- Email: [email protected]
Surface roughness effect on the metallic bipolar plates of a proton exchange membrane fuel cell
Applied Energy 104 (2013) 898–904
Contents lists available at SciVerse ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
Surface roughness effect on the metallic bipolar plates of a proton exchange membrane fuel cell Chien-Hung Lin ⇑ Department of Physics, ROC Military Academy, Feng-Shan, Kaohsiung 830, Taiwan
h i g h l i g h t s " Various degrees of roughness are caused by the sandblasting method. " An improper surface modification depletes the PEMFC performance severely. " The AC impedance are used to assess the fuel gas transfer effect. " The Warburg resistance form in the coarse flow channel surface.
a r t i c l e
i n f o
Article history: Received 24 May 2012 Received in revised form 5 December 2012 Accepted 10 December 2012 Available online 7 January 2013 Keywords: Proton exchange membrane fuel cells (PEMFCs) Contact resistance Roughness Bipolar plate
a b s t r a c t Proton exchange membrane fuel cells (PEMFCs) is a promising candidate as energy systems. However, the stability and lifetime of cells are still important issues. The effect of surface roughness on metallic bipolar plate is discussed in this paper. Various roughness on the bulk surface are obtained by the sandblasting method. The grain sizes of sand are selected as 50, 100 and 200 lm. The Ac impedance experiment results show that the bipolar plate roughness and carbon paper porosity are well matched when the surface roughness is within 1–2 lm. Superior condition decreases the contact resistance loss in the fuel cell. The high frequency resistance of the coarse surface was larger than that of the substrate by around 5 mX. Furthermore, a new arc was formed at the low frequency region. Hence, the unmatch roughness condition of the bipolar plate significantly increases the contact resistance and mass transfer resistance. This paper develops a sequential approach to study an optimum surface roughness by combining the whole performance (I–V) curve and AC impedance result. It benefits us to quantify the contact and mass transfer resistance exists in the PEMFC. The proposed surface treatment improves the surface effect and promotes the implement of potential metallic bipolar plate in near future. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Hydrogen energy is considered to be an ideal alternative energy in the foreseeable future [1–4]. Hydrogen has been used as the fuel in the fuel cells [5,6]. The major advantages of proton exchange membrane fuel cells (PEMFCs) are low-temperature operation, quick starting and high energy density. Therefore, the PEMFC can be extensively applied to power generation, portable electric equipment, ship and hybrid vehicles. The bipolar plate acts as an important component in the PEMFC [7]. The bipolar plate served as a fuel gas feed, water drain and electronics transfer medium. These properties, such as corrosion resistance, electrical resistance, flow pattern, hydrophobic surface, cost and weight of the bipolar plate, were discussed extensively [8–12]. Kanezaki et al. [13] discussed the cross-leakage from the adjacent gas flow and GDL that ⇑ Tel.: +886 7 7425024; fax: +886 7 7194170. E-mail address: [email protected] 0306-2619/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apenergy.2012.12.024
influenced the single-cell performance. This cross-leakage was useful to reduce the concentration loss and water drain in the fuel cell. Fundamentally, this mechanism occurred frequently under the larger pressure difference and flux density of gas. However, the friction and bending loss decreased the pressure difference in the flow channel. Hence, the worse fuel cell performance was observed under large friction and the rough surface in the fuel cell. The gas concentration, diffusion coefficient, pressure, viscosity and friction involved a mass transfer reaction. Cooper and Smith [14] adopted AC impedance, current interruption and high frequency resistance technology to analyze the fuel cell performance. The ohmic resistance error by three methods was within 4.3% of each other. This high frequency resistance deviation was very small for the contact resistance physically. Lee et al. [15] manufactured graphite bipolar plates by various processes. Various pressure, temperature and graphite proportion were arranged in this experiment. The wellproportioned graphite reduced the contact resistance. Furthermore, the increasing graphite particle enhanced the hydrophobicity
899
C.-H. Lin / Applied Energy 104 (2013) 898–904
of the bipolar plate. This phenomenon proven that the surface effect was significant to the whole fuel cell performance. Kraytsberg et al. [16] attributed the attributed contact resistance to the bipolar plate and the GDL. The passivation layer and morphology of the bulk influenced the performance significantly. The morphology of the bulk was polished by various grit sizes of Emery paper. The maximum contact resistance (361 mX) was measured under the 1-lm grain diamond polishing. However, the opposite trend could be measured when the minimum contact resistance (22 mX) was measured under the 127-lm grain diamond polishing. The smooth surface acted a steep descend of the contact conductivity. Yan et al. [17] observed the 2-kW PEMFC performance by AC impedance technology. Fundamentally, the gas mass transfer effect was measured under the small gas stoichiometry. The electronic transfer effect was observed under low relative humidity. The mass transfer effect to the cell stack was more important than in the single cell. Hence, the AC impedance technology could provide reliable fuel cell information for the cell preliminary design. The AC impedance technique was adopted to evaluate the PEMFC performance in many papers [18–20]. Barber et al. [21] discussed the contact surface area between the bipolar plate and the collector by the mathematical model. Three main operating factors, bulk surface roughness, compact force and the coating thickness of the collector, affected the total contact surface area significantly. Moreover, the minimum porosity of the bipolar graphite led to the maximum contact surface with the collector. Avasarala et al. [22] discussed the relation of the roughness over the complex bipolar plate and the contact resistance. The artificial roughness was polished by various grit sizes of sandpapers (#80–1000). The minimum contact resistance caused by the #600 sandpapers and the roughness was approximately 0.9 lm. Contrary to common belief, maximum contact resistance was caused by the #1000 sandpapers. Antoni et al. [23] discussed the contact resistance under various passivation layers and surface roughness. The stainless steel 316 and SS 904 were selected as the substrates. The roughness of the bipolar plate was polished by various sandpapers. The value of the contact resistance increased gradually when the roughness was modified within 1–2 lm. The contact resistance increased sharply when the surface roughness was less than the 0.5 lm. Furthermore, this experiment was set at a potentiostatic mode of 800 mV and 500 mV in the cathodic condition. The SS 316 also possessed superior performance under the polarization test. Shoyama et al. [24] discussed the relation of the hydrophobicity and water management in the bipolar plate and GDL. The roughness involved the hydrophobicity in the flow channel significantly. It was observed that the coarse surface enhanced the surface tension. Furthermore, the abundant PTFE in the GDL would enhance the hydrophobicity. A better single-cell performance curve was measured when the hydrophobic GDL accompanied with the hydrophilic bipolar plate was set in high relative humidity. It was mentioned that the roughness of the bipolar plate affected the fuel cell significantly. The contact resistance involving the bipolar plate roughness and GDL porosity was discussed in many papers. There are few papers depicting the bipolar plate roughness involved in the fuel gas transfer mechanics. However, the stability of the fuel cell should be improved before it is fully developed. Therefore, the fuel gas transfer resistance and bipolar plate roughness are discussed in this paper.
2. Experiment 2.1. Surface roughness on the bipolar plate A single serpentine gas flow bipolar plate with an active area of 25 cm2 was manufactured from an Al alloy 5052 and machined as
Fig. 1. The morphology of 5052 Al-alloy bipolar plate.
shown in Fig. 1. The flow channel was machined as 0.9 mm in both depth and width. The smooth surface was clearly observed on the substrate (Ra = 0.2131 lm) in the inset. Roughness variations on the bulk surface were caused by the sandblasting method. The sand grain sizes were selected as 50, 100, and 200 lm. Furthermore, the bipolar plate after sandblasting with medium sand was immersed in sulfuric acid for 30 min in order to analyze the characteristic of the surface roughness after the corrosion. Various sandblasting methods involved not only the contact resistance effects but also the mass transfer mechanism. The relationship between the surface roughness and the sandblasting method is listed in Table 1. 2.2. Fuel cell configurations The manufactured PEMFC used in this experiment was assembled with commercial membranes (Nafion-112), sandblasted metallic bipolar plates, and Toray carbon paper (E-Tek). 180-lmthick gas diffusion layers were used as the base material in this study. The active area of the Nafion 112 membrane was 25 cm2 (5 cm 5 cm). The schematic diagram and photo of PEMFC apparatus were shown in Fig. 2. In the fuel cell experiments, the compaction force imparted on a cell was loaded as 200 N/cm2. 2.3. Electrochemical measurements The fuel gas was supplied with oxygen (140 sccm) in the cathode and hydrogen gas (210 sccm) in the anode. The temperature of the fully humid flow gas and cell were set at 60 °C. The range of the electric load was 0.85–0.1 V (voltage scan rate was 0.1 V/ h), and the limit current was set at 25 A. The fuel cell performance
Table 1 Change in the roughness of the bipolar plate with various degrees of sand. Part no.
Name of part
1 2 3 4 5
Substrate Fine sand Medium sand Coarse sand Medium sand with etching
Sizes of sand (lm)
Roughness (Ra) (lm)
50 100 200 100
0.213 1.521 1.892 2.932 1.2920
Etching
30 min
900
C.-H. Lin / Applied Energy 104 (2013) 898–904
Fig. 2. The schematic diagram of PEMFC apparatus. Fig. 4. Change in power density curve of the PEMFC with bipolar plate roughness.
was measured using AC impedance and performance (I–V) curves. Finally, the AC impedance curve was fitted by a mathematical model. The curve-fitting coefficient was used to assess the surface roughness effect on the metallic bipolar plates of a proton exchange membrane fuel cell.
3. Results and discussion 3.1. Single cell performance The experiments were performed under ambient pressure and temperature (333 K), as shown in Fig. 3. The PEMFC assembled with the substrate bipolar plate reached 0.45 V and 6.8 W under a current density of 600 mA cm2 (Figs. 3 and 4). Simultaneously, the PEMFC assembled with the bipolar plate after sandblasting with fine sand reached 0.62 V and 8.75 W under a current density of 600 mA cm2 (Figs. 3 and 4). It was noteworthy that the PEMFC performance significantly changed with the bulk surface after sandblasting with fine sand. This fine sand sandblasting treatment caused the bipolar plate and GDL to be well matched, which effectively lowered the contact resistance. However, unlimited roughness expansion may not always be the ideal method to lower the contact resistance. The PEMFC assembled with the bulk surface after sandblasting with either medium sand or coarse sand apparently caused lower power generation than the fine sand treatment. Certainly, the bipolar plate after sandblasting with coarse sand increased the fuel gas concentration loss to the PEMFC. The PEMFC
Fig. 3. Change in performance (I–V) curve of the PEMFC with bipolar plate roughness.
assembled with bipolar plates after sandblasting with medium sand and coarse sand reached 4.8 W and 4.3 W, respectively, under a current density at 500 mA/cm2. After immersing in 0.5 M sulfuric acid for 30 min, the bulk surface after sandblasting with medium sand has a lower roughness (Ra = 1.29 lm), (Table 1). The etching effect decrease PEMFC performance severely when it operated at high current density stage (Figs. 3 and 4). The roughness effect certainly influenced the contact resistance and the fuel gas mass transfer mechanism of the PEMFC. The smooth surface decreased the fuel gas transfer shearing force. The fuel gas transfer barrier is caused by the uneven surface. Oxygen starvation occurs because of the non-uniform distribution of the fuel gas in the flow channel. Hence, the optimum surface roughness is a trade off among the various component requirements of the PEMFC. The coarse sand blasting is a crux in this paper. It is helpful for us to know the failure mechanism. As Fig. 3, the polarization curve decays rapidly when PEMFC operates at high current load. The higher exhaustion rate at fuel gas as the rapid electrochemical reaction occurs. Evidently, the surface morphology significantly changes with the bipolar plate after sandblasting with coarse sand. The hydrophobicity which varies with bipolar plate roughness involves rapid water accumulation problem in the PEMFC. Hence, the coarse sand blasting method on bipolar plate causes the disadvantage to gas transfer ability.
3.2. Surface roughness analysis in the flow channel The morphology of the Al alloy bipolar plate in the flow channel was observed by SEM (Fig. 5). The smooth surface was clearly observed on the substrate (Ra = 0.2131 lm). Moreover, the Al alloy bipolar plate after sandblasting with coarse sand was uneven (Ra = 2.932 lm). There were many cavities formed on the surface of the flow channel. The etching treatment likely caused a smoother surface (Ra = 1.29 lm) than the surface after sandblasting with medium sand (Ra = 1.892 lm). The above results correspond to the surface roughness by the different sandblasting methods listed in Table 1, where it can be seen that the bipolar plate surface significantly influences the PEMFC performance. With regard to the contact resistance, the compact force of the single cell exerts pressure on the interface, which increases the contact area between the fuel cell components and decreases the contact resistance. However, an improper compact force may deform the GDL and decrease the PEMFC lifetime. The changes in the optimal compact force and bipolar plate surface roughness that decrease the contact resistance are discussed in some papers [16,25–27]. However, certain quantitative state-
901
C.-H. Lin / Applied Energy 104 (2013) 898–904
(a) Substrate
(c) Sandblasting with medium sand
(e) Sandblasting with medium sand and etching
(b) Sandblasting with fine sand
(d) Sandblasting with coarse sand
Fig. 5. The morphology of Al-alloy bipolar plate under the SEM.
ments of the bipolar plate roughness involving the fuel gas transfer mechanism are scarcely discussed. 3.3. AC impedance test Electrochemical impedance spectroscopy (EIS) is a powerful tool for analyzing the PEMFC operational behavior. There were many analogies between the electrochemical reaction mechanism and the electrical circuit operation. For impedance spectroscopy, a small perturbation voltage or current is externally applied to the PEMFC at different frequencies. The alternating current or voltage signal is measured by an AC impedance meter. A phase shift exists between the voltage and current to observe the resistance change. The electrochemical properties and the interface reaction research are available for the PEMFC. The contact resistance, the fuel gas transfer mechanism, and the anode and cathode interface loss are identified easily from the impedance information. 3.3.1. Nyquist plot The electrochemical impedance spectroscopy experiments were performed under ambient pressure, an electric load of 0.6 V, and a temperature 333 K. The impedance change in the Nyquist plot is shown in Fig. 6. The high frequency resistance of the PEMFC assembled with the bipolar plate after sandblasting with fine sand and the etching treatment was approximately 10.97 mX. The worse performance certainly occurred with the PEMFC assembled with the substrate bipolar plate (12.4 mX). Moreover, the bipolar plate performance after sandblasting with medium sand (13.5 mX) was the worst condition. It was concluded that the roughness of the bipolar plate caused the high frequency change of the PEMFC shown in Fig. 6. However, the fuel gas transfer resistance did not change significantly when the surface roughness was within a reasonable range. The high and low frequency resistance increased with the coarse surface when the roughness was greater than 2 lm on the bipolar plate. The high frequency resistance of the coarse surface was larger than that of the substrate by around 5 mX. Furthermore, a new arc was formed at the low frequency region. This indicates that the roughness affects not only the contact resistance but also the gas mass transfer. The interception line of the low frequency position was approximately 37.5 mX. It is
Fig. 6. Change in Nyquist curve of the PEMFC with bipolar plate roughness. (a) The roughness of the bipolar plate within 2 lm. (b) Sandblasting with coarse sand.
obvious that the gas mass transfer resistance of the coarse sand is higher than that in the other surface treatment methods. This
902
C.-H. Lin / Applied Energy 104 (2013) 898–904
phenomenon cannot be neglected, especially in the large current operation. The rapid electrochemical mechanism occurs when there is high fuel gas consumption at the cathode. The single cell performance is significantly affected by the surface roughness. 3.3.2. Bode plot The frequency behavior of the fuel cell performance can be interpreted clearly by the bode plot. The amplitude of the resistance is shown in Fig. 7. Obviously, the bipolar plate after sandblasting with fine sand has a small resistance within all frequency regions. The bipolar plate after sandblasting with coarse sand is significantly affected by the gas transfer mechanism. Similar to the high frequency region, the resistance of the bipolar plate after sandblasting with coarse sand is higher than that of the substrate by 7 mX. Furthermore, the low frequency resistance of this bipolar plate after sandblasting with coarse sand (37.5 mX) is obviously higher than that of the substrate (19 mX) at 0.1 Hz. The Warburg resistance was obtained at 0.05 Hz, indicating the gas transfer barrier formation. Moreover, the resistance of the bipolar plate after sandblasting with fine sand and medium sand and etching showed the same amplitude in all frequency regions. This result corresponds to the roughness value listed in Table 1. The phase angle of the resistance is shown in Fig. 8. Most fuel cells show the same phase angle when the bipolar plate surface roughness is less than 2 lm. As with the resistance of the bipolar plate after sandblasting with coarse sand, there is another arc formed at the low frequency region (1 Hz). This arc is attributed to the gas mass transfer barrier in the flow channel of the fuel cell. The increasing roughness leads to large shear force and worse mass transfer ability. 3.3.3. Fuel cell parameter fitting The data obtained from experiments were fitted by the mathematical model shown in Fig. 9. The ohmic resistance (R1) of a single cell represented as a sum of contributions from the wire/contact resistance of entity in PEMFC [20]. Moreover, R2 is the kinetic resistance of the electrochemical reaction in the anode. The C2 shows the electric double layer over the interference in the cathode. The other first-order circuit describes the electrochemical reaction process in the cathode. It is noteworthy that the Warburg resistance describes the nonlinear electrochemical reaction process in the fuel cell. Theoretically, it is attributed to the gas transfer barrier in the flow channel. The fitting data are recorded in Table 2. R1, R2, and R3 are plotted in Fig. 10. A smaller R1 is obtained from the bipolar plate after sandblasting and the etching treatment (11 mX). An opposite trend is observed for the bipolar plate after
Fig. 7. Change in the fuel cell impedance magnitude with frequency.
Fig. 8. Change in the fuel cell impedance phase with frequency.
sandblasting with coarse sand (17.4 mX). It is proved that parameter R1 involves the roughness variation of the contact resistance. R2 and R3 describe the electrochemical kinetic resistance in the anode and cathode, respectively [17]. Both of them show increased resistance with roughness variation. At the anode side, the contact regions at the interface of bipolar plate and gas diffusion layer causes the electronics transfer resistance of a PEMFC. At the cathode side, the bumpy surface decreases the hydrophobicity. Hence, the gas transfer resistance occurs in the cathode side owing to the water accumulation and decreases the feed rate of oxygen. Parameter R3 of the bipolar plate after sandblasting with coarse sand is larger than that of the substrate by 9 mX. Hence, it is certain that the resistance of the increasing R3 value is attributed to the nonuniform distribution of the oxygen gas in the flow channel. As Fig. 5 and Table 1, the medium sand with etching method caused the smoother surface than the pure medium sand. Hence, the minor change in R2 value can be obtained between the medium sand with etching and fine sand in Fig. 10. The variation of parameters C1 and C2 are plotted in Fig. 11. The largest capacitance obtained from this bipolar plate after sandblasting with coarse sand is not plotted and that differs by one order of magnitude from the others. The C1 and C2 fitting data for this bipolar plate after sandblasting and the etching treatment are recorded as 0.81 F and 0.28 F in Table 2, respectively. The sandblasting changes the morphology on the substrate and electrochemical reaction in PEMFC. The smoother plane enhances the electronics mobility and decreases electric double layer thickness. Hence, the significant change on capacitance is attributed to the oxygen film on the bipolar plate. The abundant electronics accumulate on the electric double layer surface and enhance the capacitance effect in the anode and cathode. The fitting parameter of Warburg is also recorded in Table 2. The largest Warburg value is obtained from this bipolar plate after sandblasting with coarse sand. Obviously, the coarse roughness enhances the shear force to the fuel gas and mass transfer barrier. The larger Warburg value indicates oxygen starvation in the fuel cell flow channel.
Fig. 9. Equivalent circuit of the fuel cell used for the fitting of the impedance measurement.
903
C.-H. Lin / Applied Energy 104 (2013) 898–904 Table 2 List of coefficients for curve-fitting expression of fuel cell model. Part no.
Name of part
R1 (mX)
R2 (mX)
R3 (mX)
C1 (F)
C2 (F)
Warburg
1 2 3 4 5
Substrate Fine sand Medium sand Coarse sand Medium sand with etching
12.4 11.2 13.8 17.4 10.8
4.3 3.9 6.2 7.8 4.2
2.2 1.1 2.4 11.3 1.1
0.5622 0.6717 0.5767 15.9780 0.8092
0.1181 0.2175 0.1158 0.4678 0.2788
1.82E-06 4.81E-06 3.03E-06 3.90E-04 3.07E-05
condition could severely decrease the PEMFC performance. Even the mass transfer resistance is much lower than the contact resistance. However, the mass transfer resistance cannot be neglected under the large load operation. The high frequency resistance of the coarse surface was larger than that of the substrate by around 5 mX. Furthermore, a new arc was formed at the low frequency region. Certainly, the mass transfer resistance can reduce the fuel cell performance and the stability of the system output power. The unlimited increasing surface roughness decreases the hydrophobicity of the bipolar plate and gas mass transfer ability of the fuel cell. The surface effect is important to the performance of the entire fuel cell. Hence, the roughness of the bipolar plate needs to be well modified before it is fully developed to improve the contact resistance and mass transfer resistance. It benefits us to quantify the contact and mass transfer resistance exists in the PEMFC. The proposed surface treatment improves the surface effect and promotes the implement of potential metallic bipolar plate in near future. Fig. 10. Change in curve-fitting parameters R1, R2 and R3 of the fuel cell.
References
Fig. 11. Change in curve-fitting parameters C1 and C2 of the fuel cell.
The fitting data are obtained from the AC impedance test. The largest contact resistance and mass transfer are obtained from the bipolar plate after sandblasting with coarse sand. Even the mass transfer resistance is much lower than the contact resistance. However, the mass transfer resistance cannot be neglected under the large load operation. Certainly, the mass transfer resistance reduces the fuel cell performance and the stability of the system output power. 4. Conclusions Various degrees of roughness on the bulk surface are caused by the sandblasting method. In this paper, we discuss the effect of the surface roughness on the metallic bipolar plate. The selected grain sizes of sand are 50, 100 and 200 lm. The fitting data are obtained from the AC impedance test. An improper surface modification
[1] Salameh MG. Can renewable and unconventional energy sources bridge the global energy gap in the 21st century. Appl Energy 2003;75:33–42. [2] Khan MJ, Iqbal MT. Analysis of a small wind-hydrogen stand-alone hybrid energy system. Appl Energy 2009;86(11):2429–42. [3] Connolly D, Lund H, Mathiesen BV, Leahy M. A review of computer tools for analysing the integration of renewable energy into various energy systems. Appl Energy 2010;87:1059–82. [4] Wang Y, Chen KS, Mishler J, Cho SC, Adroher XC. A review of polymer electrolyte membrane fuel cells: technology, applications, and needs on fundamental research. Appl Energy 2011;88:981–1007. [5] Barelli L, Bidini G, Gallorini F, Ottaviano A. Dynamic analysis of PEMFC-based CHP systems for domestic application. Appl Energy 2012;91:13–28. [6] Juul N, Meibom P. Road transport and power system scenarios for Northern Europe in 2030. Appl Energy 2012;92:573–82. [7] Hermann A, Chaudhuria T, Spagnol P. Bipolar plates for PEM fuel cells: a review. J Hydrogen Energy 2005;30:1297–302. [8] Wang CT, Hu YC, Zheng PL. Novel biometric flow slab design for improvement of PEMFC performance. Appl Energy 2010;87:1366–75. [9] Nama J, Chippar P, Kim W, Ju H. Numerical analysis of gas crossover effects in polymer electrolyte fuel cells (PEFCs). Appl Energy 2010;87:3699–709. [10] Perng SW, Wub HW. Non-isothermal transport phenomenon and cell performance of a cathodic PEM fuel cell with a baffle plate in a tapered channel. Appl Energy 2011;88:52–67. [11] Wu HW, Ku HW. The optimal parameters estimation for rectangular cylinders installed transversely in the flow channel of PEMFC from a three-dimensional PEMFC model and the Taguchi method. Appl Energy 2011;88:4879–90. [12] Nishimura A, Shibuya K, Morimoto A, Tanaka S, Hirota M, Nakamura Y, et al. Dominant factor and mechanism of coupling phenomena in single cell of polymer electrolyte fuel cell. Appl Energy 2012;90:73–9. [13] Kanezaki T, Li X, Baschuk JJ. Cross-leakage flow between adjacent flow channels in PEM fuel cells. J Power Sources 2006;162:415–25. [14] Cooper KR, Smith M. Electrical test methods for on-line fuel cell ohmic resistance measurement. J Power Sources 2006;160:1088–95. [15] Lee HS, Kim HJ, Kim SG, Ahn SH. Evaluation of graphite composite bipolar plate for PEM (proton exchange membrane) fuel cell: electrical, mechanical, and molding properties. J Mater Process Technol 2007;187:425–8. [16] Kraytsberg A, Auinat M, Ein-Eli Y. Reduced contact resistance of PEM fuel cell’s bipolar plates via surface texturing. J Power Sources 2007;164:697–703. [17] Yan X, Hou M, Sun L, Liang D, Shen Q, Xu H, et al. AC impedance characteristics of a 2 kW PEM fuel cell stack under different operating conditions and load changes. Int J Hydrogen Energy 2007;32:4358–64. [18] Lee Y, Cho E, Lee JH, Kim HJ, Lim TH, Oh IH, et al. Effects of purging on the degradation of PEMFCs operating with repetitive on/off cycles. J Electrochem Soc 2007;154(2):B194–200. [19] Shahryari A, Kamal W, Omanovic S. The effect of surface roughness on the efficiency of the cyclic potentiodynamic passivation (CPP) method in the
904
[20] [21]
[22]
[23]
C.-H. Lin / Applied Energy 104 (2013) 898–904 improvement of general and pitting corrosion resistance of 316LVM stainless steel. Mater Lett 2008;62:3906–9. Ishibashi Y, Nishikata A, Tsuru T. Electrochemical Impedance spectroscopy of PEM fuel cell with metal bipolar plates. J Electrochem Soc 2008;16(24):85–9. Barber M, Sun TS, Petrach E, Wang X, Zou Q. Contact mechanics approach to determine contact surface area between bipolar plates and current collector in proton exchange membrane fuel cells. J Power Sources 2008;185:1252–6. Avasarala B, Haldar P. Effect of surface roughness of composite bipolar plates on the contact resistance of a proton exchange membrane fuel cell. J Power Sources 2009;188:225–9. Antoni L, André J, Petit JP. Optimization of the conductivity and passivity properties of passive films on stainless steels used as bipolar plates in PEFC. J Electrochem Soc 2009;17(1):383–92.
[24] Shoyama M, Tomimura T, Mizutani S. Effect of hydrophilic properties of gas channel on water management of PEM fuel cells. J Electrochem Soc 2009;17(1):461–4. [25] Wang Y, Shakhshir SA, Li X. Development and impact of sandwich wettability structure for gas distribution media on PEM fuel cell performance. Appl Energy 2011;88(6):2168–75. [26] Lin CH, Tsai SY. An investigation of coated aluminium bipolar plates for PEMFC. Appl Energy 2012;100:81–92. [27] Yuan W, Tang Y, Yang X, Wang Z. Porous metal materials for polymer electrolyte membrane fuel cells – a review. Appl Energy 2012;94:309–29.
Contents lists available at SciVerse ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
Surface roughness effect on the metallic bipolar plates of a proton exchange membrane fuel cell Chien-Hung Lin ⇑ Department of Physics, ROC Military Academy, Feng-Shan, Kaohsiung 830, Taiwan
h i g h l i g h t s " Various degrees of roughness are caused by the sandblasting method. " An improper surface modification depletes the PEMFC performance severely. " The AC impedance are used to assess the fuel gas transfer effect. " The Warburg resistance form in the coarse flow channel surface.
a r t i c l e
i n f o
Article history: Received 24 May 2012 Received in revised form 5 December 2012 Accepted 10 December 2012 Available online 7 January 2013 Keywords: Proton exchange membrane fuel cells (PEMFCs) Contact resistance Roughness Bipolar plate
a b s t r a c t Proton exchange membrane fuel cells (PEMFCs) is a promising candidate as energy systems. However, the stability and lifetime of cells are still important issues. The effect of surface roughness on metallic bipolar plate is discussed in this paper. Various roughness on the bulk surface are obtained by the sandblasting method. The grain sizes of sand are selected as 50, 100 and 200 lm. The Ac impedance experiment results show that the bipolar plate roughness and carbon paper porosity are well matched when the surface roughness is within 1–2 lm. Superior condition decreases the contact resistance loss in the fuel cell. The high frequency resistance of the coarse surface was larger than that of the substrate by around 5 mX. Furthermore, a new arc was formed at the low frequency region. Hence, the unmatch roughness condition of the bipolar plate significantly increases the contact resistance and mass transfer resistance. This paper develops a sequential approach to study an optimum surface roughness by combining the whole performance (I–V) curve and AC impedance result. It benefits us to quantify the contact and mass transfer resistance exists in the PEMFC. The proposed surface treatment improves the surface effect and promotes the implement of potential metallic bipolar plate in near future. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Hydrogen energy is considered to be an ideal alternative energy in the foreseeable future [1–4]. Hydrogen has been used as the fuel in the fuel cells [5,6]. The major advantages of proton exchange membrane fuel cells (PEMFCs) are low-temperature operation, quick starting and high energy density. Therefore, the PEMFC can be extensively applied to power generation, portable electric equipment, ship and hybrid vehicles. The bipolar plate acts as an important component in the PEMFC [7]. The bipolar plate served as a fuel gas feed, water drain and electronics transfer medium. These properties, such as corrosion resistance, electrical resistance, flow pattern, hydrophobic surface, cost and weight of the bipolar plate, were discussed extensively [8–12]. Kanezaki et al. [13] discussed the cross-leakage from the adjacent gas flow and GDL that ⇑ Tel.: +886 7 7425024; fax: +886 7 7194170. E-mail address: [email protected] 0306-2619/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apenergy.2012.12.024
influenced the single-cell performance. This cross-leakage was useful to reduce the concentration loss and water drain in the fuel cell. Fundamentally, this mechanism occurred frequently under the larger pressure difference and flux density of gas. However, the friction and bending loss decreased the pressure difference in the flow channel. Hence, the worse fuel cell performance was observed under large friction and the rough surface in the fuel cell. The gas concentration, diffusion coefficient, pressure, viscosity and friction involved a mass transfer reaction. Cooper and Smith [14] adopted AC impedance, current interruption and high frequency resistance technology to analyze the fuel cell performance. The ohmic resistance error by three methods was within 4.3% of each other. This high frequency resistance deviation was very small for the contact resistance physically. Lee et al. [15] manufactured graphite bipolar plates by various processes. Various pressure, temperature and graphite proportion were arranged in this experiment. The wellproportioned graphite reduced the contact resistance. Furthermore, the increasing graphite particle enhanced the hydrophobicity
899
C.-H. Lin / Applied Energy 104 (2013) 898–904
of the bipolar plate. This phenomenon proven that the surface effect was significant to the whole fuel cell performance. Kraytsberg et al. [16] attributed the attributed contact resistance to the bipolar plate and the GDL. The passivation layer and morphology of the bulk influenced the performance significantly. The morphology of the bulk was polished by various grit sizes of Emery paper. The maximum contact resistance (361 mX) was measured under the 1-lm grain diamond polishing. However, the opposite trend could be measured when the minimum contact resistance (22 mX) was measured under the 127-lm grain diamond polishing. The smooth surface acted a steep descend of the contact conductivity. Yan et al. [17] observed the 2-kW PEMFC performance by AC impedance technology. Fundamentally, the gas mass transfer effect was measured under the small gas stoichiometry. The electronic transfer effect was observed under low relative humidity. The mass transfer effect to the cell stack was more important than in the single cell. Hence, the AC impedance technology could provide reliable fuel cell information for the cell preliminary design. The AC impedance technique was adopted to evaluate the PEMFC performance in many papers [18–20]. Barber et al. [21] discussed the contact surface area between the bipolar plate and the collector by the mathematical model. Three main operating factors, bulk surface roughness, compact force and the coating thickness of the collector, affected the total contact surface area significantly. Moreover, the minimum porosity of the bipolar graphite led to the maximum contact surface with the collector. Avasarala et al. [22] discussed the relation of the roughness over the complex bipolar plate and the contact resistance. The artificial roughness was polished by various grit sizes of sandpapers (#80–1000). The minimum contact resistance caused by the #600 sandpapers and the roughness was approximately 0.9 lm. Contrary to common belief, maximum contact resistance was caused by the #1000 sandpapers. Antoni et al. [23] discussed the contact resistance under various passivation layers and surface roughness. The stainless steel 316 and SS 904 were selected as the substrates. The roughness of the bipolar plate was polished by various sandpapers. The value of the contact resistance increased gradually when the roughness was modified within 1–2 lm. The contact resistance increased sharply when the surface roughness was less than the 0.5 lm. Furthermore, this experiment was set at a potentiostatic mode of 800 mV and 500 mV in the cathodic condition. The SS 316 also possessed superior performance under the polarization test. Shoyama et al. [24] discussed the relation of the hydrophobicity and water management in the bipolar plate and GDL. The roughness involved the hydrophobicity in the flow channel significantly. It was observed that the coarse surface enhanced the surface tension. Furthermore, the abundant PTFE in the GDL would enhance the hydrophobicity. A better single-cell performance curve was measured when the hydrophobic GDL accompanied with the hydrophilic bipolar plate was set in high relative humidity. It was mentioned that the roughness of the bipolar plate affected the fuel cell significantly. The contact resistance involving the bipolar plate roughness and GDL porosity was discussed in many papers. There are few papers depicting the bipolar plate roughness involved in the fuel gas transfer mechanics. However, the stability of the fuel cell should be improved before it is fully developed. Therefore, the fuel gas transfer resistance and bipolar plate roughness are discussed in this paper.
2. Experiment 2.1. Surface roughness on the bipolar plate A single serpentine gas flow bipolar plate with an active area of 25 cm2 was manufactured from an Al alloy 5052 and machined as
Fig. 1. The morphology of 5052 Al-alloy bipolar plate.
shown in Fig. 1. The flow channel was machined as 0.9 mm in both depth and width. The smooth surface was clearly observed on the substrate (Ra = 0.2131 lm) in the inset. Roughness variations on the bulk surface were caused by the sandblasting method. The sand grain sizes were selected as 50, 100, and 200 lm. Furthermore, the bipolar plate after sandblasting with medium sand was immersed in sulfuric acid for 30 min in order to analyze the characteristic of the surface roughness after the corrosion. Various sandblasting methods involved not only the contact resistance effects but also the mass transfer mechanism. The relationship between the surface roughness and the sandblasting method is listed in Table 1. 2.2. Fuel cell configurations The manufactured PEMFC used in this experiment was assembled with commercial membranes (Nafion-112), sandblasted metallic bipolar plates, and Toray carbon paper (E-Tek). 180-lmthick gas diffusion layers were used as the base material in this study. The active area of the Nafion 112 membrane was 25 cm2 (5 cm 5 cm). The schematic diagram and photo of PEMFC apparatus were shown in Fig. 2. In the fuel cell experiments, the compaction force imparted on a cell was loaded as 200 N/cm2. 2.3. Electrochemical measurements The fuel gas was supplied with oxygen (140 sccm) in the cathode and hydrogen gas (210 sccm) in the anode. The temperature of the fully humid flow gas and cell were set at 60 °C. The range of the electric load was 0.85–0.1 V (voltage scan rate was 0.1 V/ h), and the limit current was set at 25 A. The fuel cell performance
Table 1 Change in the roughness of the bipolar plate with various degrees of sand. Part no.
Name of part
1 2 3 4 5
Substrate Fine sand Medium sand Coarse sand Medium sand with etching
Sizes of sand (lm)
Roughness (Ra) (lm)
50 100 200 100
0.213 1.521 1.892 2.932 1.2920
Etching
30 min
900
C.-H. Lin / Applied Energy 104 (2013) 898–904
Fig. 2. The schematic diagram of PEMFC apparatus. Fig. 4. Change in power density curve of the PEMFC with bipolar plate roughness.
was measured using AC impedance and performance (I–V) curves. Finally, the AC impedance curve was fitted by a mathematical model. The curve-fitting coefficient was used to assess the surface roughness effect on the metallic bipolar plates of a proton exchange membrane fuel cell.
3. Results and discussion 3.1. Single cell performance The experiments were performed under ambient pressure and temperature (333 K), as shown in Fig. 3. The PEMFC assembled with the substrate bipolar plate reached 0.45 V and 6.8 W under a current density of 600 mA cm2 (Figs. 3 and 4). Simultaneously, the PEMFC assembled with the bipolar plate after sandblasting with fine sand reached 0.62 V and 8.75 W under a current density of 600 mA cm2 (Figs. 3 and 4). It was noteworthy that the PEMFC performance significantly changed with the bulk surface after sandblasting with fine sand. This fine sand sandblasting treatment caused the bipolar plate and GDL to be well matched, which effectively lowered the contact resistance. However, unlimited roughness expansion may not always be the ideal method to lower the contact resistance. The PEMFC assembled with the bulk surface after sandblasting with either medium sand or coarse sand apparently caused lower power generation than the fine sand treatment. Certainly, the bipolar plate after sandblasting with coarse sand increased the fuel gas concentration loss to the PEMFC. The PEMFC
Fig. 3. Change in performance (I–V) curve of the PEMFC with bipolar plate roughness.
assembled with bipolar plates after sandblasting with medium sand and coarse sand reached 4.8 W and 4.3 W, respectively, under a current density at 500 mA/cm2. After immersing in 0.5 M sulfuric acid for 30 min, the bulk surface after sandblasting with medium sand has a lower roughness (Ra = 1.29 lm), (Table 1). The etching effect decrease PEMFC performance severely when it operated at high current density stage (Figs. 3 and 4). The roughness effect certainly influenced the contact resistance and the fuel gas mass transfer mechanism of the PEMFC. The smooth surface decreased the fuel gas transfer shearing force. The fuel gas transfer barrier is caused by the uneven surface. Oxygen starvation occurs because of the non-uniform distribution of the fuel gas in the flow channel. Hence, the optimum surface roughness is a trade off among the various component requirements of the PEMFC. The coarse sand blasting is a crux in this paper. It is helpful for us to know the failure mechanism. As Fig. 3, the polarization curve decays rapidly when PEMFC operates at high current load. The higher exhaustion rate at fuel gas as the rapid electrochemical reaction occurs. Evidently, the surface morphology significantly changes with the bipolar plate after sandblasting with coarse sand. The hydrophobicity which varies with bipolar plate roughness involves rapid water accumulation problem in the PEMFC. Hence, the coarse sand blasting method on bipolar plate causes the disadvantage to gas transfer ability.
3.2. Surface roughness analysis in the flow channel The morphology of the Al alloy bipolar plate in the flow channel was observed by SEM (Fig. 5). The smooth surface was clearly observed on the substrate (Ra = 0.2131 lm). Moreover, the Al alloy bipolar plate after sandblasting with coarse sand was uneven (Ra = 2.932 lm). There were many cavities formed on the surface of the flow channel. The etching treatment likely caused a smoother surface (Ra = 1.29 lm) than the surface after sandblasting with medium sand (Ra = 1.892 lm). The above results correspond to the surface roughness by the different sandblasting methods listed in Table 1, where it can be seen that the bipolar plate surface significantly influences the PEMFC performance. With regard to the contact resistance, the compact force of the single cell exerts pressure on the interface, which increases the contact area between the fuel cell components and decreases the contact resistance. However, an improper compact force may deform the GDL and decrease the PEMFC lifetime. The changes in the optimal compact force and bipolar plate surface roughness that decrease the contact resistance are discussed in some papers [16,25–27]. However, certain quantitative state-
901
C.-H. Lin / Applied Energy 104 (2013) 898–904
(a) Substrate
(c) Sandblasting with medium sand
(e) Sandblasting with medium sand and etching
(b) Sandblasting with fine sand
(d) Sandblasting with coarse sand
Fig. 5. The morphology of Al-alloy bipolar plate under the SEM.
ments of the bipolar plate roughness involving the fuel gas transfer mechanism are scarcely discussed. 3.3. AC impedance test Electrochemical impedance spectroscopy (EIS) is a powerful tool for analyzing the PEMFC operational behavior. There were many analogies between the electrochemical reaction mechanism and the electrical circuit operation. For impedance spectroscopy, a small perturbation voltage or current is externally applied to the PEMFC at different frequencies. The alternating current or voltage signal is measured by an AC impedance meter. A phase shift exists between the voltage and current to observe the resistance change. The electrochemical properties and the interface reaction research are available for the PEMFC. The contact resistance, the fuel gas transfer mechanism, and the anode and cathode interface loss are identified easily from the impedance information. 3.3.1. Nyquist plot The electrochemical impedance spectroscopy experiments were performed under ambient pressure, an electric load of 0.6 V, and a temperature 333 K. The impedance change in the Nyquist plot is shown in Fig. 6. The high frequency resistance of the PEMFC assembled with the bipolar plate after sandblasting with fine sand and the etching treatment was approximately 10.97 mX. The worse performance certainly occurred with the PEMFC assembled with the substrate bipolar plate (12.4 mX). Moreover, the bipolar plate performance after sandblasting with medium sand (13.5 mX) was the worst condition. It was concluded that the roughness of the bipolar plate caused the high frequency change of the PEMFC shown in Fig. 6. However, the fuel gas transfer resistance did not change significantly when the surface roughness was within a reasonable range. The high and low frequency resistance increased with the coarse surface when the roughness was greater than 2 lm on the bipolar plate. The high frequency resistance of the coarse surface was larger than that of the substrate by around 5 mX. Furthermore, a new arc was formed at the low frequency region. This indicates that the roughness affects not only the contact resistance but also the gas mass transfer. The interception line of the low frequency position was approximately 37.5 mX. It is
Fig. 6. Change in Nyquist curve of the PEMFC with bipolar plate roughness. (a) The roughness of the bipolar plate within 2 lm. (b) Sandblasting with coarse sand.
obvious that the gas mass transfer resistance of the coarse sand is higher than that in the other surface treatment methods. This
902
C.-H. Lin / Applied Energy 104 (2013) 898–904
phenomenon cannot be neglected, especially in the large current operation. The rapid electrochemical mechanism occurs when there is high fuel gas consumption at the cathode. The single cell performance is significantly affected by the surface roughness. 3.3.2. Bode plot The frequency behavior of the fuel cell performance can be interpreted clearly by the bode plot. The amplitude of the resistance is shown in Fig. 7. Obviously, the bipolar plate after sandblasting with fine sand has a small resistance within all frequency regions. The bipolar plate after sandblasting with coarse sand is significantly affected by the gas transfer mechanism. Similar to the high frequency region, the resistance of the bipolar plate after sandblasting with coarse sand is higher than that of the substrate by 7 mX. Furthermore, the low frequency resistance of this bipolar plate after sandblasting with coarse sand (37.5 mX) is obviously higher than that of the substrate (19 mX) at 0.1 Hz. The Warburg resistance was obtained at 0.05 Hz, indicating the gas transfer barrier formation. Moreover, the resistance of the bipolar plate after sandblasting with fine sand and medium sand and etching showed the same amplitude in all frequency regions. This result corresponds to the roughness value listed in Table 1. The phase angle of the resistance is shown in Fig. 8. Most fuel cells show the same phase angle when the bipolar plate surface roughness is less than 2 lm. As with the resistance of the bipolar plate after sandblasting with coarse sand, there is another arc formed at the low frequency region (1 Hz). This arc is attributed to the gas mass transfer barrier in the flow channel of the fuel cell. The increasing roughness leads to large shear force and worse mass transfer ability. 3.3.3. Fuel cell parameter fitting The data obtained from experiments were fitted by the mathematical model shown in Fig. 9. The ohmic resistance (R1) of a single cell represented as a sum of contributions from the wire/contact resistance of entity in PEMFC [20]. Moreover, R2 is the kinetic resistance of the electrochemical reaction in the anode. The C2 shows the electric double layer over the interference in the cathode. The other first-order circuit describes the electrochemical reaction process in the cathode. It is noteworthy that the Warburg resistance describes the nonlinear electrochemical reaction process in the fuel cell. Theoretically, it is attributed to the gas transfer barrier in the flow channel. The fitting data are recorded in Table 2. R1, R2, and R3 are plotted in Fig. 10. A smaller R1 is obtained from the bipolar plate after sandblasting and the etching treatment (11 mX). An opposite trend is observed for the bipolar plate after
Fig. 7. Change in the fuel cell impedance magnitude with frequency.
Fig. 8. Change in the fuel cell impedance phase with frequency.
sandblasting with coarse sand (17.4 mX). It is proved that parameter R1 involves the roughness variation of the contact resistance. R2 and R3 describe the electrochemical kinetic resistance in the anode and cathode, respectively [17]. Both of them show increased resistance with roughness variation. At the anode side, the contact regions at the interface of bipolar plate and gas diffusion layer causes the electronics transfer resistance of a PEMFC. At the cathode side, the bumpy surface decreases the hydrophobicity. Hence, the gas transfer resistance occurs in the cathode side owing to the water accumulation and decreases the feed rate of oxygen. Parameter R3 of the bipolar plate after sandblasting with coarse sand is larger than that of the substrate by 9 mX. Hence, it is certain that the resistance of the increasing R3 value is attributed to the nonuniform distribution of the oxygen gas in the flow channel. As Fig. 5 and Table 1, the medium sand with etching method caused the smoother surface than the pure medium sand. Hence, the minor change in R2 value can be obtained between the medium sand with etching and fine sand in Fig. 10. The variation of parameters C1 and C2 are plotted in Fig. 11. The largest capacitance obtained from this bipolar plate after sandblasting with coarse sand is not plotted and that differs by one order of magnitude from the others. The C1 and C2 fitting data for this bipolar plate after sandblasting and the etching treatment are recorded as 0.81 F and 0.28 F in Table 2, respectively. The sandblasting changes the morphology on the substrate and electrochemical reaction in PEMFC. The smoother plane enhances the electronics mobility and decreases electric double layer thickness. Hence, the significant change on capacitance is attributed to the oxygen film on the bipolar plate. The abundant electronics accumulate on the electric double layer surface and enhance the capacitance effect in the anode and cathode. The fitting parameter of Warburg is also recorded in Table 2. The largest Warburg value is obtained from this bipolar plate after sandblasting with coarse sand. Obviously, the coarse roughness enhances the shear force to the fuel gas and mass transfer barrier. The larger Warburg value indicates oxygen starvation in the fuel cell flow channel.
Fig. 9. Equivalent circuit of the fuel cell used for the fitting of the impedance measurement.
903
C.-H. Lin / Applied Energy 104 (2013) 898–904 Table 2 List of coefficients for curve-fitting expression of fuel cell model. Part no.
Name of part
R1 (mX)
R2 (mX)
R3 (mX)
C1 (F)
C2 (F)
Warburg
1 2 3 4 5
Substrate Fine sand Medium sand Coarse sand Medium sand with etching
12.4 11.2 13.8 17.4 10.8
4.3 3.9 6.2 7.8 4.2
2.2 1.1 2.4 11.3 1.1
0.5622 0.6717 0.5767 15.9780 0.8092
0.1181 0.2175 0.1158 0.4678 0.2788
1.82E-06 4.81E-06 3.03E-06 3.90E-04 3.07E-05
condition could severely decrease the PEMFC performance. Even the mass transfer resistance is much lower than the contact resistance. However, the mass transfer resistance cannot be neglected under the large load operation. The high frequency resistance of the coarse surface was larger than that of the substrate by around 5 mX. Furthermore, a new arc was formed at the low frequency region. Certainly, the mass transfer resistance can reduce the fuel cell performance and the stability of the system output power. The unlimited increasing surface roughness decreases the hydrophobicity of the bipolar plate and gas mass transfer ability of the fuel cell. The surface effect is important to the performance of the entire fuel cell. Hence, the roughness of the bipolar plate needs to be well modified before it is fully developed to improve the contact resistance and mass transfer resistance. It benefits us to quantify the contact and mass transfer resistance exists in the PEMFC. The proposed surface treatment improves the surface effect and promotes the implement of potential metallic bipolar plate in near future. Fig. 10. Change in curve-fitting parameters R1, R2 and R3 of the fuel cell.
References
Fig. 11. Change in curve-fitting parameters C1 and C2 of the fuel cell.
The fitting data are obtained from the AC impedance test. The largest contact resistance and mass transfer are obtained from the bipolar plate after sandblasting with coarse sand. Even the mass transfer resistance is much lower than the contact resistance. However, the mass transfer resistance cannot be neglected under the large load operation. Certainly, the mass transfer resistance reduces the fuel cell performance and the stability of the system output power. 4. Conclusions Various degrees of roughness on the bulk surface are caused by the sandblasting method. In this paper, we discuss the effect of the surface roughness on the metallic bipolar plate. The selected grain sizes of sand are 50, 100 and 200 lm. The fitting data are obtained from the AC impedance test. An improper surface modification
[1] Salameh MG. Can renewable and unconventional energy sources bridge the global energy gap in the 21st century. Appl Energy 2003;75:33–42. [2] Khan MJ, Iqbal MT. Analysis of a small wind-hydrogen stand-alone hybrid energy system. Appl Energy 2009;86(11):2429–42. [3] Connolly D, Lund H, Mathiesen BV, Leahy M. A review of computer tools for analysing the integration of renewable energy into various energy systems. Appl Energy 2010;87:1059–82. [4] Wang Y, Chen KS, Mishler J, Cho SC, Adroher XC. A review of polymer electrolyte membrane fuel cells: technology, applications, and needs on fundamental research. Appl Energy 2011;88:981–1007. [5] Barelli L, Bidini G, Gallorini F, Ottaviano A. Dynamic analysis of PEMFC-based CHP systems for domestic application. Appl Energy 2012;91:13–28. [6] Juul N, Meibom P. Road transport and power system scenarios for Northern Europe in 2030. Appl Energy 2012;92:573–82. [7] Hermann A, Chaudhuria T, Spagnol P. Bipolar plates for PEM fuel cells: a review. J Hydrogen Energy 2005;30:1297–302. [8] Wang CT, Hu YC, Zheng PL. Novel biometric flow slab design for improvement of PEMFC performance. Appl Energy 2010;87:1366–75. [9] Nama J, Chippar P, Kim W, Ju H. Numerical analysis of gas crossover effects in polymer electrolyte fuel cells (PEFCs). Appl Energy 2010;87:3699–709. [10] Perng SW, Wub HW. Non-isothermal transport phenomenon and cell performance of a cathodic PEM fuel cell with a baffle plate in a tapered channel. Appl Energy 2011;88:52–67. [11] Wu HW, Ku HW. The optimal parameters estimation for rectangular cylinders installed transversely in the flow channel of PEMFC from a three-dimensional PEMFC model and the Taguchi method. Appl Energy 2011;88:4879–90. [12] Nishimura A, Shibuya K, Morimoto A, Tanaka S, Hirota M, Nakamura Y, et al. Dominant factor and mechanism of coupling phenomena in single cell of polymer electrolyte fuel cell. Appl Energy 2012;90:73–9. [13] Kanezaki T, Li X, Baschuk JJ. Cross-leakage flow between adjacent flow channels in PEM fuel cells. J Power Sources 2006;162:415–25. [14] Cooper KR, Smith M. Electrical test methods for on-line fuel cell ohmic resistance measurement. J Power Sources 2006;160:1088–95. [15] Lee HS, Kim HJ, Kim SG, Ahn SH. Evaluation of graphite composite bipolar plate for PEM (proton exchange membrane) fuel cell: electrical, mechanical, and molding properties. J Mater Process Technol 2007;187:425–8. [16] Kraytsberg A, Auinat M, Ein-Eli Y. Reduced contact resistance of PEM fuel cell’s bipolar plates via surface texturing. J Power Sources 2007;164:697–703. [17] Yan X, Hou M, Sun L, Liang D, Shen Q, Xu H, et al. AC impedance characteristics of a 2 kW PEM fuel cell stack under different operating conditions and load changes. Int J Hydrogen Energy 2007;32:4358–64. [18] Lee Y, Cho E, Lee JH, Kim HJ, Lim TH, Oh IH, et al. Effects of purging on the degradation of PEMFCs operating with repetitive on/off cycles. J Electrochem Soc 2007;154(2):B194–200. [19] Shahryari A, Kamal W, Omanovic S. The effect of surface roughness on the efficiency of the cyclic potentiodynamic passivation (CPP) method in the
904
[20] [21]
[22]
[23]
C.-H. Lin / Applied Energy 104 (2013) 898–904 improvement of general and pitting corrosion resistance of 316LVM stainless steel. Mater Lett 2008;62:3906–9. Ishibashi Y, Nishikata A, Tsuru T. Electrochemical Impedance spectroscopy of PEM fuel cell with metal bipolar plates. J Electrochem Soc 2008;16(24):85–9. Barber M, Sun TS, Petrach E, Wang X, Zou Q. Contact mechanics approach to determine contact surface area between bipolar plates and current collector in proton exchange membrane fuel cells. J Power Sources 2008;185:1252–6. Avasarala B, Haldar P. Effect of surface roughness of composite bipolar plates on the contact resistance of a proton exchange membrane fuel cell. J Power Sources 2009;188:225–9. Antoni L, André J, Petit JP. Optimization of the conductivity and passivity properties of passive films on stainless steels used as bipolar plates in PEFC. J Electrochem Soc 2009;17(1):383–92.
[24] Shoyama M, Tomimura T, Mizutani S. Effect of hydrophilic properties of gas channel on water management of PEM fuel cells. J Electrochem Soc 2009;17(1):461–4. [25] Wang Y, Shakhshir SA, Li X. Development and impact of sandwich wettability structure for gas distribution media on PEM fuel cell performance. Appl Energy 2011;88(6):2168–75. [26] Lin CH, Tsai SY. An investigation of coated aluminium bipolar plates for PEMFC. Appl Energy 2012;100:81–92. [27] Yuan W, Tang Y, Yang X, Wang Z. Porous metal materials for polymer electrolyte membrane fuel cells – a review. Appl Energy 2012;94:309–29.