Mathematical Modeling of Mass and Charge Transfer in Anion-Exchange Membrane Direct Glycerol Fuel Cells Under Steady State and Dynamic Operations

Currently we are overly reliant on non-renewable petroleum resources, and it has been a long time desire to develop processes to cogenerate energy and chemicals from renewable sources. Anion-exchange membrane direct alcohol fuel cells (AEM-DAFC) have received great attention for their high electricity generation performance, non-pollutant emissions, low fuel cost, low operating temperatures, little alcohol crossover, and non-platinum group metal catalysts [1]. In the future they could meet the demand for production of renewable energy in the form of electricity, while also producing chemicals from biomass-derived fuels. Research challenges which have limited widespread fuel cell application are that 1) output power density remains limited by low selectivity to completely oxidized CO₂ or other deeply oxidized carboxylates [2]; 2) it is still difficult to quantify the complicated physicochemical process occurring on time scales of several different magnitudes; and 3) it remains a challenge to isolate the individual effects of mass transport, charge transport, and electrochemical kinetics on cell performance. However, mathematical modeling is a powerful and economical tool that, when combined with experimental methods, may quantify those complicated physicochemical processes [3].

We present a study of performance and behavior of an alkaline anion-exchange membrane direct glycerol fuel cell (AEM-DGFC) with Au/C anode catalyst for cogeneration of electricity and tartronic acid, a valuable chemical product. This research was conducted with the aim of numerically analyzing mass and charge distributions under steady state, dynamic, and oscillating conditions and of identifying the effects of design, reaction conditions, mass and charge transport, and electrochemical kinetics on cell performance. The results show that anode overpotential is the major source of voltage loss at middle to high current density regions, due to limited glycerol diffusion at the catalyst layer. Additionally, the dynamic response of AEM-DGFC to current density step changes was simulated by considering transient species transport and double-layer capacitance charging. Analysis of dynamic simulation reveals that the liquid-phase reactant diffusion is a key factor influencing the transient AEM-DGFC behavior and is very sensitive to diffusion layer design. This work presents a new numerical analysis of a glycerol-fed fuel cell and demonstrates that a single oxidation product model can effectively predict the steady state behavior and dynamic voltage losses and this model can be significantly important for future improvement of AEM fuel cell design and operation for the cogeneration of products and electricity from renewable sources.

[1]        Bianchini, C. and P.K. Shen, Chem. Rev., 2009, 109, 4183-4206.

[2]        Marchionni, A., Bevilacqua, M., Bianchini, C. Chen, Y.-X., Filippi, J., Fornasiero, P., Lavacchi, A., Miller, H., Wang, L. and Vizza, F., ChemSusChem, 2013, 6, 518–528.

[3]        Zawodzinski T, Wieckowski A, Mukerjee S, Neurock M. Electrochem Soc Interface, 2007, 16(2), 37-41