One-step synthesis of Ni-doped SnO₂ nanospheres with enhanced lithium ion storage performance

Abstract

In our work, Ni-doped SnO₂ nanospheres have been synthesized via a one-step hydrothermal method using glucose as the soft template. Their structure and physicochemical properties were investigated using X-ray diffraction (XRD), a transmission electron microscope (TEM), a field-emission scanning electron microscope (FE-SEM) equipped with energy-dispersive X-ray spectroscopy (EDS), high resolution transmission electron microscopy (HRTEM) and electrochemical methods. Compared with the pristine SnO₂, appropriate Ni-doped SnO₂ nanospheres showed much better rate capability and excellent cycling performance. In particular, the sample with 5 mol% Ni showed a high initial reversible capacity of 1267 mA h g⁻¹ at a charge–discharge rate of 0.2 C, and a stable reversible capacity of 674.8 mA h g⁻¹ after 35 cycles. Nickel doping could accommodate the huge volume expansion and avoid the agglomeration of nanoparticles. Thus, the electrochemistry performance was significantly improved.

---

Introduction

Lithium-ion batteries, with high reversible capacity, long cycle life, and low cost, have attracted considerable attention in science and industry fields.^1–3 Graphite, as the most popular anode material in commercial LIBs anodes, shows excellent cyclability, but low capacity (372 mA h g⁻¹) limits its industrial applications.⁴ In particular, some promising negative materials such as silicon (Si), tin (Sn), SnO₂, Fe₂O₃, Co₃O₄, and MnO₂ have attracted much interest, because of their higher theoretical capacities and higher energy densities than those of conventional graphite anodes.⁵ Among them, tin dioxide (SnO₂), with a high theoretical capacity (782 mA h g⁻¹), has attracted considerable attention from chemical engineers.⁶ Due to the maximum intake of 4.4 Li per Sn atom to form Li_4.4Sn during alloying, SnO₂ is considered as one of the most suitable materials to replace the carbon anode in the lithium-ion battery. Unfortunately, the practical application of SnO₂-based anodes has been hampered by their poor cycling stability. SnO₂ suffers from drastic volume expansion–contraction (∼300%) during the Li-ion insertion–extraction reaction, which eventually leads to mechanical failure and a loss of electrical activity of the electrode.⁷

To improve the electrochemical performance, significant research efforts have focused on producing a buffer layer or void space for SnO₂.⁷ These methods include reducing the particle size to nanoscopic dimensions by synthesizing various morphologies of SnO₂, such as zero-dimensional (0D) nanoparticles;⁸ one-dimensional (1D) nanorods, nanobelts,⁹ nanowires and nanotubes;¹⁰ two-dimensional (2D) nanosheets;¹¹ and three-dimensional (3D) hierarchical architectures.¹² For example, Lou et al. reported a one-pot synthesis of SnO₂ hollow nanospheres, which had a high capacity Li-ion storage (about 500 mA h g⁻¹ after 40 cycles at 0.2 C).¹³ Moreover, carbon supported SnO₂ nanocomposites offer an effective way to tailor their electrical and microstructural properties.¹⁴ Lou et al. prepared SnO₂–carbon composite hollow spheres by ternary self-assembly approaches, which displayed a stable reversible capacity of about 473 mA h g⁻¹ after 50 cycles.¹⁵ The carbon networks acted as a physical buffer for the huge volume change, as well as the electrical conducting path. There have also been efforts to mix SnO₂ with other metal oxides, such as ZnO–SnO₂, CuO–SnO₂, InO₂–SnO₂, SiO₂–SnO₂, and NiO–SnO₂.^16–19 Some common properties appear in these hybrid systems, such as the capability to absorb the volume change and the ability to react reversibly with a number of lithium ions.^20,21 Due to those properties, the electrochemical performance is greatly improved.

Nowadays, SnO₂ are also intentionally ‘doped’ by incorporating atoms or ions of suitable elements into host lattices to enhance the cycling stability.^22,23 For example, Wang et al. synthesized antimony-doped SnO₂ with excellent cycling performance.²⁴ A specific discharge capacity of 637 mA h g⁻¹ was found after 100 cycles at 0.2 C, corresponding to 49.5% of the initial specific capacity.

However, studies on Ni-doped SnO₂ as anode materials for lithium-ion batteries have been seldom reported. Herein, we report a simple hydrothermal method for the synthesis of nickel-doped SnO₂ nanospheres. Glucose played an important role in mediating the nanospheres. Nickel doping occurred by replacing tin in the crystal structure, and then the component Ni served as a buffer, so that the electrode avoided the usual drastic volume expansion and agglomeration of nanoparticles. Based on this, the electrochemical performance of the electrode was significantly improved. Our results indicate the potential application for Ni-doped SnO₂ nanospheres as promising anode materials for future lithium-ion batteries.

Experimental

Preparation of samples

All of the reactants were of analytical grade and used without further purification. The content of Ni was controlled according to the relation: Ni/(Ni + Sn). In a typical synthesis, 0.6 g of tin dichloride dehydrate (SnCl₂·2H₂O), 2.51 g of sodium citrate (Na₃C₆H₅O₇·2H₂O), and 5% molar ratio of nickel acetate (Ni(CH₃COO)₂·4H₂O) were dispersed in distilled water (15 ml) under stirring. After stirring for several minutes, 0.55 g of glucose was added to the solution with continuous stirring to form a transparent solution. Then, the solution was transferred into a 25 ml Teflon-lined stainless steel autoclave and kept at 180 °C for 9 h. The as-prepared precipitate was separated by centrifugation and washed with distilled water and ethanol several times. Finally, the product was dried at 60 °C overnight. The dried products were annealed at 700 °C under air for 4 h to obtain the SnO₂ nanospheres. For comparison, the control samples with 0 mol%, 2.5 mol% and 7.5% mol Ni were prepared under identical conditions.

Characterization

XRD measurements were carried out with a Bruker D8 advance X-ray diffractometer with Cu-Kα radiation (λ = 0.15418 nm) in the 2θ range from 20° to 70°. The morphology of the as-prepared samples were characterized using a field-emission scanning electron microscope (FE-SEM, Hitachi, S4800) equipped with energy dispersive X-ray spectroscopy (EDS), a transmission electron microscope (TEM, JEM 100-CXII, 80 kV) and a high-resolution transmission electron microscope (HRTEM, JEM-2100, 200 kV).

Electrochemical measurement

In order to measure the electrochemical performance of the electrodes, CR2032 coin-type cells were assembled in an argon-filled glove box. The anode electrode was prepared by dispersing 70 wt% active materials, 20 wt% of acetylene black and 10 wt% of carboxymethyl cellulose (CMC) binder in distilled water. Then, the mixed slurry was spread uniformly on a Cu foil current collector and dried at 90 °C overnight, before testing. The electrolyte was 1 M LiPF₆ in a mixture of ethylene carbonate (EC)–diethyl carbonate (DEC) (1 : 1 in volume), and the separator was a polypropylene micro-porous film (Celgard 2400). The charge–discharge tests were performed on a Land CT2001A system (Wuhan, China), at a rate of 0.2 C, and in the potential range of 0.005–2.0 V. Cycle voltammograms (CV) curves were recorded on a CHI-630B electrochemical workstation in the range of 0.0–2.50 V (vs. Li⁺/Li), at the scan rate of 0.2 mV s⁻¹. The electrochemical impedance spectroscopy (EIS) was carried out by an electrochemical workstation (MATERIALS MATES 510, ITALIA), by applying ac voltage in the frequency range of 10 mHz–100 kHz.

Results

Characterization of Ni-doped SnO₂ nanospheres

Ni-doped SnO₂ nanospheres were synthesized via a one-step hydrothermal method in our work. Their structure and physicochemical properties were investigated via a series of measurements. Fig. 1 shows the XRD patterns for the Ni-doped SnO₂ samples and the individual SnO₂. These indicate that all the peaks corresponded to a pure phase (cassiterite, JCPDS No. 41-1445) and no diffraction peaks of crystalline by-products, such as NiO, were detected.²⁵ This indicates that the nickel doping was performed by substituting tin in the crystal structure. For the undoped SnO₂, the peaks at 26.57°, 33.92°, 37.927°, 51.834°, 54.797°, 57.928° and 61.849° were attributed to the (110), (101), (200), (211), (220), (002) and (310) crystal planes, respectively.

Fig. 1 XRD patterns of (a) pristine SnO₂, (b) 2.5 mol% Ni–SnO₂, (c) 5 mol% Ni–SnO₂, and (d) 7.5 mol% Ni–SnO₂.

All the samples were also characterized by the Raman spectra, with an exciting laser wavelength of 514 nm, as illustrated in Fig. 2. For the undoped-SnO₂ sample, typical modes positioned at A_1g = 634.7 cm⁻¹ and B_2g = 778.19 cm⁻¹ were detected.²⁶ Clearly, the strongest typical A_1g mode showed a redshift and a decrease of peak intensity with the increase in the Ni²⁺ doping concentration. The main features of the other two bands, labelled as M1 and M2, were associated with surface modes.²⁷ It also could be seen that the peak values dramatically decreased due to the Ni doping. The above phenomena are powerful evidence of the nickel incorporation into the SnO₂ cassiterite lattice.^27,28 Moreover, no Ni–O characteristic peak (about 570 cm⁻¹) was detected in the spectrum of the doped samples. This is in agreement with the previous XRD results and further confirms the incorporation of bivalent Ni²⁺ to the SnO₂ cassiterite lattice.

Fig. 2 Raman spectra of (a) pristine SnO₂, (b) 2.5 mol% Ni–SnO₂, (c) 5 mol% Ni–SnO₂, and (d) 7.5 mol% Ni–SnO₂.

The morphology of the pure SnO₂ and Ni-doped SnO₂ samples were examined by TEM and SEM images. Fig. 3 displays the TEM images of the individual SnO₂ and Ni-doped SnO₂. All of them showed nanospheres composed of numerous tiny nanoparticles. The TEM images also showed that the surface of the nanospheres had many pores, indicating good accessibility to the exterior.²² The diameter size of the nanospheres for 0 mol% Ni, 2.5 mol% Ni, 5 mol% Ni and 7.5 mol% Ni samples were around 160 nm, 140 nm, 120 nm, and 100 nm, respectively. The size of the nanospheres slightly reduced with the increasing content of Ni.

Fig. 3 TEM images of (a) pristine SnO₂, (b) 2.5 mol% Ni–SnO₂, (c) 5 mol% Ni–SnO₂, and (d) 7.5 mol% Ni–SnO₂.

The schematic illustration in Scheme 1 is proposed to show the changes during the hydrothermal and calcination process. Herein, the introduction of sodium citrate and glucose were important for the preparation of SnO₂ nanospheres. The chelation between the divalent tin ions and sodium citrate could efficiently restrain the rapid precipitation of Sn(OH)₂via hydrothermal treatment, which was beneficial for the oxidation of bivalent tin to tetravalent tin with the oxygen dissolved in the solution.²⁹ Moreover, glucose was found to play an important role in mediating the spherical structure. The hydrolysis of SnCl₂ occurred in the microenvironment of glucose dehydration, and carbon nanospheres loaded with SnO₂ nanoparticles were formed in the hydrothermal reaction.³⁰ After calcination in air, the carbon in the nanospheres was removed, leaving behind the nanospheres structure.

Scheme 1 Schematic illustration of the hydrothermal and calcination process.

For further investigation of the morphology of the nanospheres, the representative SEM image, accompanied with the EDS spectrum of the 5 mol% Ni sample, are depicted in Fig. 4a and b. The nanospheres showed uniform size and good dispersion (Fig. 4a). The EDS spectrum of the nanospheres confirmed the existence of Sn, O and Ni, from which the atom ratio of Ni/(Sn + Ni) was calculated to be 5.09 mol%. By combing the Raman spectrum analysis, Ni²⁺ was successfully incorporated into the SnO₂ lattice. The crystal structures and edges of the 5 mol% Ni-doped SnO₂ were determined with HRTEM (Fig. 4c and d). Clear lattice fringes were detected in the image. An interplanar spacing of about 0.325 nm corresponded to the (110) planes of SnO₂, and another at about 0.26 nm was attributed to the (101) planes. Moreover, the selected-area electron diffraction (SAED) pattern (Fig. 4e) of the Ni-doped nanospheres showed the characteristic diffraction rings, which were assigned to the (110), (101), (200) and (211) crystal planes. These data are in agreement with the XRD results and indicated the polycrystalline nature of the powder.

Fig. 4 SEM image (a), EDS spectrum (b), HR-TEM images (c) and (d), the corresponding SAED pattern (e) of the as-prepared 5 mol% Ni-doped sample.

Electrochemical performance of Ni-doped SnO₂

Fig. 5 exemplifies the discharge cycling performance of different electrodes at 0.2 C in a voltage window of 0.005–2.0 V for 35 cycles. The initial reversible capacity of 954.1 mA h g⁻¹ and the specific discharge capacity of 296.9 mA h g⁻¹ after 35 cycles for pristine SnO₂ were measured. Because of the large volume change during the charge–discharge process, the capacity of SnO₂ decayed rapidly.⁷ In this work, the cycling performance of the SnO₂ was greatly enhanced with the appropriate content of nickel doping. For example, after 35 cycles, the reversible capacities for 2.5 mol%, 5 mol% and 7.5 mol% Ni-doped samples were 230.2 mA h g⁻¹, 674.8 mA h g⁻¹ and 447.8 mA h g⁻¹, respectively. It was found that the appropriate molar rate of Ni led to a much improved electrochemical performance.

Fig. 5 Cycling performance of SnO₂ and Ni-doped SnO₂ samples in the voltage range of 0.005–2.0 V.

The 5 mol% Ni-doped SnO₂ electrode exhibited a much better rate capability than the individual SnO₂ at various rates of 0.1 C, 0.2 C, 0.5 C and 1 C (Fig. 6). As shown in Fig. 6, at a rate of 1 C, the specific capacity of 5 mol% Ni-doped SnO₂ was about 485.5 mA h g⁻¹, and this was much higher than that of the pure SnO₂ electrode (181.9 mA h g⁻¹). When the rate returned to 0.1 C, the discharge capacity recovered to 628.9 mA h g⁻¹. This excellent capacity retention was mainly attributed to the excellent structural stability. It is hypothesized that the nickel doping could decrease the lattice volume expansion effect in the alloying and dealloying process. When bivalent nickel substituted tetravalent tin in the crystal structure, Ni was thought to be a buffer for the lattice expansion, which could relieve the mechanical stress induced by the large volume change.¹⁹

Fig. 6 Rate performance of SnO₂ and 5 mol% Ni-doped SnO₂ samples in the voltage range of 0.005–2.0 V.

To further understand the electrochemical process, the cycle voltammogram (CV) curves of different electrodes for the first two cycles were analyzed, as shown in Fig. 7. The CV curves (Fig. 7b) of 5 mol% Ni-doped SnO₂ were in good agreement with the characteristics of pristine SnO₂ (Fig. 7a). It was thus evident that the doping of Ni did not change the electrochemical reaction process. Fig. 7b clearly indicated a strong irreversible reduction peak around 0.7 V during the first cycle, which was due to the irreversible reduction of SnO₂ to Sn (eqn (1)).^22,24 In the subsequent cycles, this peak no longer appeared.

Sn_1−xNi_xO + 4Li⁺ + 4e⁻ → Sn_1−xNi_x + 2Li₂O (1)

The irreversible capacity loss during the initial cycles was mostly due to the formation of Li₂O or the solid electrolyte interface (SEI) layers.³¹ In the first cycle, the cathodic peak at about 0.06 V and the anodic peak at 0.59 V were detected, which corresponded to the reversible alloying and dealloying (Fig. 7b), as described in eqn (2).³²

Sn_1−xNi_x + yLi⁺ + ye⁻ ↔ Li_ySn_1−xNi_x (0 ≤ y ≤ 4.4) (2)

An oxidation peak around 1.3 V and a reduction peak around 0.87 V were assigned to the partial reversibility of eqn (1).³³ This implied that the overall Li storage capacity in the SnO₂ anodes increased.³⁴ Compared to pure SnO₂ (Fig. 7a), the potential difference (ΔE) between the anodic and cathodic peaks of 5 mol% Ni-doped SnO₂ (Fig. 7b) was reduced. The doping of Ni prevented SnO₂ from cracking and led to the improvement of the redox behavior.

Fig. 7 First second CV curves of SnO₂ and Ni-doped SnO₂ anodes: (a) pristine SnO₂, and (b) 5 mol% Ni-doped SnO₂.

Fig. 8 displays the first three voltage profiles for SnO₂ and the 5 mol% Ni-doped SnO₂ electrodes at a constant current density of 0.2 C. The initial discharge and charge capacity of the Ni-doped anode were much larger than the tin oxide anode. The first discharge capacity for SnO₂ and Ni-doped SnO₂ were 1392 mA h g⁻¹ and 1951 mA h g⁻¹, showing a coulombic efficiency of 67.3% and 62.6%, respectively. During the second cycle, the discharge capacity of undoped and doped electrodes decreased to 954 mA h g⁻¹ and 1267 mA h g⁻¹, leading to a much higher coulombic efficiency of 85.2% and 88.9%, respectively. After the first cycle, the 5 mol% Ni-doped sample displayed a higher coulombic efficiency than that of individual SnO₂. It was also evident that both of the electrodes exhibited a large capacity loss in the subsequent cycles, which was due to the formation of a solid electrolyte interface (SEI) layer and the irreversible reduction of SnO₂ to Sn (eqn (1)).

Fig. 8 The initial three charge and discharge curves of (a) pure SnO₂ electrode, and (b) the 5 mol% Ni-doped electrode.

To better demonstrate the improved electrochemical performance of 5 mol% Ni-doped SnO₂, Nyquist plots of the Ni-doped and pure SnO₂ are shown in Fig. 9. Before the EIS measurement, the electrodes were cycled between 0 V and 2.0 V at a 0.1 C rate for 3 cycles and allowed to stand for 10 h at 2.0 V to ensure the formation of the stable SEI films on the surface of the electroactive particles. All the electrodes exhibited Nyquist plots composed of a small intercept at high frequency, a circular arc at high to medium frequency, and a straight line at low frequency. For the Li-ion battery, the intercept in the high-frequency region was attributed to ohmic resistance, which represented the total resistance of the electrolyte, separator, and electrical contacts; the circular arc in the medium-frequency region was assigned to the charge-transfer impedance on electrode; and the inclined line corresponded to the lithium-diffusion process within the electrodes.^35,36 The ohmic resistance was similar for all the electrodes, because the electrodes were prepared by adding a conductive carbon black agent, which meant good conductivity in the electrode. From Fig. 9, it is clear that for different circular arcs, the order of the curvature from small to large arcs was 0 mol%, 2.5 mol%, 7.5 mol% and 5 mol%, respectively. The 5 mol% Ni-doped electrode provided the largest curvature compared to the others. This indicates that the charge transfer resistance at the electrode/electrolyte interface of the 5 mol% Ni-doped SnO₂ was lowest, which consequently decreases the overall battery internal resistance. As a result, the 5 mol% Ni-doped electrode could accordingly have a higher reactivity and lower polarization. Based on the above analysis of the tests, nickel doping was proven to be beneficial for enhancing the electrochemistry performance.

Fig. 9 Electrochemical impedance spectroscopy plots of SnO₂ and Ni-doped SnO₂ samples.

Conclusions

In summary, using a quite facile hydrothermal route, we succeeded in the fabrication of Ni-doped SnO₂ nanospheres with different molar ratios of Ni. In this process, the structures were mediated by glucose. Our work focused on determining the sample with the most suitable amount of nickel doping, which exhibited much better cyclic performance and rate capability than the pristine SnO₂. In particular, the sample with 5 mol% Ni had the best electrochemical performance, and with a high initial reversible capacity of 1267 mA h g⁻¹ at a charge–discharge rate of 0.2 C and a reversible capacity of 674.8 mA h g⁻¹ after 35 cycles. The doping of nickel acted as a buffer, by reducing the volume expansion effect. In addition, the decreased possibility of tin aggregation during the alloying and dealloying process led to a lower charge transfer resistance, which then brought about a higher overall ionic conductivity. It is anticipated that nickel-doped SnO₂ nanostructures could play an important role in the further improvement of lithium-ion battery anodes of SnO₂-based materials.

Acknowledgements

This study was supported by the Natural Science Foundation of China (Grant 21276142).

Notes and references

1. S. Ding, J. S. Chen and X. W. Lou, Adv. Funct. Mater., 2011, 21, 4120.
2. S. P. Wang, H. X. Yang, L. J. Feng, S. M. Sun, J. X. Guo, Y. Z. Yang and H. Y. Wei, J. Power Sources, 2013, 233, 43.
3. W. J. Zhang, Q. J. Liu, L. J. Feng, S. P. Wang, Y. Z. Yang and H. Y. Wei, New J. Chem., 2014, 38, 2265.
4. S. Ding, D. Luan, F. Y. C. Boey, J. S. Chen and X. W. Lou, Chem. Commun., 2011, 47, 7155.
5. C. M. Wang, W. Xu, J. Liu, J. G. Zhang, L. V. Saraf and B. W. Arey, Nano Lett., 2011, 11, 1874.
6. C. Wang, Y. Zhou, M. Ge, X. Xu, Z. Zhang and J. Jiang, J. Am. Chem. Soc., 2010, 132, 46.
7. P. G. Bruce, B. Scrosati and J. M. Tarascon, Angew. Chem., Int. Ed., 2008, 47, 2930.
8. D. F. Zhang, L. D. Sun, J. L. Yin and C. H. Yan, Adv. Mater., 2003, 15, 1022.
9. J. Z. Wang, N. Du, H. Zhang, J. X. Yu and D. Yang, J. Phys. Chem. C, 2011, 115, 11302.
10. Y. Wang, H. C. Zeng and J. Y. Lee, Adv. Mater., 2006, 18, 645.
11. Y. M. Li, X. J. Lv, J. Lu and J. H. Li, J. Phys. Chem. C, 2010, 114, 21770.
12. H. Wang, Y. M. Wu, Y. S. Bai, W. Zhou, Y. R. An, J. H. Li and L. Guo, J. Mater. Chem., 2011, 21, 10189.
13. X. W. Lou, Y. Wang, C. Yuan, J. Y. Lee and L. A. Archer, Adv. Mater., 2006, 18, 2325.
14. X. W. Lou, J. S. Chen, P. Chen and L. A. Archer, Chem. Mater., 2009, 21, 2868.
15. X. W. Lou, D. Deng, J. Y. Lee and L. A. Archer, Chem. Mater., 2008, 5, 6562.
16. P. Sun, L. You, Y. Sun, N. Chen, X. Li, H. Sun, J. Ma and G. Lu, CrystEngComm, 2012, 14, 1701.
17. C. Li, W. Wei, S. Fang, H. Wang, Y. Zhang, Y. Gui and R. Chen, J. Power Sources, 2010, 195, 2939.
18. H. B. Wu, J. S. Chen, H. H. Hng and X. W. D. Lou, Nanoscale, 2012, 4, 2526.
19. J. S. Chen and X. W. Lou, Small, 2013, 9, 1877.
20. P. Lavela and J. L. Tirado, J. Power Sources, 2007, 172, 379.
21. P. Lavela, J. L. Tirado and C. Vidal-Abarca, Electrochim. Acta, 2007, 52, 7986.
22. Y. D. Wang and T. Chen, Electrochim. Acta, 2009, 54, 3510.
23. H. El-Shinawi, M. Böhm, T. Leichtweiß, K. Peppler and J. Janek, Electrochem. Commun., 2013, 36, 33.
24. Y. D. Wang, I. Djerdj, B. Smarsly and M. Antonietti, Chem. Mater., 2009, 21, 3202.
25. M. F. Hassan, M. M. Rahman, Z. P. Guo, Z. X. Chen and H. K. Li, J. Mater. Chem., 2010, 20, 9707.
26. J. Hays, A. Punnoose, R. Baldner, M. H. Engelhard, J. Peloquin and K. M. Reddy, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 72, 075203.
27. F. H. Aragón, J. A. H. Coaquira, P. Hidalgo, S. W. da Silva, S. L. M. Brito, D. Gouvêa and P. C. Morais, J. Raman Spectrosc., 2011, 42, 1081.
28. X. F. Liu, J. Iqbal, Z. B. Wu, B. He and R. H. Yu, J. Phys. Chem. C, 2010, 114, 4790.
29. R. Yang, Y. G. Gu, Y. Q. Li, J. Zheng and X. G. Li, Acta Mater., 2010, 9, 864.
30. D. Deng and J. Y. Lee, Chem. Mater., 2008, 20, 1841.
31. D. Fattakhova, L. Kavan and P. Krtil, J. Solid State Electrochem., 2001, 5, 196.
32. I. A. Courtney, W. R. McKinnon and J. R. Dahna, J. Electrochem. Soc., 1999, 59, 146.
33. Y. Idota, T. Kubota, A. Matsufuji, Y. Maekawa and T. Miyasaka, Science, 1997, 276, 1395.
34. Y. S. Lin, J. G. Duh and M. H. Hung, J. Phys. Chem. C, 2010, 114, 13136.
35. M. M. Liu, X. W. Li, H. Ming, J. Adkins, X. M. Zhao, L. Su, Q. Zhou and J. W. Zheng, New J. Chem., 2013, 37, 2096.
36. Q. J. Liu, F. Yang, S. P. Wang, L. J. Feng, W. J. Zhang and H. Y. Wei, Electrochim. Acta, 2013, 111, 903.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4nj00989d

---