2007 Annual Meeting

(564c) Oscillatory Flow in the Direct Methanol Fuel Cell: Effects on Mass Transfer, Bubble Management, and Cell Performance



The direct methanol fuel cell suffers performance decay at high current densities because of mass transfer limitations at the cell anode. The catalyst instantaneously consumes any methanol which arrives at the surface and the reaction becomes diffusion-limited. Simultaneously, a significant amount of carbon dioxide is generated by the reaction and gas bubbles are formed. The bubbles, if they are not swept out of the fuel cell by the flow, can interfere with the diffusion of methanol to the catalyst by physically blocking diffusion pathways.

In previous work, we have shown that the mass transfer limitations in the direct methanol fuel cell can be overcome by superimposing an oscillatory flow on the steady anode feed. An oscillatory flow which is tangential to the catalyst surface was induced in the porous diffusion layer and enhanced the mass transport of methanol, resulting in improvements of up to 100% for the limiting current density and 30% for the peak power density. The enhancement is observed only if the stroke volume and frequency of the oscillations are chosen such that the Peclet number in the diffusion layer exceeds one. The conditions of these experiments were such that carbon dioxide gas was not a significant mass transfer inhibitor.

In this work, the operational parameters of a 5 cm2 direct methanol fuel cell were probed in order to determine the most advantageous and efficient use of fluid oscillations. The membrane electrode assembly was a commercially available product (anode: 4.0 mg/cm2 50:50 Pt:Ru, cathode: 2.0 mg/cm2 Pt, membrane: Nafion 117). The variables examined in this study include the concentration of methanol in the feed, the steady feed rate of methanol/water solution, and the thickness of the diffusion layer. The performance of the cell with and without oscillation was measured to observe its effect. For example, a thicker diffusion layer, in conjunction with a higher methanol concentration in the feed and superimposed fluid oscillations, was observed to generate a larger peak power density than a cell with a thinner diffusion layer. Significant carbon dioxide bubble generation was also observed at low steady feed rates (e.g. 1.0 cm3 min-1) which rendered the fuel cell unstable. The degradation in performance at low flow rates can be partially reversed by fluid oscillations, but some bubble volume always remains. If the fluid oscillations are anti-symmetric, however, the bubbles can be given a net convective displacement out of the cell despite the low steady flow rate. Recommendations for optimum use of oscillatory flow in the anode of a direct methanol fuel cell will be made based on experimentally observed phenomena.