Dynamic Characteristics of a Direct Methanol Fuel Cell

Transcription

1 Maohai Wang 1 Hang Guo Chongfang Ma Enhanced Heat Transfer and Energy Conservation Key Lab of Ministry of Education, College of Environmental and Energy Engineering, Beijing University of Technology, Beijing , China and Heat Transfer and Energy Conversion Key Lab of Beijing Municipality, College of Environmental and Energy Engineering, Beijing University of Technology, Beijing , China Dynamic Characteristics of a Direct Methanol Fuel Cell The detailed dynamic characteristics of direct methanol fuel cells need to be known if they are used for transportable power sources. The dynamic response of a direct methanol fuel cell to variable loading conditions, the effect of cell temperature and oxygen flow rate on the cell response, and the cell response to continuously varying cell temperatures were examined experimentally. The results revealed that the cell responds rapidly to variable current cycles and to continuously varying cell temperatures. The increasing rate of gradual loading significantly influences the dynamic behavior. The effects of cell temperature and oxygen flow rate on the cell dynamic responses are considerable, but the cell voltage differences over the range of cell temperatures and oxygen flow rates are small for gradual loading. The cell response value to cell temperature during decreasing temperature is lower than that during increasing temperature. DOI: / Keywords: liquid feed direct methanol fuel cell, dynamic characteristics, operating conditions, continuously varying temperature 1 Introduction The direct methanol fuel cell DMFC is a promising transportable power source 1. In practice, however, the output load and environmental conditions under which a DMFC operates are not constant. Thus, in order to provide references for the practical manufacture and operation of DMFC, it is necessary to know the dynamic response of the DMFC under variable loading and different operating conditions. K. Scott and his partners 2 8 investigated experimentally and numerically the influences of variable loadings and different operational conditions on dynamic responses of a single DMFC and small DMFC stack. J. Kallo et al. reported the transient behaviors of a gas-fed direct methanol fuel cell 9. But, the research on the transient performances of DMFC is still sparse when compared to research on the steady-state performances. In published papers, the types of applied current cycles are limited and the effects of temperature and oxygen flux on the dynamic response were not reported. More detailed study on the dynamic characteristics of DMFC is needed if the DMFC is to be used for practical application. In this paper, we report on the cell dynamic characteristics of a single DMFC under different loading cycles and under different operating conditions. Also, the cell dynamic responses to continuously varying cell temperatures were explored. Except for anode concentration, flow rate, and cathode pressure 3, the cell temperature and oxygen flow rate also significantly affect cell performance Thus, the influences of cell temperature and oxygen flow rate on the cell dynamic response under variable current cycles were also investigated in our work. 1 Corresponding author. maohaiwang@gmail.com Manuscript received August 11, 2005; final manuscript received October 19, Review conducted by Prabhakar Singh. 2 Experimental The liquid-feed DMFC test system is shown in Fig. 1. A methanol/water solution was used for the anode fuel, and oxygen was used for the cathode oxidant. The flow rates of the methanol solution and oxygen were controlled by pumps and mass flow controllers, respectively. A dew point humidifier was used to humidify oxygen and a bypass loop with a change valve was designed to supply dry oxygen. Both the methanol/water solution and oxygen were preheated with heating tapes in inlet tubes. Two buffers were used in cell cathode inlet and cell anode inlet to alleviate the flow and pressure fluctuation of reactants. Two gasliquid separators were placed in the anode and cathode outlets of the cell. A temperature and heating control unit was applied to heat and control the cell temperature. The pressures of cell inlet and cell outlet were measured with both pressure transmitters and differential pressure transmitters at both the anode and cathode sides. The operating pressures of the cell could be maintained by regulating back pressure valves. An ARBIN FCTS LNR was used to supply variable electronic loads to the cell and to record cell responses, cell temperatures, flow rates, and inlet temperatures of reactants. The materials of the loop and connector were stainless steel and perfluoroalkoxy. The single DMFC used in the experiments has an active area of 50 cm 2. The membrane electrode assembly MEA, the core part of a DMFC, consisted of two catalyst layers, two diffusion layers, and one polymer electrolyte membrane. The single cell was fitted with one MEA sandwiched between two graphite plates. Singlepath serpentine channels were cut in graphite plates for methanol and oxygen distribution. The hydraulic diameter of the channel was 0.8 mm. An alloy consisting of platinum and ruthenium was used for the anode catalyst Pt/Ru loading: 0.4 mg/cm 2 and carbon-supported platinum was adopted for the cathode catalyst Pt loading: 0.4 mg/cm 2. Teflonized carbon cloth acted as the diffusion layer. The polymer electrolyte membrane was Nafion Results and Discussion As discussed in published papers, the transient characteristics of a DMFC are mainly caused by the interaction between electrochemical reactions and transient mass transfer inside the cell. The factors related to dynamic response of DMFC include. response of electrochemical reactions, methanol crossover through the solid polymer electrolyte, the mass transfer of reactants to catalyst layers, the production of carbon dioxide and water and their release from anode and cathode catalyst layers, respectively, the two-phase flow of methanol solution and carbon dioxide gas through the anode diffusion layer, 202 / Vol. 3, MAY 2006 Copyright 2006 by ASME Transactions of the ASME

2 Fig. 1 Test rig of liquid-feed direct methanol fuel cell the hydrodynamics of the two-phase flow of methanol solution and carbon dioxide gas in the flow bed, the variation in temperature response of the DMFC, the two-phase flow of liquid water and oxygen gas through cathode diffusion layer 2 4, the double layer capacity and the charge transfer resistance of the anode and cathode 9. The factors presented above will interact and their influences on the dynamic response of the DMFC depend on the shape of the load change and the operating conditions applied to the cell Cell Response to Variable Current Cycles Under Fixed Operating Conditions. Loading cycles, which have not been discussed in published papers, were applied to the direct methanol fuel cell under constant operating conditions cell temperature: 80 C, methanol solution concentration: 1.0 M, flux of methanol solution: 10 ml min 1, flux of pure oxygen: 70 ml min 1. Each cycle was tested several times to ensure repeatability and each constant loading inside a load cycle was repeated three times to validate the influence of the loading history on the cell dynamic response. A current cycle, shown in Fig. 2, was used to examine the effect of a constant loading period on the cell response under pulsated loading. The loading periods were 60, 120, and 180 s, respectively and the rest time was kept at 60 s. As can be seen in Fig. 2, the voltage decays slightly with the loading period. The reason is that the time is not enough for the cell to reach the steady-state balance between the electrochemical reactions and heat/mass transfer. Once the cell is loaded, more and more carbon dioxide and water are produced on the anode and cathode sides, respectively, and will impede the transportation of methanol and oxygen to the catalyst layers. Figure 3 shows the cell response to a triangle -shaped current followed by a rest period of 10 s and the influence of increasing rate of current on the cell response. The voltage responds rapidly to current change. During the period of increasing current, the decline of responded cell voltage becomes slow suddenly, followed by an initially linear fall. There is a clear turning point in responded voltage. The turning point stands for the transition from a region of activation polarization to a region of ohmic polarization in the galvanostatic polarization curves 13. On the other hand, the cell responds to the current cycle more smoothly during decreasing current and no clear turning point can be found. Three rates of steady increase in current were applied: 0.5, 0.2, and 0.1 A/s. As shown in Fig. 3 a, the open circuit voltage decreases slightly with the decrease of change rate in the current. The reason is that the cell has more time at a lower rate to set up a balance Fig. 2 The influence of constant loading period on cell voltage response under pulsated loading Journal of Fuel Cell Science and Technology MAY 2006, Vol. 3 / 203

3 Fig. 5 Cell voltage response to an instantaneous loading followed by gradual unloading Fig. 3 Effect of increasing/decreasing loading slope on the cell voltage response Fig. 4 Cell voltage response under trapeziform current loading between the electrochemical reactions and the mass transfer. The voltage value at the turning point is lower at higher slope of increasing current. This can be seen distinctly from Fig. 3 b. Figure 3 b also shows that the lower slope will lead to a smaller current and longer time where turning takes place. As predicted, a lower rate of change of load results in a slower fall or rise of the cell voltage. Figure 4 depicts cell response to a trapezium-shaped current load. The slope of increasing/decreasing current is 0.1 A/ s, the period of constant loading is 60 s, and the rest time is 20 s. Compared with constant loading after instantaneous loading, the cell voltage response is more stable under constant loading after gradual loading. One of the causes is that a gradual increasing current gives the reactants more time to be transferred to catalyst layers and achieve a transient balance. The same phenomena as depicted by Fig. 3 occurred during continuously loading and unloading period. Figure 5 shows the response of the cell to a sudden loading followed by gradual unloading and a relaxation period of 30 s prior reloading. As can be seen, the cell voltage elevates to a higher value promptly after an initial decline, and then responds smoothly to gradual unloading of the cell current. This behavior is due to the complex interaction between the electrochemical reactions and mass transfer inside the fuel cell. The electrochemical processes are faster than the mass transfer processes. Continuously varying load results in the mass transfer processes being unable to keep up with the electrochemical processes. So, a mass transfer limitation takes place in the catalyst layer when the cell is loaded suddenly, which causes a lower initial value of cell voltage. The open circuit voltage of the cell decreases slightly with the decrease of slope of unloading. 3.2 Influence of Cell Temperature on the Cell Dynamic Response. Figure 6 shows the influences of cell temperature on the cell voltage response. Three different cell temperatures 80 C, 70 C, 60 C were examined under variable types of loads methanol solution concentration: 1.0 M, flux of methanol solution: 10 ml min 1, flux of pure oxygen: 700 ml min 1. Generally, the higher cell temperature gives higher voltage response. Increasing the cell temperature can boost the activity of the catalysts, can enhance the mass transfer of reactants, and can benefit proton transportation through the Nafion membrane 14. Figure 6 a demonstrates the effect of cell temperature on the cell voltage responses under pulsating load conditions. An important point is that the voltage reaches the steady state faster at a higher cell temperature after unloading, which follows from the fact that the mass transfer processes inside the cell are faster at the higher cell temperature. The time to reach the balance between electrochemical reactions and physics processes is also shortened. Figure 6 b shows the cell voltage responses to continually increasing loading followed by a sudden unloading under different cell temperatures. Three increasing slopes of current load 0.5, 0.2, and 0.1 A/s were set up in the load cycle. As shown, the differences in voltage over the range of cell temperatures increase with the magnitude of current during continuously loading and increase with a decreasing slope of increasing current. Figure 6 c illustrates the cell voltage responses at different cell temperatures under gradual loading followed by constant loading 204 / Vol. 3, MAY 2006 Transactions of the ASME

4 Fig. 6 The effect of cell temperature on the voltage response under variable loads Fig. 7 The effect of oxygen flux on the voltage response under variable loads and a sudden unloading. Different increasing slopes of current were also investigated in Fig. 6 c. The affect of increasing the slope of the current on the dynamic behavior is in accord with that shown in Fig. 6 b. As can be seen, the difference of the responding voltage under constant loading is greater than that under gradually increasing loading. 3.3 Influence of Oxygen Flux on the Cell Dynamic Response. The effect of oxygen flux on the cell voltage response to variable current cycles is demonstrated in Fig. 7. Three oxygen fluxes 1000, 700, and 400 ml min 1 were tested under different current cycles. Operating parameters, except for the oxygen flux, were kept constant: methanol solution concentration: 1.0 M, flux of methanol solution: 10 ml min 1, cell temperature: 80 C. Generally, the cell response voltage during loading is greater at a higher oxygen flux. Elevating the oxygen flux can enhance the transportation of oxygen to the catalyst layer, can improve the ability of its reaction with the catalyst and impede the reaction of the methanol crossed over from the anode side, and can speed up the discharge of water produced on the cathode side. The open circuit voltage of the cell increases with the oxygen flow rate. One of the factors contributing to this phenomenon is that the higher oxygen flux reduces the mixed potential caused by the methanol crossover. Figure 7 a depicts the cell voltage responses at different oxygen fluxes under pulsated loading conditions. As can be seen from Fig. 7 a, the best performance is achieved at the oxygen flux of 1000 ml min 1. The oxygen flux seems to have a posi- Journal of Fuel Cell Science and Technology MAY 2006, Vol. 3 / 205

5 Fig. 8 Cell open circuit voltage response to continue change of cell temperature Fig. 9 Cell response to continue change of cell temperature under constant voltage of 0.2 V tive influence on the cell response. The cell performs with a faster and more stable response when operated with a higher oxygen flow rate. Figure 7 b shows the effect of oxygen flux on the dynamic response under gradual loading followed by a sudden unloading. As shown in Fig. 7 b, the difference in cell voltage over the range of oxygen flow rates is small. Compared with constant loading, the effect of oxygen flux on the cell response is small for gradual loading. This is clearly described in Fig. 7 c, in which the differences in voltage under constant loading are bigger than that under gradual loading. 3.4 Cell Response to Continuously Varying Cell Temperature. The cell dynamic response to continuously varying cell temperatures is depicted in Figs. 8 and 9. When direct methanol fuel cells are used for transportable power sources, it is impossible to keep the cell temperature constant. So, it is essential to know the cell performance at varying cell temperatures. The cell was heated to a value of temperature about 72 C, then the heating was removed and the cell was cooled in static air. The cell was always in operation and the data were recorded when the cell temperature varied. Figure 8 shows the open circuit voltage response to varying cell temperatures. As can be seen from Fig. 8 a, the cell voltage responds quickly to cell temperature and follows the cell temperature. But the open circuit voltage during decreasing cell temperature is about 20 mv lower than that at the same cell temperature during increasing cell temperature. This is demonstrated in Fig. 8 b. Figures. 9 illustrates the cell current responses to varying cell temperatures when the cell operates with constant voltages of 0.2 V. As shown in Fig. 9 a, the cell has a fast and smooth response to continuously varying cell temperature. As shown in Fig. 9 b, the cell current during decreasing temperature is lower than that at the same cell temperature during increasing temperature. So, it is necessary to monitor the change trend of cell temperature when the DMFC is operating. 4 Conclusions The experimental results lead to the following conclusions: The cell voltage responds rapidly to variable loading change but in a different way depending on the magnitude and the slope of change load. An increasing slope of gradual loading significantly influences the dynamic behavior. The cell temperature and oxygen flow rate significantly affect the cell dynamic response. The cell voltage difference over the range of cell temperatures and oxygen flow rates is small for gradual loading. The cell responds rapidly to continuously varying cell temperatures, but the cell response value during decreasing temperature is lower than that during increasing temperature. Acknowledgment The authors would like to acknowledge the following supports of this research: National Nature Science Foundation of China Grant Nos and , Sino-German Center for Research Promotion of DFG and NSFC Grant No. GZ / 7. The authors are grateful to Yan Wang, Jie Lin Jia, and Xuan Liu for the cooperation and helpful discussions and to Peter King and P. Eng for editing. 206 / Vol. 3, MAY 2006 Transactions of the ASME

6 References 1 Aricó, A. S., Srinivasan, S., and Antonucci, V., 2001, DMFCs: From Fundamental Aspects to Technology Development, Fuel Cells, 1 2, pp Argyropoulos, P., Scott, K., and Taama, W. M., 2000, Dynamic Response of the Direct Methanol Fuel Cell Under Variable Load Conditions, J. Power Sources, , pp Argyropoulos, P., Scott, K., and Taama, W. M., 2000, The Effect of Operating Conditions on the Dynamic Response of the Direct Methanol Fuel Cell, Electrochim. Acta, 45 12, pp Argyropoulos, P., Scott, K., and Taama, W. M., 2001, An Investigation of Scale-Up on the Response of the Direct Methanol Fuel Cell Under Variable Load Conditions, J. Appl. Electrochem., 31 1, pp Zhou, S., Schultz, T., Peglow, M., and Sundmacher, K., 2001, Analysis of the Nonlinear Dynamics of a Direct Methanol Fuel Cell, Phys. Chem. Chem. Phys., 3 3, pp Sunmacher, K., Schultz, T., Zhou, S., Scott, K., Ginkel, M., and Gilles, E. D., 2001, Dynamics of the Direct Methanol Fuel Cell DMFC : Experiments and Model-Based Analysis, Chem. Eng. Sci., 56 2, pp Simoglou, A., Argyropoulos, P., Martin, E. B., Scott, K., Moris, A. J., and Taama, W. M., 2001, Dynamic Modelling of the Voltage Response of Direct Methanol Fuel Cells and Stacks Part I: Model Development and Validation, Chem. Eng. Sci., 56 23, pp Simoglou, A., Argyropoulos, P., Martin, E. B., Scott, K., Moris, A. J., and Taama, W. M., 2001, Dynamic Modelling of the Voltage Response of Direct Methanol Fuel Cells and Stacks Part II: Feasibility Study of Model-Based Scale-Up and Scale-Down, Chem. Eng. Sci., 56 23, pp Kallo, J., Kamara, J., Lehnert, W., and Helmolt, R. V., 2004, Cell Voltage Transients of a Gas-Fed Direct Methanol Fuel Cell, J. Power Sources, , pp Wang, M. H., Guo, H., Ma, C. F., Ye, F., Yu, J., Liu, X., and Wang, C. Y., 2004, in 3rd Int. Symp. Heat Transfer Enhancement and Energy Conservation, Guangzhou, China, B. Hua, Z. Y. Guo, and C. F. Ma, eds. South China University of Technology Press, Guangzhou, China, pp Qi, Z. G., and Kaufman, A., 2002, Open Circuit Voltage and Methanol Crossover in DMFCs, J. Power Sources, 110 1, pp Guo, H., Ma, C. F., Wang, M. H., Ye, F., Wang, Y., Yu, J., and Wang, C. Y., 2004, Influence of Thermophysical Parameters on the Mass Transfer in Fuel Cells, Chin. J. Eng. Thermophys., 25, pp Mench, M. M., Wang, C. Y., and Thynell, S. T., 2001, An Introduction to Fuel Cells and Related Transport Phenomena, Int. J. Transp. Phenom., 3 3, pp Mennola, T., Mikkola, M., Noponen, M., Hottinen, T., and Lund, P., 2002, Measurement of Ohmic Voltage Losses in Individual Cells of a PEMFC Stack, J. Power Sources, 112 1, pp Journal of Fuel Cell Science and Technology MAY 2006, Vol. 3 / 207