A phosphotungstic acid self-anchored hybrid proton exchange membrane for direct methanol fuel cells

Abstract

A phosphotungstic acid (HPW) self-anchored hybrid proton exchange membrane (PES/PVP-HPW) is prepared and evaluated in direct methanol fuel cells (DMFCs). The proton conductivity of the hybrid membrane is 0.045 S cm⁻¹ at 25 °C, and reaches 0.078 S cm⁻¹ at 80 °C. The hybrid membrane shows a methanol permeability of 1.65 × 10⁻⁶ cm² s⁻¹. The stability test for the hybrid membrane in 2 M methanol at 50 °C for about 100 h reveals that HPW is well immobilized in the hybrid membrane. The DMFC based on the PES/PVP-HPW hybrid membrane with a thickness of 55 μm exhibits comparable performance of 132 mW cm⁻² to that of Nafion212 at 80 °C. The effects of the hybrid membrane thickness and methanol concentration on DMFCs performance are evaluated. The optimal methanol concentration and thickness of the membrane is about 1–2 M and 55 μm, respectively. Furthermore, a 130 h stability test for DMFC with PES/PVP-HPW demonstrates that the hybrid membrane is quite stable, which indicates that PES/PVP-HPW is an attractive low-cost alternative proton exchange membrane to Nafion® for portable power sources.

---

Introduction

Direct methanol fuel cells (DMFCs) have gained great attention as a candidate for portable power devices because of their advantages, such as high energy density, low maintenance cost, quick refuelling, easy fuel handling and simple operation systems. A proton exchange membrane (PEM) is one of the key components in DMFCs.^1–3 The commonly used PEM in DMFCs is the state-of-the-art Nafion® membrane. However, Nafion® suffers from a high cost, complicated synthesis procedure, high methanol crossover and environmental incompatibility.^4,5 These drawbacks in addition to the high catalyst cost are the major obstacles that hinder the commercialization of DMFCs.⁶

Great efforts have been made to develop alternative PEMs for DMFCs in terms of cost reduction and improvement of functionality, including (i) modified Nafion® membrane, (ii) alternative sulfonated polymers and their composite membrane. The objective of modification for Nafion® membrane is to reduce the methanol crossover mainly through blending of inorganic components like silica,^7,8 zeolite^9,10 cesium phosphotungstate salt,^11,12 graphene oxide,¹³ and various polymers, such as poly(vinylidene fluoride),¹⁴ polybenzimidazole,¹⁵ and purple membrane,¹⁶ Also, layer-by-layer self-assembly has been introduced for surface modification of Nafion® membrane.¹⁷ Though the modification can improve the performance of Nafion®, it complicates the membrane synthesis procedure with increased cost. In another hand, the alternative sulfonated polymers, sulfonated aromatic polymers have been considered as the most attractive alternatives. However, due to the low compatibility with Nafion®-bonded electrodes, promising alternatives such as sulfonated poly(ether ether ketone),^18,19 and chitosan sulfate blending membrane²⁰ with low methanol crossover and high proton conductivity still exhibits poor performance in application of DMFCs.

Organic–inorganic hybrid membranes have attracted much attention because of their potential advantages over conventional materials due to coexistence of characteristic inorganic and organic properties of their component.^21,22 Phosphotungstic acid (H₃PW₁₂O₄₀, HPW), as one of the most attractive inorganic proton conductors, has been introduced into inorganic materials or polymer matrices either to modify the membrane or as the sole proton conductor to develop alternative PEM with improved proton conductivity and significantly reduced cost.^23–25 The major problem for application of HPW in DMFCs is the high water-solubility of HPW, which would result in its leakage and decay the conductivity of PEMs gradually. In our previous work, a novel inorganic HPW/MCM-41 PEM with high conductivity of 0.045 S cm⁻¹ at 150 °C was successfully developed.²³ However, the stability of proton conductivity needs to be improved. The HPW/MCM-41 PEM was then improved by Chan et al. to achieve an improved proton conductivity of 0.098 S cm⁻¹ at 90 °C, but the proton conductivity exhibited a rapid decrease of ∼70% for 4 h.^24,26 To overcome the water-solubility issue of HPW, the polymer with –NH₂ or –NR₂ functional groups (such as chitosan) was used to immobilize HPW molecules and obtain organic–inorganic hybrid membranes.^27–29 The composite membranes had good stability of proton conductivity, however, the DMFCs based on them showed very poor output performance due to the low proton conductivity of the membranes.

Recently, our group successfully demonstrated a highly stable self-anchored hybrid proton exchange membrane, consisting of phosphotungstic acid (HPW), polyvinylpyrrolidone (PVP) and polyethersulfones (PES), donated as PES/PVP-HPW.³⁰ The hybrid membrane exhibits high proton conductivity of 0.066 S cm⁻¹ at 60 °C and impressive performance in both conductivity stability and PEMFCs output test for about 500 h, respectively. In the hybrid membrane, PES is added as a skeleton to improve the formation and enhance the mechanical strength of the membrane. PVP that contains N-heterocycles not only plays a key role in the formation of membrane but offers self-anchored sites for HPW. The self-anchored HPW in the membrane acts as the sole proton conductor. During membrane formation, the proton of HPW jumps to an N atom of PVP, forming the NR₃H⁺ species. The left PW₁₂O₄₀³⁻ with negative charge will be surrounded by the NR₃H⁺ of PVP. Thus, HPW is well immobilized in the PVP net-like structure due to the electrostatic force.³⁰

In this work, PES/PVP-HPW hybrid membrane was prepared and evaluated for DMFC application. The proton conductivity and stability of the hybrid membrane in methanol solution were studied. The effects of the hybrid membrane thickness and methanol concentration on DMFCs performance were evaluated. The PES/PVP-HPW hybrid membrane with thickness of 55 μm exhibits comparable fuel cell performance of 132 mW cm⁻² to that of Nafion212 at 80 °C. Furthermore, a 130 h stability test for PES/PVP-HPW based single cell demonstrates that the hybrid membrane is quite stable in DMFCs, making it an attractive low-cost alternative proton exchange membrane to Nafion® for portable power sources.

Experimental

Materials

PES (M_w ≈ 480 000) was purchased from Changchun Jida New Material Co. and PVP K90 (M_w ≈ 630 000) was obtained from Gobekie New Material Co. The HPW was purchased from Tianjin Jinke Fine Chemical Research Institute. The commercial Nafion212 (EW 2100, thickness ∼50 μm) membrane was purchased from Du Pont Co. The Pt C⁻¹ (40 wt%) and PtRu black used for preparing catalyst ink were from the E-TEK. All other solvents and chemicals were purchased from commercial sources and used without further purification.

Membrane preparation

PVP, PES (mass ratio PVP : PES = 90 : 10) and HPW with desired content (30 wt% in total composite membrane) were mixed in N,N-dimethylformamide (DMF) and stirred for 12 h at room temperature to obtain a transparent and homogeneous casting solution.³⁰ The total polymer concentration was controlled to be 3.0–8.0 wt%. Then the blend solution with different polymer concentration was cast onto a glass plate and dried at 70 °C for 24 h to remove the solvent and obtain a PES/PVP-HPW hybrid membrane.

Water uptake and swelling ratio

Water uptake and swelling ratio were determined from the weight and volume differences between the dry and wet membranes. In details, the membranes with thickness about 55 μm were dried at 120 °C for several hours until constant weights and volumes were obtained. And then the dried membranes were placed in water with methanol concentration from 0 M to 10 M at room temperature (25 °C) for 24 h. The wet membranes were wiped with filter paper and quickly weighed and measured. The water uptake and swelling ratio of the membranes were calculated from the following equations:where W_dry and W_wet are the weights of the dry and wet membranes, respectively;where V_dry and V_wet are the volumes of the dry and wet membranes, respectively.

Membrane stability test

The stability of self-anchored HPW in the hybrid membrane was evaluated by immersing the hybrid membrane with thickness about 55 μm in 2 M methanol at 50 °C and followed by characterizing the HPW in leach liquor by UV-vis. The UV-vis spectra was performed on GBC Cintra 10e instrument with scanning range from 200 nm to 800 nm. The scanning speed was 400 nm min⁻¹ and slit width was 0.8 nm.

Conductivity test

The through-plane proton conductivity was measured in a single cell testing set-up via electrochemical impedance spectroscopy using IviumStat (Ivium Technologies, Netherlands). The measurement was carried out in the constant voltage mode over a frequency range of 100 kHz to 100 Hz with an oscillating voltage of 10 mV. Membranes were sandwiched between two gas diffusion electrodes, and resistance measurements were carried out under a continuous flow of 2 M methanol solution on both sides of the electrodes within a temperature range of 25–80 °C. The through-plane conductivity of the membrane was calculated by eqn (3):where l is the thickness of the membrane, S is the active area of the testing membrane, and R is the membrane resistance.

Methanol permeability measurement

The methanol permeability was determined using a diffusion cell consisting of two half-cells.³¹ Methanol solution (2 M) was placed at one side of the diffusion cell (compartment A), and deionized water was placed at the other side (compartment B). The membrane was fixed between two half-cells to separate methanol solution and deionized water. Magnetic stirrers were used in each compartment to ensure uniformity. The concentration of methanol solution was measured by a SHIMADZU GC-2014C series gas chromatograph. Peak areas were converted into methanol concentration with a calibration curve. The methanol permeability was calculated by eqn (4):where C_A and C_B are the methanol concentration in feed and in permeate, respectively. A, L and V_B are the effective area, the thickness of membrane and the volume of permeated compartment, respectively. DK is defined as the methanol permeability and t₀ is the time lag.

DMFC single fuel cell test

The PES/PVP-HPW hybrid membranes and Nafion212 were used to fabricate membrane electrolyte assemblies (MEAs) respectively. The PtRu catalyst loading of the anode was 4.0 mg cm⁻² and the Pt catalyst loading of cathode was 2.0 mg cm⁻². Catalyst ink was prepared by mixing PtRu black or Pt C⁻¹ powers and Nafion solution (5 wt%) under ultrasonic vibration. And then the catalyst ink was brushed onto the gas diffusion layer (GDL) to prepare the electrodes, and dried at 40 °C. The membrane was sandwiched between two electrodes to obtain a MEA with 4 cm² active area. The performances of the single fuel cell were tested at 25 °C, 50 °C or 80 °C by a fuel cell testing system (Greenlight G20, Canada). The anode input methanol aqueous solution flow rate was 2 mL min⁻¹ with a methanol concentration of 2 M, and the cathode input O₂ flow rate was 150 mL min⁻¹ without humidification. Before recording the polarization curve, the cell was activated at 0.4 V for 4 h at 80 °C. The methanol crossover current generated by methanol oxidation at the fuel cell cathode was measured at 80 °C with a voltage range from 0.1 V to 1.0 V at a scan rate of 5 mV s⁻¹. During the test, the N₂ flow rate at cathode was 150 mL min⁻¹.

Results and discussion

The prepared PES/PVP-HPW hybrid membrane appears transparent and homogeneous with mechanical flexibility. HPW is uniformly dispersed in the hybrid membrane, which is proved by transmission electron microscopy and X-ray diffraction. The detailed physical properties of PES/PVP-HPW hybrid membrane can be obtained in our previous work.³⁰ This work mainly focuses on the performance of the hybrid membrane in DMFC application.

Membrane proton conductivity and methanol permeation

The proton conductivity and methanol permeability of the hybrid membrane are listed in Table 1. The corresponding data of Nafion212 under the same testing conditions is also given for comparison. The through-plane proton conductivity is measured in a single cell testing set-up by sandwiching membrane between two gas diffusion electrodes, and resistance measurements are carried out under a continuous flow of 2 M methanol solution on both sides of the electrodes. The proton conductivity of the hybrid membrane increases from 0.045 S cm⁻¹ at 25 °C to 0.078 S cm⁻¹ at 80 °C, which is comparable with that of Nafion212. Methanol crossover in DMFC would result in fuel loss and a decrease of energy efficiency due to the side reaction of permeated methanol with oxygen at the cathode. Therefore the methanol permeability of membrane is an important factor affecting the cell performance of DMFC. As shown in Table 1, the methanol permeability of Nafion212 is 1.31 × 10⁻⁶ cm² s⁻¹, which is consistent with those reported in literature.⁷ Though the PES/PVP-HPW hybrid membrane has a higher permeability of 1.65 × 10⁻⁶ cm² s⁻¹, it meets the methanol permeability requirement for a PEM in DMFC (less than 5.6 × 10⁻⁶ cm² s⁻¹).³² It is thought that methanol transport in the proton exchange membrane mainly through three kinds of paths: (i) ion-cluster pores; (ii) the connecting ion channels; (iii) polymer backbone.⁷ The addition of HPW can reduce the methanol permeability of the hybrid membrane, because the hydrophilic HPW nanoparticles mainly exist around the hydrophilic ion-clusters and the ion channels, which would act as good methanol barriers.¹¹

Table 1 Proton conductivity and methanol permeability of the hybrid membrane, Nafion212 is used as control

Membrane type	Proton conductivity (S cm^−1)		Methanol permeability (10^−6 cm^2 s^−1)
	25 °C	80 °C	25 °C
PES/PVP-HPW	0.045	0.078	1.65
Nafion212	0.052	0.103	1.31

---

Water uptake and swelling ratio

Water uptake and dimensional stability of the hybrid membrane were measured in methanol water solution with methanol concentration from 0 M to 10 M, shown in Fig. 1. Both the water uptake and swelling ratio of the hybrid membrane increase with the increasing of methanol concentration. The water uptake increases from 116% to 278% with the methanol concentration increasing to 6 M. While the swelling ratio increases from 191% to 468%. However, the water uptake and swelling ratio increase rapidly when methanol concentration increases to 6 M. These results indicate that the hybrid membrane is relatively stable in low methanol concentration water solution (less than 6 M).

Fig. 1 Water uptake and swelling ratio of the hybrid membrane in methanol water solution.

Proton conductivity stability

The biggest challenge for HPW-based PEM is the leakage of HPW for its high water-solubility, which would lead to a decrease of proton conductivity and decay the DMFCs performance gradually. So it is necessary to evaluate the stability of proton conductivity and self-anchored HPW in the hybrid membrane. The proton conductivity stability of the hybrid membrane is measured in a single cell testing set-up with a continuous flow of 2 M methanol solution at both sides of a PES/PVP-HPW hybrid membrane at 25 °C (Fig. 2a). There is no obvious reduction of the proton conductivity after 120 h testing. The stability of self-anchored HPW in the hybrid membrane is tested by UV-vis spectrum, as shown in Fig. 2b. The UV-vis absorption peak of HPW in 2 M methanol solution is obtained at the wavelength of 256 nm. If the self-anchored HPW lost from the hybrid membrane, the UV-vis spectrum for the leach liquor of hybrid membrane would have an absorption peak at 256 nm. However, no characteristic absorption peak of HPW is found for the leach liquor at different time, which means HPW is extremely stable in the hybrid membrane. Comparison the characteristic absorption peak of HPW in methanol and PES/PVP-HPW hybrid membrane, it has a redshift from 256 nm to 264 nm, which means that there exists interaction between HPW and PVP in the hybrid membrane. That's why HPW is extremely stable in the hybrid membrane. These results demonstrate that the hybrid membrane is particularly suitable for long-term operation under practical working conditions.

Fig. 2 (a) Conductivity as a function of test time under a continuous flow of 2 M methanol solution at both sides of a PES/PVP-HPW hybrid membrane at 25 °C; (b) UV-vis spectra for HPW in 2 M methanol solution and the leach liquor of the PES/PVP-HPW hybrid membrane in 2 M methanol solution (after 0 h, 50 h, 100 h) at 50 °C respectively, PES/PVP-HPW is used as a control.

Fuel cell performance

In order to evaluate the feasibility of PES/PVP-HPW in DMFCs, single fuel cell with PES/PVP-HPW-55 (with thickness of 55 μm) or Nafion212 hybrid membrane was fabricated. The fuel cells are operated at 80 °C with 2 M methanol solution supply to anode. Fig. 3 shows the typical polarization and power density curves for DMFCs with Nafion212 and PES/PVP-HPW hybrid membrane respectively. The maximum power density is 132 mW cm⁻² for the hybrid membrane, which is significantly better than that of HPW-based composite membranes in other works (see Table 2). Even though the single cell based on Nafion212 gets a maximum power density of 148 mW cm⁻², which is 12% higher than that based on PES/PVP-HPW-55 hybrid membrane, it shows no distinguished advantages under the normal working cell voltage range of over 0.4 V. Considering the low cost of the hybrid membrane, it would have great potential to be a promising alternative for Nafion® membrane in DMFCs.

Fig. 3 Comparison of fuel cell performance based on the hybrid membrane and Nafion212 at the same thickness.

Table 2 Comparison of fuel cells performance in different literature

Membrane	Test conditions	Power density (mW cm^−2)	Ref.
HPW/MCM-41	150 °C, 2 M	90	23
HPW/MCM-41	150 °C, 2 M	125	24
PTA-CS-HEC	70 °C, 2 M	58	28
HPW-meso-silica	80 °C, 2 M	75	25
CS/PWA	No data	No data	27
PES/PVP-HPW	80 °C, 2 M	132	This work

---

The hybrid membrane with different thickness, denoted as PES/PVP-HPW-x (x represented the thickness, e.g., 55, 110, 170 μm) is also prepared to study the membrane thickness on DMFCs performance. The performance of DMFCs based on hybrid membrane with different thickness is shown in Fig. 4. It can be seen that the peak power density decreases from 132 mW cm⁻² to 107 mW cm⁻² when the membrane thickness increases from 55 μm to 170 μm. This is mainly due to the increased membrane resistance of thicker membrane. However, the increased membrane thickness would reduce the methanol crossover. Thus, the open circuit voltage (OCV) of DMFCs increases from 0.73 V to 0.78 V for the one with PES/PVP-HPW-170 hybrid membrane.

Fig. 4 Effect of membrane thickness on the DMFCs output performance.

Methanol crossover limiting current

In DMFCs, methanol crossover will poison the cathode Pt catalyst and reduce the OCV due to methanol oxidation at cathode. Methanol crossover current densities at 80 °C in cathode of the single cell with Nafion212 or PES/PVP-HPW-x hybrid membrane are shown in Fig. 5. PES/PVP-HPW-55 hybrid membrane has a higher limiting current density of 433 mA cm⁻² than that of Nafion212 (382 mA cm⁻²). It has been indicated experimentally that the methanol crossover rate decreases with an increase in membrane thickness. With the membrane thickness increased from 55 to 170 μm, the limiting current density decreases from 433 mA cm⁻² to 290 mA cm⁻². However, increasing of membrane thickness would increase the membrane resistance, which gradually decays the fuel cell output performance. Thus, an appropriate membrane thickness is needed to balance the methanol crossover and fuel cell performance.

Fig. 5 Comparison of methanol crossover current densities of the hybrid membranes with different thickness, Nafion212 is used as control.

Effects of the methanol concentration and operating temperature

The effects of methanol concentration on the performance of the fuel cell based on PES/PVP-HPW-55 hybrid membrane are investigated by feeding the DMFC with different methanol concentrations, as shown in Fig. 6. The peak power density increases from 80 mW cm⁻² to 132 mW cm⁻² with the increase of methanol concentration from 0.5 M to 2 M. However, the cell performance degradation is serious in both peak power density and OCV with methanol concentration of 4 M. This suggests that methanol crossover does not affect the fuel cell performance until the methanol concentration is up to 2 M. At high methanol concentration, the cathode would be drastically poisoned which results in the reduced OCV. Thus, high methanol concentration is not preferred for the DMFC based on the hybrid membrane.

Fig. 6 Effect of methanol concentration on the performance of the DMFCs based on PES/PVP-HPW-55 hybrid membrane.

The effect of operating temperature is investigated by controlling the temperature increased from 25 °C to 80 °C with 2 M methanol feeding to the anode, as shown in Fig. 7. The fuel cell performance increases significantly with higher temperature. The OCV increases from 0.7 V to 0.73 V when the temperature increases from 25 °C to 80 °C, while the increase of peak power density from 23.3 mW cm⁻² to 132 mW cm⁻² is more noticeable. Such an improvement can be attributed to the enhanced proton conductivity and acceleration in the kinetics of methanol oxidation reaction in the anode and oxygen reduction reaction in the cathode. The data indicates that the practical operating temperature for the DMFCs based on PES/PVP-HPW-x hybrid membrane should be around 80 °C.

Fig. 7 Effect of operating temperatures on the performance of the DMFCs based on PES/PVP-HPW-55 hybrid membrane.

Fuel cell stability test

To evaluate the stability of the hybrid membrane in application of DMFCs, a prolonged discharge around 130 h is performed with PES/PVP-HPW-55 at 0.4 V, 80 °C and a methanol concentration of 2 M, as shown in Fig. 8. It is found that the current density decays a little at the initial 30 h, which could be attributed to the deterioration of the Pt-based catalysts. After 30 h, the average current density becomes relatively stable around 120 mA cm⁻² without distinguished performance degradation. The stability test is very encouraging and demonstrates that the PES/PVP-HPW hybrid membrane has great potential to be a good candidate for DMFCs.

Fig. 8 A 130 h stability test for the DMFC based on PES/PVP-HPW-55 at 80 °C.

Conclusions

A HPW self-anchored hybrid membrane is prepared and evaluated as PEMs in DMFCs. The hybrid membrane shows a proton conductivity of 0.078 S cm⁻¹ at 80 °C and impressive stability in 2 M methanol at 50 °C for 100 h. Compared with Nafion212, the PES/PVP-HPW hybrid membrane shows a little higher methanol permeability of 1.65 × 10⁻⁶ cm² s⁻¹. Membrane thickness, methanol concentration and operation temperature are studied to evaluate the hybrid membrane in DMFCs. The results indicated that thinner thickness (55 μm), lower methanol concentration (1–2 M) and the temperature (∼80 °C) are suitable for the application of this hybrid membrane in DMFCs. A single cell based on PES/PVP-HPW-55 exhibits cell performance of 132 mW cm⁻², which is comparable to the one with Nafion212. In addition, the single cell shows impressive stability in a durability test for about 130 h. These results demonstrate that the PES/PVP-HPW hybrid membrane with low-cost and high stability would be a promising candidate for DMFCs applications.

Acknowledgements

The authors thank the financial support by grants from the National High Technology Research and Development Program of China (863 program, 2013AA031902), National Natural Science Foundation of China (No. 21576007, 51422301).

Notes and references

1. X. Zhao, M. Yin, L. Ma, L. Liang, C. Liu, J. Liao, T. Lu and W. Xing, Energy Environ. Sci., 2011, 4, 2736–2753.
2. J. Chang, L. Feng, C. Liu, W. Xing and X. Hu, Energy Environ. Sci., 2014, 7, 1628–1632.
3. Z. Jiang, Y. Shi, Z.-J. Jiang, X. Tian, L. Luo and W. Chen, J. Mater. Chem. A, 2014, 2, 6494–6503.
4. M. M. Hasani-Sadrabadi, E. Dashtimoghadam, F. S. Majedi, S. Wu, A. Bertsch, H. Moaddel and P. Renaud, RSC Adv., 2013, 3, 7337–7346.
5. M. E. Scofield, H. Liu and S. S. Wong, Chem. Soc. Rev., 2015, 44, 5836–5860.
6. X. Li and A. Faghri, J. Power Sources, 2013, 226, 223–240.
7. R. Jiang, H. R. Kunz and J. M. Fenton, J. Membr. Sci., 2006, 272, 116–124.
8. P. Staiti, A. S. Aricò, V. Baglio, F. Lufrano, E. Passalacqua and V. Antonucci, Solid State Ionics, 2001, 145, 101–107.
9. W. Han and K. Lun Yeung, Chem. Commun., 2011, 47, 8085–8087.
10. M. Lavorgna, L. Sansone, G. Scherillo, R. Gu and A. P. Baker, Fuel Cells, 2011, 11, 801–813.
11. S. Rao, R. Xiu, J. Si, S. Lu, M. Yang and Y. Xiang, ChemSusChem, 2014, 7, 822–828.
12. Y. Xiao, Y. Xiang, R. Xiu and S. Lu, Carbohydr. Polym., 2013, 98, 233–240.
13. C. W. Lin and Y. S. Lu, J. Power Sources, 2013, 237, 187–194.
14. S. W. Choi, Y. Z. Fu, Y. R. Ahn, S. M. Jo and A. Manthiram, J. Power Sources, 2008, 180, 167–171.
15. R. Wycisk, J. Chisholm, J. Lee, J. Lin and P. N. Pintauro, J. Power Sources, 2006, 163, 9–17.
16. X. Yan, Z. Jin, L. Yang, Z. Guo and S. Lu, J. Membr. Sci., 2011, 367, 325–331.
17. Y. Xiang, S. Lu and S. P. Jiang, Chem. Soc. Rev., 2012, 41, 7291–7321.
18. M. Li, G. Zhang, S. Xu, C. Zhao, M. Han, L. Zhang, H. Jiang, Z. Liu and H. Na, J. Power Sources, 2014, 255, 101–107.
19. Y. Liang, C. Gong, Z. Qi, H. Li, Z. Wu, Y. Zhang, S. Zhang and Y. Li, J. Power Sources, 2015, 284, 86–94.
20. X. Yan, Y. Meng, Z. Guo and C. Zheng, J. Membr. Sci., 2009, 337, 318–323.
21. F. Xu and S. Mu, J. Nanosci. Nanotechnol., 2014, 14, 1169–1180.
22. C. Laberty-Robert, K. Valle, F. Pereira and C. Sanchez, Chem. Soc. Rev., 2011, 40, 961–1005.
23. S. Lu, D. Wang, S. P. Jiang, Y. Xiang, J. Lu and J. Zeng, Adv. Mater., 2010, 22, 971–976.
24. L. Bai, L. Zhang, H. Q. He, R. K. S. O. A. Rasheed, C. Z. Zhang, O. L. Ding and S. H. Chan, J. Power Sources, 2014, 246, 522–530.
25. J. Lu, H. Tang, S. Lu, H. Wu and S. P. Jiang, J. Mater. Chem., 2011, 21, 6668–6676.
26. L. Zhang, H. Q. He, R. K. Abdul Rasheed, W. J. Zhou, Y. H. Xue, O. L. Ding and S. H. Chan, J. Power Sources, 2013, 221, 318–327.
27. Z. Cui, W. Xing, C. Liu, J. Liao and H. Zhang, J. Power Sources, 2009, 188, 24–29.
28. S. Mohanapriya, S. D. Bhat, A. K. Sahu, S. Pitchumani, P. Sridhar and A. K. Shukla, Energy Environ. Sci., 2009, 2, 1210–1216.
29. N. A. Choudhury, J. Ma and Y. Sahai, J. Power Sources, 2012, 210, 358–365.
30. S. Lu, X. Xu, J. Zhang, S. Peng, D. Liang, H. Wang and Y. Xiang, Adv. Energy Mater., 2014, 4, 1400842.
31. Q. He, J. Zheng and S. Zhang, J. Power Sources, 2014, 260, 317–325.
32. A. S. Aricò, S. Srinivasan and V. Antonucci, Fuel Cells, 2001, 1, 133–161.

---