Journal of Membrane Science 417418 (2012) 210216

Contents lists available at SciVerse ScienceDirect

Journal of Membrane Science

journal homepage: www.elsevier.com/locate/memsci

Effect of pressure on through-plane proton conductivity of polymer electrolyte membranes

Sung-Hyun Yun a, Sung-Hee Shin a, Ju-Young Lee a, Seok-Jun Seo a, Se-Hun Oh a, Young-Woo Choi b, Seung-Hyeon Moon a,n
a b

School of Environmental Science and Engineering, Gwangju Institute of Science and Technology (GIST), 123 Cheomdan-gwagiro, Buk-gu, Gwangju 500-712, Republic of Korea Fuel Cell Research Center, Korea Institute of Energy Research (KIER), PO Box 103, Yusong, Daejeon 305-600, Republic of Korea

a r t i c l e i n f o
Article history: Received 4 March 2012 Received in revised form 20 June 2012 Accepted 22 June 2012 Available online 3 July 2012 Keywords: Polymer electrolyte membranes Electrochemical impedance spectroscopy Through-plane conductivity Anisotropic conduction Contact resistance

a b s t r a c t
This study reports an electrochemical analogy with the invention of a 2-electrode impedance cell oriented to analyze through-plane ionic conductivity of polymer electrolyte membranes. The membranes are utilized under varied pressures in assembled membrane stacks for water treatment, energy conversion, and storage devices. In addition, polymeric membranes are a viscoelastic material, which allow their shape to change according to the external pressure. Therefore, the impedance under controlled pressure is crucial information for analyzing their electrochemical properties. In this regards, an apparatus is designed to indicate the absolute pressure in the sample and the resulting thickness of the sample while measuring the impedance. As a model analysis, we employ a proton conducting membrane, Naons117, Naons115, and Naons112. The membrane thickness in a wet state was found to be signicantly variable with the pressure, and the impedance spectra showed a clear dependence on the pressure. Therefore, this new approach facilitates a precise impedance analysis, since the exact sample thickness is indicated, along with the absolute pressure in the sample. As such, this technique could be a useful tool for analyzing the through-plane conductivity of polymer electrolyte membranes with high accuracy. & 2012 Elsevier B.V. All rights reserved.

1. Introduction Polymer electrolyte membranes (PEMs) have been used in a diverse range of electrochemical water treatment and energy conversion-storage systems, including capacitive deionization [1], electrodialysis [2], reverse electrodialysis [3], fuel cells [4], secondary batteries [5], redox ow batteries [6], dye sensitized solar cells [7], and others. In such electrochemical systems, the general requirements of these membranes, such as ion-exchange capacity, water contents, swelling behaviors, mechanical properties, and thermal and chemical durability should be satised according to electrochemical specications in the target system. In addition, the ionic conductivity of a PEM has been considered a primary parameter for estimating system performance, as PEM resistivity is the predominant reason for the potential drop in particular for a fuel cell [8,9]. To estimate proton conductivity under specic conditions, a number of galvanostatic alternating current (ac) electrochemical impedance spectroscopic (EIS) techniques have been well documented, based on equivalent circuits corresponding to the electrochemical impedance cells [1023].

Corresponding author. Tel.: 82 62 715 2435; fax: 82 62 715 2434. E-mail address: shmoon@gist.ac.kr (S.-H. Moon).

To date, conventional 4-electrode cells have been universally used and analyzed for in-plane proton conductivity, due to advantageous features such as the geometrically large cell constant, negligible electronic resistance, and the interfacial contact between the membrane surface and thin platinum wires [12,13]. In addition, recent efforts have evolved analytical approaches for through-plane conductivity due to the increasing concerns of anisotropic proton conduction in the membranes [14,20,24]. It has also been posited that measuring proton conductivity in the through-plane direction may enable the precise estimation of PEM fuel cell performance, since the practical proton conduction occurs in the through-plane direction of a PEM. Indeed, several electrode arrangements have been reported for through-plane conductivity analyses [12,14,20,21,23] however, systematic investigations of through-plane impedance data under the inuence of controlled pressure on a polymeric membrane and its viscoelastic behavior under pressure are sparse, and have yet to be strongly emphasized with the exploration of anisotropic proton conducting materials. In an effort to precisely obtain the through-plane resistance, we report here the invention and application of a technically improved 2-electrode electrochemical cell [25] and a novel analogy of impedance data for the through-plane conductivity of Naons membranes. This cell is designed to indicate the

0376-7388/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.memsci.2012.06.041

S.-H. Yun et al. / Journal of Membrane Science 417418 (2012) 210216

211

absolute pressure applied to the sample as well as the corresponding thickness of the sample during the through-plane EIS analysis. Within the best of our knowledge, previous works have neglected to include the thickness changes of a PEM under pressure, though this change may signicantly affect the hydration rate (or water content) and proton conductivity of a suppressed PEM.

designed to indicate the membrane thickness directly by the distance between the two electrodes. The dial indicator was mechanically set-up to monitor the vertical position of electrode 1, and thus could readily monitor the distance of electrode 1 from the xed electrode 2. The details on the technical reliability were discussed in the operational data. 2.3. Equivalent circuit model for through-plane EIS measurement

2. Experimental 2.1. Materials Two disc type stainless steel electrodes were fabricated, each having an effective area of 1 cm2. A dial indicator (No. 2113S-10, Mitutoyo Manufacturing Co., Ltd., Japan) was used for the thickness measurement. A load sensor (SML-30L, Curiosity Technology Co., Ltd., Korea) having a maximum 66 pound-force capacity was calibrated (and communicates) with a digital indicator (MI-830, MiGun ST Co., LTD, Korea) as the pressure sensor. The polymer electrolyte membranes (Naons117, Naons115, and Naons112, DuPont, USA) were prepared after being treated with a 1 mol/l HCl solution for 5 h. Then, in order to examine the practical compression pressure for a membrane in a fuel cell and the apparatus system, two types of pressure sensitive lm (Pressurexs, SPI Inc., USA) Superlow for the range of 70 350 psi and Low for 3501400 psi were assembled instead of using a traditional membrane. A video microscope (SV-32, Sometech Co., Korea) was temporarily used to visually observe the thickness change of the membrane under specic pressures, and the photographed thickness was obtained using image tracer software (IT-Pro3.0, Sometech Co., Korea). 2.2. Cell conguration and instrumentation Two disc electrodes were placed horizontally (see Fig. 1) for a through-plane EIS analysis of the polymeric membranes. As illustrated, electrode 2 on the bottom was xed to the apparatus body (made by aluminum pillar), whereas electrode 1 was designed to allow only vertical motion, induced by the externally applied vertical compression force. Therefore, the electrode 1 provides a certain compression pressure to the target sample placed on electrode 2, and the apparatus was mechanically designed to exert the force to the load sensor through a static bypass aluminum frame. Finally, the absolute pressure was indicated by the calibrated digital indicator. Concurrently, the membrane thickness was monitored via the dial gauge. The device is

Two critical advances of this 2-electrode cell are the ability to exactly control the pressure applied to a PEM and to monitor the reduced thickness of a viscoelastic PEM under a stationary pressure. The pressure prole of a PEM obtained by the sandwiched electrodes signicantly affects the impedance data, particularly due to variation of the physical interfacial contact resistance (Rinter) between the electrode and PEM surfaces [20]. Then, at the interface of the PEM surfaces, an electrochemical double layer accompanies the polarization resistance (Rp) in parallel with a constant phase element (CPEdl) within the outer Helmholtz plane (OHP) [19], and Warburg impedance (W) is exhibited by a diffusion boundary layer (DBL) at the PEM surfaces. These electrochemical phenomena are described in Fig. 2(a), in which the electronic components include the electronic cell resistance (Relectro) and membrane resistance (Rmembr). Among the components, only Relectro is independent to the pressure load; hence, the equivalent circuit can be modelled at a pressure i (Pi), as shown in Fig. 2(b). 2.4. Through-plane proton conductivity via an electrochemical impedance measurement Each swollen Naons membrane in deionized water was assembled between the two electrodes. The EIS analyses were carried out using a potentiostat/galvanostat (Autolab PGSTAT 30, Eco Chemie, Netherlands), and operated via a galvanostatic method under a 0.1 mA ac amplitude over a frequency range of 1 MHz to 10 Hz by using a frequency response analyzer. The EIS analysis was carried out under ambient conditions at room temperature and there was no additional humidication. Also, every membrane was analyzed at the same time and inside of a copper shield net in order to minimize the variation of temperature and humidity, and the signal interruption of high frequency from other electrical equipments, respectively. As

Fig. 1. Schematic of electrode structure for the 2-electrode cell used for the through-plane EIS analysis. A target sample is compressed by the external pressure from P0 to Pi, and the thickness of a viscoelastic PEM is reduced from t Po to t Pi .

Fig. 2. (a) Impedance components based on electrochemical behaviors between the electrode and a swollen PEM. (b) Simplied equivalent circuit model as a function of pressure on a PEM.

212

S.-H. Yun et al. / Journal of Membrane Science 417418 (2012) 210216

noted above, the impedance behaviors were measured by increasing the absolute pressure from the initial pressure of zero, and corresponding thickness changes were observed related to the pressure. Based on the equivalent circuit model in Fig. 2(b), the experimental responses were analyzed by tting the Nyquist plot and the electric components such as Rhf, Rp, and CPEdl were analyzed from imperfect semicircle simulation [8]. Note that the high frequency resistance (Rhf) be obtained from the Nyquist plot at the phase angle zero includes the three serially combined resistances Rmembr, Rinter, and Relectro, as described by Eq. (1).
i i i RP RP RP Relectro inter hf membr

Eventually, the through-plane conductivity is calculated from the Rmembr, the thickness under Pi (t Pi ), and the effective contact area (Aeffect). k i
P

t Pi
i RP membr

Aef f ect

3. Results and discussion 3.1. Practical operating pressure of a PEM in a fuel cell In addition to the anisotropic conduction nature of polymer electrolyte membranes, interest pertaining to the through-plane resistance and conductivity has recently grown, since ionic conduction occurs perpendicular to the membrane in actual systems. Accordingly it has become important to precisely estimate the performance via an ex-situ analysis of conductivity under representative conditions. In this regard, it would be quite valuable to reproduce the system pressure in the ex-situ analytical apparatus, since the membranes suffer variation of the pressure during the stack assembly. To determine the approximate pressure that compresses the membrane in a practical system, we chose a single cell that is typically used in fuel cell applications. The pressure prole by clamping the cell could then be estimated by using a pressure sensitive lm. In this study, a lm was installed instead of a membrane, as schematically described in Fig. 3(a) for a pressure range of 70 350 psi and in Fig. 3(b) for 3501400 psi, and it was exposed for 10 min under a relative humidity of 2025% at room temperature. In order to reproduce the pressure on the membrane, 5 cm2 of catalyst- coated electrodes were assembled with a silicon gasket, as previously reported [4], and the target effective area of the cell was then analyzed. The resultant lms reveal the pressure distribution (Fig. 3(a2) and (b2)) within the effective area and further digital analysis show well-dened pressure proles, based on the colored pressure density bar shown in Fig. 3(a3) and (b3). The effective minimum pressure of Fig. 3(a3) was about 100 psi, and the exceeded pressure domains over 350 psi were analyzed in Fig. 3(b3), which shows the effective maximum pressure of around 600 psi. Therefore, it was conrmed that a membrane in this system can be tested under an approximate pressure range of 100600 psi.

Fig. 3. Estimation of an approximate pressure prole obtained from a practical fuel cell design: (a1) and (b1) present the schematic of the method; (a2) and (b2) show the resultant pressure sensitive data; and (a3) and (b3) illustrate the subsequent digital analysis, with corresponding colored pressure density bars. Once the 2-electrode apparatus was designed to facilitate a uniformly distributed pressure control, maintenance of the pressure is essential for performing a precise ac EIS analysis, due to the signicant effects of compression pressure on the impedance data [20]. In this regard, the absolute pressure on the sample and corresponding sample thickness were observed in real-time by using a pressure indicator and thickness gauge during the analysis. Fig. 5 shows the stability of pressure loading and thickness observations for 6 h. This test was repeated twice and conrmed that the apparatus technically facilitates a very stable pressure loading to the sample and accurately reads the sample thickness.

3.3. Thickness changes of a PEM by the pressure load It is well recognized that the compression pressure can impart a signicant variation on the impedance data; however, previous analyses did not consider the viscoelastic behavior of ionic conducting polymers. Since a polymer electrolyte membrane is a relatively soft and viscoelastic material, the thickness is variable with the external pressure applied by the electrodes. Therefore, the ability to simultaneously measure the absolute pressure and corresponding absolute dimensions of the membrane is essential information for a precise analysis of the electrochemical properties of polymer electrolyte membranes. In this apparatus, the membrane thickness is indicated by the distance between the electrodes, and the thickness was monitored using a dial gauge. Further, the thickness changes of Naons117 were photographed by a video microscope, and compared with the thickness indicated by the dial indicator as shown in Fig. 6. The effective contact area of the electrode was 0.4 cm2, which increased the absolute pressure on the membrane for this experiment only. The initial membrane thickness was 204 mm at zero pressure, and was compressed to 160.5 mm at 781 psi; i.e., the absolute compression pressure reduced the thickness by 21%.

3.2. Pressure load to a PEM of the 2-electrode cell Fig. 4(a) shows the system conguration equipped with a thickness indicator, a load sensor, and a pressure indicator. A membrane sample was placed between the electrodes and pressure was loaded to the sample as shown in Fig. 4(b). In order to control the pressure to the membrane imparted by electrode 1, as described in Fig. 1, an aluminum frame was designed to apply a regular force on the membrane sample and load sensor. As shown in Fig. 4(c), the frame structure was mechanically designed to deliver an external force F to electrode 1, and F was then directly used to compress the sample placed on electrode 2, which was separately xed to the frame without spinning of the electrode to avoid damage on the membrane. A Teon body was used to hold electrode 1, which maintains its position by using a stopper on the frame. Since there is no vertical motion of the Teon body, it is considered a single body with an aluminum frame. Therefore, the pressure distribution could be analyzed based on a static state at the starting point (zero pressure at the initial sample thickness). Eventually, the target force F was again delivered via the in-plane mechanical pillars to the load sensor based on the assumption of no sheer stress. Here, the effective contact area of the electrode was 1 cm2 and the absolute force was mechanically loaded and indicated; as such, the maximum capacity of the pressure sensor was calculated as 426 psi. Additionally, the pressure distribution between electrodes 1 and 2 was examined (Fig. 4(d)) and a uniformly distributed pressure prole was obtained after a mechanical adjustment, implying that the apparatus is suitable for reproducing the system pressure and applying a uniform pressure to the sample membrane.

3.4. Through-plane EIS measurements of Naon membranes As a reference study, Naons membranes such as Naon117, Naon115, and Naon112 were analyzed. Swollen membranes 1 cm2 in size (same as electrode area) were employed in the 2-electrode cell, and Nyquist plots were obtained at room temperature (2025 1C). Fig. 7 shows the Nyquist plots obtained while increasing compression pressure from 0 psi to 426 psi, and the corresponding thicknesses of the suppressed membranes. After obtaining the experimental Nyquist plots, a simulation of the circuit model was conducted in order to conrm that the proposed equivalent circuit is valid or not. As a result, it was conrmed

S.-H. Yun et al. / Journal of Membrane Science 417418 (2012) 210216

213

Fig. 4. (a) Photographs show system conguration of the 2-electrode apparatus and (b) detailed sample membrane assembling between the electrodes. (c) A schematic describes a static force distribution frame of the apparatus used to bypass the absolute force F from electrode 1 to the load sensor, and (d) uniform pressure distribution was obtained after a mechanical adjustment.

Fig. 5. Technical stability during a long-term test 1 () and test 2 (J) showing the stable pressure loading and thickness monitoring of the apparatus. that the proposed circuit was suitable to model the experimental responses as shown in Fig. 7(d). Each electric component was obtained after matching these plots based on the equivalent circuit model (Fig. 2(b)), and the predicted Rhf, Rp, and CPEdl are summarized in Table 1. Specically, Rhf was the required target impedance, though the other impedances accompanying the electrochemical double layer are also important for verifying the suggested equivalent circuit. In

Fig. 6. Validity test for the indicated thickness of the apparatus by additionally observing the membrane thickness relative to the step-wisely increased pressure. particular, for the interfaces between an electrode and a solid electrolyte, electrochemical polarization takes place at the surface of the membranes within the electrical double layer [19] and varies on the effective surface area of the solid electrolyte beyond the materials polarity. As previously described in Fig. 2(a), because the physical surfaces of the electrode and the membranes are not

214

S.-H. Yun et al. / Journal of Membrane Science 417418 (2012) 210216

Fig. 7. Nyquist plots measured at room temperature by the 2-electrode cell from 0 psi to 426 psi with the monitored thicknesses of (a) Naon117; (b) Naon115; (c) Naon112, respectively. (d) Simulated experimental responses corresponding to the model circuit (input parameters were Rhf 50 O, Rp 150 O, Waburg parameter (Y0 0.1 10 3 O s0.5), and CPEdl parameters (Q0 0.1 10 5 O s n) for n 1 and n 0.8, respectively). Table 1 Summary of the predicted electric components based on the equivalent circuit for the Nyquist plots of Naon117, Naon115, and Naon112, respectively. Pi [Psi] ([kgf/cm2]) Naon117 Thickness [mm] 0 71 142 213 284 355 426
a

Naon115 Rp CPEdla Rhf 7 F] [O cm2] [O cm2] [ 10 7.064 4.527 4.479 4.315 4.306 4.165 3.700 21.549 25.769 21.853 15.697 14.988 8.249 13.620 9.223 40.763 18.338 4.424 3.740 0.099 3.670 n Thickness [mm] 143.0 134.2 127.8 121.0 114.8 109.2 103.6 Rp CPEdla Rhf 7 F] [O cm2] [O cm2] [ 10 6.431 4.332 4.246 4.230 3.918 3.889 3.565 43.462 36.940 28.203 21.065 17.927 14.498 12.751 2.948 10.227 5.390 2.039 1.202 0.518 0.430 n

Naon112 Thickness [mm] 53.0 46.0 39.0 32.2 27.0 17.0 13.5 Rp CPEdla Rhf 7 [O cm2] [O cm2] [ 10 F] 5.710 4.261 4.055 3.664 3.361 2.915 2.854 43.039 33.525 26.587 23.149 20.952 17.015 19.362 3.807 6.184 2.743 2.118 1.857 1.215 2.493 n

(0) (5) (10) (15) (20) (25) (30)

205.0 197.0 189.4 183.0 177.0 169.5 164.8

0.861 0.760 0.839 0.956 1.005 1.260 1.011

0.901 0.841 0.907 0.999 1.036 1.108 1.133

0.870 0.862 0.938 0.975 0.993 1.045 0.984

Double layer capacitance was calculated from the CPE parameters including Qo and n. extrapolated to derive the y-intercept, denoting Rhf at innite pressure with no membrane. As expected, the y-intercept displayed a similar value, since the electrical electrode resistance is independent to the pressure. From the y-intercept in Fig. 8, Relectro was obtained (aver. 1.877 O cm2) and then used to extract Rinter. Note that the interfacial contact resistance is governed by the physical contact situation [27], so Rinter is expected to be the same for an equivalent pressure load on the same membrane. Base on this assumption, Rhf was redrawn as a function of the thickness change at each stationary pressure (71, 142, 213, 284, and 355 psi) for Naons membranes (Fig. 9). It was subsequently found that a linear relation could be employed, and another y-intercept was obtained. Here, the Rhf derived by y-intercept at each pressure indicates the sum of the constant Relectro and Rinter, because only Rmembr is zero at the zero thickness. Therefore, Rinter could be calculated based on the following analogy.
Pi i RP inter Rhf , t 0 Relectro

completely even, the effective polarization surface area differs depending on the roughness of the membrane surface. On the other hand, this area of the electrical double layer may decrease when the membrane surfaces are suppressed by the electrode, thereby reducing the polarization resistance. As expected, in this study, Rp signicantly reduced with increasing pressure, and the double layer capacitance calculated from the CPEdl decreased as the polarization resistance decreased excepting the data at 0 and 426 psi. Therefore, it could be conrmed that the predicted electrical components based on the equivalent circuit are valid for analyzing the target Rhf, and that further analyses can be based on these highfrequency resistances and thickness changes. 3.5. Analogy to obtain Relectro and Rinter from Rhf In order to derive Rmembr for the through-plane conductivity, the extraction of Relectro and Rinter from Rhf is possible under the controlled pressure and thickness [21,26]. If the compression pressure Pi is innitely increased (i-N), then the membrane thickness is innitely reduced (t P1 -0). In this situation, Rmembr and Rinter can be neglected at the zero thickness of the membrane. Therefore, we can assume that the Rhf at the innite pressure is equal to Relectro as follows.
1 Relectro RP hf , t 0

3.6. Through-plane area specic resistance and conductivity of the membranes Once Relectro and Rinter were obtained, Rmembr at a given pressure could be calculated using Eq. (1) as follows.
i i i RP RP Relectro RP inter membr hf

Based on Eq. (3), we can extrapolate Rhf with increasing pressure from 71 355 psi (excepting the data at 0 and 426 psi due to high error from the extrapolated curve) in terms of the thickness change as shown in Fig. 8. In this case, with increasing pressure, Rhf displayed a linear relationship with the decreased membrane thickness. As such, a linear curve was subsequently

Fig. 10 shows the contribution of each resistance to the through-plane Rhf of the Naons membranes. With increasing pressure, the Rinter signicantly

S.-H. Yun et al. / Journal of Membrane Science 417418 (2012) 210216

215

Fig. 8. Extrapolation of high frequency resistance with increasing pressure for extraction of electronic cell resistance (Relectro).

Fig. 9. Extrapolation of high frequency resistance at each constant pressure for extraction of interfacial contact resistance (Rinter). decreased, implying that the compression pressure reduces the contact resistance of the electrode-membrane interface. In contrast, it was found that Rmembr slightly increased with increasing pressure, partially due to the reduced water at the compressed membrane thickness. Based on Rmembr and the corresponding thickness under each pressure, the through-plane conductivity was calculated using Eq. (2), as summarized in Table 2. Compared to the in-plane conductivity of Naon membranes (0.091, 0.093, and 0.083 S/cm for Naon117, Naon115, and Naon112, respectively) measured by using a 4-electrode cell [4,8] at room temperature, the analyzed through-plane conductivity was notably low, which implies that the proton conduction of Naons membranes is anisotropic. Interestingly, the through-plane conductivity decreased with increasing pressure, which was physically attributed to the reduced free water (hydration rate) through the water channel (proton conducting channel). Moreover, the pressure prole of the through-plane conductivity may enable the precise estimation of fuel cell performance since the membranes are operated at high pressure in a fuel cell. Besides the anisotropic characteristic, it was expected that the conductivity at a constant pressure would show the same value since the Naon membranes have the same equivalent weight. Except the through-plane conductivity at 284 psi, the calculated conductivity at each pressure showed little different values, partially due to the difference of temperature and humidity during the EIS measurement. Therefore, further examination under precisely controlled temperature and humidity is recommended.

Fig. 10. Quantitative comparison of Relectro, Rinter, and Rmembr included in the total Rhf for (a) Naon117; (b) Naon115; (c) Naon112 which was obtained from the Nyquist plots by using the 2-electrode cell.

4. Conclusions In conclusion, a new electrochemical approach for throughplane proton conductivity under various pressures was

demonstrated. With the growing concerns about anisotropic proton conduction through polymer electrolyte membranes, this research may thus enable researchers to precisely analyze through-plane conductivity based on the EIS analogy. Furthermore, analysis of through-plane conductivity under various pressures was recommended for improving and characterizing proton conducting nanochannels, since practical proton conduction occurs in the through-plane direction under the given system pressure.

216

S.-H. Yun et al. / Journal of Membrane Science 417418 (2012) 210216

Table 2 Through-plane conductivitya of Naons membranes under different pressure loads. Pi [psi] k
Pi

[9] [10]

[S/cm] Naon115 0.0777 0.0406 0.0183 0.0158 0.0101 Naon112 0.0453 0.0314 0.0342 0.0160 0.0154

Naon117 71 142 213 284 355

a

[11]

0.0535 0.0345 0.0246 0.0159 0.0125

[12]

[13]

Calculated using Eq. (2).

[14]

Acknowledgments This work was supported by a project, 10037748, of the Korea Institute of Industrial Technology (KEIT) grant funded by the Korea government Ministry of Knowledge Economy. References

[15]

[16]

[17]

[18] [1] J.-S. Kim, J.-H. Choi, Fabrication and characterization of a carbon electrode coated with cation-exchange polymer for the membrane capacitive deionization applications, J. Membr. Sci. 355 (2010) 8590. [2] H.-J. Lee, M.-K. Hong, S.-D. Han, S.-H. Moon, Inuence of the heterogeneous structure on the electrochemical properties of anion exchange membranes, J. Membr. Sci. 320 (2008) 549555. [3] D.A. Vermaas, M. Saakes, K. Nijmeijer, Power generation using proled membranes in reverse electrodialysis, J. Membr. Sci. 385386 (2011) 234242. [4] S.-H. Yun, J.-J. Woo, S.-J. Seo, L. Wu, D. Wu, T.W. Xu, S.-H. Moon, Sulfonated poly(2,6-dimethyl-1,4-phenylene oxide) (SPPO) electrolyte membranes reinforced by electrospun nanober porous substrates for fuel cells, J. Membr. Sci. 367 (2011) 296305. [5] F.-M. Wang, S.-C. Lo, C.-S. Cheng, J.-H. Chen, B.-J. Hwang, H.-C. Wu, Selfpolymerized membrane derivative of branched additive for internal short protection of high safety lithium ion battery, J. Membr. Sci. 368 (2011) 165170. [6] M. Vijayakumar, M.S. Bhuvaneswari, P. Nachimuthu, B. Schwenzer, S. Kim, Z. Yang, J. Liu, G.L. Graff, S. Thevuthasan, J. Hu, Spectroscopic investigations of the fouling process on Naon membranes in vanadium redox ow batteries, J. Membr. Sci. 366 (2011) 325334. [7] S.-J. Seo, S.-H. Yun, J.-J. Woo, D.-W. Park, M.-S. Kang, A. Hinsch, S.-H. Moon, Preparation and characterization of quasi-solid-state electrolytes using a brominated poly(2,6-dimethyl-1,4-phenylene oxide) electrospun nanober mat for dye-sensitized solar cells, Electrochem. Commun. 13 (2011) 13911394. [8] S.-J. Seo, J.-J. Woo, S.-H. Yun, H.-J. Lee, J.-S. Park, T.W. Xu, T.-H. Yang, J. Lee, S.-H. Moon, Analyses of interfacial resistances in a membrane-electrode assembly [19] [20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

for a proton exchange membrane fuel cell using symmetrical impedance spectroscopy, Phys. Chem. Chem. Phys. 12 (2010) 1529115300. B.S. Pivovar, Y.S. Kim, The membrane-electrode interface in PEFCs, J. Electrochem. Soc. 154 (2007) B739B744. A.L. Ocampo, J.C. Aguilar, E.R.D. Miguel, M. Monroy, P. Roquero, J. de Gyves, Novel proton-conducting polymer inclusion membranes, J. Membr. Sci. 326 (2009) 382387. P. Dlugolecki, P. Ogonowski, S.J. Metz, M. Saakes, K. Nijmeijer, M. Wessling, On the resistances of membrane, diffusion boundary layer and double layer in ion exchange membrane transport, J. Membr. Sci. 349 (2010) 369379. C.H. Lee, H.B. Park, Y.M. Lee, R.D. Lee, Importance of proton conductivity measurement in polymer electrolyte membrane for fuel cell application, Ind. Eng. Chem. Res. 44 (2005) 76177626. Y. Sone, P. Ekdunge, D. Simonsson, Proton conductivity of Naon 117 as measured by a four-electrode AC impedance method, J. Electrochem. Soc. 143 (1996) 12541259. R.F. Silva, M. De Francesco, A. Pozio, Tangential and normal conductivities of Naons membranes used in polymer electrolyte fuel cells, J. Power Sources 134 (2004) 1826. B.P. Ladewig, R.B. Knott, D.J. Martin, J.C.D. da Costa, G.Q. Lu, Naon-MPMDMS nanocomposite membranes with low methanol permeability, Electrochem. Commun. 9 (2007) 781786. M. Monroy-Barreto, J.C. Aguilar, E.R.D. Miguel, A.L. Ocampo, M. Munoz, J. de Gyves, Novel semi-interpenetrating polymer network hybrid membranes for proton conduction, J. Membr. Sci. 344 (2009) 92100. T.C. Chilcott, H.G.L. Coster, E.P. George, A novel method for the characterisation of the double xed charge (bipolar) membrane using impedance spectroscopy, J. Membr. Sci. 108 (1995) 185197. T. Uma, M. Nogami, Proton-conducting glass electrolyte, Anal. Chem. 80 (2008) 506508. S.-M. Park, J.-S. Yoo, Electrochemical impedance spectroscopy for better electrochemical measurements, Anal. Chem. 75 (2003) 455a461a. T.Y. Soboleva, Z. Xie, Z.Q. Shi, E. Tsang, T.C. Navessin, S. Holdcroft, Investigation of the through-plane impedance technique for evaluation of anisotropy of proton conducting polymer membranes, J. Electroanal. Chem. 622 (2008) 145152. K.R. Cooper, Progress toward accurate through-plane ion transport resistance measurement of thin solid electrolytes, J. Electrochem. Soc. 157 (2010) B1731B1739. I. Radev, G. Georgiev, V. Sinigersky, E. Slaucheva, Proton conductivity measurements of PEM performed in easytest cell, Int. J. Hydrogen Energy 33 (2008) 48494855. A. Kishi, M. Umeda, Proton conductivity measurement of polymer electrolyte membrane in lm thickness direction using small-size electrodes, Electrochemistry 75 (2007) 130132. X. Zhang, S. Chen, J. Liu, Z. Hu, S. Chen, L. Wang, Preparation and properties of sulfonated poly(phenylene arylene)/sulfonated polyimide (SPA/SPI) blend membranes for polymer electrolyte membrane fuel cell applications, J. Membr. Sci. 371 (2011) 276285. S.-H. Moon, S.-H. Yun, J.-J. Woo, S.-J. Seo, Apparatus for Measuring Thickness and Pressure of Ion Exchange Membrane and Method of Measurement therefore, Korea Patent Pending, No. 10-2010-0085577, Sep. 01, (2010). H.-L. Lin, T.L. Yu, F.-H. Han, A method for improving ionic conductivity of Naon membranes and its application to PEMFC, J. Polym. Res. 13 (2006) 379385. Y.L. Zhu, C. Liu, J.S. Liang, L.D. Wang, Investigation of the effects of compression pressure on direct methanol fuel cell, J. Power Sources 196 (2011) 264269.