Ge@C three-dimensional porous particles as high-performance anode materials of lithium-ion batteries

Abstract

We demonstrate the synthesis of Ge@C three-dimensional porous particles (Ge@C TPP) via the decomposition of magnesium germanide (Mg₂Ge) and subsequent deposition of a carbon layer. Briefly, Ge TPP is first synthesized by the annealing of a Mg₂Ge precursor in air and a subsequent acid pickling process. Then, the carbon layer is deposited onto the Ge TPP by the pyrolysis of acetylene to form Ge@C TPP. When used as anode materials in lithium-ion batteries, the Ge@C TPP shows higher reversible capacity and better cycling performance than bulk Ge and bare Ge TPP. It is believed that the porous and core–shell structures can accommodate the volume change, give more lithiation sites, and stabilize the structure during the charge/discharge process, which may be responsible for the enhanced performance.

---

Introduction

Lithium-ion batteries (LIBs) have been widely used as power sources for portable electronic devices such as laptops, cell phones and digital cameras due to the advantages such as high energy density, long cyclic life, low self-discharge and lack of memory effect.^1–3 Currently the commercial anode material in LIBs is graphite which shows good cyclic life and low working voltage, however, the limited volumetric capacity (372 mA h g⁻¹) cannot meet the growing demand for LIBs with high energy density in high-power devices such as electrical vehicles.^4,5 Therefore, new alternative anode materials for LIBs have been pursued.

Group IV elements, such as silicon (Si), germanium (Ge) and tin (Sn), are considered to be most promising candidates for the anode materials of next-generation LIBs because of the low working voltage and high specific capacity.^6–8 For example, Si has the highest specific capacity of ∼4200 mA h g⁻¹ and low working potential (0.37 mV vs. Li/Li+). However, Si undergoes a huge volume change up to 400% during the insertion/extraction of lithium ions, which brings cracking and crumbling of the electrode.^9,10 Moreover, the low diffusivity of Li in Si and the low conductivity limits its future application in high-power devices. Compared with Si, Ge not only exhibits the high specific capacity (1600 mA h g⁻¹), but also the high Li diffusivity (400 times faster than Si),⁷ and the good electrical conductivity (4 orders higher than Si).¹¹ Moreover, Ge forms less native oxide layer on the surface than Si, leading to the relatively high initial coulombic efficiency.¹² Besides, it is known that the lithiation of Ge is isotropic while the lithiation of Si is highly anisotropic, indicating that Ge is more stable than Si during the charge/discharge process.¹³ However, Ge still suffers a volume expansion of ∼370% during the insertion/extraction of Li, which leads to pulverization, isolation of active materials from conductive network in electrodes, and ultimately fading of the capacity.¹⁴ Therefore, extensive research has been conducted to improve the cycling performance of Ge anode. Typically, there are two strategies. The first is the design of composite materials in which Ge dispersing in a matrix.^15–24 The matrix is used to buffer the stress induced by the volume expansion and enhance the conductivity. The other strategy is the design of the nano/micro-structure of Ge materials^14,25–31 and porous structures,^32–36 which can sustain the physical stains during the lithiation/de-lithiation process. Among them, the Ge porous structure is the hot spot because of its pores and relative large surface area, which can enhance the diffusion of lithium ions and accommodate the volume change. For example, Park et al. synthesized Ge@C three-dimensional porous particles by using the SiO₂ spheres as the template.³² When used as anode materials of LIBs, the three-dimensional particles show good cycling performance, which is extremely better than Ge@C isolated hollow spheres. Liu et al. demonstrated the synthesis of Ge@C mesoporous spheres via a chelation reaction and extended stöber method.³⁴ The Ge@C mesoporous spheres delivered an initial discharge capacity of 1653 mA h g⁻¹ and slightly decrease to 1099 mA h g⁻¹ after 100 cycles at 0.1C. However, the above-mentioned synthetic method is complicated and the exploration of large-scale and convenient way for the preparation of Ge porous materials is a great challenge.

Herein, we report the synthesis of Ge@C three-dimensional porous particles (TPP) via the simple decomposition of Mg₂Ge and subsequent pyrolysis of acetylene. The porous and core–shell structure can accommodate large volume changes, give more lithiation sites, and stabilize the structure during the alloying/dealloying process, which contributes to high capacity and stable cycling performance.

Experimental section

Synthesis of Ge@C TPP

Ge@C TPP were synthesized as follows: firstly, Mg₂Ge were annealed at 700 °C for 3 h under air in a tube furnace; after cooling down to room temperature, the resulted substance was washed with 1 M HCl solution for 30 min to remove MgO, then centrifuged by deionized water and ethanol for several times, and dried in the vacuum box at 80 °C for 10 h; the carbon layer was then deposited onto bare Ge TPP by pyrolysis of acetylene in a tube furnace; the furnace was purged with Ar for 30 min before a gas flow of Ar : C₂H₂ = 10 : 1 was introduced, and then was heated to 650 °C at a heating rate of 5 °C min⁻¹ for 3 h.

Characterization

The phase of the products were characterized by X-ray Diffraction (XRD) using a Rigaku D/max-ga X-ray diffractometer with graphite monochromatized Cu Kα radiation (λ = 1.54178 Å). The Brunauer–Emmett–Teller (BET) test was performed by a Micrometrics Tristar II 3020 analyzer system. The morphology of the products was characterized with scanning electron microscopy (SEM HITACHI SU70) and transmission electron microscopy (TEM FEI F200). The Raman experiments were performed with a HR800 Raman spectrometer using the 514 nm line of an Ar ion laser operated at 10 mW.

The electrochemical performance

The electrochemical measurements were carried out by using coin type (CR2025) half cells with the lithium foils as the counter electrode. A mixture of Ge@C TPP, Super P conductive carbon black (Timcal) and polyvinylidene fluoride with a weight ratio of 8 : 1 : 1 was loaded onto copper foils as electrodes. 1 M LiPF₆ in a mixture of ethylene carbonate (EC) and dimethyl carbonate (DMC) (1 : 1, v/v) were used as the electrolyte. The half cells were assembled in a glovebox (Mbraun, Labstar, Germany) under argon atmosphere and aged for 24 h before measurements. A galvanostatic cycling test of the assembled cells was carried out on a Land CT2001A system with a potential range of 0.01–1.5 V at a current density of 0.2C (1C = 1600 mA g⁻¹) mA g⁻¹. Cyclic Voltammetry (CV) of the same potential range was recorded on an Arbin BT2000 system at a scanning rate of 0.1 mV s⁻¹. Electrochemical impedance spectroscopy (EIS) measurements were performed on CHI660D electrochemical workstation over a frequency range of 10 mHz to 1 MHz by applying an AC signal of 5 mV in amplitude throughout the tests.

Results and discussion

Fig. 1 shows the schematic illustration for the synthetic procedure of Ge three-dimensional porous particles (TPP). It can be seen that the primary Mg₂Ge can transform into MgO and Ge via the annealing process under air. After acid pickling process, the MgO is removed and Ge TPP is left. The porous structure is due to the removal of MgO. The XRD analysis is employed to confirm the transformation process (Fig. 2). It can be seen that the initial source material is mainly Mg₂Ge with the trace Ge. It should be mentioned that the trace Ge doesn't affect the transformation process. After the annealing process, the peaks of MgO and Ge emerge as well as the disappear of Mg₂Ge, indicating the transformation from Mg₂Ge to MgO and Ge. After reacting with hydrochloric acid (HCl) solution, MgO was removed and only the peaks of Ge are left. After the carbon-coated process, no obvious peaks corresponding to the carbon layer were observed, indicating that the carbon layer is amorphous.

Fig. 1 Schematic illustration for the synthetic procedure of Ge TPP.

Fig. 2 XRD patterns of products along with the synthetic procedure (a) raw materials; (b) after annealing process; (c) after aid pickling and (d) after the carbon coating process.

In order to further confirm the transformation process, morphological characterizations were employed (Fig. 3). It can be seen from Fig. 3a and b (SEM images) that the initial Mg₂Ge shows the morphology of microparticles range from hundreds of nanometers to several micrometers, and the surface of the Mg₂Ge microparticle is smooth. After the annealing and acid pickling process, the size of the particles is maintained. However, the surface of the particle turns into rough and many pores emerge (Fig. 3c). The sizes of the pores range from tens to hundreds of nanometers due to the removal of MgO particles (Fig. 3d). Fig. 3e shows the TEM image of an individual Ge TPP. As observed, the pores not only exist on the surface, but also in the particle inside, indicating the formation of homogenous Ge TPP. The sizes of the pores are roughly consistent with the result of SEM image. In HRTEM image (Fig. 3f), the lattice spacing of 0.2 nm can correspond to the {220} plane of Ge, further confirm the synthesis of Ge TPP. It should be mentioned that almost no amorphous oxide layer can be detected on the Ge particles, which is different from Si.^9,10 Considering the oxide layer would lead to the low initial coulombic efficiency, Ge can benefit from the minimal amount of native oxide on the surface.¹² Fig. 4a and b show the SEM images of the Ge@C TPP via the pyrolysis of acetylene for 2 h. As can be seen, not only the surface of the Ge TPP, but also the partial pores have been deposited by carbon layer. Fig. 4c and d show the TEM and HRTEM (inset) images of an individual Ge@C TPP. As can be seen, an amorphous carbon layer formed on the surface of the particle, which is different from the bare Ge TPP. Moreover, both the outer surface and the inner pore are covered with a carbon layer, indicating the homogenous deposition of carbon layer via the pyrolysis of acetylene. In the Raman pattern (Fig. 4f), the peak located at ∼285 cm⁻¹ corresponds to Ge, while two typical peaks at ∼1340 and ∼1580 cm⁻¹ (ref. 30) represent the amorphous feature of the carbon layer. The above-mentioned analysis confirms the synthesis of homogenous Ge@C TPP.

Fig. 3 Low-(a) and high-(b) magnified SEM images of Mg₂Ge particles; low-(c) and high-(d) magnified SEM images of bare Ge TPP; TEM (e) and HRTEM (f) images of bare Ge TPP.

Fig. 4 Morphological and compositional characterization of Ge@C TPP: (a) SEM image; (b) high-magnified SEM images; (c) TEM image; (d) HRTEM image; (e) EDX pattern; (f) Raman spectrum.

To further confirm the content of the carbon in the Ge@C TPP, TGA analysis was employed in air from room temperature to 700 °C and the results are shown in Fig. 5. As can be seen, a small weight lose between 100 and 400 °C were found in Ge@C TPP, which may be attributed to the loss of adsorbed water. The largest weight loss occurs between 400 and 600 °C due to the complete oxidation reaction of the carbon layer. It is indicated that the carbon content of the Ge@C TPP is ∼19.9%. Fig. 6 shows the Brunauer–Emmett–Teller (BET) spectra for Mg₂Ge (a) and (b), bare Ge TPP (c) and (d), Ge@C TPP (e) and (f), respectively. As can be seen, the BET surface area of the Mg₂Ge is only 0.01 m² g⁻¹, which increase to 10.4 m² g⁻¹ after the annealing and acid pickling process. It is obvious that the removal of the MgO in the acid pickling process is responsible for the increased BET surface area, indicating the formation of the porous structures. After the coating of the carbon layer, the BET surface area decreases to 6.8 m² g⁻¹ because the carbon may cover partial pores in the particles, which is consistent with the TEM and SEM result. The peak of the pore diameter distribution is ∼42 (Fig. 6d) and 40 nm (Fig. 6f) for bare Ge and Ge@C TPP, respectively, which is also due to the deposition of carbon layer in pores.

Fig. 5 TGA analysis of the Ge@C TPP.

Fig. 6 Brunauer–Emmett–Teller (BET) spectra for Mg₂Ge (a), (b), bare Ge TPP (c) and (d), and Ge@C TPP (e) and (f).

Motivated by the novel porous structures, the products were tested as anode materials of LIBs. Fig. 7a shows the 1^st, 5^th, 10^th and 100^th discharge and charge curves of Ge@C TPP with a testing current density of 0.2C. The first discharge and charge capacity is 1195 and 879 mA h g⁻¹, leading to the initial coulombic efficiency of 73.5%. The initial capacity loss may be due to the formation of the irreversible solid electrolyte interphase (SEI) film, as well as to the consumption of lithium ions in the Ge@C TPP via structure defects in the first lithiation process. The 5^th, 10^th and 100^th discharge capacity is 935, 923, 839 mA h g⁻¹, respectively. The capacity loss per cycle is ∼0.1% from 5^th to 100^th cycle, indicating the good reversibility of Ge@C TPP. Fig. 7b shows the capacities of bulk Ge, bulk Ge@C, bare Ge TPP, Ge@C TPP versus cycle numbers with a testing current density of 0.2C (1C = 1600 mA g⁻¹) at a voltage range of 0.01–1.5 V. As can be seen, the initial coulombic efficiency of bulk Ge, bulk Ge@C, bare Ge TPP, Ge@C TPP is 54.3%, 69.7%, 60.7% and 73.6%, respectively. Generally, low surface area favors a small extent of side reactions between the electrode and electrolyte.²⁸ However, the initial coulombic efficiency of bulk Ge is lower than bare Ge and Ge@C TPP. The low initial coulombic efficiency of bulk Ge may be attributed to the loss of contact to conductive network caused by volume expansion during the initial lithiation/delithiation process, leading to the trapping of lithium ions in the electrodes. In addition, the initial coulombic efficiency of Ge@C TPP is higher than bare Ge TPP because of the relatively low surface area. After the first cycle, the coulombic efficiency of Ge@C TPP increases up to 98% after 10 cycles. The bulk Ge shows a first discharge capacity of 1249 mA h g⁻¹, and decreases to 679 mA h g⁻¹ in the second cycle due to the low initial coulombic efficiency. After 100 cycles, the discharge capacity of bulk Ge is 257 mA h g⁻¹, indicating the poor cycling performance. For comparison, the bulk Ge@C shows a first reversible capacity of 1193.2 mA h g⁻¹ and it decrease to 434 mA h g⁻¹ after 75 cycles. The bulk Ge@C show better cycling performance than bulk Ge, which is mainly due to the carbon layer, however, the discharge capacity is still fading fast. The cycling performance is improved to a limited extent by the carbon layer since the particles would still pulverize because of the huge volume expansion. At the same time, bare Ge TPP shows a first reversible capacity of 962 mA h g⁻¹, and it decreases to 395.5 mA h g⁻¹ after 100 cycles. The bare Ge TPP shows better capacity retention than bulk Ge, which may be due to the porous structure that can buffer the volume change to some extent. However, the capacity fade still can't be prevented because the large surface area would lead to significant side reaction with the electrolyte and badly electronic contact between the Ge and the Cu current, resulting in poor cyclic performance. After the coating of carbon layer, the Ge@C TPP shows an initial reversible capacity of 955 mA h g⁻¹, which is lower than bare Ge TPP. The existence of carbon should be responsible for the relatively low initial capacity. After 100 cycles, the discharge capacity of Ge@C TPP is 839 mA h g⁻¹ with a capacity retention of 87.8%, which is extremely higher than bulk Ge (37.8%) and bare Ge TPP (41.1%). The better cycling performance is due to the three-dimensional porous and uniform carbon layer that can enhance the electrical conductivity, alleviate the large volume expansion, and stabilize the SEI film during the charge/discharge process. Fig. 7c shows the rate performance of Ge and Ge@C TPP. It can be seen the Ge@C TPP anode shows the capacity of 945, 811, 735, 623, 451 and 413 mA h g⁻¹ at 0.2, 1, 2, 4, 8 and 16C, respectively. When the current density reset to 0.2C, the capacity is 838 mA h g⁻¹. For comparison, the bare Ge TPP shows the higher initial capacity of 1007 mA h g⁻¹ at 0.2C. However, it decreases to 257 mA h g⁻¹ when the current density increases to 2C. It can be concluded that the Ge@C TPP shows better rate performance than bare Ge TPP. The carbon layer can stabilize the SEI film, buffer the volume change, and enhance the conductivity, which may be responsible for the enhanced rate performance.

Fig. 7 (a) The 1^st, 5^th, 10^th, 50^th and 100^th voltage profile of Ge@C TPP; (b) cycling performance and coulombic efficiency versus cycle numbers of bulk Ge, bulk Ge@C, bare Ge TPP, Ge@C TPP at a current density of 0.2C; (c) the rate performance of bare Ge TPP and Ge@C TPP; (d) electrochemical impedance spectra of bulk Ge, bare Ge TPP and Ge@C TPP after 20 cycles.

To further identify the advantage of Ge@C TPP anode, the AC impedance measurements for bulk Ge, bare Ge TPP and Ge@C TPP is employed (Fig. 7d), respectively. It can be seen that both spectra consist of a depressed semicircle in the high-medium frequency range and a straight line in the low frequency range. At high frequencies, the spectrum is dominated by external cell connections, electrical conduction between the current collector and sputtered substance, and the ionic conduction through the electrolyte.³⁷ As the frequency decreased to high-medium range, the depressed semicircle in the spectrum is attributed to the interfacial charge transfer impedance and SEI film.³⁸ And in the low frequency range, the straight line can be attributed to the diffusion of Li in the electrodes.³⁹ As observed, the diameter of the semicircle in the Ge@C TPP anode is significantly smaller than the other two anodes, revealing lower charge-transfer impedances. The EIS spectrums of 20^th, 50^th and 100^th cycles are also compared (Fig. 8). As can be seen, the conductivity of Ge@C TPP at 20^th cycle is better than Ge TPP, while the difference is small. When the cycles increase, both the charge-transfer impedances of the Ge TPP and Ge@C TPP electrodes are increasing. However, from 20^th to 100^th, the charge-transfer impedances of the Ge TPP electrode increase very quickly, while the Ge@C TPP sample is relatively stable, especially for the 50^th and 100^th cycle. These results indicate the carbon-coated sample has good electrical contact during long-term cycles, which may result in the good long-term cycling performance.

Fig. 8 EIS spectrum of Ge TPP and Ge@C TPP in different cycle number (a) 20^th; (b)50^th; (c)100^th.

In order to confirm the stability of the three-dimensional porous structure during the lithiation/delithiation process, the morphology of the bulk Ge, bare Ge TPP, and Ge@C TPP before and after cycling was characterized by SEM analysis (Fig. 9). It can be seen that the structure of the Ge@C TPP can be retained after cycling, while the bulk Ge and bare Ge TPP seems broken and pulverizing. Therefore, it can be concluded that the structure of the Ge@C TPP is more stable than the bulk Ge and bare Ge TPP, which can also explain the enhanced performance of the Ge@C TPP anode.

Fig. 9 SEM images of bulk Ge particles (a) and (b), Ge TPP (c) and (d), and Ge@C TPP (e) and (f) before and after cycling.

Conclusions

Ge@C three-dimensional porous particles were synthesized via the decomposition of magnesium germanide (Mg₂Ge) and subsequent deposition of carbon layer. The as-synthesized Ge@C TPP was applied as anode materials of Li-ion batteries, which show a discharge capacity of 839 mA h g⁻¹ after 100 cycles at 0.2C. The performance of Ge@C TPP anode is extremely better than bulk Ge and bare Ge TPP anodes. The porous and core–shell structures can accommodate the volume change, give more lithiation sites, and stabilize the structure during the charge/discharge process, which are responsible for the enhanced performance.

Acknowledgements

The authors would like to appreciate the financial supports from the 863 Project (No. 2011AA050517), Natural Science Foundation of China (No. 51002133), Zhejiang Provincial Natural Science Foundation of China (LY13E020003) and the Fundamental Research Funds for the Central Universities.

References

1. J. Tarascon and M. Armand, Nature, 2001, 414, 359–367.
2. G. Derrien, J. Hassoun, S. Panero and B. Scrosati, Adv. Mater., 2007, 19, 2336–2340.
3. S. Han, B. Jang, T. Kim, S. M. Oh and T. Hyeon, Adv. Funct. Mater., 2005, 15, 1845–1850.
4. J. Chen, L. Xu, W. Li and X. Gou, Adv. Mater., 2005, 17, 582–586.
5. X. Zhu, Y. Zhu, S. Murali, M. Stoller and R. Ruoff, ACS Nano, 2011, 5, 3333–3338.
6. M. Winter, J. O. Besenhard, M. E. Spahr and P. Novak, Adv. Mater., 1998, 10, 725–763.
7. G. Cui, L. Gu, L. Zhi, N. Kaskhedikar, P. Aken, K. Müllen and L. Maier, Adv. Mater., 2008, 20, 3079–3083.
8. G. Derrien, J. Hassoun, S. Panero and B. Scrosati, Adv. Mater., 2007, 19, 2336–2340.
9. J. Yoo, J. Kim, Y. Jung and K. Kang, Adv. Mater., 2012, 24, 5452–5456.
10. W. Li, Y. Yin, S. Xin, W. Song and Y. Guo, Energy Environ. Sci., 2012, 5, 8007–8013.
11. D. Wang, Y. Chang, Q. Wang, J. Cao, D. Farmer, R. Gordon and H. Dai, J. Am. Chem. Soc., 2004, 126, 11602–11611.
12. I. Hwang, J. Kim, S. Seo, S. Lee, J. Lee and D. Kim, Chem. Commun., 2012, 48, 7061–7063.
13. X. Liu, Y. Liu, A. Kushima, S. Zhang, T. Zhu, J. Li and J. Huang, Adv. Energy Mater., 2012, 2, 722–741.
14. C. Chan, X. Zhang and Y. Cui, Nano Lett., 2008, 8, 307–309.
15. K. Seng, M. Park, Z. Guo, H. Liu and J. Cho, Angew. Chem., Int. Ed., 2012, 51, 5657–5661.
16. J. Wang, N. Du, H. Zhang, J. Yu and D. Yang, J. Power Sources, 2012, 208, 434–439.
17. J. Yu, N. Du, H. Zhang and D. Yang, RSC Adv., 2013, 3, 7713–7717.
18. J. Wang, N. Du, H. Zhang, Z. Song, H. Wu, H. Zhang and D. Yang, J. Power Sources, 2013, 229, 185–189.
19. S. Woo, S. Choi, J. Park, W. Yoon, S. Hwang and D. Whang, J. Electrochem. Soc., 2013, 160(1), A112–A116.
20. J. Ren, Q. Wu, H. Tang, G. Hong, W. Zhang and S. Lee, J. Mater. Chem. A, 2013, 1, 1821–1826.
21. W. Li, Z. Yang, J. Cheng, X. Zhong, L. Gu and Y. Yu, Nanoscale, 2014, 6, 4532–4537.
22. S. Fang, L. Shen, H. Zheng and X. Zhang, J. Mater. Chem. A, 2015, 3, 1498–1503.
23. D. Li, C. Feng, H. Liu and Z. Guo, J. Mater. Chem. A, 2015, 3, 978–981.
24. L. Ouyang, L. Guo, W. Cai, J. Ye, R. Hu, J. Liu, L. Yang and M. Zhu, J. Mater. Chem. A, 2014, 2, 11280–11285.
25. Y. Ko, J. Kang, G. Lee, J. Park, K. Park, Y. Jin and D. Kim, Nanoscale, 2011, 3, 3371–3375.
26. M. Park, Y. Cho, K. Kim, J. Kim, M. Liu and J. Cho, Angew. Chem., Int. Ed., 2011, 50, 9647–9650.
27. G. Yue, X. Zhang, Y. Zhao, Q. Xie, X. Zhang and D. Peng, RSC Adv., 2014, 4, 21450.
28. T. Kim, J. Park, S. Chang, S. Choi, J. Ryu and H. Song, Adv. Energy Mater., 2012, 2, 860–872.
29. F. Yuan, H. Yang and H. Tuan, ACS Nano, 2012, 6, 9932–9942.
30. H. Kim, Y. Son, C. Park, J. Cho and H. Choi, Angew. Chem., Int. Ed., 2013, 52, 5997–6001.
31. J. Wang, N. Du, H. Zhang, J. Yu and D. Yang, J. Mater. Chem. A, 2012, 22, 1511–1515.
32. M. Park, K. Kim, J. Kim and J. Cho, Adv. Mater., 2010, 22, 415–418.
33. L. Yang, Q. Gao, L. Li, Y. Tang and Y. Wu, Electrochem. Commun., 2010, 12, 418–421.
34. M. Liu, X. Ma, L. Gan, Z. Xu, D. Zhu and L. Chen, J. Mater. Chem. A, 2014, 2, 17107–17114.
35. J. Cheng, J. Wang, W. Li, X. Liu and Y. Yu, RSC Adv., 2014, 4, 37746–37751.
36. H. Jia, R. Kloepsch, X. He, P. Badillo, P. Gao, O. Fromm, T. Placke and M. Winter, Chem. Mater., 2014, 26, 5683–5688.
37. G. Wang, Z. Lu, X. Gao, X. Liu and J. Wang, J. Power Sources, 2009, 189, 655.
38. R. Ruffo, S. Hong, C. Chan, R. Huggins and Y. Cui, J. Phys. Chem. C, 2009, 113, 11390.
39. C. Wang, A. Appleby and F. Little, J. Electroanal. Chem., 2001, 497, 33.

---