CHAPTER 4 TRANSPORT COMPONENT DESIGN EFFECTS ON
5.4 Cell Performance for Channel Design with Various S/C Ratios
In order to provide a quantitative comparison of the novel flow channel design, the bar chart of average current density at various cell voltages is shown in Fig. 5.9 for five cathode outlet port S/C ratios. This figure indicates the transition of better cell channel design with operating voltage. At cell voltage of 0.22V, divergent channel with S/C ratio of 0.67 creates the greatest current density than other cases. This is because that at such a high reaction rate, more oxygen can be transported to the reaction sites under the channel region for this channel configuration. However, as the average reaction slows down and cell voltage increases, this characteristic gradually lost its importance. Contrarily, a smaller S/C design which offers wider shoulder region becomes more beneficial for the cell reaction because it facilities the electrons transport to the catalyst layer and enhances the activation overpotetnial.
5.5 Effect of Ractant Stoichiometry on Cell Performance
As the reactant flow rates have significant impact on cell performance, it is instructive to examine the effect of reactant stoichiometry on reaction rate. Figure 5.10 shows the cell output current densities for three values of reactant stoichiometry at cell voltages of 0.22V and 0.62V. It is found that the influence of reactant stoichiometry is obvious at lower cell voltage such as 0.22V. When the reactant stoichiometry is low, less fuel and oxygen are supplied into the cell so the corresponding output current density is small. In such circumstance, the role of channel geometry on cell performance is vital. Fig. 5.10 reveals that when the stoichiometry is 2, larger variations of output current exist among the three different channel shapes. This is attributed to the fact that at 0.22V, more reactants are required to fulfill electrochemical reactions. Therefore, the reductions of stoichiometry and reactant flow rate enhance the important role of channel geometry design. However, such tendency decreases as the stoichiometry increases to 4 because the fertile reactant flow rate makes the role of channel geometry less important than that at stoichiometry of 2.
At a medium cell voltage, the reactant stoichiometry effect on current density is minor as the electrochemical reaction is dominated by conduction and activation overpotentials. However, a close inspection of Fig. 5.10 indicates that it presents a
relatively larger influence on current density for the convergent shape channel.
Despite the greater shoulder region of such channel configuration facilitates electrons transport to reaction sites and enhances output current, the cell reaction rate is reduced at lower stoichiometry as the reactant concentration is relatively small. This phenomenon reflects the need of detail inspection of the impact for various operating parameters on such novel flow channel.
5.6 Summary
An investigation of the effects of novel flow channel with various outlet port S/C ratios in PEM fuel cell on the transport phenomena and catalyst reaction has been performed through a three-dimensional multi-component model. The influence of channel geometry on local oxygen, potential and current density distributions are examined in detail. According to the results and discussion, the following conclusions can be drawn.
1. The configuration of flow channel influences the distribution of various transport variables such as reactant concentration, saturation level, potential field and activation overpotential.
2. For a narrower channel with large S/C ratio at outlet port, the average passage for electron conduction to the reaction site is shorter, such that it can generate more
current at medium cell voltage where the activation overpotential dominates the catalyst reaction.
3. With the increase of cell reaction rate, the requirement of high reactant concentration becomes more important. Therefore, the best channel geometry shifts toward the design with small S/C outlet port ratio, which offers a shorter transport passage for oxygen around channel region.
4. The saturation level of liquid water also has an important effect on oxygen transport and cell reaction rate, especially at low cell voltage. Consequently, the divergent channel design creates higher current density at such operating condition as the liquid water is easy to transport out of the channel from the outlet port.
5. Transverse plots of local current density and oxygen concentration demonstrate the dominate mechanism of local cell reaction. When the variation trends between these two parameters are consistent, the cell reaction is dominated by concentration overpotential. Otherwise, it is controlled by conduction and activation overpotentials.
6. The reactant stoichiometry effect on cell performance is quite obvious at lower cell voltage where a great amount of fuel and oxygen are required to fulfill the
electrochemical reaction. Furthermore, it bears greater influence on output current among different channel geometries at a value of 2.
Table 5.1 Main cell parameters, properties and operating conditions
gas channel length, 6.0E-2 m diffusion layer porosity, 0.4 gas channel thickness, 1.0E-3 m catalyst layer porosity, 0.28 diffusion layer thickness, 2.54E-4 m membrane porosity, 0.28
catalyst layer thickness, 1.0E-5 m diffusion and catalyst layer permeability, 2.3E-11 m2
membrane thickness, 1.75E-4 m membrane permeability, 1.0E-18 m2
gas channel half width (inlet port), 5.0E-4 m reactant relative humidity, 100 %
shoulder width (inlet port), 5.0E-4 m oxygen mass fraction at channel inlet port, 0.196 cathode inlet velocity, 0.84 m/sec vapor mass fraction at cathode inlet port, 0.160 anode inlet velocity, 0.35 m/sec vapor mass fraction at anode inlet port, 0.733 cell back pressure, 2 atm hydrogen mass fraction at anode inlet port, 0.267 cell temperature, 80 ℃ electrode conductivity, 114 S/m
Table 5.2 Simulation cases for various channel configurations used in this study
case A B C D E
divergent divergent straight convergent convergent description
channel
channel channel channel channel
Ws(S) 0.4mm 0.45mm 0.5mm 0.55mm 0.6mm
Wc(C) 0.6mm 0.55mm 0.5mm 0.45mm 0.4mm
S/C ratio
(Ws/Wc) 0.67 0.82 1.00 1.22 1.5
Figure 5.1 (a) Computational Domain (b) Cathode Channel Configuration
DimensionlessChannelLength
Figure 5.2 (a) Oxygen mass fraction contours at cell voltage of 0.62V for outlet port S/C values of 0.67, 1 and 1.5
DimensionlessChannelLength
region DimensionlessChannelLength
0
Figure 5.2 (b) Oxygen mass fraction contours at cell voltage of 0.22V for outlet port S/C values of 0.67, 1 and 1.5
DimensionlessChannelLength
Figure 5.3 (a) Liquid water saturation contours at cell voltage of 0.62V for outlet port S/C values of 0.67, 1 and 1.5
DimensionlessChannelLength
Figure 5.3 (b) Liquid water saturation contours at cell voltage of (a) 0.62V (b) 0.22V for outlet port S/C values of 0.67, 1 and 1.5
DimensionlessChannelLength
Figure 5.4 (a) Solid and membrane phase potential contours at cell voltage of 0.62V for straight channel
DimensionlessChannelLength
region DimensionlessChannelLength
0
Figure 5.4 (b) Solid and membrane phase potential contours at cell voltage of 0.22V for straight channel
DimensionlessChannelLength
Figure 5.5 (a) Activation overpotential contours at cell voltage of 0.62V for outlet port S/C values of 0.67, 1 and 1.5
DimensionlessChannelLength
Figure 5.5 (b) Activation overpotential contours at cell voltage of 0.22V for outlet port S/C values of 0.67, 1 and 1.5
DimensionlessChannelLength
Figure 5.6 (a) Current density contours at cell voltage of 0.62V for outlet port S/C values of 0.67, 1 and 1.5
DimensionlessChannelLength
Figure 5.6 (b) Current density contours at cell voltage of 0.22V for outlet port S/C values of 0.67, 1 and 1.5
1 0.8 0.6 0.4 0.2 0 Dimensionless Width
2000 2400 2800 3200 3600 4000
Current Density (A/m2)
0.08 0.12 0.16
Oxygen Mass Fraction
S/C 0.67 Diverget S/C 1.50 Convergent O2 Mass Fraction Current Density
Figure 5.7 (a) Transverse oxygen mass fraction and current density at x=0.012m of CL middle plane with cell voltage of 0.62V and S/C values of 0.67, 1.50
1 0.8 0.6 0.4 0.2 0 Dimensionless Width
2000 2400 2800 3200 3600 4000
Current Density (A/m2)
0 0.04 0.08 0.12 0.16
Oxygen Mass Fraction
S/C 0.67 Divergent S/C 1.50 Convergent
O2 Mass Fraction Current Density
Figure 5.7 (b) Transverse oxygen mass fraction and current density at x=0.048m of CL middle plane with cell voltage of 0.62V and S/C values of 0.67, 1.50
1 0.8 0.6 0.4 0.2 0 Dimensionless Width
2000 4000 6000 8000 10000 12000
Current Density (A/m2)
0 0.02 0.04 0.06 0.08 0.1 0.12
Oxygen Mass Fraction
S/C 0.67 Divergent S/C 1.50 Convergent O2 Mass Fraction Current Density
Figure 5.8 (a) Transverse oxygen mass fraction and current density at x=0.012m of CL middle plane with cell voltage of 0.22V and S/C values of 0.67, 1.50
1 0.8 0.6 0.4 0.2 0 Dimensionless Width
0 2000 4000 6000 8000 10000
Current Density (A/m2)
0 0.02 0.04 0.06 0.08
Oxygen Mass Fraction
S/C 0.67 Divergent S/C 1.50 Convergent
O2 Mass Fraction Current Density
Figure 5.8 (b) Transverse oxygen mass fraction and current density at x=0.048m of CL middle plane with cell voltage of 0.22V and S/C values of 0.67, 1.50
0.22 0.32 0.42 0.52 0.62 0.72 0.82 0.92
Cell Voltage (V)
0 0.4 0.8 1.2
Current Density (A/cm2)
S/C 1.50 S/C 1.22 S/C 1.00 S/C 0.82 S/C 0.67
Figure 5.9 Comparison of cell performance for various cathode channel geometry designs
2 3 4 Reactant Stoichiometry
0.6 0.8 1 1.2
Current Density (A/cm2)
S/C 1.50 Convergent S/C 1.00 Straight S/C 0.67 Divergent 0.22V
0.62V
Figure 5.10 Effect of reactant stoichiometry on cell current density at operating voltages of 0.22V and 0.62V for three cathode channel configurations
CHAPTER 6
CONCLUSIONS AND FUTURE PERSPECTIVES
6.1 Concluding Remarks
In this dissertation, comprehensive descriptions on the multi-component, multi-phase transport phenomena as well as the processes of electrochemical kinetics, local current transfer and thermal flow have been presented. A multi-physics model with computational fluid dynamics technique is successfully employed to investigate complex behaviors in the PEMFCs. The model is able to resolve local activation overpotential which is the actual driving force of the cell reaction. In addition, it is validated through comparison with several previous works including numerical and experimental ones and considerable consistency of the trends of performance variations is achieved.
In the analysis of the effects of temperature and humidification level, a series of numerical predictions of cell performance subjected to various gradients of temperature or humidification level have been conducted. It is found that cell performance can be enhanced by the scheme of cathode reduced humidification because the reason of mass transport passage is unlikely to be plugged at high reaction rate. However, cell output current decays at higher or lower anode humidification
levels with the reasons of mass transport overpotential or poor membrane conductivity respectively. The analysis also reveals that larger temperature gradient is harmful to cell performance at high current density for a larger temperature gradient. Despite of this result, at medium reaction rate and low anode temperature, cell temperature gradient imposes positive effect on its output current. Furthermore, local physical properties distributions are also presented in the analysis. It exposes a close relation between these distribution and cell performance.
Subsequently, the developed model is used to investigate the impacts of cell transport component designs on the performance and various transport properties. The effects of channel aspect ratio on local oxygen and potential distributions are demonstrated. Furthermore, the relationships of these properties with cell polarization curves are well discussed. It is found that the higher the current density, the slender the optimal channel aspect ratio is. Also, position of highest activation overpotential shifts toward shoulder central at various AR designs. However, the largest current density location is subjected to the competition effect between reactants and electrons transport. In this content, its values exhibits different trend for medium and high reaction rates. As a result, optimal AR can be drawn depending on the operating voltage.
Following the previous works, the influence of GDL thickness on various parameters and cell performance is examined. It is demonstrated that there exists different variation tendencies between membrane conductivity and phase potential at low reaction rate. However, at high reaction rate they coincide with each other.
Furthermore, this factor imposes limitation on the transport of electrons and fluids in either transverse or vertical directions of model domain. For a thinner thickness, these species are easy to deliver in vertical direction but difficult in transverse direction. So that thinner value of GDL thickness is beneficial at medium cell voltage. On the contrary, at high cell reaction rate, the requirement of high concentration reactants results in the increase of GDL thickness which draws the largest cell current. However, this trend of maximum output current thickness reverses its direction with the reasons of low oxygen concentration and mass transport clogging at higher reaction rate.
In the final part of this dissertation, the developed model is employed to quest the electrochemical reaction and performance of PEMFCs with a novel cathode flow channel. Through the assignment of S/C ratio at cathode outlet port with various values, the configurations of this novel channel can be cataloged into divergent, straight and convergent ones. Contour plots of the essential model variables such as reactant concentration, potential fields, activation overpotential as well as local reaction rate are presented and discussed.
Numerical results show that flow channel configuration influences the local distribution of the aforementioned model variables as well as the cell performance.
For a convergent shape channel, the electrons transport passages are enhanced and the cell performance increases at medium cell voltage where activation overpotential dominated the cell reaction. However, when the reaction rate increases, the divergent channel is able to provide sufficient oxygen to the reaction sites and the output current increases accordingly. Furthermore, the water saturation level in this channel configuration is smaller then that of other channel configurations. This is also beneficial to the performance at low cell voltage.
The transverse plots of local current density and oxygen concentration provide a further insight to the dominant mechanism of cell reaction. When these two plots have consistent variation trends, the cell reaction is controlled by the mass overpotential. This is usually seen at scenarios when cell reaction rate is higher or where the oxygen concentration is low especially at shoulder region of channel down stream. Moreover, the discussion of the reactant stoichiometry effect shows that at a value of 2, it bears greater influence on output current among the various channel configurations. With the knowledge of the well comprehended determining reasons of global performance variation, these results can further offer the explanations, which are of great important to the researchers and engineers, of the different dominant
mechanisms resulting from the characteristic differentials of transport phenomena for the various species inside the fuel cell.
6.2 Future Perspectives
The investigations and findings in previous chapters offer an essential knowledge base for the subsequent study. The transport component design effects study in chapter 4 focus on the discussion at a representative position and a constant AR along channel direction. In an actual cell the species concentrations vary continuously along channel direction. At upper region of channels, oxygen and hydrogen have higher mass fraction and the cell reaction rate is faster than downstream location. As a result, there exists an extreme un-uniform electrochemical reaction and current density, especially at low cell voltage. This phenomenon is harmful for the durability of a fuel cell. It is possible that overall cell performance could be degenerated due to local fatigue from long term operation at higher current density. To ensure the reaction uniformity and enlarge the cell life time, special efforts should be addressed on this problem. A novel configuration of cell channel geometry is investigated in chapter 5. However, the model for these works contains only one single cell and unit channel. In actual situation, the cell system could be made from several unit cells stacked together to obtain a higher voltage output. The phenomenon of non-uniformity of reactant concentration as well as cell reaction could be more
serious in this situation. Therefore, it is necessary to quest this issue in the future by expanding previously developed model into multi-cell stack.
Another recommendation of alleviating previous problem would be focused on a new reactant delivery system that has several layered channels stacked together in channel height direction with a common inlet port and different lengths. The lowest channel has the shortest length for the reactant delivery. In such a way, only limited amount of reactant can participate the cell reaction of the upper channel region and the remaining high concentration reactant is transported to the downstream region.
With this design, the objective of uniform cell reaction can be achieved by way of adjusting the ratio of sub channel length to the total length as well as the number of sub channels. The prediction of cell performance for these novel designs is feasible from the previous developed model. Only a small modification is required for the implementation of numerical calculation of domain properties and cell polarization curve.
REFERENCES
Barbir, F., PEM Fuel Cells: Theory and Practice, Elsevier Academic Press, San Diego, 2005.
Baschuk, J.J. and Li, X., “Modeling of polymer electrolyte membrane fuel cells with variable degree of water flooding,” J. Power Sources, vol. 86, pp. 181-196, 2000.
Bernardi, D.M., “Water-balance calculation for solid-polymer-electrolyte fuel cells,” J.
Electrochem. Soc., vol. 137, no. 11, pp. 3344-3350, 1990.
Bernardi, D.M. and Verbrugge, M.V., “Mathematical model of a gas diffusion electrode bounded to a polymer electrolyte,” AIChE Journal, vol. 37, no. 8, pp.
1151-1163, 1991.
Bernardi, D.M. and Verbrugge, M.W., “A mathematical model of the solid-polymer-electrolyte fuel cell,” J. Electrochem. Soc., vol. 139, no. 9, 2477-2491, 1992.
Berning, T., Lu, D.M. and Djilali, N., “Three-dimensional computational analysis of transport phenomena in a PEM fuel cell,” J. Power Sources, vol. 106, pp. 284-294, 2002.
Berning, T. and Djilali, N., “Three-dimensional computational analysis of transport phenomena in a PEM fuel cell – a parametric study,” J. Power Sources, vol. 124, pp.
440-452, 2003.
CFD-ACE(U) TM User Manual, CFD Research Corp., Huntsville, AL, 2004.
Cussler, E.L. Diffusion Mass Transfer in Fluid System, second ed., Cambridge, 1997.
Djilali, N and Lu, D., “Influence of heat transfer on gas and water transport in fuel cells,” Int. J. Therm. Sci., vol. 41, pp. 29-40, 2000.
Dutta, S., Shimpalee, S. and Van Zee, J.V., “Three-dimensional numerical simulation
of straight channel PEM fuel cells,” J. Applied Electrochem., vol. 30, pp. 135-146, 2000.
Dutta, S., Shimpalee, S. and Van Zee, J.V., “ Numerical prediction of mass-exchange between cathode and anode channels in a PEM fuel cell,” Int. J. Heat and Mass Trans., vol. 44 pp. 2029-2042, 2001.
Fuller, T.F. and Neuman, J., “Water and thermal management in solid-polymer-electrolyte fuel cells,” J. Electrochem. Soc., vol. 140, no. 5, pp.
1218-1225, 1993.
Gurau, V., Liu, H.T. and Kakac, S., “Two-dimensional model for proton exchange membrane fuel cells,” AIChE Journal, vol.44, pp.2410-2422, 1998.
Giorgi, L., Antolini, E., Pozio, A. and Passalacqua, E., “Influence of the PTFE content in the diffusion layer of low-Pt loading electrodes for polymer electrolyte fuel cell,”
Electrochimica Acta, vol. 43, pp. 3675-3680, 1998.
He, W., Yi, J.S. and Nguyen, T.V., “Two-phase flow model of the cathode PEM fuel cell using interdigitated flow fields,” AIChE Journal, vol. 46, pp. 2053-2064, 2000.
Hum, B. and Li, X., “Two-dimensional analysis of PEM fuel cells,” J. Applied Electrochem., vol. 34, pp. 205-215, 2004.
http://www.etek-inc.com/home.php.
Jeng, K.T., Lee, S.F., Tsai, G.F. and Wang, C.H., “Oxygen mass transfer in PEM fuel cell gas diffusion layers,” J. Power Sources, vol. 138, pp. 41-50, 2004.
Jordan, L.R., Shukla, A.K., Behrsing, T., Avery, N.R., Muddle, B.C. and Forsyth, M.,
“Diffusion layer parameters influencing optimal fuel cell performance,” J. Power Sources, vol. 86, pp. 250-254, 2000.
Ju, H., Meng, H. and. Wang, C.Y., “A Single-phase, non-isothremal model for PEM
fuel cells,” Int. J. Heat and Mass Trans., vol. 48, pp. 1303-1315, 2005.
Kulikovsky, A.A., Divisek, J. and Kornyshev, A.A., “Modeling the cathode compartment of PEMFC dead and active reaction zones,” J. Electrochem. Soc., vol.
146, pp. 3981-3991, 1999.
Lee, W.K., Ho, C.H., Van Zee, J.W. and Murthy, M., “The effects of compression and gas diffusion layers on the performance of a PEM fuel cell,” J. Power Sources, vol. 84, pp. 45-51, 1999a.
Lee, W.K., Van Zee, J.W., Shimpalee and Dutta, S., “Effect of humidity on PEM fuel cell performance Part 1 – experiments,” 1999 Internation Mechnical Engineering Congress & Expostion, Nashville, TN, 1999b.
Liu, H.C., Yan, Y.M., Soong, C.Y. and Chen, Falin, “Effects of baffle-blocked flow channel on reactant transport and cell performance of a proton exchange membrane fuel cell,” J. Power Sources, vol. 142, pp. 125-133, 2005.
Liu, H.C., Yan, Y.M., Soong, C.Y., Chen, Falin and Chu, H.S., “Reactant gas transport and cell performance of proton exchange membrane fuel cells with tapered flow field design,” J. Power Sources, vol. 158, pp.78-87, 2006.
Marr, C. and Li, X., “Composition and performance modeling of catalyst layer in a proton exchange membrane fuel cell,” J. Power Sources, vol. 77, pp. 17-27, 1999.
Mazumder, S. and Cole, J.S., “Rigorous 3-D mathematical modeling of PEM fuel cells I. Model prediction without liquid water transport,” J. Electrochem. Soc., vol.
150, pp. A1503-A1509, 2003.
Meng. H. and Wang. Chao-Yang, “Electron Transport in PEFCs,” J. Electrochem.
Soc., vol 151, pp. A358-A367, 2004.
Natarajan, D. and Nguyen, T.V., “ A two-dimensional two-phase multicomponent transient model for the cathode of a proton exchange membrane fuel cell using
conventional gas distributors,” J. Electrochem. Soc., vol. 148, pp. A1342-A1335, 2001.
Nguyen, T. and White, R.E., “A water and heat management model for proton-exchange-membrane fuel cells,” J. Electrochem. Soc., vol. 140, no. 8, pp.
2178-2186, 1993.
Nguyen, P.T., Berning, T. and Djilali, N., “Computational model of a PEM fuel cell with serpentine gas flow channels,” J. Power Sources, vol. 130, pp. 149-157, 2004.
Parthasarathy, A., Srinivasan, S. and Appleby, A.J., “Temperature dependence of the electro kinetics of oxygen reduction at the platinum/nafion interface- a microelectrode
Parthasarathy, A., Srinivasan, S. and Appleby, A.J., “Temperature dependence of the electro kinetics of oxygen reduction at the platinum/nafion interface- a microelectrode