Germanium-based multiphase material as a high-capacity and cycle-stable anode for lithium-ion batteries

Abstract

Copper germanate has been used in the electrical field and in lithium-ion battery anode applications. The known bonding energies between Cu and O enable us to control the chemical reduction process by thermal decomposition of hydro-carbon gas at high temperature. Herein, we demonstrate the synthesis of germanium-based multiphase materials by a carbothermic reduction process, in which the copper germanate (CuGeO₃) single phase material is transformed to multiphase composite materials including Cu, Cu₃Ge, GeO_x, and Ge by controlling the reaction temperatures. The resulting Ge-based anodes exhibit a reversible capacity of ∼600 mA h g⁻¹, stable capacity retention (80% after 100 cycles at 25 °C and 78% after 200 cycles at 60 °C) at a rate of C/5 with low electrode swelling (23% after 100 cycles). This is a one-step reaction process simultaneously involving reduction, phase transformation, and carbon coating. During the thermolysis reaction, various phases in the material are formed at the interface between Cu and Ge, which can affect the electrochemical performance of the resulting Ge-based anodes.

---

Introduction

Nowadays, lithium-ion batteries (LIBs) have become commonly used as the power sources of various portable electric devices including smart mobile phones, tablet computers, and wearable devices.^1–3 In addition, the market for LIBs will very likely allow progress from small to large size energy storage devices for electric vehicles and smart grid energy storage systems to store electricity generated by renewable energy conversion systems.^4,5 The performance of present LIB technology, however, still falls short of the requirements for future transportation and large-size energy storage applications in terms of energy density, power density, safety, charging time, cycle life, and price per unit.^6,7

In an effort to increase the energy density of LIBs, Li-alloying materials (Si, Sn, and Ge) have been widely studied for the anodes since they meet two requirements of high energy LIBs: higher theoretical specific capacity than that of conventional graphite electrodes (LiC₆: ∼372 mA h g⁻¹) and a low working voltage that is close to that of graphite (<0.5 V vs. Li/Li⁺).^8,9 Among these alloying materials, germanium (Ge) has gained much attention as an alternative anode material for advanced LIBs with high energy density; high theoretical capacity (Li₂₂Ge₅: 1600 mA h g⁻¹), low operating voltage (0.3 V vs. Li/Li⁺).¹⁰ Furthermore, Ge exhibits ∼400 times higher Li-ion diffusion coefficient at room temperature and higher electronic conductivity (1.45 S m⁻¹), compared to Si.¹¹ However, two major problems of Ge anodes should be solved for practical applications of this Li-alloying anode material. One is a large volume change of electrodes upon alloying/de-alloying, and the other is the formation of unstable solid electrolyte interphase (SEI) layers on the surfaces of active material.¹²

Several approaches have been made to solve or at least to mitigate the volume-change problem; down-sizing alloying materials,^13,14 shape-control of Ge-based materials including one-dimensional (1D) shape,^15,16 hollow structures,¹⁷ 3D porous structure,¹⁸ and various synthetic methods (e.g., sol–gel method,¹⁹ hydrothermal,^17,20 chemical deposition,²¹ hydrolysis processes²² and composites^23–25).

Recently, numerous researchers have demonstrated that various Ge-based alloying materials could improve the performance of LIBs.^26,27 For example, Lu's group reported that core–shell type Ge nanofibers made by electro-spinning exhibited a reversible capacity of 1050 mA h g⁻¹ after 100 cycles.²⁶ Kang's group demonstrated the fabrication of long-term stable battery anodes showing a capacity retention of over 90% after 1600 cycles by combining reduced graphene oxide and amorphous GeO_x composite material.²⁷

As another strategy, metal germanate battery anodes have been developed, because metal oxide matrix in the metal germanate could provide an elastic buffer to accommodate the volume changes and result in improving the electrochemical performance.^28–30 However, a poor initial coulombic efficiency and unsatisfied capacity retention at high temperature should be resolved.

Herein, we demonstrate thermolysis processes that simultaneously facilitate a chemical reduction and a phase transformation of copper germanate (CuGeO₃), in which copper germanate (CuGeO₃) single phase material is transformed to multiphase composite materials including Cu, Cu₃Ge, GeO_x, and Ge by controlling reaction temperatures. The resulting Ge-based anodes exhibit a reversible capacity of ∼600 mA h g⁻¹, stable capacity retention (80% after 100 cycles at 25 °C and 78% after 200 cycles at 60 °C) at a rate of C/5 with low electrode swelling (23% after 100 cycles). CuGeO₃ material with electromagnetic properties has been used in electrical field applications.³¹ The known bonding energies between Cu and O enable us to control the chemical reduction process by thermal decomposition of hydro-carbon gas at high temperature. This is one-step reaction process simultaneously involving reduction, phase transformation, and carbon coating. During the thermolysis reaction, various phases in material are formed at the interface between Cu and Ge, which can affect the electrochemical performance of the resulting Ge-based anodes.

Experimental

Synthesis of copper germanate

Nanocrystalline CuGeO₃ was synthesized by a hydrothermal process according to the procedure reported by Feng et al.³² with some modifications. In a typical synthesis, 0.364 g of cetyltrimethylammonium bromide (CTAB, Aldrich, >99.9%), 0.52 g of GeO₂ (Aldrich, 99.999%), and 0.85 g of CuCl₂·H₂O (Aldrich, >99.9%) were added to 15.0 mL of distilled water. The pH value of the resulting mixture was adjusted to 8 by adding NaOH (30 wt%). The mixture was stirred for 1 h and then transferred to a stainless Teflon-lined autoclave of 25 mL inner volume with 80% fill ratio. The autoclave was heated to 180 °C and held for 24 h, followed by cooling to room temperature. The product was centrifuged, filtered, and rinsed with alcohol and deionized water for several times. Finally, the product was dried at 80 °C for 12 h in the air. A blue colored CuGeO₃ powder was obtained with a yield of about 90%.

Carbothermic reduction of copper germanate

The as-synthesized CuGeO₃ was placed inside an alumina boat. Subsequently, the hydro-carbon chemical reduction was performed at 600–900 °C (a rate of 5 °C min⁻¹) for 20 min under an argon environment to synthesize a shape-preserving Cu and GeO_x composite materials and porous structured Cu–Ge alloy and Ge-based composite materials.

Physical characterization

The surface morphologies of nano crystalline CuGeO₃ and as-synthesized Cu–Ge composites were characterized using a field-emission scanning electron spectroscopy (FESEM, Hitachi S-4800), which was performed at 10 keV. The microstructures of the CuGeO₃ and as-synthesized Cu–Ge composites were investigated using a X-ray diffractormeter (XRD, Bruker D8-Advance), which were performed at 3 kW using Cu Ka radiation.

Electrochemical test

The electrochemical properties of the CuGeO₃ and Cu₃Ge–Ge based composites electrodes were evaluated using coin-type half cells (2016R-type) at 25 °C and 60 °C. The CuGeO₃ and Cu₃Ge–Ge based electrodes were composed of the each active material, super-P carbon black, and poly (acrylic acid)/sodium carboxymethyl cellulose (1 : 1, w/w) as binder in a weight ratio of 7 : 1.5 : 1.5.

The capacity of battery cell was calculated by total mass of active materials in electrode. The proportion of Ge material in active materials was estimated by specific capacities of electrodes. The coin cell consisted of lithium metal as the counter electrode, a polyethylene separator, and 1.3 M LiPF₆ containing ethylene carbonate/diethyl carbonate (PANAX starlyte, Korea, 3/7 (v/v)) and 10 wt% fluoroethylene carbonate as the electrolytes. The half cells were tested galvanostatically between 0.005 and 3.0 V (versus Li/Li⁺) for C rate 0.05–3C. The cell performance was examined using a cycle tester (WBCS 3000 battery systems, Wonatech). For impedance measurement, AC complex impedance analysis with an IVIUM frequency response analyzer was used over a frequency range of 10 mHz to 1 MHz.

Results and discussion

Synthesis and characterization of Ge-based multiphase materials

Fig. 1a shows the schematic illustration of the synthetic routes of various Cu-incorporated Ge-based composite materials at three different reaction temperatures. As-synthesized CuGeO₃ nanorods were obtained through a hydrothermal reaction at 180 °C for 24 h. The crystalline CuGeO₃ (JCPDS card no. 01-070-6183) showed an average diameter of 70 nm and a length of 0.5–2 μm (ESI, Fig. S1†).

Fig. 1 Schematic illustration showing various Ge-based multiphase composite materials and microscopy images of (a) as-prepared CuGeO₃ nanorods, (b) Cu and GeO_x composite materials synthesized at 600 °C (Cu–Ge-600), (c) Cu₃Ge, GeO_x and Ge composite materials synthesized at 700 °C (Cu–Ge-700), (d) Cu₃Ge and Ge composite materials synthesized at 900 °C (Cu–Ge-900).

Meanwhile, the CuGeO₃ can also be used a source material to make other multi-phase materials consisting of Cu, Ge, O and C via thermal decomposition of acetylene gas (C₂H₂). At a low reaction temperature of 600 °C, shape preserved GeO_x/Cu/C composites were synthesized by a carbothermic reduction process. The resulting materials maintained an original nanorod shaped structure with rough surfaces and a thin carbon layer on the surface as shown in the scanning electron microscopy (SEM) image (left in Fig. 1b). Furthermore, the transmission electron microscopy (TEM) image (middle in Fig. 1b) and the energy-dispersive X-ray spectroscopy (EDS) mapping image (right in Fig. 1b) show the decorated Cu nano particles in GeO_x nanorods. This result indicates that only Cu and O bonding in CuGeO₃ is broken owing to a thermal energy that is not enough to reduce the GeO₂ phase. As the reaction temperature is increased to 700 °C, Cu₃Ge was formed at the interface of Cu and Ge and GeO_x partially reduced to pure Ge³³ (Fig. 2a). The resulting Ge/GeO_x/Cu₃Ge/C composite materials had a porous structure with spherical primary particles all covered by a thin carbon layer on the outer surface (left in Fig. 1c). The TEM image (middle in Fig. 1c) and the EDS mapping image (right in Fig. 1c) show a porous Ge matrix including Cu₃Ge particles for high electrical conductivity. Further increase of the reaction temperature (900 °C) led to a full chemical reduction of CuGeO₃ to Ge/Cu₃Ge/C composite (Fig. 2a). At this temperature, all oxygen elements in the original materials were completely removed by the carbothermic reduction process due to evacuation of by-products such as CO and CO₂.³⁴ However, the resulting Ge/Cu₃Ge/C composite showed collapsed and aggregated morphologies owing to the high thermal shock as shown in TEM image and EDS mapping images (Fig. 1d).

Fig. 2 Characterization of as-synthesized Ge-based multiphase composite materials. (a) XRD patterns of Cu–Ge-600 (red), Cu–Ge-700 (blue), and Cu–Ge-900 (black), assigned to Cu (yellow), GeO₂ (red), Ge (black), Cu₃Ge (purple). (b) The plot of each component in composite materials as a function of reaction temperature: GeO_x phase, Ge, Cu phases and Cu₃Ge alloy phase. XRD patterns of (c) Cu–Ge-600 and (d) Cu particles before (top, black) and after (bottom, red) thermal oxidation testing at 200 °C for 6 h.

A Brunauer–Emmett–Teller (BET) analysis indicated that the porous Cu–Ge-600 composite materials had a BET surface area of 12.74 m² g⁻¹, indicating that numerous mesopores were formed during the carbothermic reduction process. The pore size distribution, obtained using the Barrett–Joyner–Halenda (BJH) method, suggested that the Cu–Ge-600 materials had an average pore size of 14 nm and a narrow pore size distribution (less than 10 nm pore diameter) (ESI, Fig. S2†).

Fig. 2a presents the X-ray diffraction (XRD) patterns of chemically reduced Ge-based multiphase composites obtained at various reaction temperatures. At 600 °C, Cu (JCPDS card no. 01-070-3038) and low intensity GeO₂ (JCPDS card 01-074-5906) peaks are clearly observed, respectively. The remarkable reduction of GeO₂ peak intensity indicates that amorphous GeO_x (oxygen-deficient GeO₂) is formed by the carbothermic reduction. As the reaction temperature increases, the peak intensity of GeO₂ gradually decreases and simultaneously new phases, Cu₃Ge (JCPDS card no. 01-070-3058) and Ge (JCPDS card no. 01089-5011), are developed (Fig. 2a). Based on the results of XRD patterns, all the resulting components have been plotted as a function of reaction temperature (Fig. 2b).

Meanwhile, we investigated the thermal stability of Cu–Ge-600 samples in air at 200 °C for 6 h. The Cu–Ge-600 was not oxidized, owing to carbon coating layers formed on the Cu–Ge surface (Fig. 2c). In contrast, when pure Cu particles were treated in the same oxidation conditions, they were completely converted to copper oxides (Fig. 2d). These results indicate that Ge-based multiphase composites synthesized by the carbothermic reduction process are quite stable under thermal oxidation conditions.

Electrochemical performance of Ge-based multiphase composites

Ge-based multiphase composites (synthesized at three different temperatures) and pristine CuGeO₃ materials were tested as lithium-ion battery anodes. The voltage profile and cyclic voltammogram (CV) curves of CuGeO₃ nanorod anodes were obtained in the potential window of 0.01–3.0 V (ESI, Fig. S3†). These results show typical cathodic and anodic peaks of both CuO and GeO₂, which are in good agreement with previous reports.³¹ A sharp cathodic peak appeared at ∼1.5 V, which is related to the formation of SEI layers and amorphous Li₂O and Cu^II_1−xCu^I_xO_1−x/2 originating from the decomposition of CuGeO₃ as the follow equations:³⁵

CuGeO₃ + 2Li → Cu + GeO₂ + Li₂O (1.7 V) (i)

Cu₂O + 2Li → 2Cu + Li₂O/2CuO + 2Li → 2Cu + Li₂O at 1.5 V (∼0.5 V) (ii)

GeO₂ + 4Li → Ge + 2Li₂O (0.4 V) (iii)

Ge + XLi → GeLi_X (0.4 V) (iv)

As a result, the CuGeO₃ anode shows a poor initial columbic efficiency of ∼68%. When we consider practical full-cell systems, multiple lithiation processes (ranging from 1.75 to 0.005 V) of the CuGeO₃ will make the energy density of such full cell to be lower. In this aspect, the reduction process of CuGeO₃ is necessary to increase the initial coulombic efficiency and to bring the lithiation voltage as low as possible.

Fig. 3a shows the first cycle voltage profiles of discharging (lithiation) and charging (delithiation) for three Cu–Ge composite anodes at a rate of C/20 in the potential window 0.01–3.0 V. The first discharge capacities of Cu–Ge-600, Cu–Ge-700, and Cu–Ge-900 anodes are 1129, 756, and 388 mA h g⁻¹ with significantly enhanced initial coulombic efficiencies of 77%, 79% and 77%, respectively (Fig. 3a). Increasing reaction temperature leads to the reduction of specific capacity owing to an increased amount of inactive Cu₃Ge phase. The CV curve of Cu–Ge-600 electrode shows typical cathodic and anodic peaks of GeO_x while Cu–Ge-700 and Cu–Ge-900 profiles are similar to those of typical Ge electrodes^36,37 (Fig. 3b).

Fig. 3 Electrochemical performance of chemically reduced Cu–Ge-600 (red), Cu–Ge-700 (blue), Cu–Ge-900 (black) anodes. (a) The first cycle voltage profiles of three electrodes obtained at a rate of C/20, (b) cyclic voltammograms of three electrodes at a scan rate of 1 mV s⁻¹, (c) cycle performance of three electrodes tested at a rate of C/5, and (d) rate capability of three electrodes tested at various C-rates. The same charge and discharge rate was used.

Three Cu–Ge composite anodes exhibit highly stable cycling performance, corresponding to the capacity retention of 80% (Cu–Ge-600), 91% (Cu–Ge-700), and 90% (Cu–Ge-900) at a rate of C/5 after 100 cycles, respectively (Fig. 3c). Further, both Cu–Ge-600 and Cu–Ge-700 electrodes exhibit superior rate capabilities, which correspond to a charge capacity of ∼480 mA h g⁻¹ at 1C rate (Fig. 3d). Outstanding capacity retention and excellent rate capability of Cu–Ge electrodes may be explained as follows: (i) mesoporous structures have many void spaces which can act as a buffer layer to alleviate a large volume change. (ii) Good mechanical properties of Cu and Cu₃Ge retain the structural integrity of mesoporous Cu–Ge during the repeated cycles. (iii) The electrical conductivity of Cu phase in materials will significantly increase the conductivity of Cu–Ge composites. And (iv) the outer carbon layers can form SEI layers on the Cu–Ge composite surface. In contrast, the CuGeO₃ electrode showed a poor cycling performance (capacity retention of 82% after 100 cycles) and rate capabilities due to a large volume expansion of the electrode (ESI, Fig. S3c†).

As aforementioned, Cu–Ge anodes with porous structure containing void spaces and Cu phases in composite materials provide an additional buffer matrix for a large volume expansion of Ge particles, which is clearly seen in the cross-sectional SEM images of the electrodes after 100 cycles at a rate of C/5 discharge/charge in Fig. 4. As expected, the Cu–Ge-600 anode showed a remarkably suppressed volume expansion (23%), compared to that of the CuGeO₃ anode (320%), as shown in Fig. 4a and b. In addition to the Cu–Ge-600, both Cu–Ge-700 and Cu–Ge-900 also showed a low electrode swelling of 24% and 14%, respectively, owing to the buffer matrix of Cu₃Ge and carbon (ESI, Fig. S4†).

Fig. 4 Characterization of CuGeO₃ and Cu–Ge-600 electrodes. Cross-sectional SEM images of (a) CuGeO₃ and (b) Cu–Ge-600 electrodes before and after 100 cycles. (c) Electrochemical impedance spectra of CuGeO₃ (black) and Cu–Ge-600 (red) electrodes obtained after the first cycle (inset is an equivalent circuit model).

To further understand the superior electrochemical performance of Cu–Ge-600, electrochemical impedance spectroscopy (EIS) measurements were performed. The Cu–Ge-600 electrodes showed a slightly lower impedance value (R_ct: 33.9 ohm) than that of CuGeO₃ (R_ct: 84.5 ohm) electrodes due to the reduced Cu phases and oxygen-deficient GeO_x (Fig. 5).

Fig. 5 Electrochemical performance of CuGeO₃ (black) and Cu–Ge-600 (red) electrodes operated at 60 °C. (a) The first cycle voltage profiles of two electrodes obtained at a rate of C/20 and (b) cycle performance of two electrodes tested at rates of C/5.

We investigated the electrochemical performance of conducting agent-free systems. The Cu–Ge-600 and CuGeO₃ anodes were composed of 95% active material and 5% binder without any super P carbon black conducting agent. The Cu–Ge-600 electrode showed a higher specific capacity of 290 mA h g⁻¹ after 20 cycles at a rate of C/2, compared to that of the CuGeO₃ electrode (50 mA h g⁻¹). Even though the reversible capacity of Cu–Ge-600 is low, Cu and C matrix in the composite materials may enhance its electrical conductivity in a composite electrode (ESI, Fig. S5†).

For practical LIB applications, an electrochemical test at high temperature is required. We investigated the electrochemical performance of CuGeO₃ and Cu–Ge-600 electrodes at 60 °C. The Cu–Ge-600 anode showed a high specific capacity of ∼800 mA h g⁻¹ in the first cycle, and maintained the capacity of ∼500 mA h g⁻¹ (capacity retention of 63%) after 200 cycles at a rate of C/2. In contrast, the CuGeO₃ electrode exhibited a charge capacity of ∼1050 mA h g⁻¹ in the first cycle, but a poor capacity retention of 47% after 200 cycles. This result indicates that Cu and carbon matrix in the composite material may induce the formation of stable SEI layers, even though serious electrolyte decomposition occurs at 60 °C.

Conclusion

We successfully synthesized various Ge-based multiphase composite materials via one-step carbothermic reduction processes from CuGeO₃. Cu, Cu₃Ge, and carbon phases derived from carbothermic reduction of CuGeO₃. Also, porous structures of Ge-based composite materials gave additional void spaces which can alleviate large volume changes in Ge-based anodes during lithiation–delithiation processes. As a result, the Ge-based composite anodes showed a high reversible capacity of 590 mA h g⁻¹ after 100 cycles at C/2, excellent rate capability (∼480 mA h g⁻¹ at a rate of 1C), and remarkably suppressed electrode swelling (23%) at 25 °C. Furthermore, it exhibited highly stable cycle retention (63%) after 200 cycles even at 60 °C. We believe that this strategy of Ge-based multiphase composites can establish a milestone for fabricating practical alloy-based materials.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and future Planning (2015R1A2A1A01003143).

Notes and references

1. L. Ji, Z. Lin, M. Alcoutlabi and X. Zhang, Energy Environ. Sci., 2011, 4, 2682–2699.
2. V. Etacheri, R. Marom, R. Elazari, G. Salitra and D. Aurbach, Energy Environ. Sci., 2011, 4, 3243–3262.
3. W.-J. Zhang, J. Power Sources, 2011, 196, 13–24.
4. B. Scrosati, Nature, 1995, 373, 557–558.
5. R. F. Nelson, J. Power Sources, 2000, 91, 2–26.
6. Y. Nishi, J. Power Sources, 2001, 100, 101–106.
7. J. Xu, S. Dou, H. Liu and L. Dai, Nano Energy, 2013, 2, 439–442.
8. A. Magasinski, P. Dixon, B. Hertzberg, A. Kvit, J. Ayala and G. Yushin, Nat. Mater., 2010, 9, 353–358.
9. B. Scrosati, J. Hassoun and Y.-K. Sun, Energy Environ. Sci., 2011, 4, 3287–3295.
10. J. Graetz, C. Ahn, R. Yazami and B. Fultz, J. Electrochem. Soc., 2004, 151, A698–A702.
11. C. Fuller and J. Severiens, Phys. Rev., 1954, 96, 21–24.
12. P. B. Balbuena and Y. Wang, Lithium-Ion Batteries: Solid Electrolyte Interphase, Imperial College Press: World Scientific Publishing Co. Pte. Ltd., London, 2004.
13. J. R. Szczech and S. Jin, Energy Environ. Sci., 2011, 4, 56–72.
14. A. S. Arico, P. Bruce, B. Scrosati, J.-M. Tarascon and W. Van Schalkwijk, Nat. Mater., 2005, 4, 366–377.
15. M.-H. Seo, M. Park, K. T. Lee, K. Kim, J. Kim and J. Cho, Energy Environ. Sci., 2011, 4, 425–428.
16. C. K. Chan, X. F. Zhang and Y. Cui, Nano Lett., 2008, 8, 307–309.
17. T. Lv, X. Li and J. Ma, RSC Adv., 2014, 4, 49942–49945.
18. M. H. Park, K. Kim, J. Kim and J. Cho, Adv. Mater., 2010, 22, 415–418.
19. J. Hwang, C. Jo, M. G. Kim, J. Chun, E. Lim, S. Kim, S. Jeong, Y. Kim and J. Lee, ACS Nano, 2015, 9, 5299–5309.
20. J. Feng, M. Lai and L. Lu, Electrochem. Commun., 2011, 13, 287–289.
21. J. L. Lensch-Falk, E. R. Hemesath, F. J. Lopez and L. J. Lauhon, J. Am. Chem. Soc., 2007, 129, 10670–10671.
22. H. Miguez, F. Meseguer, C. Lopez, M. Holgado, G. Andreasen, A. Mifsud and V. Fornés, Langmuir, 2000, 16, 4405–4408.
23. L. Yuan, H. Yuan, X. Qiu, L. Chen and W. Zhu, J. Power Sources, 2009, 189, 1141–1146.
24. L. Shen, X. Zhang, E. Uchaker, C. Yuan and G. Cao, Adv. Energy Mater., 2012, 2, 691–698.
25. J. Liu, W. Li and A. Manthiram, Chem. Commun., 2010, 46, 1437–1439.
26. X. Wang, L. Fan, D. Gong, J. Zhu, Q. Zhang and B. Lu, Adv. Funct. Mater., 2016, 26, 1104–1111.
27. S. H. Choi, K. Y. Jung and Y. C. Kang, ACS Appl. Mater. Interfaces, 2015, 7, 13952–13959.
28. J. Wang, C.-Q. Feng, S.-L. Chou, H.-K. Liu and J.-Z. Wang, Sci. Rep., 2014, 4, 7030.
29. C. H. Kim, Y. S. Jung, K. T. Lee, J. H. Ku and S. M. Oh, Electrochim. Acta, 2009, 54, 4371–4377.
30. W. Li, Y.-X. Yin, S. Xin, W.-G. Song and Y.-G. Guo, Energy Environ. Sci., 2012, 5, 8007–8013.
31. M. Hase, I. Terasaki and K. Uchinokura, Phys. Rev. Lett., 1993, 70, 3651.
32. J. Feng, M. O. Lai and L. Lu, Mater. Res. Bull., 2012, 47, 1693–1696.
33. R. W. Olesinski and G. J. Abbaschian, J. Phase Equilib., 1986, 7, 28–35.
34. S. Choi, J. Kim, D. Y. Hwang, H. Park, J. Ryu, S. K. Kwak and S. Park, Nano Lett., 2016, 16, 1179–1185.
35. S. Wu, R. Wang, Z. Wang and Z. Lin, Nanoscale, 2014, 6, 8350–8358.
36. X. Liu, J. Hao, X. Liu, C. Chi, N. Li, F. Endres, Y. Zhang, Y. Li and J. Zhao, Chem. Commun., 2015, 51, 2064–2067.
37. W. He, H. Tian, X. Wang, F. Xin and W. Han, J. Mater. Chem. A, 2015, 3, 19393–19401.

---

Footnotes

† Electronic supplementary information (ESI) available: Characterization of copper germanate, BET surface area, and electrochemical test. See DOI: 10.1039/c6ra19811b

‡ These authors equally contributed to this work.

---