Effect of Membrane Electrode Assembly Fabrication ... - Springer Link

Macromolecular Research, Vol. 22, No. 11, pp 1214-1220 (2014) DOI 10.1007/s13233-014-2167-x

www.springer.com/13233 pISSN 1598-5032 eISSN 2092-7673

Effect of Membrane Electrode Assembly Fabrication Method on the Single Cell Performances of Polybenzimidazole-Based High Temperature Polymer Electrolyte Membrane Fuel Cells JuYeon Lee1,2, Jeawoo Jung1,3, Jun Young Han1, Hyoung-Juhn Kim*,1,2, Jong Hyun Jang*,1,2, Hye-Jin Lee1, Eun Ae Cho1, Dirk Henkensmeier1, Jin Young Kim1, Sung Jong Yoo1, Seong-Ahn Hong1,4, and Sang Yong Nam5 1

Fuel Cell Research Center, Korea Institute of Science and Technology, Seoul 136-791, Korea Clean Energy and Chemical Engineering, University of Science and Technology, Daejeon 305-333, Korea 3 Energy and Environmental Policy and Technology, Green School, Korea University, Seoul 136-701, Korea 4 Department of Advanced Materials Chemistry, Korea University, Sejong 339-700, Korea 5 Department of Materials Engineering and Convergence Technology, Engineering Research Institute, Gyeongsang National University, Gyeongnam 660-701, Korea

2

Received April 16, 2014; Revised July 14, 2014; Accepted July 29, 2014 Abstract: Membrane electrode assemblies (MEAs) for a high temperature polymer electrolyte membrane fuel cell (HTPEMFC) were fabricated using acid-doped polybenzimidazole (PBI) as the electrolyte membrane and polytetrafluoroethylene (PTFE) as the electrode binder. PTFE concentrations of 20, 30, and 45 wt% in the electrode were evaluated to determine the optimal binder content. Additionally, the influence of applying a pressing process during MEA fabrication on the electrode performance was examined. When MEA was prepared without the pressing process, the electrode containing 20 wt% PTFE exhibited the best cell performance (338 mA cm-2 at 0.6 V). However, when MEA was prepared with the pressing process, the electrode containing 45 wt% PTFE exhibited the best cell performance (281 mA cm-2 at 0.6 V). This result is because of the inclusion of the pressing process, as gas permeability is hindered by the transfer of excess phosphoric acid from the electrolyte membrane to the electrodes. Keywords: high temperature polymer electrolyte membrane fuel cell, polybenzimidazole, polytetrafluoroethylene binder.

operation of PEMFC at temperatures of 120 oC or higher. First, the necessity of the moisture supply and removal process is eliminated because of operation at high temperatures, which simplifies the fuel cell system. Further, the size of the cooling system for the fuel cell stack is reduced. Finally, the HTPEMFC exhibits an improved CO tolerance of up to 1% in H2.4-7 Since Wainright et al.8 demonstrated the proton conductivity of PBI doped with a strong acid such as phosphoric acid in the mid-1990s, several studies on the operation of fuel cells using acid-doped PBI electrolyte membranes have been conducted. Research was accelerated further upon the development of a method for manufacturing electrolyte membranes from polymer solutions without the need for isolation of the polymer after the synthesis of PBI.9 This enabled the preparation of PBI derivatives with a diversity of functional groups. Benicewizc et al. synthesized dihydroxy-PBI (2OHPBI), which exhibted a higher proton conductivity than previously reported PBIs.10 Incorporation of various functional groups such as pyridine rings into the polymer backbone or side chains was examined for potential improvement in the conductivity of the polymer electrolyte.11,12 Incorporation of various inor-

Introduction A polymer electrolyte membrane fuel cell (PEMFC) is a type of fuel cell with an electrolyte membrane constructed from a polymer that shows proton exchange capability. Because of several advantages compared to other fuel cell types, including low operating temperatures and small performance reduction by the repetition of on/off operation,1,2 the applications of PEMFCs are expanding to include automobile power sources and power supplies for households. PEMFCs can be classified as low temperature PEMFCs (LTPEMFCs) and high temperature PEMFCs (HTPEMFCs) according to their respective operating temperatures.3 LTPEMFCs operate at temperatures below 100 oC using an electrolyte membrane with sulfonic acid as the functional group. On the other hand, HTPEMFCs mainly operate at temperatures above 120 oC using an acid-doped polybenzimidazole (PBI) electrolyte membrane. There are several advantages for the *Corresponding Authors. E-mails: [email protected] or [email protected] The Polymer Society of Korea

1214

Effect of Membrane Electrode Assembly Fabrication Method on the Single Cell Performances of PBI Based HTPEMFCs

ganic compounds in PBIs was also examined to improve proton conductivity.13 Research has been carried out in our laboratory to characterize the influential factors associated with the performance of HTPEMFCs. Studies on the influence of casting conditions, polymer structure, and changes in electrode manufacturing conditions on the performance of PBI electrolyte membranes were carried out.14-16 Recently, the method of manufacturing the membrane electrode assembly (MEA) was modified to improve the long-term stability of PBI electrolyte membranes. This MEA was used to review the applicability of the HTPEMFC stack.17 Despite its numerous advantages, the HTPEMFC is still underutilized because of its insufficient performance compared to existing LTPEMFCs. Though improvement in the performance of HTPEMFCs requires improved proton conductivity of the electrolyte membrane, optimization of the electrode composition and electrode structure must first be accomplished. Also, it is necessary to reduce the amount of the expensive electrode catalyst, Pt, in order to reduce cost of the fuel cells. The number of reaction-active sites in the electrode must also be increased for improved performance. These issues have not been investigated for HTPEMFCs as extensively as in the case of LTPEMFCs. As in LTPEMFC, the electrode used in HTPEMFC is composed of Pt catalyst and binder. Various materials are currently being used as binders for HTPEMFCs.18,19 Among them, polytetrafluoroethylene (PTFE) are most commonly used.20-24 HTPEMFC electrodes composed of Pt/C and PTFE have the ability to transfer protons by receiving phosphoric acid from the acid-doped PBI electrolyte membrane. PTFE does not have proton conductivity. Therefore, PTFE acts as the resistance when it is used excessively. On the contrary, insufficient amount of PTFE in the electrodes results in poor performance due to rapid reduction in the ability to bind catalyst. Therefore, in order to obtain high performance of HTPEMFC, it is important to adjust the amount of PTFE in the electrode. Also, MEA fabrication method has to be established for the membrane and electrodes. In this study, the electrochemical characteristics of MEA with varying amounts of PTFE in the electrode were analyzed. In particular, the performance of HTPEMFCs when employing a pressing process during MEA fabrication was evaluated to determine the correlation between the electrochemical characteristics of the MEA and the performance of the fuel cell.

Experimental Chemicals. 3,3′-Diaminobenzidine (DAB), terephthalic acid (TPA), and polyphosphoric acid (PPA) (115% phosphoric acid equivalent) were purchased from Aldrich and used without purification. 45.9 wt% Pt/C catalyst (Tanaka, Japan), carbon cloth (with a microporous layer, WIS1005 from CeTech Co., Ltd), and polytetrafluoroethylene (60 wt% dispersion in H2O, Macromol. Res., Vol. 22, No. 11, 2014

Aldrich chemical co.,) were used to fabricate the MEA. Synthesis of Poly[2,2′( p-phenylene)-5,5′-bibenzimidazole] (p-PBI) and Membrane Fabrication. p-PBI was synthesized as described in our previous paper.13 DAB (3.0 g, 0.014 mol), TPA (2.3 g, 0.014 mol), and PPA (125 g) were mixed in a round-bottom flask and stirred for 15 h at 150 oC under argon atmosphere. The reaction temperature was then increased to 220 oC, and the mixture was maintained for 5 h. Upon completion of the reaction, the polymerized solution was applied to a glass plate, and the electrolyte membrane was prepared with uniform thickness using a Doctor blade. The electrolyte membrane was then placed in a temperature and humidity chamber set at 50 oC and 50% humidity for one day to manufacture the acid-doped PBI for MEA fabrication. Characterization of PBI Membrane. An acid-doped PBI membrane (2 cm×2 cm) was immersed in water (15 mL) for 3 days to wash away phosphoric acid. The phosphoric acid in the water was titrated with 0.1 M sodium hydroxide. The de-doped PBI membrane was dried under vacuum at 80 oC for 1 day to obtain the weight of the dry membrane. The H3PO4 doping level was defined as the mole number of phosphoric acid per mole of PBI repeat unit with the following equation: VNaOH  CNaOH Doping level = ---------------------------------------Wdry membrance /MPBI

(1)

VNaOH and CNaOH are the volume and concentration of sodium hydroxide solution to neutralize the phosphoric acid, Wdry membrane is the dry weight of the PBI membrane, MPBI is the molecular weight of p-PBI repeat unit. Inherent viscosity was measured at 30 oC using Ubbelode viscometer and concentration of the solution was 0.2 g/dL in 96% sulfuric acid. Fabrication of MEA. Catalyst slurry for anode and cathode was prepared according to the following procedures. In case of MEA which has 20 wt% PTFE in the electrodes, catalyst slurry was prepared by mixing Pt/C catalyst powder (1.0 g), PTFE (0.42 g (60 wt% dispersion in H2O)), isopropyl alcohol (44.1 g), and water (11.0 g). The MEAs with 30 and 45 wt% PTFE in the electrodes were also prepared as the same method by varying the amounts of PTFE. For the gas diffusion medium, carbon cloth, which contains 10 wt% PTFE in the gas diffusion layer and 30 wt% PTFE in the microporous layer, was used. The catalyst slurry was uniformly sprayed onto the microporous layer. The Pt catalyst loading was fixed at 1 mg cm-2 in the anode and cathode, respectively. The manufactured electrode was thermally treated for 5 min at 350 oC at argon atmosphere. The acid-doped p-PBI electrolyte membrane (synthesized as described in Synthesis of p-PBI and Membrane Fabrication Section) was placed between two electrodes to prepare the MEA. For the preparing the MEA, pressing was effected for 1 minute at room temperature at 400 kPa. The active area of the prepared MEAs was 10.24 cm2. Single Cell Test. A single cell was assembled with a homemade MEA, gasket, serpentine graphite plates, and aluminum 1215

J. Lee et al.

end plates, as described in a previous study.16 After installation of the single cell at a PEMFC test station, non-humidified hydrogen (100 mL/min) and air (300 mL/min) were supplied to the anode and cathode, respectively. The cell temperature was maintained at 150 oC. Prior to electrochemical analysis, the single cell was stabilized for 50 h under a constant voltage of 0.6 V. Electrochemical Analysis. Polarization curves were obtained by measuring cell voltages at various current densities using an electric loader (ELPTEK CO., ESL-300Z). Impedance spectroscopy measurements for each MEA were performed at a DC potential of 0.85 V with an amplitude of 5 mV, over the frequency range 10 mHz to 100 kHz (IM6, ZAHNER Elektrik Inc.). The experimental impedance data were analyzed by Z-MAN software (version 2.2, WonATech). For this, the electrical characteristics of the single cells were represented by an equivalent circuit. Then, the total impedance function was generated by the software, and for each impedance data, fitting parameters of circuit elements were simultaneously optimized through complex nonlinear least squares analysis. The weighing factor was set to unity in order to employ every data point with equal impact.

Results and Discussion Using homemade acid-doped PBI membranes, MEAs were fabricated with three different PTFE concentrations in the catalyst layer (20, 30, and 45 wt%). PBI membrane was obtained by in situ method. The inherent viscosity of p-PBI membrane indicates 3.1 dL/g and acid doping level of p-PBI membrane indicates 28 mol H3PO4/PBI unit.16 Scheme I shows the polymerization reaction of DAB and TPA in the PPA to produce p-PBI.13 Each MEA was named according to its respective PTFE content: F20, F30, and F45. In addition, the prepared MEAs were pressed at room temperature to evaluate the effect of phosphoric acid supply from the membranes to the catalyst layers. The corresponding pressed MEAs were named F20p, F30p, and F45p. Figure 1 shows the polarization curves of the acid-doped MEAs with various PTFE concentrations in the catalyst layers. The open-circuit voltages (OCV) were above 0.97 V for the

Scheme I. Synthesis of PBI. 1216

Figure 1. I-V characteristics of non-pressed MEAs with different PTFE contents in the electrode.

three tested MEAs, which indicates that gas crossover through the polymer electrolyte membranes or gaskets in single cell fabrication is not significant. As the current density increased, the voltage drop was more severe for the MEA with a higher PTFE content. As a result, increased cell performances were observed when the PTFE content was lower. For example, the following was observed in the voltage at 0.4 A/cm2: 0.58 V (F20) > 0.55 V (F30) > 0.50 V (F45). At higher concentrations of PTFE, which is highly hydrophobic, the phosphoric acid content in the catalyst layers is expected to be reduced. As proton conductivity in the catalyst layers occurs via the phosphoric acid molecules in the pores, a larger PTFE content will induce a decrease in proton conductivity in the catalyst layers. In such cases, some areas of the Pt catalyst in the cathode layers cannot be accessed by the protons transferred through the PEMs, which results in lower catalytic activity. Even though the effect of PTFE addition will be similar for cathodes and anodes, the influence on cell performance will mainly depend on the overpotential variation in the cathodes, since the cathodic overpotential is more dominant in hydrogen-fueled PEMFCs.19 In Figure 2, Nyquist plots of the electrochemical impedance data are presented for MEAs with varying PTFE concentrations. In the case of F45, a sloped line is clearly observed at high frequencies, along with the typical semicircular shape at lower frequencies. As the PTFE content decreased, the sloped line in the high-frequency region became less pronounced and the diameter of the semicircle gradually decreased. As a result, the total resistance, determined from the x-axis intercept at low frequencies, gradually increased with the PTFE content: 11.4 Ω cm2 (F20) < 13.2 Ω cm2 (F30) < 16.1 Ω cm2 (F45). This characteristic feature in the impedance spectra seems to be the result of the different phosphoric acid concentrations, based on previous reports that the linear shape with a ca. 45o slope originated from ion transport in thick elecMacromol. Res., Vol. 22, No. 11, 2014


Figure 2. Impedance spectra of non-pressed MEAs with different PTFE contents in the electrode, at 0.85 V.

Figure 3. Schematic and equivalent circuit for the cathode layers in HTPEMFCs.

trodes.25 When the impedance data of electrochemical systems are analyzed using equivalent circuits, the charge transfer process and double-layer charging process at electrode/electrolyte interface are generally described by a resistance and a capacitance, respectively. However, in the porous electrodes of HTPEMFCs, the interfacial sites are highly distributed26,27 and their electrochemical characteristics are not equivalent due to the ionic resistance in electrodes. Therefore, in this study, the electrochemical characteristics of cathodes were represented by five sets of R1Q1 parallel combinations that are connected through R2 (Figure 3). Here, the R1, Q1, and R2 are the equivalent elements that correspond to the electrochemical processes of charge transfer, double-layer charging, and proton conduction, respectively. Also, an inductance (Ls) and a resistance (Rohm) were serially connected to describe the total conductance and total ohmic resistance of the HTPEMFC single cells, respectively. By using the equivalent circuit, the impedance data of HTPEMFC MEAs could be well fitted, as shown by the simulated lines with the fitted parameters (Figure 2). From the Macromol. Res., Vol. 22, No. 11, 2014

Figure 4. (a) Ion resistance of non-pressed and pressed MEAs with different PTFE contents in the electrode; (b) Charge transfer resistance of non-pressed and pressed MEAs with different PTFE contents in the electrode.

fitted parameters, the proton transfer resistance (RH+=R2×4) and total charge transfer resistance (Rct=R1/5) were calculated and plotted as a function of PTFE content (Figure 4). The observed increase in the proton transfer resistance with the PTFE content suggests a decreased amount of phosphoric acid. An increase in the charge transfer resistance, which is inversely related to the number of available Pt active sites, could be also confirmed at higher PTFE contents in the catalyst layers. The MEAs with varying PTFE concentrations were pressed at ambient temperature for 1 min to supply more phosphoric acid to the catalyst layers. Figure 5 shows the I-V curves of the pressed MEAs (F20p, F30p, and F45p). In this case, the voltage decrease with an increase in current to 0.1 A/cm2 was more severe at low PTFE contents in the catalyst layers. Therefore, the best performance among the pressed MEAs was achieved with F45p MEA, while a lower PTFE content gave the best results for the non-pressed MEAs. Since the phosphoric acid content in the catalyst layer plays an important 1217

J. Lee et al.

Figure 5. I-V characteristics of pressed MEAs with different PTFE contents in the electrode.

Figure 6. Impedance spectra of pressed MEAs with different PTFE contents in the electrode, at 0.85 V.

role in determining the cell performance, as confirmed for the non-pressed MEAs, it seems that the amount of phosphoric acid added to the catalyst layers by pressing the MEAs is influenced by the PTFE content. Figure 6 shows Nyquist plots for the impedance data of the pressed MEAs. Compared to the case of the non-pressed MEAs, the sloped region at high frequencies was significantly diminished, reflecting that proton conductivity in the catalyst layers was enhanced by the additional phosphoric acid present as a result of MEA pressing.28 However, the sloped line was most clearly observed for F45p among the three pressed MEAs. Using the same equivalent circuit as that for the non-pressed MEAs (Figure 4), all the impedance data were analyzed by non-linear fitting; the resulting fitted parameters are presented in Figure 4 for comparison against those of the non-pressed MEAs. As shown by the Nyquist plots, proton resistance was largely decreased by pressing, while the values were higher with increasing PTFE content. 1218

Figure 7. (a) Current density at 0.6 V of non-pressed and pressed MEAs with different PTFE contents in the electrode; (b) Current density at 0.6 V for non-pressed and pressed MEAs with ion resistance.

Interestingly, for the pressed MEAs, the charge transfer resistance at 0.85 V gradually decreased with PTFE content, which is the reverse trend as compared to that for the non-pressed MEAs. Figure 7(a) summarizes the single cell performances of non-pressed and pressed MEAs. Please be noted that the PEMFCs are typically evaluated at the cell voltage of 0.6 V, considering the usual operating voltage of practical systems. With higher PTFE content, the cell performance gradually decreased in the case of the non-pressed MEAs, whereas the pressed MEAs showed the opposite behavior. As identical amount of Pt/C catalysts were utilized, the amount of phosphoric acid in catalyst layers is expected to play an important role in determining cell performances. In Figure 7(b), the current density at 0.6 V was plotted as a function of proton transfer resistance (RH+), which is inversely proportional to the phosphoric acid amount. It shows a peakshaped curve with a maximum at the RH+ of ca. 0.4 Ω cm2. To act as active sites in PEMFC cathodes, the Pt catalyst surface should be accessed by both protons and oxygen gas through Macromol. Res., Vol. 22, No. 11, 2014


the networks of phosphoric acid and pores, respectively. In the case of HTPEMFCs, the larger amount of phosphoric acid will promote the proton transport; however, at the same time, oxygen transport will be more hindered as pore volume is decreased. Therefore, maximum Pt utilization is expected when the proton and oxygen transport is well balanced with adequate amount of phosphoric acid in the catalyst layer, which explains the reverse trend in the cell performance vs. PTFE amount (Figure 7(a)). In the case of non-pressed MEAs, the proton conduction limits the amount of active sites and cell performances, and therefore the cell performances gradually decreased with higher PTFE amount (lower phosphoric acid amount). In contrast, when the catalysts layers contain more phosphoric acid after pressing, the cell performance will be more dependent on the oxygen transport characteristics, instead of the proton conduction, which is in accordance with the gradual performance increase with higher PTFE amount (lower phosphoric acid amount). The complex effect of phosphoric acid amount also can be noticed by observing the pressing effect. The current density of the MEA with 20 wt% PTFE decreased significantly as a result of pressing, whereas, when the MEA with 45 wt% PTFE was pressed, the cell performance increased. As shown in Figure 4(a), the RH+ significantly decreased for both MEAs: 0.033 Ω cm2 (F20) to 0.387 Ω cm2 (F20p) and 0.627 Ω cm2 (F45) to 2.144 Ω cm2 (F45p), indicating that, for MEAs with different PTFE content, the phosphoric acid amount in cathodes was increased by pressing. If an MEA has a large amount of phosphoric acid in the catalyst layers (F20), protons can be supplied to most of the Pt catalyst, but oxygen transport might be limited as a large proportion of the pores become unavailable for gas supply. Therefore, the overall catalytic activity will be determined by the accessibility of the gas reactant. Accordingly, after pressing, the F20p showed lower cell performance compared to the F20, due to the reduced Pt utilization by severe blocking of oxygen transport with increased amount of phosphoric acid. In contrast, when an MEA contains low quantities of phosphoric acid (F45), utilization of the Pt catalyst will be mainly determined by the proton supply, whereas gas transport is relatively facile through the well-developed pore networks. In such a case, when the MEA is pressed (F45p), the Pt utilization will be increased with enhanced proton conduction by larger amount of phosphoric acid.

Conclusions The effect of the amount of PTFE in the catalyst layers and the influence of utilizing a pressing process were investigated using electrochemical techniques. For non-pressed MEAs, the cell performance gradually decreased as the PTFE content increased in the catalyst layers. Based on polarization and impedance data, it could be concluded that the amount of phosphoric acid was reduced at higher PTFE contents, resulting in larger proton transfer resistance and lower Pt Macromol. Res., Vol. 22, No. 11, 2014

utilization. In contrast, when the MEAs were pressed and more phosphoric acid was supplied from the membranes to the catalyst layers, optimal cell performance was achieved at the highest PTFE concentration. For pressed MEAs that contained a large amount of phosphoric acid, oxygen transport was a more significant factor than proton conduction in influencing the resultant catalytic activities and cell performance. Therefore, it was experimentally confirmed that the cell performance significantly depends on the amount of phosphoric acid, which was controlled by the PTFE content and pressing, and that balance between hydrogen and oxygen transport is required for high Pt utilization. Acknowledgments. This work was financially supported by the projects “KIST Institutional Program” and “COE” from the Korea Institute of Science and Technology. This work was partially supported by the New & Renewable Energy Technology Development Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Knowledge Economy (No. 2009T100200046).

References (1) S. J. Peighambardoust, S. Rowshanzamir, and M. Amjadi, Int. J. Hydrogen Energy, 35, 9349 (2010). (2) T. Freya and M. Linardi, Electrochim. Acta, 50, 99 (2004). (3) S. Bose, T. Kuilaa, T. X. H. Nguyenb, N. H. Kim, K.-T. Lau, and J. H. Lee, Prog. Polym. Sci., 36, 813 (2011). (4) Q. Li, J. O. Jensen, R. F. Savinell, and N. J. Bjerruma, Prog. Polym. Sci., 34, 449 (2009). (5) C.-P. Wang, H.-S. Chua, Y.-Y. Yan, and K.-L. Hsueh, J. Power Sources, 170, 235 (2007). (6) C. Pan, R. He, Q. Li, J. O. Jensen, N. J. Bjerrum, H. A. Hjulmand, and A. B. Jensen, J. Power Sources, 145, 392 (2005). (7) A. Chandan, M. Hattenberger, A. El-kharouf, S. Du, A. Dhir, V. Self, B. G. Pollet, A. Ingrama, and W. Bujalski, J. Power Sources, 231, 264 (2013). (8) J. S. Wainright, J. T. Wang, D. Weng, R. F. Savinell, and M. Litt, Electrochem. Soc., 142, L121 (1995). (9) L. Xiao, H. Zhang, E. Scanlon, L. S. Ramanathan, E.-W. Choe, D. Rogers, T. Apple, and B. C. Benicewicz, Chem. Mater., 17, 5328 (2005). (10) S. Yu and B. C. Benicewicz, Macromolecules, 42, 8640 (2009). (11) P. Mustarelli, E. Quartarone, S. Grandi, S. Angioni, and A. Magistris, Solid State Ionics, 225, 228 (2012). (12) M. Hazarika and T. Jana, ACS Appl. Mater. Interfaces, 4, 5256 (2012). (13) P. Staiti, M. Minutoli, and S. Hocevar, J. Power Sources, 90, 231 (2000). (14) J.-H. Kim, H.-J. Kim, T.-H. Lim, and H.-I. Lee, J. Power Sources, 170, 275 (2007). (15) J. W. Lee, D. Y. Lee, H.-J. Kim, S. Y. Nam, J. J. Choi, J.-Y. Kim, J. H. Jang, E. A. Cho, S.-K. Kim, S.-A. Hong, and T.-H. Lim, J. Membr. Sci., 357, 130 (2010). (16) H.-J. Lee, D. H. Lee, D. Henkensmeier, J. H. Jang, E. A. Cho, 1219

J. Lee et al.

(17)

(18)

(19) (20) (21) (22)

1220

H.-J. Kim, and H. Kim, Bull. Korean Chem. Soc., 33, 103279 (2012). H.-J. Lee, B. G. Kim, D. H. Lee, S. J. Park, Y. Kim, J. W. Lee, D. Henkensmeier, S. W. Nam, H.-J. Kim, H. Kim, and J.-Y. Kim, Int. J. Hydrogen Energy, 36, 5521 (2011). H.-J Kim, S. J. An, J.-Y. Kim, J. K. Moon, S. Y. Cho, Y. C. Eun, H.-K. Yoon, Y. Park, H.-J. Kweon, and E.-M. Shin, Macromol. Rapid Commun., 25, 1410 (2004). D. Ergun, Y. Devrim, N. Bac, and I. Eroglu, J. Appl. Polym. Sci., 124, 267 (2012). C. Wannek, W. Lehnert, and J. Mergel, J. Power Sources, 192, 258 (2009). M. Mamlouk and K. Scott, Int. J. Energy Res., 35, 507 (2011). H. Su, S. Pasupathi, B. Bladergroen, V. Linkov, and B. G. Pol-

let, J. Power Sources, 242, 510 (2013). (23) J. O. Park, S. G. Hong, T. Kim, K. Kwon, S. Suh, M. Cho, and D. Yoo, ECS Trans., 3, 447 (2006). (24) C. Pan, Q. Li, J. O. Jensen, R. Heb, L. N. Cleemann, M. S. Nilsson, N. J. Bjerrum, and Q. Zenga, J. Power Sources, 172, 278 (2007). (25) R. Makharia, M. F. Mathias, and D. R. Bakerb, J. Electrochem. Soc., 152, A970 (2005). (26) P. M. Gomadam and J. W. Weidnern, Int. J. Energy Res., 29, 1133 (2005). (27) A. Lasia, Modern Aspects of Electrochemistry, 32, 143 (2002). (28) J. Zhang, Z. Xie, J. Zhang, Y. Tang, C. Song, T. Navessin, Z. Shi, D. Song, H. Wang, D.-P. Wilkinson, Z.-S. Liu, and S. Holdcroft, J. Power Sources, 160, 872 (2006).

Macromol. Res., Vol. 22, No. 11, 2014