- Email: [email protected]
Design and investigation of dual-layer electrodes for proton exchange membrane fuel cells
SOSI-13003; No of Pages 6 Solid State Ionics xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Solid State Ionics journal homepage: www.elsevier.com/locate/ssi
Design and investigation of dual-layer electrodes for proton exchange membrane fuel cells Bote Zhao a, Liangliang Sun a, Ran Ran a, Zongping Shao a,b,⁎ a State Key Laboratory of Materials-Oriented Chemical Engineering, College of Chemistry & Chemical Engineering, Nanjing University of Technology, No.5 Xin Mofan Road, Nanjing 210009, PR China b College of Energy, Nanjing University of Technology, No.5 Xin Mofan Road, Nanjing 210009, PR China
a r t i c l e
i n f o
Article history: Received 8 May 2013 Received in revised form 20 August 2013 Accepted 25 August 2013 Available online xxxx Keywords: Catalyst-coated membranes Proton exchange membrane fuel cell Spray deposition Catalyst layers Water management
a b s t r a c t With an aim to develop a proton-exchange-membrane fuel cell (PEMFC) with improved water management, catalyst-coated membranes based upon Nafion 212 membrane with electrodes of dual-layer structure which consist of one hydrophilic layer of Pt/C + Nafion and one hydrophobic layer of Pt/C + PTFE arranged in a proper order, is specifically designed and successfully fabricated by a facile high-temperature spray deposition technique. Dual-layer structured anode and cathode are separately evaluated by electrochemical performance in single cells. Effect of relative thickness of the dual layers in the electrode on the cell performance is investigated. No improvement in cell performance is observed by adopting the dual-layer structure for the anode as compared to conventional anode with single hydrophilic catalyst layer. However, better cell performance is observed for the cell with dual-layer structured cathode, and the optimal cell reaches a peak power density of about 800 mW cm−2 at 50 °C with humidified hydrogen and oxygen as fuel and oxidant respectively. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Proton exchange membrane fuel cells (PEMFCs) have received considerable attention recently as a potential power source for portable, mobile, and stationary applications because of their numerous inherent advantages, such as high efficiency, high energy density, low emissions, and fast start-up and shut-down capability [1]. Nafion-type perfluorosulfonic acid polymers are the most widely used electrolytes in PEMFCs due to their high chemical, mechanical and thermal stability [2]. However, the currently used perfluorosulfonic acid polymers in PEMFCs must be well hydrated to maintain high proton conductivity. As a result, the proper water management is of critical importance for achieving high cell performance. For the catalyst layers (electrodes), both water drying and flooding will exert unfavorable influence on the cell performance. Triple phase boundaries (TPBs) reduce and proton transfer becomes difficult as water drying in the catalyst layers, while the effective catalytic sites will also decrease for the water flooding in catalyst layers due to blocked gas passages. There are several ways to control the water balance in PEMFCs including development of self-humidifying composite membranes [3–8], improvement of catalyst layer [9–20], and optimization of the gas diffusion layers (GDLs) [21–23]. Up to now, several researchers
⁎ Corresponding author at: State Key Laboratory of Materials-Oriented Chemical Engineering, College of Chemistry & Chemical Engineering, Nanjing University of Technology, No.5 Xin Mofan Road, Nanjing 210009, PR China. Tel.:+ 86 25 83172256; fax: +86 25 83172242. E-mail address: [email protected] (Z. Shao).
have successfully improved the water balance in PEMFCs by optimizing the component and structure of catalyst layers. Adding hydrophilic oxides to the anode catalyst layer turns out to be an effective way in improving cell performance under low humidification and high temperature operation conditions [9–12]. Han et al. designed a self-humidified anode by adding silica (20–30 nm) into Nafion matrix of anode [10], and they observed that the cell performances increased with the increase of silica content in anodes from 0 to 6 wt.% at 60 °C under the condition of without external humidification. Chao et al. investigated the performance of PEMFCs by adding Pt/TiO2 particle into the anode catalyst layer and they demonstrated the best cell performance was obtained with 5 wt.% Pt/TiO2 particle addition at temperatures of anode humidifier ranging from 25 to 75 °C [11]. Since water is produced at the cathode side during the fuel cell process, the water flooding at the cathode may happen. To improve water management of the cathode, some hydrophobic materials such as polytetrafluoroethylene (PTFE) [13] and dimethyl silicone oil (DSO) [14,15], were added into the cathode catalyst layer. Optimizing the structure of electrode is also effective in mediating the water drying or flooding problem of PEMFCs. For example, it was confirmed that low Nafion content near the GDLs is beneficial for oxygen diffusion and water removal [17]. Su et al. also demonstrated that better Pt utilization and mass transfer thus cell performance can be realized by an appropriate cathode structure. Previously, Zhang et al. prepared a gas diffusion electrode (GDE) that contains a cathode with the structure of dual-bonded catalyst layers [18,19], and better cathode performance was achieved than conventional cathode with single layer PTFE or ionomer-bonded catalyst. They also investigated the Nafion content in the ionomer-bonded layer and Nafion loading
0167-2738/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ssi.2013.08.025
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025
2
B. Zhao et al. / Solid State Ionics xxx (2013) xxx–xxx
in recast film between hydrophilic layer and hydrophobic layer, and found no obvious improvement in cell performance by changing Nafion content in the hydrophilic layer, while finding the cell without the recast film exhibited the best performance. However, they did not get optimal thicknesses of the respective layer for the best cell performance. Besides, there are two typical fabricating methods for PEMFC, the deposition of catalyst onto the GDL and the direct deposition of catalyst onto the electrolyte membrane surface. Due to their different structures, different water management may be required for them. The direct deposition of catalyst onto electrolyte surface has the advantages of high catalyst utilization. Previously we have successfully prepared a catalyst-coated membrane by direct high-temperature spray deposition of catalyst ink onto Nafion electrolyte membrane with high performance [24,25]. In this study, in order to optimize water management, catalyst-coated membranes (CCMs) with dual-layer electrodes including a hydrophobic Pt/C + PTFE catalyst layer and a hydrophilic Pt/C + Nafion catalyst layer were prepared by the high-temperature spray deposition technique. We evaluated the dual layer-structured anode and cathode separately by examining the electrochemical performance in single cells. A systematic study was also made on the effect of relative thickness of the dual layers while fixing Pt loading in whole electrode on cell performance, and an optimal layer thickness was developed. 2. Experimental 2.1. Preparation of the gas diffusion layer The gas diffusion layer (GDL), composed of a PTFE modified carbon paper substrate and a PTFE bonded carbon micro-porous layer (MPL), was fabricated as follows. Firstly, the commercial carbon paper (Toray Inc., Japan) was modified with PTFE to increase its water-proof performance. An aqueous solution with 1 wt.% PTFE was prepared by diluting 60 wt.% PTFE emulsion (DuPont-TE-3893) in proper amount of deionized water, the carbon paper was then soaked in the PTFE aqueous solution for 1–2 min and dried on a hot-plate at 120 °C, the steps were repeated until the weight of PTFE reached around 10 wt.% of the carbon paper, the carbon paper was thermally treated at 250 °C for 30 min to remove the dispersion agent contained in PTFE and further sintered in air at 340 °C for 30 min. Secondly, the MPL was deposited on the carbon paper to increase the gas diffusion properties. To deposit the MPL onto the carbon paper substrate, a homogeneous slurry of 60 wt.% PTFE, carbon powder (Vulcan XC-72), alcohol was first prepared by ultrasonic dispersing and ball milling, then the slurry was coated onto the carbon paper substrate. Finally, the whole GDL was finally treated at 250 °C for 30 min and then 340 °C for 30 min. The amount of MPL on the carbon paper substrate was approximately 3 mg cm−2 with the carbon to PTFE weight ratio of 4:1. 2.2. Fabrication of catalyst-coated membranes Catalyst-coated membranes were prepared by a technique based on high-temperature spray deposition of catalyst directly onto the Nafion membrane [24–26]. Before spray deposition of the catalyst layers, polymer electrolyte membranes (Nafion 212) were pre-treated under the standard procedure of 1 h in 5 wt.% H2O2 solution at 85 °C, 1 h in deionized water at 85 °C, 1 h in 0.5 M H2SO4 solution at 85 °C, and 1 h in deionized water at 85 °C, in sequence. After the treatment, the membranes were stored in deionized water before use. A catalyst ink was prepared by ultrasonic dispersing of commercial 40 wt.% Pt/C (Pearl Hydrogen Technology Co., Ltd, Shanghai) and 5 wt.% Nafion solution (DuPont 520) at the catalyst to Nafion weight ratio of 3:1 in isopropanol media for about 1 h. To deposit the catalyst layers onto the electrolyte membrane to form CCMs, the pretreated Nafion 212 membrane was fixed onto a flat Pyrex glass by high-temperature adhesive tape over a hot plate. After the hot plate was heated up to around 150 °C, the
Fig. 1. Schematic preparing process and structure of dual-layer structured electrode CCM.
catalyst ink was air-driven spray deposited onto the Nafion 212 membrane surface. The catalyst ink was wetted by a small amount of deionized water to prevent burning during the spray deposition. Most of the liquid in catalyst ink was vaporized before reaching the membrane, thus just allowing the catalyst particle deposited onto the electrolyte membrane surface. The spray was conducted in a zigzag manner at a deposition rate of 0.3–0.5 mL min−1. The same catalyst loading of 0.4 mg cm −2 is adopted for both electrodes. To prepare the CCMs with dual-layer structure of the electrode, as schematically shown in Fig. 1, a catalyst ink with Pt/C and PTFE was further sprayed on the hydrophilic catalyst layer (Pt/C and Nafion) on electrolyte membrane as the hydrophobic catalyst layer. The hydrophobic catalyst ink was prepared by ultrasonic dispersing 40 wt.% Pt/C with 60 wt.% PTFE suspension at the catalyst to PTFE weight ratio of 4:1 in isopropanol/water (4:1) media for 1 h. CCMs with different ratios of hydrophilic catalyst layer to hydrophobic catalyst layer thicknesses with fixed total Pt content were prepared and the corresponding parameters are listed in Table 1. 2.3. Fuel cell performance test and characterization CCMs with 3 × 3 cm2 active electrode area for both anode and cathode were sandwiched between two gas diffusion layers to form the PEMFC and assembled with Teflon gaskets and graphite blocks with a serpentine flow field for performance test. Hydrogen with purity higher than 99.99% was applied as the fuel at the flow rate of 250 mL min−1 [STP] and industrial grade oxygen as oxidant at the cathode side at the flow rate of 300 mL min−1 [STP]. Typically, the single cell was operated at 50 °C under ambient pressure and the reactant gas was humidified by Table 1 CCMs with different ratios of hydrophilic catalyst layer to hydrophobic catalyst layer thicknesses.
CCM-N CCM-A1 CCM-A2 CCM-A3 CCM-C1 CCM-C2 CCM-C3
Pt loading of anode (mg cm−2)
Pt loading of cathode (mg cm−2)
PTFE-bonded catalyst layer
Ionomer-bonded catalyst layer
PTFE-bonded catalyst layer
Ionomer-bonded catalyst layer
0 0.1 0.2 0.3 0 0 0
0.4 0.3 0.2 0.1 0.4 0.4 0.4
0 0 0 0 0.1 0.2 0.3
0.4 0.4 0.4 0.4 0.3 0.2 0.1
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025
B. Zhao et al. / Solid State Ionics xxx (2013) xxx–xxx
3
Fig. 2. SEM images of (a) surface morphology, (b) cross-sectional view of single catalyst layer CCM; (c) surface morphology, (d) cross-sectional view of dual-layer structured electrode CCM; and (e) surface morphology of MPL on GDL.
passing through a water bottle at 55 °C. When the cells were tested under no humidified conditions, they were first operated with dry N2 for 2–3 h. The gas flow rates were controlled by mass flow controllers while the gas flow directions on both the anode and cathode are of counter-flow mode. The I–V polarization curves were measured by applying an electronic load (ITECH, IT8514F) at potentiostat mode, and the data acquisition was conducted once the stable performance was reached. Electrochemical impedance spectroscopy (EIS) and cyclic
voltammetry (CV) were conducted using an advanced electrochemical system model PARSTAT 2273 from Princeton Applied Research. The EIS test was made at a cell voltage of 0.8 V, and the impedance spectra were recorded in the 100 mHz to 100 kHz frequency range with amplitude of 10 mV. The CV analysis was performed on single cells at 50 °C by supplying humidified nitrogen and hydrogen to the cathode (serving as working electrode) and anode (serving as counter and reference electrode), respectively. The gases were humidified by passing through a
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025
4
B. Zhao et al. / Solid State Ionics xxx (2013) xxx–xxx
water bottle at 55 °C. The CV test was recorded from 0.05 to 1.2 V at a scan rate of 50 mV s−1. The surface and cross-sectional morphology of the CCMs were characterized by an Environmental Scanning Electron Microscope (ESEM, QUATA-2000). The samples for cross-sectional morphology were prepared by assisting with liquid nitrogen. 3. Results and discussion 3.1. Cell morphology High-temperature spray deposition was turned out to be a facile way to prepare catalyst-coated membranes. The as-obtained CCMs with single layer catalyst or with dual-layer structured electrodes are flat without the appearance of local wrinkle. Fig. 2a is the surface morphology of a typical CCM with a single hydrophilic catalyst layer (CCM-N) observed by SEM. The deposited Pt/C + Nafion catalyst layer on the Nafion electrolyte membrane showed very homogenous in microstructure without the appearance of any crack throughout the entire catalyst layer. A flat surface is important to obtain a crack-free coating of the second hydrophobic Pt/C + PTFE layer. Fig. 2b shows the SEM image of a typical single catalyst layer CCM from the crosssectional view. The top Pt/C + Nafion electrode layer adhered to the Nafion electrolyte membrane pretty well without the appearance of any delamination, ensuring a low charge transfer polarization resistance during the operating process. In comparison, there were obvious cracks or delamination which appeared between the electrode and electrolyte membrane layers for the cells with the catalyst layer deposited onto GDL, as demonstrated by Zhang and Shi [18]. The thickness of the single catalyst layer with 0.4 mg cm−2 Pt loading is about 7 μm (Fig. 2b). The small electrode thickness is beneficial for water and gas diffusion in the electrode. In order to prepare the electrode with dual-layer structure, the hydrophobic Pt/C + PTFE layer was further deposited onto the Pt/C + Nafion layer surface in different thickness ratios by high-temperature spray. Fig. 2c & d show surface and cross-sectional view of a typical CCM with the dual-layer structure of the electrode. Here the two layers have the same Pt loading, i.e. 0.2 mg cm−2. The microstructure of the hydrophobic PTFE-bonded catalyst layer was still highly homogenous without the appearance of any crack observed from the surface. Furthermore, the hydrophobic Pt/C + PTFE catalyst layer (top layer) and the Pt/C + Nafion hydrophilic catalyst layer adhered to each other pretty well. The whole thickness of electrode with the dual-layer structure is also about 7 μm. The thicknesses of hydrophobic catalyst layer and hydrophilic catalyst layer are almost equal as around 3.5 μm. It was reported that modification of GDL layer could also achieve effective water management within the cell [21–23]. Fig. 2e shows the surface morphology of the coated micro-porous layer over the GDL observed by SEM. There are cracks formed across the hydrophobic layer with the size of larger than 100 μm. The inhomogeneous distribution and irregular shapes of the pores within the GDL layer alongside with the unchanged dimension of carbon paper caused large internal stress within the functional layer during the evaporation of organics. If Pt/C + PTFE layer was deposited on GDL, the local defects will also form in the catalyst layer. The appearance of large cracks would significantly affect the gas and liquid distribution within the electrode and thus the cell performance. 3.2. Cell performance with dual-layer structured anode Up to now, proper water management is still one of the biggest challenges for PEMFCs [27,28], due to the complicated water transportation in MEA. As illustrated in Fig. 3, many aspects contribute to the water transportation and production in PEMFCs: (1) water production at the cathode due to electrode reaction, (2) water supply from the humidified gas and water removal/evaporation by the gases at operating temperature, (3) water electro-osmotic drag due to the proton transfer, (4) water
Fig. 3. Various water transport phenomena that may occur in a typical PEMFC, symbol “Z” is a mark here.
back diffusion from the cathode to the anode because of the water concentration gradient, and (5) water transfer due to the difference of pressure between anode and cathode. When the water balance is disturbed, drying or flooding of the electrode will happen, leading to the decay of the cell performance. The anode with dual-layer structure was designed for two expectative functions. Firstly, the hydrophobic layer plays the same role as MPL to keep the liquid water in the inner hydrophilic layer from removing, especially under low humidification operation (as marked in the Fig. 3, symbol “Z”), which is helpful to increase the protonic conductivity of the anode. Secondly, the hydrophobic layer benefits gas transfer. However, as gas transfer becomes easy in the hydrophobic layer, the gases that go through the anode will also take water away in the hydrophilic layer easily in water vapor form to reduce the water content in the hydrophilic layer. Thus, the overall effect of the hydrophobic layer on the water retention in the hydrophilic layer is dependent on the competition of the above two functions. Since the hydrophobic layer itself has poor water retention, thus lower proton conductivity, it is not beneficial for the electrode reaction. The effect of the hydrophobic layer on the overall performance of the electrode is thus very complicated. It is very hard to predict the effect simply by modeling, thus we prepared different thicknesses of the two layers in anode as listed in Table 1 and the CCMs were evaluated by the electrochemical performance in a single cell. Fig. 4 shows the I–V and I–P curves of the cells with dual-layer structured anode of different thicknesses for the respective layer, the results of cell with the anode composed of single hydrophilic layer are also presented. For reasonable comparison, all the anodes have the same Pt loading of 0.4 mg cm−2. Under the externally humidifying condition at 55/55 °C (Tanode humidifier/Tcathode humidifier), similar cell performance was observed for the cells of CCM-N, CCM-A1 and CCM-A2 over the whole investigated current density range (0–2200 mA cm−2). The maximum power density of 720 mW cm−2 was reached for all three cells. However, the cell of CCM-A3 showed the worst performance, and its peak power density was only about 610 mW cm−2. The poor performance of CCM-A3 can be attributed to the poor proton conductivity of the thick hydrophobic layer as PTFE could not provide proper proton passages while the hydrophilic layer did not provide sufficient reaction sites due to the small thickness. As a whole, the dual catalyst structure of the anode had little effect on the cell performance under humidifying condition. It can be explained by the fact that hydrogen transfer in anode is easy due to high diffusivity of small hydrogen molecule, and water flooding in the anode is negligible under the anode operation conditions.
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025
B. Zhao et al. / Solid State Ionics xxx (2013) xxx–xxx
5
Fig. 4. I–V and I–P curves of the related cells (CCM-N, CCM-A1, CCM-A2 and CCM-A3) with different anodes and the same single hydrophilic layer cathode at 55/50/55 °C (Tanode humidifier/Tcell/Tcathode humidifier).
Fig. 6. I–V and I–P curves of the related cells (CCM-N, CCM-C1, CCM-C2 and CCM-C3) with different cathodes and the same single hydrophilic layer anode at 55/50/55 °C (Tanode humidifier/Tcell/Tcathode humidifier).
Under the non-humidifying condition, the water in the anode is mainly back diffused from the cathode due to the water concentration difference between the anode and the cathode. If ideal dual layers are designed and the first function (retaining the water in the anode hydrophilic layer) played the leading role, self-humidifying anode could be achieved. We then tested the cells with different thickness of the respective layers with dry gases. The cell performance of CCM-N, CCM-A1, and CCM-A2 operating at 50 °C with dry H2/O2 is shown in Fig. 5. Before the operation on dry gases, the cells were stabilized by first operating at 50 °C under the externally humidifying condition until a stable performance was obtained, then the anode gas was changed to dry N2 at 500 mL min−1 and flowed for 2–3 h. The fuel cells of CCM-N and CCM-A1 exhibited similar performance and a maximum power density of about 670 mW cm−2 was reached. The cell of CCM-A2 shows slightly worse performance with a peak power density of 630 mW cm−2. The above results suggest the second function is actually more competitive in the water management.
Among the electrode reactions, the oxygen reduction reaction (ORR) has a high reduction overpotential due to the difficulty of oxygen reduction and the low kinetic rate of this reaction, while the high reaction rate of the hydrogen oxidation reaction leads to a low oxidation overpotential in the performance condition of PEMFCs. Therefore, usually more Pt/C is required in catalyst layer of cathode in conventional
PEMFC. Optimization of the cathode structure can result in improvement of power output and catalyst utilization for ORR. As shown in Fig. 3, water continuously produces at the cathode during polarization due to the electrode reaction, which may cause the water flooding in the cathode layer. If excessive water fills the pores of the catalyst layer, the electrode reaction becomes mass transport limiting. From the above experimental results, it was demonstrated that the hydrophobic layer may facilitate the water released from the cathode; it is unfavorable for the water retention in the cathode. However, it is beneficial for the cathode to avoid water flooding. Thus, the dual-layer structure was also applied to the cathode. Fig. 6 shows the I–V and I–P curves of the related cells composed of single hydrophilic cathode and dual-layer structured cathode with different relative thickness of the hydrophobic and hydrophilic catalyst layers. The cells were operated under the externally humidifying condition at 55/55 °C (Tanode humidifier/Tcathode humidifier). The cells of CCM-C1 and CCM-C2 with dual-layer structured cathodes, in which the ratio of the thickness of the hydrophilic inner layer to the hydrophobic outer layer thicknesses is of 3:1 and 2:2 respectively, showed better performance than the normal fuel cell with single hydrophilic cathode. In particular, the cell of CCM-C1 exhibited current density of 1106 mA cm−2 at 0.6 V and the peak power density of about 800 mW cm−2, while the peak power density of the cell with single layer cathode was about 720 mW cm−2. With proper hydrophobic catalyst layer coated on the inner hydrophilic catalyst layer, the dual layer cathode optimizes the water balance effectively. The presence of hydrophobic layer near the
Fig. 5. I–V and I–P curves of the related cells (CCM-N, CCM-A1, CCM-A2) with different anodes and the same single hydrophilic layer cathode at 50 °C using dry H2/O2.
Fig. 7. Cyclic voltammograms of the related cells (CCM-N, CCM-C1, CCM-C2 and CCM-C3) with different cathodes and the same single hydrophilic layer anode.
3.3. Cell performance with dual-layer structured cathode
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025
6
B. Zhao et al. / Solid State Ionics xxx (2013) xxx–xxx
inner to outer layer thicknesses of 3:1 is turned out to the optimal structure, which exhibit the best cell performance. 4. Conclusions
Fig. 8. Impedance spectra of the related cells (CCM-N, CCM-C1, CCM-C2 and CCM-C3) with different cathodes and the same single hydrophilic layer anode at 55/50/55 °C (Tanode humidifier/Tcell/Tcathode humidifier) at 0.8 V.
GDL may enhance the oxygen diffusion inside the cathode layer. As a result, the cell performance improved. However, the peak power density of CCM-C3 was about 690 mW cm−2, which was slightly lower than the cell of CCM-N without a hydrophobic layer. This indicated that the hydrophobic Pt/C + PTFE layer in the cathode of CCM-C3 was too thick, and the active sites in Pt/C + Nafion layer were not sufficient due to its small layer thickness. The cyclic voltammograms of the cells with dual-layer structured cathode as working electrodes are shown in Fig. 7. ECA stands for the electrochemical active surface area where the catalyst contacts with proton conductor and the electron conductor [16]. The ECA decreased with the decrease of total Nafion content in the dual-layer structured electrode, which is consistent with the previous report [17,20]. Thus, we suggest that the high performance of CCM-C1, which has a lower ECA, is mainly attributed to the easy diffusion of oxygen through the cathode catalyst layer. To get further information about the effect of hydrophobic layer in the cathode on the cell performance, we measured the EIS of the cells. Fig. 8 shows the EIS of the corresponding cells in Nyquist plots, measured at 0.8 V at 50 °C by applying pure hydrogen and oxygen as the anode and cathode atmosphere respectively. All cells show only one impedance arc in the spectra. The high-frequency intercept (left-end) of the single impedance arc on the horizontal axis, RΩ, represents the total ohmic resistance of the cell. The diameter of the kinetic loop, Rct, corresponds to the charge transfer resistance of the ORR [29]. From the EIS plots, the high-frequency intercepts of horizontal axis indicated that ohmic resistance of the membrane with dual-layer cathode was less than that of the fuel cell with single hydrophilic cathode. This phenomenon could be explained by the increased back diffusion of water in the cathode to the membrane due to the hydrophobic catalyst layer [14]. By comparing the impedance arcs in the spectra of the corresponding cells, we found the charge transfer resistances Rct were in the sequence of Rct-CCM-C3 N Rct-CCM-C2 ≈ Rct-CCM-N N Rct-CCM-C1. The results were in accordance with the cell power output. The hydrophilic catalyst layer of CCM-C3 was too thin which may reduce the overall triple phase boundary length. The cathode structure of CCM-C1 with the ratio of
In this work, catalyst-coated membranes with dual-layer structured electrode were successfully fabricated by direct high-temperature spray deposition of catalyst ink onto Nafion membrane. The microstructure of the as-prepared electrodes is homogenous and free of cracks throughout the catalyst layer. The hydrophobic catalyst layer and the hydrophilic catalyst layer show pretty well adherence to each other. With the dual catalyst layers applied to the anode, there was no improvement in cell performance as compared to the cell with normal anode composed of single hydrophilic catalyst layer, under both humidifying and nonhumidifying operation conditions. But the dual catalyst layers cathode structures of CCM-C1 and CCM-C2 improve the cell performance. The maximum power density of about 800 mW cm−2 is achieved with the dual catalyst layers cathode of CCM-C1 (with the ratio of inner to outer layer thicknesses of 3:1). Acknowledgements This work was supported by the “National Science Foundation for Distinguished Young Scholars of China” under contract No. 51025209. References [1] B.C.H. Steele, A. Heinzel, Nature 414 (2001) 345–352. [2] F. Lufrano, V. Baglio, P. Staiti, A.S. Arico, V. Antonucci, J. Power Sources 179 (2008) 34–41. [3] Y.H. Liu, B.L. Yi, Z.G. Shao, L. Wang, D.M. Xing, H.M. Zhang, J. Power Sources 163 (2007) 807–813. [4] C. Wang, Z.X. Liu, Z.Q. Mao, J.M. Xu, K.Y. Ge, Chem. Eng. J. 112 (2005) 87–91. [5] M. Watanabe, H. Uchida, Y. Seki, M. Emori, P. Stonehart, J. Electrochem. Soc. 143 (1996) 3847–3852. [6] F.Q. Liu, B.L. Yi, D.M. Xing, J.R. Yu, Z.J. Hou, Y.Z. Fu, J. Power Sources 124 (2003) 81–89. [7] B. Yang, Y.Z. Fu, A. Manthiram, J. Power Sources 139 (2005) 170–175. [8] H. Uchida, Y. Ueno, H. Hagihara, M. Watanabe, J. Electrochem. Soc. 150 (2003) A57–A62. [9] U.H. Jung, K.T. Park, E.H. Park, S.H. Kim, J. Power Sources 159 (2006) 529–532. [10] M. Han, S.H. Chan, S.P. Jiang, Int. J. Hydrogen Energy 32 (2007) 385–391. [11] W.K. Chao, R.H. Huang, C.J. Huang, K.L. Hsueh, F.S. Shieu, J. Electrochem. Soc. 157 (2010) B1012–B1018. [12] J. Tian, G.Q. Sun, M. Cai, Q. Mao, Q. Xin, J. Electrochem. Soc. 155 (2008) B187–B193. [13] Z.Q. Tian, X.L. Wang, H.M. Zhang, B.L. Yi, S.P. Jiang, Electrochem. Commun. 8 (2006) 1158–1162. [14] A. Li, S.H. Chan, N.T. Nguyen, Electrochem. Commun. 11 (2009) 897–900. [15] A. Li, M. Han, S.H. Chan, N.T. Nguyen, Electrochim. Acta 55 (2010) 2706–2711. [16] W. Song, H.M. Yu, L.X. Hao, Z.L. Miao, B.L. Yi, Z.G. Shao, Solid State Ionics 181 (2010) 453–458. [17] H.N. Su, S.J. Liao, Y.N. Wu, J. Power Sources 195 (2010) 3477–3480. [18] X.W. Zhang, P.F. Shi, Electrochem. Commun. 8 (2006) 1229–1234. [19] X.W. Zhang, P.F. Shi, Electrochem. Commun. 8 (2006) 1615–1620. [20] K.H. Kim, H.J. Kim, K.Y. Lee, J.H. Jang, S.Y. Lee, E. Cho, I.H. Oh, T.H. Lim, Int. J. Hydrogen Energy 33 (2008) 2783–2789. [21] G.G. Park, Y.J. Sohn, T.H. Yang, Y.G. Yoon, W.Y. Lee, C.S. Kim, J. Power Sources 131 (2004) 182–187. [22] M. Ahn, Y.H. Cho, Y.H. Cho, J. Kim, N. Jung, Y.E. Sung, Electrochim. Acta 56 (2011) 2450–2457. [23] M.V. Williams, H.R. Kunz, J.M. Fenton, J. Power Sources 135 (2004) 122–134. [24] L.L. Sun, R. Ran, G.X. Wang, Z.P. Shao, Solid State Ionics 179 (2008) 960–965. [25] L.L. Sun, R. Ran, Z.P. Shao, Int. J. Hydrogen Energy 35 (2010) 2921–2925. [26] B.T. Zhao, J. Song, R. Ran, Z.P. Shao, Int. J. Hydrogen Energy 37 (2012) 1133–1139. [27] T. Van Nguyen, M.W. Knobbe, J. Power Sources 114 (2003) 70–79. [28] N. Yousfi-Steiner, P. Mocoteguy, D. Candusso, D. Hissel, A. Hernandez, A. Aslanides, J. Power Sources 183 (2008) 260–274. [29] X.Z. Yuan, H.J. Wang, J.C. Sun, J.J. Zhang, Int. J. Hydrogen Energy 32 (2007) 4365–4380.
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025
Contents lists available at ScienceDirect
Solid State Ionics journal homepage: www.elsevier.com/locate/ssi
Design and investigation of dual-layer electrodes for proton exchange membrane fuel cells Bote Zhao a, Liangliang Sun a, Ran Ran a, Zongping Shao a,b,⁎ a State Key Laboratory of Materials-Oriented Chemical Engineering, College of Chemistry & Chemical Engineering, Nanjing University of Technology, No.5 Xin Mofan Road, Nanjing 210009, PR China b College of Energy, Nanjing University of Technology, No.5 Xin Mofan Road, Nanjing 210009, PR China
a r t i c l e
i n f o
Article history: Received 8 May 2013 Received in revised form 20 August 2013 Accepted 25 August 2013 Available online xxxx Keywords: Catalyst-coated membranes Proton exchange membrane fuel cell Spray deposition Catalyst layers Water management
a b s t r a c t With an aim to develop a proton-exchange-membrane fuel cell (PEMFC) with improved water management, catalyst-coated membranes based upon Nafion 212 membrane with electrodes of dual-layer structure which consist of one hydrophilic layer of Pt/C + Nafion and one hydrophobic layer of Pt/C + PTFE arranged in a proper order, is specifically designed and successfully fabricated by a facile high-temperature spray deposition technique. Dual-layer structured anode and cathode are separately evaluated by electrochemical performance in single cells. Effect of relative thickness of the dual layers in the electrode on the cell performance is investigated. No improvement in cell performance is observed by adopting the dual-layer structure for the anode as compared to conventional anode with single hydrophilic catalyst layer. However, better cell performance is observed for the cell with dual-layer structured cathode, and the optimal cell reaches a peak power density of about 800 mW cm−2 at 50 °C with humidified hydrogen and oxygen as fuel and oxidant respectively. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Proton exchange membrane fuel cells (PEMFCs) have received considerable attention recently as a potential power source for portable, mobile, and stationary applications because of their numerous inherent advantages, such as high efficiency, high energy density, low emissions, and fast start-up and shut-down capability [1]. Nafion-type perfluorosulfonic acid polymers are the most widely used electrolytes in PEMFCs due to their high chemical, mechanical and thermal stability [2]. However, the currently used perfluorosulfonic acid polymers in PEMFCs must be well hydrated to maintain high proton conductivity. As a result, the proper water management is of critical importance for achieving high cell performance. For the catalyst layers (electrodes), both water drying and flooding will exert unfavorable influence on the cell performance. Triple phase boundaries (TPBs) reduce and proton transfer becomes difficult as water drying in the catalyst layers, while the effective catalytic sites will also decrease for the water flooding in catalyst layers due to blocked gas passages. There are several ways to control the water balance in PEMFCs including development of self-humidifying composite membranes [3–8], improvement of catalyst layer [9–20], and optimization of the gas diffusion layers (GDLs) [21–23]. Up to now, several researchers
⁎ Corresponding author at: State Key Laboratory of Materials-Oriented Chemical Engineering, College of Chemistry & Chemical Engineering, Nanjing University of Technology, No.5 Xin Mofan Road, Nanjing 210009, PR China. Tel.:+ 86 25 83172256; fax: +86 25 83172242. E-mail address: [email protected] (Z. Shao).
have successfully improved the water balance in PEMFCs by optimizing the component and structure of catalyst layers. Adding hydrophilic oxides to the anode catalyst layer turns out to be an effective way in improving cell performance under low humidification and high temperature operation conditions [9–12]. Han et al. designed a self-humidified anode by adding silica (20–30 nm) into Nafion matrix of anode [10], and they observed that the cell performances increased with the increase of silica content in anodes from 0 to 6 wt.% at 60 °C under the condition of without external humidification. Chao et al. investigated the performance of PEMFCs by adding Pt/TiO2 particle into the anode catalyst layer and they demonstrated the best cell performance was obtained with 5 wt.% Pt/TiO2 particle addition at temperatures of anode humidifier ranging from 25 to 75 °C [11]. Since water is produced at the cathode side during the fuel cell process, the water flooding at the cathode may happen. To improve water management of the cathode, some hydrophobic materials such as polytetrafluoroethylene (PTFE) [13] and dimethyl silicone oil (DSO) [14,15], were added into the cathode catalyst layer. Optimizing the structure of electrode is also effective in mediating the water drying or flooding problem of PEMFCs. For example, it was confirmed that low Nafion content near the GDLs is beneficial for oxygen diffusion and water removal [17]. Su et al. also demonstrated that better Pt utilization and mass transfer thus cell performance can be realized by an appropriate cathode structure. Previously, Zhang et al. prepared a gas diffusion electrode (GDE) that contains a cathode with the structure of dual-bonded catalyst layers [18,19], and better cathode performance was achieved than conventional cathode with single layer PTFE or ionomer-bonded catalyst. They also investigated the Nafion content in the ionomer-bonded layer and Nafion loading
0167-2738/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ssi.2013.08.025
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025
2
B. Zhao et al. / Solid State Ionics xxx (2013) xxx–xxx
in recast film between hydrophilic layer and hydrophobic layer, and found no obvious improvement in cell performance by changing Nafion content in the hydrophilic layer, while finding the cell without the recast film exhibited the best performance. However, they did not get optimal thicknesses of the respective layer for the best cell performance. Besides, there are two typical fabricating methods for PEMFC, the deposition of catalyst onto the GDL and the direct deposition of catalyst onto the electrolyte membrane surface. Due to their different structures, different water management may be required for them. The direct deposition of catalyst onto electrolyte surface has the advantages of high catalyst utilization. Previously we have successfully prepared a catalyst-coated membrane by direct high-temperature spray deposition of catalyst ink onto Nafion electrolyte membrane with high performance [24,25]. In this study, in order to optimize water management, catalyst-coated membranes (CCMs) with dual-layer electrodes including a hydrophobic Pt/C + PTFE catalyst layer and a hydrophilic Pt/C + Nafion catalyst layer were prepared by the high-temperature spray deposition technique. We evaluated the dual layer-structured anode and cathode separately by examining the electrochemical performance in single cells. A systematic study was also made on the effect of relative thickness of the dual layers while fixing Pt loading in whole electrode on cell performance, and an optimal layer thickness was developed. 2. Experimental 2.1. Preparation of the gas diffusion layer The gas diffusion layer (GDL), composed of a PTFE modified carbon paper substrate and a PTFE bonded carbon micro-porous layer (MPL), was fabricated as follows. Firstly, the commercial carbon paper (Toray Inc., Japan) was modified with PTFE to increase its water-proof performance. An aqueous solution with 1 wt.% PTFE was prepared by diluting 60 wt.% PTFE emulsion (DuPont-TE-3893) in proper amount of deionized water, the carbon paper was then soaked in the PTFE aqueous solution for 1–2 min and dried on a hot-plate at 120 °C, the steps were repeated until the weight of PTFE reached around 10 wt.% of the carbon paper, the carbon paper was thermally treated at 250 °C for 30 min to remove the dispersion agent contained in PTFE and further sintered in air at 340 °C for 30 min. Secondly, the MPL was deposited on the carbon paper to increase the gas diffusion properties. To deposit the MPL onto the carbon paper substrate, a homogeneous slurry of 60 wt.% PTFE, carbon powder (Vulcan XC-72), alcohol was first prepared by ultrasonic dispersing and ball milling, then the slurry was coated onto the carbon paper substrate. Finally, the whole GDL was finally treated at 250 °C for 30 min and then 340 °C for 30 min. The amount of MPL on the carbon paper substrate was approximately 3 mg cm−2 with the carbon to PTFE weight ratio of 4:1. 2.2. Fabrication of catalyst-coated membranes Catalyst-coated membranes were prepared by a technique based on high-temperature spray deposition of catalyst directly onto the Nafion membrane [24–26]. Before spray deposition of the catalyst layers, polymer electrolyte membranes (Nafion 212) were pre-treated under the standard procedure of 1 h in 5 wt.% H2O2 solution at 85 °C, 1 h in deionized water at 85 °C, 1 h in 0.5 M H2SO4 solution at 85 °C, and 1 h in deionized water at 85 °C, in sequence. After the treatment, the membranes were stored in deionized water before use. A catalyst ink was prepared by ultrasonic dispersing of commercial 40 wt.% Pt/C (Pearl Hydrogen Technology Co., Ltd, Shanghai) and 5 wt.% Nafion solution (DuPont 520) at the catalyst to Nafion weight ratio of 3:1 in isopropanol media for about 1 h. To deposit the catalyst layers onto the electrolyte membrane to form CCMs, the pretreated Nafion 212 membrane was fixed onto a flat Pyrex glass by high-temperature adhesive tape over a hot plate. After the hot plate was heated up to around 150 °C, the
Fig. 1. Schematic preparing process and structure of dual-layer structured electrode CCM.
catalyst ink was air-driven spray deposited onto the Nafion 212 membrane surface. The catalyst ink was wetted by a small amount of deionized water to prevent burning during the spray deposition. Most of the liquid in catalyst ink was vaporized before reaching the membrane, thus just allowing the catalyst particle deposited onto the electrolyte membrane surface. The spray was conducted in a zigzag manner at a deposition rate of 0.3–0.5 mL min−1. The same catalyst loading of 0.4 mg cm −2 is adopted for both electrodes. To prepare the CCMs with dual-layer structure of the electrode, as schematically shown in Fig. 1, a catalyst ink with Pt/C and PTFE was further sprayed on the hydrophilic catalyst layer (Pt/C and Nafion) on electrolyte membrane as the hydrophobic catalyst layer. The hydrophobic catalyst ink was prepared by ultrasonic dispersing 40 wt.% Pt/C with 60 wt.% PTFE suspension at the catalyst to PTFE weight ratio of 4:1 in isopropanol/water (4:1) media for 1 h. CCMs with different ratios of hydrophilic catalyst layer to hydrophobic catalyst layer thicknesses with fixed total Pt content were prepared and the corresponding parameters are listed in Table 1. 2.3. Fuel cell performance test and characterization CCMs with 3 × 3 cm2 active electrode area for both anode and cathode were sandwiched between two gas diffusion layers to form the PEMFC and assembled with Teflon gaskets and graphite blocks with a serpentine flow field for performance test. Hydrogen with purity higher than 99.99% was applied as the fuel at the flow rate of 250 mL min−1 [STP] and industrial grade oxygen as oxidant at the cathode side at the flow rate of 300 mL min−1 [STP]. Typically, the single cell was operated at 50 °C under ambient pressure and the reactant gas was humidified by Table 1 CCMs with different ratios of hydrophilic catalyst layer to hydrophobic catalyst layer thicknesses.
CCM-N CCM-A1 CCM-A2 CCM-A3 CCM-C1 CCM-C2 CCM-C3
Pt loading of anode (mg cm−2)
Pt loading of cathode (mg cm−2)
PTFE-bonded catalyst layer
Ionomer-bonded catalyst layer
PTFE-bonded catalyst layer
Ionomer-bonded catalyst layer
0 0.1 0.2 0.3 0 0 0
0.4 0.3 0.2 0.1 0.4 0.4 0.4
0 0 0 0 0.1 0.2 0.3
0.4 0.4 0.4 0.4 0.3 0.2 0.1
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025
B. Zhao et al. / Solid State Ionics xxx (2013) xxx–xxx
3
Fig. 2. SEM images of (a) surface morphology, (b) cross-sectional view of single catalyst layer CCM; (c) surface morphology, (d) cross-sectional view of dual-layer structured electrode CCM; and (e) surface morphology of MPL on GDL.
passing through a water bottle at 55 °C. When the cells were tested under no humidified conditions, they were first operated with dry N2 for 2–3 h. The gas flow rates were controlled by mass flow controllers while the gas flow directions on both the anode and cathode are of counter-flow mode. The I–V polarization curves were measured by applying an electronic load (ITECH, IT8514F) at potentiostat mode, and the data acquisition was conducted once the stable performance was reached. Electrochemical impedance spectroscopy (EIS) and cyclic
voltammetry (CV) were conducted using an advanced electrochemical system model PARSTAT 2273 from Princeton Applied Research. The EIS test was made at a cell voltage of 0.8 V, and the impedance spectra were recorded in the 100 mHz to 100 kHz frequency range with amplitude of 10 mV. The CV analysis was performed on single cells at 50 °C by supplying humidified nitrogen and hydrogen to the cathode (serving as working electrode) and anode (serving as counter and reference electrode), respectively. The gases were humidified by passing through a
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025
4
B. Zhao et al. / Solid State Ionics xxx (2013) xxx–xxx
water bottle at 55 °C. The CV test was recorded from 0.05 to 1.2 V at a scan rate of 50 mV s−1. The surface and cross-sectional morphology of the CCMs were characterized by an Environmental Scanning Electron Microscope (ESEM, QUATA-2000). The samples for cross-sectional morphology were prepared by assisting with liquid nitrogen. 3. Results and discussion 3.1. Cell morphology High-temperature spray deposition was turned out to be a facile way to prepare catalyst-coated membranes. The as-obtained CCMs with single layer catalyst or with dual-layer structured electrodes are flat without the appearance of local wrinkle. Fig. 2a is the surface morphology of a typical CCM with a single hydrophilic catalyst layer (CCM-N) observed by SEM. The deposited Pt/C + Nafion catalyst layer on the Nafion electrolyte membrane showed very homogenous in microstructure without the appearance of any crack throughout the entire catalyst layer. A flat surface is important to obtain a crack-free coating of the second hydrophobic Pt/C + PTFE layer. Fig. 2b shows the SEM image of a typical single catalyst layer CCM from the crosssectional view. The top Pt/C + Nafion electrode layer adhered to the Nafion electrolyte membrane pretty well without the appearance of any delamination, ensuring a low charge transfer polarization resistance during the operating process. In comparison, there were obvious cracks or delamination which appeared between the electrode and electrolyte membrane layers for the cells with the catalyst layer deposited onto GDL, as demonstrated by Zhang and Shi [18]. The thickness of the single catalyst layer with 0.4 mg cm−2 Pt loading is about 7 μm (Fig. 2b). The small electrode thickness is beneficial for water and gas diffusion in the electrode. In order to prepare the electrode with dual-layer structure, the hydrophobic Pt/C + PTFE layer was further deposited onto the Pt/C + Nafion layer surface in different thickness ratios by high-temperature spray. Fig. 2c & d show surface and cross-sectional view of a typical CCM with the dual-layer structure of the electrode. Here the two layers have the same Pt loading, i.e. 0.2 mg cm−2. The microstructure of the hydrophobic PTFE-bonded catalyst layer was still highly homogenous without the appearance of any crack observed from the surface. Furthermore, the hydrophobic Pt/C + PTFE catalyst layer (top layer) and the Pt/C + Nafion hydrophilic catalyst layer adhered to each other pretty well. The whole thickness of electrode with the dual-layer structure is also about 7 μm. The thicknesses of hydrophobic catalyst layer and hydrophilic catalyst layer are almost equal as around 3.5 μm. It was reported that modification of GDL layer could also achieve effective water management within the cell [21–23]. Fig. 2e shows the surface morphology of the coated micro-porous layer over the GDL observed by SEM. There are cracks formed across the hydrophobic layer with the size of larger than 100 μm. The inhomogeneous distribution and irregular shapes of the pores within the GDL layer alongside with the unchanged dimension of carbon paper caused large internal stress within the functional layer during the evaporation of organics. If Pt/C + PTFE layer was deposited on GDL, the local defects will also form in the catalyst layer. The appearance of large cracks would significantly affect the gas and liquid distribution within the electrode and thus the cell performance. 3.2. Cell performance with dual-layer structured anode Up to now, proper water management is still one of the biggest challenges for PEMFCs [27,28], due to the complicated water transportation in MEA. As illustrated in Fig. 3, many aspects contribute to the water transportation and production in PEMFCs: (1) water production at the cathode due to electrode reaction, (2) water supply from the humidified gas and water removal/evaporation by the gases at operating temperature, (3) water electro-osmotic drag due to the proton transfer, (4) water
Fig. 3. Various water transport phenomena that may occur in a typical PEMFC, symbol “Z” is a mark here.
back diffusion from the cathode to the anode because of the water concentration gradient, and (5) water transfer due to the difference of pressure between anode and cathode. When the water balance is disturbed, drying or flooding of the electrode will happen, leading to the decay of the cell performance. The anode with dual-layer structure was designed for two expectative functions. Firstly, the hydrophobic layer plays the same role as MPL to keep the liquid water in the inner hydrophilic layer from removing, especially under low humidification operation (as marked in the Fig. 3, symbol “Z”), which is helpful to increase the protonic conductivity of the anode. Secondly, the hydrophobic layer benefits gas transfer. However, as gas transfer becomes easy in the hydrophobic layer, the gases that go through the anode will also take water away in the hydrophilic layer easily in water vapor form to reduce the water content in the hydrophilic layer. Thus, the overall effect of the hydrophobic layer on the water retention in the hydrophilic layer is dependent on the competition of the above two functions. Since the hydrophobic layer itself has poor water retention, thus lower proton conductivity, it is not beneficial for the electrode reaction. The effect of the hydrophobic layer on the overall performance of the electrode is thus very complicated. It is very hard to predict the effect simply by modeling, thus we prepared different thicknesses of the two layers in anode as listed in Table 1 and the CCMs were evaluated by the electrochemical performance in a single cell. Fig. 4 shows the I–V and I–P curves of the cells with dual-layer structured anode of different thicknesses for the respective layer, the results of cell with the anode composed of single hydrophilic layer are also presented. For reasonable comparison, all the anodes have the same Pt loading of 0.4 mg cm−2. Under the externally humidifying condition at 55/55 °C (Tanode humidifier/Tcathode humidifier), similar cell performance was observed for the cells of CCM-N, CCM-A1 and CCM-A2 over the whole investigated current density range (0–2200 mA cm−2). The maximum power density of 720 mW cm−2 was reached for all three cells. However, the cell of CCM-A3 showed the worst performance, and its peak power density was only about 610 mW cm−2. The poor performance of CCM-A3 can be attributed to the poor proton conductivity of the thick hydrophobic layer as PTFE could not provide proper proton passages while the hydrophilic layer did not provide sufficient reaction sites due to the small thickness. As a whole, the dual catalyst structure of the anode had little effect on the cell performance under humidifying condition. It can be explained by the fact that hydrogen transfer in anode is easy due to high diffusivity of small hydrogen molecule, and water flooding in the anode is negligible under the anode operation conditions.
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025
B. Zhao et al. / Solid State Ionics xxx (2013) xxx–xxx
5
Fig. 4. I–V and I–P curves of the related cells (CCM-N, CCM-A1, CCM-A2 and CCM-A3) with different anodes and the same single hydrophilic layer cathode at 55/50/55 °C (Tanode humidifier/Tcell/Tcathode humidifier).
Fig. 6. I–V and I–P curves of the related cells (CCM-N, CCM-C1, CCM-C2 and CCM-C3) with different cathodes and the same single hydrophilic layer anode at 55/50/55 °C (Tanode humidifier/Tcell/Tcathode humidifier).
Under the non-humidifying condition, the water in the anode is mainly back diffused from the cathode due to the water concentration difference between the anode and the cathode. If ideal dual layers are designed and the first function (retaining the water in the anode hydrophilic layer) played the leading role, self-humidifying anode could be achieved. We then tested the cells with different thickness of the respective layers with dry gases. The cell performance of CCM-N, CCM-A1, and CCM-A2 operating at 50 °C with dry H2/O2 is shown in Fig. 5. Before the operation on dry gases, the cells were stabilized by first operating at 50 °C under the externally humidifying condition until a stable performance was obtained, then the anode gas was changed to dry N2 at 500 mL min−1 and flowed for 2–3 h. The fuel cells of CCM-N and CCM-A1 exhibited similar performance and a maximum power density of about 670 mW cm−2 was reached. The cell of CCM-A2 shows slightly worse performance with a peak power density of 630 mW cm−2. The above results suggest the second function is actually more competitive in the water management.
Among the electrode reactions, the oxygen reduction reaction (ORR) has a high reduction overpotential due to the difficulty of oxygen reduction and the low kinetic rate of this reaction, while the high reaction rate of the hydrogen oxidation reaction leads to a low oxidation overpotential in the performance condition of PEMFCs. Therefore, usually more Pt/C is required in catalyst layer of cathode in conventional
PEMFC. Optimization of the cathode structure can result in improvement of power output and catalyst utilization for ORR. As shown in Fig. 3, water continuously produces at the cathode during polarization due to the electrode reaction, which may cause the water flooding in the cathode layer. If excessive water fills the pores of the catalyst layer, the electrode reaction becomes mass transport limiting. From the above experimental results, it was demonstrated that the hydrophobic layer may facilitate the water released from the cathode; it is unfavorable for the water retention in the cathode. However, it is beneficial for the cathode to avoid water flooding. Thus, the dual-layer structure was also applied to the cathode. Fig. 6 shows the I–V and I–P curves of the related cells composed of single hydrophilic cathode and dual-layer structured cathode with different relative thickness of the hydrophobic and hydrophilic catalyst layers. The cells were operated under the externally humidifying condition at 55/55 °C (Tanode humidifier/Tcathode humidifier). The cells of CCM-C1 and CCM-C2 with dual-layer structured cathodes, in which the ratio of the thickness of the hydrophilic inner layer to the hydrophobic outer layer thicknesses is of 3:1 and 2:2 respectively, showed better performance than the normal fuel cell with single hydrophilic cathode. In particular, the cell of CCM-C1 exhibited current density of 1106 mA cm−2 at 0.6 V and the peak power density of about 800 mW cm−2, while the peak power density of the cell with single layer cathode was about 720 mW cm−2. With proper hydrophobic catalyst layer coated on the inner hydrophilic catalyst layer, the dual layer cathode optimizes the water balance effectively. The presence of hydrophobic layer near the
Fig. 5. I–V and I–P curves of the related cells (CCM-N, CCM-A1, CCM-A2) with different anodes and the same single hydrophilic layer cathode at 50 °C using dry H2/O2.
Fig. 7. Cyclic voltammograms of the related cells (CCM-N, CCM-C1, CCM-C2 and CCM-C3) with different cathodes and the same single hydrophilic layer anode.
3.3. Cell performance with dual-layer structured cathode
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025
6
B. Zhao et al. / Solid State Ionics xxx (2013) xxx–xxx
inner to outer layer thicknesses of 3:1 is turned out to the optimal structure, which exhibit the best cell performance. 4. Conclusions
Fig. 8. Impedance spectra of the related cells (CCM-N, CCM-C1, CCM-C2 and CCM-C3) with different cathodes and the same single hydrophilic layer anode at 55/50/55 °C (Tanode humidifier/Tcell/Tcathode humidifier) at 0.8 V.
GDL may enhance the oxygen diffusion inside the cathode layer. As a result, the cell performance improved. However, the peak power density of CCM-C3 was about 690 mW cm−2, which was slightly lower than the cell of CCM-N without a hydrophobic layer. This indicated that the hydrophobic Pt/C + PTFE layer in the cathode of CCM-C3 was too thick, and the active sites in Pt/C + Nafion layer were not sufficient due to its small layer thickness. The cyclic voltammograms of the cells with dual-layer structured cathode as working electrodes are shown in Fig. 7. ECA stands for the electrochemical active surface area where the catalyst contacts with proton conductor and the electron conductor [16]. The ECA decreased with the decrease of total Nafion content in the dual-layer structured electrode, which is consistent with the previous report [17,20]. Thus, we suggest that the high performance of CCM-C1, which has a lower ECA, is mainly attributed to the easy diffusion of oxygen through the cathode catalyst layer. To get further information about the effect of hydrophobic layer in the cathode on the cell performance, we measured the EIS of the cells. Fig. 8 shows the EIS of the corresponding cells in Nyquist plots, measured at 0.8 V at 50 °C by applying pure hydrogen and oxygen as the anode and cathode atmosphere respectively. All cells show only one impedance arc in the spectra. The high-frequency intercept (left-end) of the single impedance arc on the horizontal axis, RΩ, represents the total ohmic resistance of the cell. The diameter of the kinetic loop, Rct, corresponds to the charge transfer resistance of the ORR [29]. From the EIS plots, the high-frequency intercepts of horizontal axis indicated that ohmic resistance of the membrane with dual-layer cathode was less than that of the fuel cell with single hydrophilic cathode. This phenomenon could be explained by the increased back diffusion of water in the cathode to the membrane due to the hydrophobic catalyst layer [14]. By comparing the impedance arcs in the spectra of the corresponding cells, we found the charge transfer resistances Rct were in the sequence of Rct-CCM-C3 N Rct-CCM-C2 ≈ Rct-CCM-N N Rct-CCM-C1. The results were in accordance with the cell power output. The hydrophilic catalyst layer of CCM-C3 was too thin which may reduce the overall triple phase boundary length. The cathode structure of CCM-C1 with the ratio of
In this work, catalyst-coated membranes with dual-layer structured electrode were successfully fabricated by direct high-temperature spray deposition of catalyst ink onto Nafion membrane. The microstructure of the as-prepared electrodes is homogenous and free of cracks throughout the catalyst layer. The hydrophobic catalyst layer and the hydrophilic catalyst layer show pretty well adherence to each other. With the dual catalyst layers applied to the anode, there was no improvement in cell performance as compared to the cell with normal anode composed of single hydrophilic catalyst layer, under both humidifying and nonhumidifying operation conditions. But the dual catalyst layers cathode structures of CCM-C1 and CCM-C2 improve the cell performance. The maximum power density of about 800 mW cm−2 is achieved with the dual catalyst layers cathode of CCM-C1 (with the ratio of inner to outer layer thicknesses of 3:1). Acknowledgements This work was supported by the “National Science Foundation for Distinguished Young Scholars of China” under contract No. 51025209. References [1] B.C.H. Steele, A. Heinzel, Nature 414 (2001) 345–352. [2] F. Lufrano, V. Baglio, P. Staiti, A.S. Arico, V. Antonucci, J. Power Sources 179 (2008) 34–41. [3] Y.H. Liu, B.L. Yi, Z.G. Shao, L. Wang, D.M. Xing, H.M. Zhang, J. Power Sources 163 (2007) 807–813. [4] C. Wang, Z.X. Liu, Z.Q. Mao, J.M. Xu, K.Y. Ge, Chem. Eng. J. 112 (2005) 87–91. [5] M. Watanabe, H. Uchida, Y. Seki, M. Emori, P. Stonehart, J. Electrochem. Soc. 143 (1996) 3847–3852. [6] F.Q. Liu, B.L. Yi, D.M. Xing, J.R. Yu, Z.J. Hou, Y.Z. Fu, J. Power Sources 124 (2003) 81–89. [7] B. Yang, Y.Z. Fu, A. Manthiram, J. Power Sources 139 (2005) 170–175. [8] H. Uchida, Y. Ueno, H. Hagihara, M. Watanabe, J. Electrochem. Soc. 150 (2003) A57–A62. [9] U.H. Jung, K.T. Park, E.H. Park, S.H. Kim, J. Power Sources 159 (2006) 529–532. [10] M. Han, S.H. Chan, S.P. Jiang, Int. J. Hydrogen Energy 32 (2007) 385–391. [11] W.K. Chao, R.H. Huang, C.J. Huang, K.L. Hsueh, F.S. Shieu, J. Electrochem. Soc. 157 (2010) B1012–B1018. [12] J. Tian, G.Q. Sun, M. Cai, Q. Mao, Q. Xin, J. Electrochem. Soc. 155 (2008) B187–B193. [13] Z.Q. Tian, X.L. Wang, H.M. Zhang, B.L. Yi, S.P. Jiang, Electrochem. Commun. 8 (2006) 1158–1162. [14] A. Li, S.H. Chan, N.T. Nguyen, Electrochem. Commun. 11 (2009) 897–900. [15] A. Li, M. Han, S.H. Chan, N.T. Nguyen, Electrochim. Acta 55 (2010) 2706–2711. [16] W. Song, H.M. Yu, L.X. Hao, Z.L. Miao, B.L. Yi, Z.G. Shao, Solid State Ionics 181 (2010) 453–458. [17] H.N. Su, S.J. Liao, Y.N. Wu, J. Power Sources 195 (2010) 3477–3480. [18] X.W. Zhang, P.F. Shi, Electrochem. Commun. 8 (2006) 1229–1234. [19] X.W. Zhang, P.F. Shi, Electrochem. Commun. 8 (2006) 1615–1620. [20] K.H. Kim, H.J. Kim, K.Y. Lee, J.H. Jang, S.Y. Lee, E. Cho, I.H. Oh, T.H. Lim, Int. J. Hydrogen Energy 33 (2008) 2783–2789. [21] G.G. Park, Y.J. Sohn, T.H. Yang, Y.G. Yoon, W.Y. Lee, C.S. Kim, J. Power Sources 131 (2004) 182–187. [22] M. Ahn, Y.H. Cho, Y.H. Cho, J. Kim, N. Jung, Y.E. Sung, Electrochim. Acta 56 (2011) 2450–2457. [23] M.V. Williams, H.R. Kunz, J.M. Fenton, J. Power Sources 135 (2004) 122–134. [24] L.L. Sun, R. Ran, G.X. Wang, Z.P. Shao, Solid State Ionics 179 (2008) 960–965. [25] L.L. Sun, R. Ran, Z.P. Shao, Int. J. Hydrogen Energy 35 (2010) 2921–2925. [26] B.T. Zhao, J. Song, R. Ran, Z.P. Shao, Int. J. Hydrogen Energy 37 (2012) 1133–1139. [27] T. Van Nguyen, M.W. Knobbe, J. Power Sources 114 (2003) 70–79. [28] N. Yousfi-Steiner, P. Mocoteguy, D. Candusso, D. Hissel, A. Hernandez, A. Aslanides, J. Power Sources 183 (2008) 260–274. [29] X.Z. Yuan, H.J. Wang, J.C. Sun, J.J. Zhang, Int. J. Hydrogen Energy 32 (2007) 4365–4380.
Please cite this article as: B. Zhao, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.08.025