 membrane was used as a separator. The electrolyte is 1 M LiPF₆ in ethylene carbonate (EC)/diethyl carbonate (DEC) (1 : 1 by volume). The coin cells were assembled in an argon-filled glove box (Vigor, SG1200/750TS) where the oxygen and moisture contents were less than 1 ppm. The cells were galvanostatically discharged and charged using a battery testing system (LAND CT2001A) with a voltage window of 1–3 V at various current rates. Cyclic voltammetric (CV) test was performed in an electrochemical workstation (Shanghai Chenhua, CHI600E) with 1–3 V (vs. Li/Li⁺) at 0.2 mV s⁻¹.

Results and discussion

Design and surface display of INP-SiliSila

To facilitate expression of silicatein and silaffin on E. coli surface, a plasmid harboring the DNA sequence encoding segments of INP, silicatein-α and silaffin was constructed (Fig. 1, panel A). The functional domain of INP was based on the gene segment of ice nucleation protein (inaK)³⁵ (accession number AF013159) from the bacterium Pseudomonas syringae. The functional domain of silicatein-α was based on the gene segment of silicatein-α¹⁵ (accession number AJ272013) from the sponge Suberites domuncula. The functional domain of silaffin was based on the gene segment of silaffin precursor protein sil1p (sil1)²⁹ (accession number AF191634.1) from the sponge Cylindrotheca fusiformis. The resultant recombinant protein was named as INP-SiliSila (“Sili” stands for silicatein-α; “Sila” stands for silaffin) that is designed to induce deposition and mineralization of TiO₂ on E. coli surface.

Fig. 1 The expression vector pXSH1 and SDS-PAGE analysis of INP-SiliSila expression by E. coli cells containing plasmid pXSH1. Panel (A) shows expression vector pXSH1 containing the gene encoding INP-SiliSila. Panel (B) shows SDS-PAGE analysis of INP-SiliSila. Lane M, molecular weight marker; lane 1, lysate of IPTG induced cells containing plasmid pXSH1; lane 2, lysate of uninduced cells containing plasmid pXSH1; lane 3; inclusion bodies of IPTG induced cells containing plasmid pXSH1; lane 4, mixed membranes of IPTG induced cells containing plasmid pXSH1.

Expression and subcellular location of INP-SiliSila were analyzed via SDS-PAGE (Fig. 1, panel B). Expression of INP-SiliSila (with theoretic molecular weight of 58.3 kDa) was confirmed by comparing the protein expression in whole cell extracts between IPTG induced E. coli and uninduced cells. The presence of INP-SiliSila in cell membranes was observed. This implies that displaying INP-SiliSila on E. coli surface has been achieved. It was also found a significant amount of INP-SiliSila in the inclusion bodies that was due to the fast expression and misfolding of INP-SiliSila.³⁷

Deposition and mineralization of TiO₂ precursors on INP-SiliSila modified E. coli surface

E. coli cells with INP-SiliSila or INP (as control) on surface were used to induce deposition of TiO₂ precursors on E. coli surface. Zeta potential measurement indicated that the surface of the two types of cells were negatively charged in a range of pH 5.0 to 8.0 (Fig. S1, see ESI†). This facilitates interactions between the cell surface and positively charged substances. At pH 5.0 and 6.0, the two types of cells exhibited similar surface charge. At pH 7.0 and 8.0, cells displaying INP-SiliSila had lower surface charge than cells displaying INP. The morphology of E. coli cells was examined after interacting with hydrolysed titanium precursors (Ti-BALDH) in the same pH range (Fig. 2). At pH 5.0 and pH 6.0, SEM revealed a change in the roughness of the INP-SiliSila modified E. coli cell surface from smooth to rough, suggesting TiO₂ deposition on cell surface (Fig. 2, panels A and B). EDS data showed that Ti (Kα 4.5 keV and 4.9 keV) and O (Kα 0.52 keV) are two of the four major inorganic compositions of cell surface deposits (Fig. S2, panel A, see ESI†). At pH 7.0 and pH 8.0, the unchanged smooth cell surface suggested less or no TiO₂ deposition on cell surface (Fig. 2, panels C and D). At pH 5.0 and pH 6.0, there was also deposition of TiO₂ on INP modified cell surface (Fig. 2, panels E and F). However, the amount of deposited TiO₂ significantly decreased. This implies that although surface charge of cells may promote interactions between negatively charged TiO₂ precursors and bacterial surface, INP-SiliSila makes major contributions in inducing TiO₂ deposition. When titanium precursors interacted with the control group (i.e., E. coli cells with INP modified surface), it led to cell shrinkages. Although TiO₂ on cell surface was also detected via EDS, the smooth cell surface indicated TiO₂ deposition was much less than that on INP-SiliSila modified surface. These observations suggest that INP-SiliSila mediated and enhanced specific interactions between E. coli surface and TiO₂. It facilitated the formation of TiO₂ shells surrounding individual cells and prevented cell bodies from shrinkages.

Fig. 2 SEM images of TiO₂ depositing on INP-SiliSila or INP modified E. coli surface at different pH. Panels (A–D), TiO₂ depositing on INP-SiliSila modified E. coli cells at pH 5.0, pH 6.0, pH 7.0, and pH 8.0; panels (E–H), TiO₂ depositing on INP modified E. coli cells at pH 5.0, pH 6.0, pH 7.0, and pH 8.0.

To complete TiO₂ mineralization, the TiO₂ coating on E. coli surface was subjected to calcination at 600 °C for 4 hours. SEM revealed that TiO₂ depositing and mineralizing on INP modified E. coli cells was fragmentized and in an inhomogeneous morphology after calcination (Fig. 3, panel A). However, TiO₂ depositing on INP-SiliSila modified cell surface exhibited well-ordered morphology after calcination (Fig. 3, panel B). SEM revealed that the post-heating TiO₂ on INP-SiliSila modified cell surface was composed of numerous well-defined rod-shaped particles with 800–1200 nm in length and 200–400 nm in width. A high magnification image of a broken particle showed that the inside of rod-shaped particles can be hollow (Fig. 3, inset of panel B). However, TEM showed that these post-heating TiO₂ may not form a hollow structure (Fig. 3, panel C). EDS data showed that Ti and O are the two major inorganic compositions of cell surface deposits (Fig. S2, panel B, see ESI†). Other inorganic compositions such as C, P, Si, Ca were also evidenced. It is possible that titanium materials coating on E. coli is helpful in maintaining the rod-shaped structures of E. coli cells and prevents cell bodies from shrinkage. While the TiO₂ on INP modified cells was not able to serve as frameworks to maintain cell shapes as that on INP-SiliSila modified E. coli surface.

Fig. 3 Characterization of calcined TiO₂ coating on E. coli surface. (A) SEM image of calcined TiO₂ coating on INP modified E. coli surface. (B) SEM image of calcined TiO₂ coating on INP-SiliSila modified E. coli surface. (C) TEM image of calcined TiO₂ coating on INP-SiliSila modified E. coli surface. (D) HRTEM image of calcined TiO₂ coating on INP-SiliSila modified E. coli surface.

Crystalline nature of the rod-shaped TiO₂ structure was examined and confirmed by HRTEM (Fig. 3, panel D) and XRD (Fig. 4, panel A). It was observed the interplanar spacing of 0.35 nm which is assigned to (101) plane of anatase. X-ray diffraction pattern showed characteristic peaks at 25.4°, 38.0°, 48.1°, 54.4°, 62.8°, 70.5° and 75.1°. These peaks correspond to the spacing of the (101), (004), (200), (105), (204), (220), and (215) crystal planes of anatase (JCPDS card no. 21-1272). The nitrogen adsorption–desorption isotherm indicates the mesoporous structure of the Ti material with the main pore size in the range of 3–4 nm (Fig. 4, inset of panel B). The specific surface area and pore volume are determined to be 64.12 m² g⁻¹ and 0.12 cm³ g⁻¹ (Fig. 4, panel B). These self-assembled porous structure may facilitate Li⁺ and electron transportation in LIBs.^6,18

Fig. 4 XRD spectra of TiO₂ on INP or INP-SiliSila modified E. coli surface before and after calcination (panel A). Nitrogen adsorption and desorption isotherms and pore size distribution (inset) of the calcined TiO₂ on INP-SiliSila modified E. coli surface (panel B).

It is a challenge in controlling the material synthesis for deliberate nanostructures with specific functionalities.^2,16 Learning from the nature, scientists took advantages of proteins or their analogues for synthesizing nanostructured materials.^15,18,26 These proteins or analogues can function as templates or catalysts during material synthesis process. The present study proposed to control the synthesis of anatase TiO₂ on E. coli surface that had been genetically modified with a recombinant protein INP-SiliSila. Two factors may contribute to the deposition and mineralization of TiO₂ on E. coli surface. One factor is the recombinant protein INP-SiliSila that was based on functional domains of silicatein-α and silaffin. Recent reports showed that recombinant proteins can combine functions from different mineral proteins and play various roles sequentially or simultaneously to complete a specific multistage mineralization process.^22,23 In the present study, functional domains of silicatein-α and silaffin facilitated the recombinant protein INP-SiliSila in inducing and organizing TiO₂ deposition and mineralization on E. coli surface. The other factor is the E. coli surface itself. The surface energy of phospholipid membranes can determines polymorph selection during mineralization.^21,38 It may also contribute to the TiO₂ mineralization on E. coli surface. Therefore, under functions of INP-SiliSila on E. coli surface, a nanostructured anatase TiO₂ was produced with potential applications in lithium ions battery.

Electrochemical performance of TiO₂

The electrochemical performance of the as-prepared nanostructured TiO₂ was tested as the anode of LIBs. The lithium storage was measured using a standard TiO₂/Li half-cell configuration. The representative cyclic voltammograms (CVs) of the TiO₂ electrode at 0.2 mV s⁻¹ for the first few cycles at 1–3 V (vs. Li/Li⁺) displayed redox reaction at 1.7 V during discharge and at 2.05 V during charge (Fig. 5, panel A). This was attributed to insertion and extraction of Li⁺ ions in anatase phase. The ratio of cathodic to anodic peak current intensity is close to one due to the equal extent of insertion/extraction of Li ions, indicating superior charge transfer and ion diffusion kinetics behaviour.

Fig. 5 Electrochemical performance of as-prepared nanostructured anatase TiO₂ electrode. (A) Cyclic voltammogram at 0.2 mV s⁻¹ scan rates. (B) Charge–discharge voltage profile at a current rate of 0.2C. (C) and (D) Rate capability at various current rates.

Panel B in Fig. 5 presents the discharge–charge voltage profiles of TiO₂ at a current of 0.2C (1C = 167 mA h g⁻¹). The discharge capacity was 266 mA h g⁻¹ in the first cycle. The corresponding charge capacity was 234 mA h g⁻¹. The irreversible capacity loss was due to the formation of solid electrolyte interface (SEI) layer.³⁹ These observations indicated that the TiO₂ was able to accommodate Li (Li_0.8TiO₂, 266 mA h g⁻¹) during the first discharge process. Subsequently, 0.7 mol of Li per 1 mol TiO₂ (Li_0.7TiO₂, 234 mA h g⁻¹) were cycled reversibly. The coulombic efficiency was 87.8%. In the second cycle, the discharge capacity decreased to 253 mA h g⁻¹ with a corresponding increased charge capacity of 246 mA h g⁻¹, leading to a coulombic efficiency of 97%. In the third discharge–charge cycle, the coulombic efficiency increased up to 98.6%.

The rate capability of the TiO₂ electrode was measured at various current densities (Fig. 5, panel C). The high initial discharge capacity (317.8 mA h g⁻¹) is attributed to surface phenomena, such as the pseudocapacitance effect in the porous amorphous phase. The discharge–charge specific capacities were detected as 243, 224, 201, 175 mA h g⁻¹ at current rates of 1C, 2C, 5C and 10C. Superior capacity retention (retained 234.8 mA h g⁻¹) of the TiO₂ electrode was observed after 350 cycles. The cycling stability and capacity retention was more than 95% at 1C current rate. The high capacity of 243 mA h g⁻¹ at 1C can lead to a lithium insertion coefficient of 0.73 that is much higher than the theoretical value of 0.5. Porous channels provide the TiO₂ electrode with a discharge capacity of 175 mA h g⁻¹ even at a rate of 10C. Panel D in Fig. 5 depicts the rate capacity of the TiO₂ electrode at a range of rates from 0.2C to 10C after 380 cycles. The discharge–charge specific capacities were detected as 310, 241, 225, 225, 200 and 174 mA h g⁻¹ at current rates of 0.2C, 0.5C, 1C, 2C, 5C and 10C. It was found that the electrode recovered to the initial capacity of 1C after high rate of 10C, manifesting their high reversibility and stability.

The as-prepared nanostructured TiO₂ exhibited excellent electrochemical performance as the anode of lithium ions battery. The assembly of TiO₂ nanoparticles on E. coli surface efficiently reduces the stacking or aggregation of nanoparticles, which is beneficial to the high capacity and enhanced stability dramatically. Nanosized particles and mesoporous structures provide sufficient active sites for active particles for Li⁺ insertion and contacting area with liquid electrolyte, leading to the high Li⁺ storage capability. It provides spacious channels for lithium ions insertion and diffusion speed of Li⁺ and electron. The mesoporous structure not only tolerates volume changes during lithium ion insertion and extraction, but also prevents the collapse of electrode. Therefore, electrode made with the as-prepared TiO₂ facilitates rapid lithium ion and electron diffusion, leads to high Li⁺ storage capability and exhibits more superior stability than that of most reported anatase TiO₂ electrode.^40–43 Residues such as carbon in the as-prepared TiO₂ may contribute to the electrochemical performance (Fig. S2, panel B, see ESI†).

Panel A in Fig. S3 (see ESI†) shows charge–discharge voltage profiles of the TiO₂ electrode at different current rates (1C, 2C, 5C and 10C). Table S1 (see ESI†) shows the relevant discharge capacities, charge capacities and irreversible capacity losses (ICL). The ICL values of the TiO₂ electrode are lower than that of most of reported anatase TiO₂ electrodes (Tables S2 and S3, see ESI†). The lower ICL values indicate a higher reversible capacity of the TiO₂ electrode that meets requirements of practical applications. The cyclic stability of the TiO₂ electrode was further investigated at 1C and 2C rate, as shown in Fig. S3, panel B (see ESI†). The TiO₂ electrode exhibits excellent cycling stability with a capacity of 200 mA h g⁻¹ after 200 cycles at 2C, even presenting increase tendency. After that, the same electrode was evaluated at 1C for another 1000 cycles, which delivered capacity of 225 mA h g⁻¹ throughout first 500 cycles and remained 200 mA h g⁻¹ after 1000 cycles, keeping a capacity retention of 89%. The superior electrochemical performance of the TiO₂ is promising in using as lithium battery anode to meet the requirements for long-term high-power applications.

Conclusions

In the present study, we established an organism based system for controlling and synthesizing nanostructured anatase TiO₂ on genetically modified E. coli surface. An artificial recombinant protein INP-SiliSila was constructed and expressed on the surface of E. coli. The INP-SiliSila facilitates deposition and arrangement of TiO₂ precursors on E. coli surface. Upon calcination, E. coli cells serve as mineralization templates or frameworks and allow transformation of amorphous TiO₂ precursors to well assembled anatase TiO₂ nanoparticles. These TiO₂ nanoparticles are composed of nanocrystals and form mesoporous structure with high surface area. The electrochemical performance of the anatase TiO₂ as anode electrodes was better than that of most reported ones. The present study not only provides an organism based approach for fabricating a nanostructured anatase TiO₂ with enhanced electrochemical performance, but also opens a new avenue to take advantages of genetically modified bacterial surface in the synthesis and structure control of materials.

Acknowledgements

This work has been performed as part of the program for Changjiang Scholars and Innovative Research Team (IRT_15R52) of Chinese Ministry of Education. This work was also supported by the Fundamental Research Funds for the Central Universities (WUT 2016IB006) and the International Science & Technology Cooperation Program of China (2015DFE52870). The authors thank Mr H. Zhao and Miss S.-M. Wu from the Research and Test Center of Materials at Wuhan University of Technology for TEM analysis.

Notes and references

1. M. Armand and J. M. Tarascon, Nature, 2009, 451, 652.
2. X. B. Chen and S. S. Mao, Chem. Rev., 2007, 107, 2891.
3. X. B. Chen, Chin. J. Catal., 2009, 30(8), 839.
4. Z. Chen, I. Belharouak and Y. K. Sun, Adv. Funct. Mater., 2013, 23, 959.
5. L. Xin, Y. Liu, B. J. Li, X. Zhou, H. Shen, W. X. Zhao and C. L. Liang, Sci. Rep., 2014, 4, 4479.
6. J. Jin, S. Z. Huang, J. Liu, Y. Li, L. H. Chen, Y. Yu, H. E. Wang, C. P. Grey and B. L. Su, Adv. Sci., 2015, 2, 1500070.
7. M. V. Reddy, G. V. S. Rao and B. V. R. Chowdari, Chem. Rev., 2013, 113, 5364.
8. X. D. Yan, Z. H. Wang, M. He, Z. H. Hou, T. Xia, G. Liu and X. B. Chen, Energy Technol., 2015, 3, 801.
9. F. D. Lupo, A. Tuel, V. Mendez, C. Francia, G. Meligrana, S. Bodoardo and C. Gerbaldi, Acta Mater., 2014, 69, 60.
10. B. Wang, H. L. Xin, X. D. Li, J. L. Cheng, G. C. Yang and F. D. Nie, Sci. Rep., 2014, 4, 3729.
11. H. Ren, Y. B. Yu, J. Y. Wang, Q. Jin, M. Yang, D. Mao, D. Kisailus, H. J. Zaho and D. Wang, Nano Lett., 2014, 14, 6679.
12. S. L. Yang, C. Y. Cao, P. P. Huang, L. Peng, Y. B. Sun, F. Wei and W. G. Song, J. Mater. Chem. A, 2015, 3, 8701.
13. H. Liu, W. Li, D. Shen, D. Zhao and G. Wang, J. Am. Chem. Soc., 2015, 137, 13161.
14. W. Li, F. Wang, Y. Liu, J. Wang, J. Yang, L. Zhang, A. A. Elzatahry, D. Al-Dahyan, Y. Xia and D. Zhao, Nano Lett., 2015, 15, 2186.
15. Y. J. Wang, A. S. Angelatos and F. Caruso, Chem. Mater., 2008, 20, 848.
16. W. Li, Z. X. Wu, J. X. Wang, A. A. Elzatahry and D. Y. Zhao, Chem. Mater., 2014, 26, 287.
17. A. I. Laskin, S. Sariaslani and G. M. Gadd, Adv. Appl. Microbiol., 2012, 78, 55.
18. Y. C. Chang, C. Y. Lee and H. T. Chiu, ACS Appl. Mater. Interfaces, 2014, 6, 31.
19. Y. J. Lee, H. Yi, W. J. Kim, K. Kang, D. S. Yun, M. S. Strano, G. Ceder and A. M. Belcher, Science, 2009, 324, 1051.
20. G. Chen, M. Li, F. Li, S. R. Sun and D. G. Xia, Adv. Mater., 2010, 22, 1258.
21. S. H. Xue, H. Xie, H. Ping, Q. C. Li, B. L. Su and Z. Y. Fu, RSC Adv., 2015, 5, 71844.
22. H. Ping, H. Xie, B.-L. Su, Y.-B. Cheng, W. Wang, H. Wang, Y. Wang, J. Zhang, F. Zhang and Z. Fu, J. Mater. Chem. B, 2015, 3, 4496.
23. X.-L. Wang, H. Xie, B.-L. Su, Y.-B. Cheng, J.-J. Xie, H. Ping, M.-H. Wang, J.-Y. Zhang, F. Zhang and Z.-Y. Fu, J. Mater. Chem. B, 2015, 3, 5951.
24. K. Shimizu, J. Cha, G. D. Stucky and D. Morse, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 6234.
25. W. D. Marner, A. S. Shaikh and S. J. Muller, Biotechnol. Prog., 2009, 25, 417.
26. R. L. Brutchey and D. E. Morse, Chem. Rev., 2008, 108, 4915.
27. W. E. G. Muller, A. Krasko, G. L. L. Pennec and H. C. Schroder, Microsc. Res. Tech., 2003, 62, 368.
28. P. Curnow, P. H. Bessette, D. Kisailus, M. M. Murr, P. S. Daugherty and D. E. Morse, J. Am. Chem. Soc., 2005, 127, 15749.
29. D. H. Nam, K. Won, Y. H. Kim and B. I. Sang, Biotechnol. Prog., 2009, 25, 1643.
30. M. Sumper and N. Kroger, J. Mater. Chem., 2004, 14, 2059.
31. E. V. Bloois, R. T. Winter, H. Kolmar and M. W. Fraaije, Trends Biotechnol., 2011, 29, 79.
32. B. Wang, P. Liu and R. K. Tang, Bioessays, 2010, 32, 698.
33. M. Shimazu, A. Mulchandani and W. Chen, Biotechnol. Prog., 2001, 17, 76.
34. G. Georgiou, C. Stathopoulos, P. S. Daugherty, A. R. Nayak, B. L. Iverson and R. Curtiss, Nat. Biotechnol., 1997, 15, 29.
35. L. Li, D. G. Kang and H. J. Cha, Biotechnol. Bioeng., 2004, 85, 214.
36. H. Xie, S. G. Patching, M. P. Gallagher, G. J. Litherland, A. R. Brough, H. Venter, S. Y. M. Yao, A. M. L. Ng, J. D. Young, R. B. Herbert, P. J. F. Henderson and S. A. Baldwin, Mol. Membr. Biol., 2004, 21, 323.
37. S. M. Singh and A. K. Panda, J. Biosci. Bioeng., 2005, 99, 303.
38. J. Xiao, Z. Wang, Y. Tang and S. Yang, Langmuir, 2009, 26, 4977.
39. S. H. Yu, D. E. Conte, S. Baek, D. C. Lee, S. K. Park and K. J. Lee, Adv. Funct. Mater., 2013, 23, 4293.
40. Y. X. Tang, Y. Y. Zhang, J. Y. Deng, D. P. Qi, W. R. Leow and J. Q. Wei, Angew. Chem., Int. Ed., 2014, 53, 13488.
41. Y. Q. Zhang, F. Du, X. Yan, Y. M. Jin, K. Zhu, X. Wang, H. M. Li, G. Chen, C. Z. Wang and Y. J. Wei, ACS Appl. Mater. Interfaces, 2014, 6, 4458.
42. M. V. Reddy, G. V. Subba and B. V. R. Chowdari, J. Mater. Chem., 2011, 21, 10003.
43. M. V. Reddy, R. Jose, T. H. Teng, B. V. R. Chowdari and S. Ramakrishna, Electrochim. Acta, 2010, 55, 3109.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra09974b

---