Confined germanium nanoparticles in an N-doped carbon matrix for high-rate and ultralong-life lithium ion batteries

Abstract

In this study, a relatively simple and direct method is used to prepare germanium nanoparticles (Ge NPs) embedded in the pore tunnels of an N-doped mesoporous carbon matrix. In the Ge/CMK-3 nanocomposite, the highly ordered porous structure and large pore volume guarantee a sufficient Ge loading and buffer the large volume changes of Ge during the discharge/charge cycles. More specifically, the mesoporous carbon matrix can supply sufficient pathways for Li⁺ and electron transport to the encapsulated nanometer-sized Ge, as well as restrain the agglomeration and growth of Ge during the crystallization process. Accordingly, the electrode of Ge/CMK-3 attained a capacity as high as 755.7 mA h g⁻¹ at 500 mA g⁻¹ after 420 cycles with a capacity retention of 93.3% based on the 11th cycle. The study shows that the electrochemical properties of Ge/CMK-3 are significantly improved compared to that of the bulk Ge anode, and it demonstrates that Ge/CMK-3 could potentially show promise as an anode material for energy storage.

---

Introduction

Among the recently developed anode materials, group IVA (Si, Ge, Sn) elements have been intensively studied due to their relatively high theoretical specific capacities when alloyed with lithium, leading to their consideration as promising candidate anode materials to take the place of graphite in lithium-ion batteries (LIBs).^1–6 Ge has attracted less attention because of its lower capacity and higher cost when compared with Si. However, Ge has several advantages that outperform Si. Although the gravimetric capacity of Ge at room temperature (1384 mA h g⁻¹ on the basis of Li₁₅Ge₄) is much lower than that of silicon (3579 mA h g⁻¹), the volumetric capacity of Ge (7366 A h cm⁻³) is second only to that of Si (8334 A h cm⁻³).⁷ Moreover, the diffusivity of Li ion in Ge is 400 times faster than that in Si, and the electrical conductivity of Ge is 10⁴ higher than that of Si at room temperature,^3,8,9 which allows Ge to be a more promising anode for high-power LIBs. Furthermore, Ge-based anodes show isotropic lithiation behavior, but Si-based ones exhibit highly anisotropic behavior, as well as nonhomogeneous alloying pathways with Li ions.^10–12 The phenomenon of isotropic lithiation can minimize fracturing in anodes, thus may lead to Ge-based anodes exhibiting highly reversible capacities.

Despite these promising characteristics, Ge also suffers from large volume changes upon the lithiation and de-lithiation processes of prolonged cycling, similar to Si.^13,14 The increasing mechanical stress results in significant electrode pulverization, which leads to rapid capacity fading and hampers its practical usage. To solve this problem, various strategies have been implemented in the design of Ge-based materials to accommodate such huge volume changes. These include reducing the particle size,^15,16 dispersing Ge in an active/inactive matrix,¹⁷ forming one-dimensional nanostructures,^18,19 constituting porous structures,^20,21 and coating materials with the Ge anode in order to improve their electrochemical performances. Among these numerous methods, a carbon buffer layer is the best choice for Ge, with several benefits such as high electrical conductivity, excellent rate performance and long cycle life. For example, Cui’s group successfully synthesized n-C/Ge by a solid-state pyrolysis of tetraallylgermane.²² Ngo et al. produced Ge nanocrystallites interconnected by carbon by thermal decomposition of a Ge–citrate complex.²³ Hwang and co-workers prepared Ge–MWCNT by a CVD method, this Ge–MWCNT nanocomposite exhibits a high reversible capacity of over 800 mA h g⁻¹ at 1C even after 200 cycles.²⁴ We have also presented a Ge–RGO–CNT nanocomposite by using a facile method, which showed enhanced electrochemical performance.²⁵ Compared with graphene and CNTs, CMK-3 as a mesoporous carbon has attracted much consideration for use as a porous matrix and conductive material for energy storage due to its many favourable characteristics.^26–28 One especially favourable feature of CMK-3 is that it can act as a conductive framework to load other materials, which results in both components of the composite making a contribution to control the overall characteristics.

Here, we demonstrate a simple method for the preparation of mesoporous Ge/CMK-3 nanocomposite for high-performance LIBs using CMK-3 as a template by a nanocasting technique, in which the Ge NPs are embedded in an N-doped carbon matrix. The CMK-3, with abundant mesoporosity, affords sufficient inner space for the volume changes of Ge and acts as a rigid support for the Ge NPs that prevents growth and agglomeration of Ge NPs during annealing. Additionally, the internal porosity provides good access to the electrolyte, allowing fast charge transfer and full lithiation. With the help of the structures, the capacity of the Ge/CMK-3 composite electrode could be stabilized at 755.7 mA h g⁻¹ at 500 mA g⁻¹ after 420 cycles with a capacity retention of 93.3%, while the capacity of the bulk Ge rapidly decreases in the first ten cycles under the same experimental conditions.

Experimental section

Preparation of materials

SBA-15 was used as the CMK-3 template, which was prepared according to the previous literature.²⁶ Briefly, 100 mg of CMK-3 was treated in 6 M HNO₃ solution under stirring for 60 min at 80 °C to render it more easily dispersible in deionized water (DI). Then, 100 mg of the above-mentioned CMK-3 and 800 mg GeO₂ were ultrasonically dispersed in 19 mL DI water and 1 mL ethanediamine (EDA) for 2 h. Subsequently, the above solution was stirred under vacuum for 48 h to ensure a full infiltration of the pores. The solution was then freeze dried and ground. Following this, the products were reduced at 600 °C for 5 h in Ar with 5% H₂ to the obtain Ge/CMK-3 nanocomposite. The contrasting sample was fabricated under the same conditions without the addition of CMK-3.

Characterization of materials

The microstructure of the composites was measured by scanning electron microscopy (SEM, HITACHI S-4800) and transmission electron microscopy (TEM, JEOL JEM-2010). X-ray diffraction (XRD) (Bruker D8 advance) with Cu Kα radiation was used to characterize the crystal structures of the samples. Nitrogen adsorption/desorption was characterized by Brunauer–Emmett–Teller (BET) measurements using an ASAP-2010 surface area analyzer. Thermogravimetric analysis was performed on a TG instrument (NETZSCH STA 409 PC) with a heating rate of 10 °C min⁻¹ from 30 to 750 °C under ambient conditions. The X-ray photoelectron spectroscopy (XPS) analyses were carried out using a Perkin-Elmer PHI 550 spectrometer with Al Kα (1486.6 eV) as the X-ray source.

Electrochemical tests

Electrochemical characterization was performed by galvanostatic charge/discharge in a 2016 coin-type cell. The mass ratio of samples, carbon black, polyacrylic acid and carboxymethyl cellulose binder was 70 : 15 : 7.5 : 7.5, with DI water as solvent, and a working electrode was made by uniformly coating the slurry on a copper foil. The average mass loading of active material was about 0.96 mg cm⁻². The cells were assembled with Li metal as the counter electrode and the polypropylene film as separator in an Ar filled glovebox. The electrolyte solution was composed of 1 M LiPF₆ solution in a mixture of dimethyl carbonate (DMC) and ethylene carbonate (EC) (1 : 1, V) containing 2% (V) fluoroethylene carbonate (FEC). Galvanostatic charge/discharge cycles were tested at voltages of 0.01–1.5 V (vs. Li/Li⁺) using a CT2001A cell test instrument (LAND Electronic Co.).

Results and discussion

Fig. 1 illustrates the synthetic procedures of Ge/CMK-3 composites. The SBA-15 was used as a hard template and sucrose as the carbon source, then, a surface oxidation was performed in concentrated HNO₃ to obtain functional oxygenated groups in the carbon matrix such as carboxyl, hydroxy and carbonyl to generate a hydrophilic surface. Afterwards, the ethanediamine (EDA) was added to the CMK-3 aqueous solution; due to the electrostatic interaction between EDA(+) and CMK-3(−), EDA is first enriched in the channels of CMK-3. Then, the GeO₂ hydrolyze to form Ge(OH)₄ is linked to EDA by hydrogen-bond interactions, facilitating the generation of a Ge(OH)₄/CMK-3 solution, followed by freeze drying to form the precursor. The interaction with metal ions drives the facile infiltration of precursors in the pores, resulting in a full loading of precursors into the pore system. Finally, the Ge/CMK-3 nanocomposite was achieved by annealing in an Ar atmosphere with 5% H₂. More details of the experiments are given in the Experimental section.

Fig. 1 The synthetic procedure of the Ge/CMK-3 nanocomposite.

A typical X-ray diffraction (XRD) pattern of mesoporous Ge/CMK-3 composites was shown in Fig. 2a, and all the diffraction peaks were indexed as cubic Ge. The sharp diffraction peaks implied that the achieved Ge was well crystallized. Presumably due to the low graphitization degree, a weak and broad peak located at 20–28° was observed, which should be ascribed to a typical characteristic peak of carbon. The small angle X-ray diffraction (SAXRD) patterns of the Ge/CMK-3 nanocomposite and CMK-3 carbon template are shown in Fig. 2b. The (200), (110) and (100) diffraction peaks of the hexagonal structure can be detected in the SAXRD pattern (inset in Fig. 2b) of the CMK-3 mesoporous carbon matrix which can be indexed to show the 2D hexagonal mesostructure, indicating a highly ordered and long range regularity feature of the mesoporous structures.^29,30 For the mesoporous Ge/CMK-3 composite, the intensity of (200) and (110) diffraction peaks decreased after the complete insertion of Ge NPs, suggesting that during the impregnation process, the ordered pore tunnels of CMK-3 are partly destroyed or disturbed because of filling effects. Raman spectroscopy was employed to further confirm the crystalline and the carbon phase of this composite (Fig. 2c). A weak and sharp peak at 298 cm⁻¹ is ascribed to Ge–Ge vibration,^22,31 the small bulge at 445 cm⁻¹ corresponds to the symmetric Ge–O–Ge stretching,¹⁴ due to the surface oxidation of Ge NPs. Two peaks located at approximately 1348 and 1588 cm⁻¹ exhibit typical characteristics of amorphous carbon, which correspond to the D and G bands, respectively. Moreover, the high intensity ratio of the D band to the G band (I_D/I_G = 0.904) demonstrates the existence of many defects and non-graphitic carbon in the hybrid, which is consistent with the XRD results. A thermogravimetric (TGA) profile was used to evaluate the mass ratio of CMK-3 in the Ge/CMK-3 composite. A large mass loss was observed in the TGA profile between 400–600 °C which can be attributed to the consumption of CMK-3 (Fig. 2d). Thus, it can be concluded that the Ge/CMK-3 composite is composed of 32.2 wt% CMK-3 and 67.8% Ge according to the weight loss upon the carbon decomposition, and that Ge is fully oxidized to GeO₂ in air.

Fig. 2 (a) X-ray diffraction pattern of Ge/CMK-3. (b) SAXRD pattern of Ge/CMK-3 nanocomposites; the inset in (b) is the SARXD pattern of the original CMK-3 carbon template. (c) Raman spectrum of Ge/CMK-3. (d) Thermogravimetric analysis (TGA) of Ge/CMK-3 composite.

The chemical composition of the surface layer of the Ge/CMK-3 was further investigated by X-ray photoelectron spectroscopy (XPS). A full XPS spectrum revealed the presence of Ge, C, O, and N species in the composite, as shown in Fig. 3a. Fig. 3b shows a high-resolution section of the Ge 3d XPS spectrum, three major binding energies of 29.8 eV, 30.4 eV and 33.4 are attributed to Ge–Ge, Ge–C, and Ge–O bonds respectively.^23,34 The relatively low intensity of Ge–Ge is possibly because of the partial surface oxidization of the Ge NPs, and the fact that the test depth of XPS is less than 2.2 nm.^32,33 However, in the XRD pattern no diffraction peaks of germanium oxides were observed, meaning they were of an amorphous structure. Besides that, the existence of germanium oxides was in agreement with the Raman result, and this phenomenon is also observed in other Ge-based anode materials.^34,35 Fitting the N 1s peaks in Fig. 3c, strong peaks at 400.6 eV and 398.85 eV indicate the presence of pyrrolic N as well as pyridinic N in the N-doped carbon frameworks, and the peak located at 397.77 eV is attributed to the Ge–N bonds in the Ge/CMK-3.³⁶ These results demonstrate that the composite precursor with EDA transformed to N-doped carbon with Ge particles after annealing. The nitrogen atom has additional lone pairs of electrons, and contributes additional electrons and provides electron carriers for the conduction band,^37,38 thereby, N-doped CMK-3 shows high electron conductivity. As a consequence, the high electrical conductivity of N-doped CMK-3 decreases the charge transfer resistance of anode materials.^39,40 This is beneficial for Li⁺ intercalation/deintercalation into the anodes.⁴¹ The high resolution spectrum of C 1s is shown in Fig. 3d, besides the main peaks located at 284.6 (graphite, C=C) and 285.1 (C=N & C–O), the weak peaks centered at 286.40 eV could be ascribed to C–N & C=O, and that at 289.2 eV attributed to C=O.⁴²

Fig. 3 XPS spectra of a typical Ge/CMK-3 nanocomposite: (a) survey spectrum, (b) Ge 3d, (c) N 1s, and (d) C 1s.

Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were used to characterize the morphology and structure of the as prepared samples. Fig. 4a shows that typical SEM images of the CMK-3 template display a columnar-like morphology with a worm-like mesoporous structure, which is similar to that of the mesoporous silica SBA-15. From Fig. 4b, the [100] directions of the mesoporous tunnels of CMK-3 can be clearly observed. Inset in Fig. 4b, the 2-D hexagonal symmetrical mesostructure with united pore size of about 4 nm has been confirmed by the image taken from [001] direction, which shows good consistency with the Brunauer–Emmett–Teller (BET) results. After impregnation with the Ge precursor and thermal treatment at 600 °C, the Ge/CMK-3 nanocomposite displays a columnar-like shape and no discernible Ge NPs were found. TEM was conducted to further confirm the dispersion of Ge NPs in the CMK-3. From Fig. 4d, the presence of Ge NPs can be recognized in the ordered tunnels of the CMK-3 for the Ge/CMK-3 composites when compared with the initial CMK-3. Probably due to the weak contrast between the CMK-3 templates, they were not very distinct. The Ge NPs had a uniform size of about 6 nm (inset in Fig. 4d), and no bulk aggregation of nanoparticles was observed on the outer surface. Based on the above analysis, we believe that the Ge NPs were encapsulated into CMK-3 ordered nanoholes. In contrast, the bulk Ge obtained without adding the CMK-3 template reveals aggregates of around micron size in diameter (Fig. S1, ESI†), demonstrating that the CMK-3 matrix prevents the agglomeration of Ge and restricts the growth during the crystallization process.

Fig. 4 (a) SEM image and (b) TEM image of the CMK-3 template, (c) SEM image and (d) TEM image of Ge/CMK-3.

In order to further testify the presence of Ge in the CMK-3 host, a nitrogen isothermal adsorption experiment was carried out. The nitrogen adsorption desorption isotherms and pore size distribution profiles of the CMK-3 and Ge/CMK-3 nanocomposite were shown in Fig. 5. A distinct step for both samples between relative pressures (P/P₀) of 0.4 and 0.5 was observed, demonstrating typical characteristics of mesoporous materials.^43,44 The specific surface areas, mean pore diameters, and pore volumes of CMK-3 and Ge/CMK-3 are listed in Table 1, and these parameters for pristine SBA-15 and bulk Ge were also given in Fig. S2, ESI.† As shown in Table 1, the surface area is found to decrease significantly from 837.4 m² g⁻¹ for CMK-3 to 269.03 m² g⁻¹ for Ge/CMK-3 nanocomposite, accompanied by a decrease in pore volume, from 0.98 cm³ g⁻¹ to 0.27 cm³ g⁻¹, illustrating that the pores in the CMK-3 samples were filled with Ge nanocrystals. This result provides strong evidence that Ge NPs were indeed incorporated into the nanoholes of CMK-3, which is in good agreement with the TEM analysis.

Fig. 5 N₂ adsorption–desorption isotherm of (a) CMK-3 and (b) Ge/CMK-3. The inset shows the pore size distribution calculated using the BJH method.

Table 1 Properties of the CMK-3 and Ge/CMK-3 samples

Sample	BET surface area (m^2 g^−1)	Pore volume (cm^3 g^−1)	Pore size (nm)
CMK-3	837.4	0.988	4.718
Ge/CMK-3	269.0	0.270	4.020

---

The electrochemical properties of the Ge/CMK-3 composite were measured by galvanostatic cycling in a Li half-cell. For comparison, bulk-Ge was also synthesized and tested under the same conditions. Specific capacity was calculated based on the total mass of the Ge/CMK-3 composite. The charge–discharge curves of bulk Ge at a current density of 100 mA g⁻¹ between 0.01 and 1.50 V are shown in Fig. 6a. The first lithiation/de-lithiation capacities of bulk Ge were 1524.5/944.3 mA h g⁻¹ with the coulombic efficiency (CE) of 61.9%. The low initial CE of bulk Ge can be attributed to the disruption and pulverization of the microstructure of the electrode, leading to a large irreversible capacity loss. However, distinctly different cycling behavior was displayed when the same experimental protocols were carried out on the Ge/CMK-3 composite sample (Fig. 6b). The initial discharge/charge capacity of Ge/CMK-3 was 1505.1/1017.5 mA h g⁻¹, corresponding to a higher CE of 67.6%. The high irreversible capacity in the first cycle may be attributed to the formation of a SEI layer on the surface of the material and the decomposition of the electrolyte. In addition, the voltage–capacity profile of bare CMK-3 was also tested (Fig. S3, ESI†), the large irreversible capacity of this carbon matrix may also result in the low initial CE of the composite. But after the first cycle, the reversibility was improved in the charge/discharge process, indicating better stability of the electrode by virtue of the volume buffer effect of the pore structure and the good electrical conductivity of the N-doped carbon matrix.

Fig. 6 Charge/discharge profiles of (a) Ge/CMK-3 nanocomposite and (b) bulk Ge at a current density of 100 mA g⁻¹. (c) Rate capability tests for Ge/CMK-3 nanocomposite and bulk Ge at various current densities. (d) Cycling performance of Ge/CMK-3 nanocomposite and bulk Ge measured at 500 mA g⁻¹.

To identify the electrochemical reactions, the corresponding differential charge–discharge capacity versus voltage profiles of Ge/CMK-3 and Ge are presented in Fig. S4, ESI.† The differential curves of bulk-Ge show the presence of characteristic peaks, corresponding to Li_xGe at 0.18 and 0.31 V, during the delithiation, the peak at about 0.47 and 0.61 V showing the phase transformation to a-Li–Ge.⁴⁵ The differential curves of Ge/CMK-3 have similar characteristic peaks to those of bulk Ge, but during the lithiation, another peak was observed at 0.58 V, corresponding to the formation of Li₉Ge₄.^46,47 The higher CE in the first cycle indicates that the carbon matrix plays a key role in decreasing the side reactions with the electrolytes.

The rate capability of Ge/CMK-3 and the bulk Ge electrode were evaluated as shown in Fig. 6c. The electrodes were tested at various current densities from 100 mA g⁻¹ to 6400 mA g⁻¹ for each of 10 cycles. The Ge/CMK-3 nanocomposites obtained a high reversible capacity of 997.4 mA h g⁻¹ at 100 mA g⁻¹. A small decrease in discharge capacity can be observed when the cycle rate was varied from 997.4 at 100 mA g⁻¹ to 765.8 mA h g⁻¹ at 800 mA g⁻¹. Upon further increase to a current density of 6400 mA g⁻¹, a capacity of 125.6 mA h g⁻¹ could still be obtained. When returning back to 100 mA g⁻¹ after 80 cycles, the discharge capacity recovered to 934.6 mA h g⁻¹, corresponding to 93.7% of the initial capacity attained at 100 mA g⁻¹ and it should be noted that the CE varied little upon changing the current densities. In contrast, the bulk Ge exhibited lower average capacities of 280.9 mA h g⁻¹ at 100 mA g⁻¹, and only 14.6 mA h g⁻¹ at 6400 mA g⁻¹, which clearly demonstrated the strong collaborative effect between the mesoporous carbon and Ge NPs. In the Ge/CMK-3 composite, the Ge NPs were confined in the ordered mesoporous carbon, which is favourable for good adherence with a large contact surface area, helping to establish good electrical contact with the conducting matrix. Additionally, the ordered nanoholes can buffer the volume expansions during the lithiation process. The cycling performance of bulk Ge and Ge/CMK-3 electrodes during charging and discharging at 500 mA g⁻¹ at the voltage range of 0.01–1.5 V (vs. Li/Li⁺ ratio) was shown in Fig. 6d. We can observe that the capacity of bulk Ge, from an initial 1524.5 mA h g⁻¹ to 21 mA h g⁻¹ after 200 cycles, displays a rapid capacity loss. This was due to the large volume changes, which caused the continuous formation of an SEI layer that consumed the active materials and the disconnection between Ge particles and the current collector. On the contrary, the Ge/CMK-3 composite demonstrated remarkable discharge capacity retention during the cycling. After 420 cycles the discharge capacity of the nanocomposite was 755.7 mA h g⁻¹ at a high current density of 500 mA g⁻¹, with only 6.7% capacity loss (based on the 11th cycle, the initial 10 cycles were carried out at 100 mA g⁻¹). The CE was close to 100%, indicating that the perfectly encapsulated Ge NPs in the N-doped carbon matrix form a stable structure and that the electrochemical lithiation/de-lithiation is completely reversible even at high current density. The SEM images before and after 200 charge/discharge cycles are shown in Fig. S5, ESI.† Some cracks in the electrode after 200 cycles can be observed, but the structure of the composite electrode still remains; significant electrode pulverization caused by Ge volume changes did not occur in this electrode. This turned out to be a proof of the structural stability of the composite electrode.

We believe that the enhanced electrochemical properties of the composite electrode can be mainly ascribed to the following three factors as illustrated in Fig. 7. First of all, the carbon matrix avoids the aggregation of Ge NPs and provides free space to buffer the volume expansion during the charge/discharge cycles. Secondly, the electrode consists of Ge NPs embedded in an N-doped ordered mesoporous carbon matrix, providing the probability of combining the high Li storage of Ge with the good electronic conductivity of the carbon matrix into one incorporated entity. Finally, the large surface areas of the porous structures also can permit large lithium-ion flux across the interface, and facilitate electron transfer to the current collector through the mesoporous structure. All the above factors contributed to the good cycling performance, and high rate capability of the composites in LIBs.

Fig. 7 Schematic representation of the electrochemical reaction pathway of the mesoporous Ge/CMK-3 nanocomposite.

Conclusions

In summary, we reported a high-performance Ge based anode via encapsulation of Ge NPs into a highly ordered N-doped mesoporous carbon. In this composite, the Ge NPs were uniformly confined in the ordered carbon channels with high electronic conductivity. The mesoporous structure can not only accommodate the large volume changes of the Ge particles during charge and discharge cycles, but can also ensure infiltration of the electrolyte into mesopores in lithium ion batteries, leading to the excellent electrochemical properties regarding reversible capacity, cycling performance, and rate capability in the lithium ion batteries. This approach provides a means of improving the electrochemical performance of Ge-based anode materials for use in lithium storage.

Acknowledgements

This work was supported by the National Basic Research Program of China (973 Program) (No. 2014CB239701), the National Natural Science Foundation of China (No. 21173120, 51372116), the Natural Science Foundation of Jiangsu Province (BK2011030, BK2011740), the Fundamental Research Funds for the Central Universities of NUAA (NP2014403) and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

References

1. D. M. Piper, J. H. Woo, S. B. Son, S. C. Kim, K. H. Oh and S. H. Lee, Adv. Mater., 2014, 26(21), 3520–3525.
2. X. L. Li, M. Gu, S. Y. Hu, R. Kennard, P. F. Yan, X. L. Chen, C. M. Wang, M. J. Sailor, J. G. Zhang and J. Liu, Nat. Commun., 2014, 5, 4105–4111.
3. J. Liu, K. P. Song, C. B. Zhu, C. C. Chen, P. A. van Aken, J. Maier and Y. Yu, ACS Nano, 2014, 8(7), 7051–7059.
4. A. Jahel, C. M. Ghimbeu, L. Monconduit and C. Vix-Guterl, Adv. Energy Mater., 2014, 4(11), 1400025–1400031.
5. L. Zhang, H. B. Wu, B. Liu and X. W. Lou, Energy Environ. Sci., 2014, 7(3), 1013–1017.
6. R. Y. Zhang, Y. J. Du, D. Li, D. K. Shen, J. P. Yang, Z. P. Guo, H. K. Liu, A. A. Elzatahry and D. Y. Zhao, Adv. Mater., 2014, 26(39), 6749–6755.
7. X. L. Wu, Y. G. Guo and L. J. Wan, Chem.–Asian J., 2013, 8(9), 1948–1958.
8. D. W. Wang, Y. L. Chang, Q. Wang, J. Cao, D. B. Farmer, R. G. Gordon and H. J. Dai, J. Am. Chem. Soc., 2004, 126(37), 11602–11611.
9. C. K. Chan, X. F. Zhang and Y. Cui, Nano Lett., 2008, 8(1), 307–309.
10. J. M. Yuk, H. K. Seo, J. W. Choi and J. Y. Lee, ACS Nano, 2014, 8(7), 7478–7485.
11. J. H. Cho and S. T. Picraux, Nano Lett., 2014, 14(6), 3088–3095.
12. M. Gu, H. Yang, D. E. Perea, J. G. Zhang, S. L. Zhang and C. M. Wang, Nano Lett., 2014, 14(8), 4622–4627.
13. T. Kennedy, E. Mullane, H. Geaney, M. Osiak, C. O’Dwyer and K. M. Ryan, Nano Lett., 2014, 14(2), 716–723.
14. K. H. Seng, M. H. Park, Z. P. Guo, H. K. Liu and J. Cho, Angew. Chem., Int. Ed., 2012, 51(23), 5657–5661.
15. D. J. Xue, S. Xin, Y. Yan, K. C. Jiang, Y. X. Yin, Y. G. Guo and L. J. Wan, J. Am. Chem. Soc., 2012, 134(5), 2512–2515.
16. S. Choi, J. Kim, N. S. Choi, M. G. Kim and S. Park, ACS Nano, 2015, 9(2), 2203–2212.
17. G. Jo, I. Choi, H. Ahn and M. J. Park, Chem. Commun., 2012, 48(33), 3987–3989.
18. H. Kim, Y. Son, C. Park, J. Cho and H. C. Choi, Angew. Chem., Int. Ed., 2013, 52(23), 5997–6001.
19. Y. Sun, S. X. Jin, G. W. Yang, J. Wang and C. X. Wang, ACS Nano, 2015, 9(4), 3479–3490.
20. Y. Xiao and M. H. Cao, Chem.–Asian J., 2014, 9(10), 2859–2865.
21. H. P. Jia, R. Kloepsch, X. He, J. P. Badillo, P. F. Gao, O. Fromm, T. Placke and M. Winter, Chem. Mater., 2014, 26(19), 5683–5688.
22. G. L. Cui, L. Gu, L. J. Zhi, N. Kaskhedikar, P. A. van Aken, K. Mullen and J. Maier, Adv. Mater., 2008, 20(16), 3079–3083.
23. D. T. Ngo, R. S. Kalubarme, H. T. T. Le, J. G. Fisher, C. N. Park, I. D. Kim and C. J. Park, Adv. Funct. Mater., 2014, 24(33), 5291–5298.
24. I. S. Hwang, J. C. Kim, S. D. Seo, S. Lee, J. H. Lee and D. W. Kim, Chem. Commun., 2012, 48(56), 7061–7063.
25. S. Fang, L. F. Shen, H. Zheng and X. G. Zhang, J. Mater. Chem. A, 2015, 3(4), 1498–1503.
26. L. F. Shen, X. G. Zhang, E. Uchaker, C. Z. Yuan and G. Z. Cao, Adv. Energy Mater., 2012, 2(6), 691–698.
27. Z. Li, Y. Jiang, L. X. Yuan, Z. Q. Yi, C. Wu, Y. Liu, P. Strasser and Y. H. Huang, ACS Nano, 2014, 8(9), 9295–9303.
28. H. Qiao, J. Li, J. P. Fu, D. Kumar, Q. F. Wei, Y. B. Cai and F. L. Huang, ACS Appl. Mater. Interfaces, 2011, 3(9), 3704–3708.
29. S. Jun, S. H. Joo, R. Ryoo, M. Kruk, M. Jaroniec, Z. Liu, T. Ohsuna and O. Terasaki, J. Am. Chem. Soc., 2000, 122(43), 10712–10713.
30. J. S. Lee, S. H. Joo and R. Ryoo, J. Am. Chem. Soc., 2002, 124(7), 1156–1157.
31. A. Arrais, P. Benzi, E. Bottizzo and C. Demaria, J. Phys. D: Appl. Phys., 2009, 42, 105406.
32. D. W. Wang, Y. L. Chang, Z. Liu and H. J. Dai, J. Am. Chem. Soc., 2005, 127(33), 11871–11875.
33. H. Adhikari, P. C. McIntyre, S. Y. Sun, P. Pianetta and C. E. D. Chidsey, Appl. Phys. Lett., 2005, 87, 263109.
34. J. W. Qin, X. Wang, M. H. Cao and C. W. Hu, Chem.–Eur. J., 2014, 20(31), 9675–9682.
35. W. Wang, Y. Xiao, X. Wang, B. Liu and M. H. Cao, ChemSusChem, 2014, 7(10), 2914–2922.
36. P. C. McIntyre, Y. Oshima, E. Kim and K. C. Saraswat, Microelectron. Eng., 2009, 86(7–9), 1536–1539.
37. M. Terrones, P. M. Ajayan, F. Banhart, X. Blase, D. L. Carroll, J. C. Charlier, R. Czerw, B. Foley, N. Grobert, R. Kamalakaran, P. Kohler-Redlich, M. Rühle, T. Seeger and H. Terrones, Appl. Phys. A: Mater. Sci. Process., 2002, 74(3), 355–361.
38. G. Liu, H. Zheng, A. S. Simens, A. M. Minor, X. Song and V. S. Battaglia, J. Electrochem. Soc., 2007, 154(12), A1129–A1134.
39. H. J. Kim, J. M. Kim, W. S. Kim, H. J. Koo, D. S. Bae and H. S. Kim, J. Alloys Compd., 2011, 509(18), 5662–5666.
40. N. H. Idris, M. M. Rahman, J. Z. Wang, Z. X. Chen and H. K. Liu, Compos. Sci. Technol., 2011, 71(3), 343–349.
41. Z. J. Ding, L. Zhao, L. M. Suo, Y. Jiao, S. Meng, Y. S. Hu, Z. X. Wang and L. Q. Chen, Phys. Chem. Chem. Phys., 2011, 13(33), 15127–15133.
42. J. Peng, W. Gao, B. K. Gupta, Z. Liu, R. Romero-Aburto, L. H. Ge, L. Song, L. B. Alemany, X. B. Zhan, G. H. Gao, S. A. Vithayathil, B. A. Kaipparettu, A. A. Marti, T. Hayashi, J. J. Zhu and P. M. Ajayan, Nano Lett., 2012, 12(2), 844–849.
43. E. Kang, Y. S. Jung, A. S. Cavanagh, G. H. Kim, S. M. George, A. C. Dillon, J. K. Kim and J. Lee, Adv. Funct. Mater., 2011, 21(13), 2430–2438.
44. W. M. Zhang, X. L. Wu, J. S. Hu, Y. G. Guo and L. J. Wan, Adv. Funct. Mater., 2008, 18(24), 3941–3946.
45. L. Y. Lim, N. Liu, Y. Cui and M. F. Toney, Chem. Mater., 2014, 26(12), 3739–3746.
46. M. H. Park, K. Kim, J. Kim and J. Cho, Adv. Mater., 2010, 22, 415–418.
47. D. Li, K. H. Seng, D. Q. Shi, Z. X. Chen, H. K. Liu and Z. P. Guo, J. Mater. Chem. A, 2013, 1, 14115–14121.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra17432e

---