Draft: Optimization of Purge Cycle for Dead-ended Anode Fuel Cell Operation

This paper focuses on the optimization of the purge cycle for dead-ended anode (DEA) operation of a proton exchange membrane (PEM) fuel cell. Controling the purge interval at given operating conditions can optimize the fuel cell efficiency and hydrogen loss during the purge. For this optimization, a model capturing the liquid water and nitrogen accumulation in the anode and the purge flow behavior is presented. A target range of purge interval is then defined based on the minimal purge time that removes the plug of liquid and nitrogen in the channel end and the maximum purge interval beyond which hydrogen is wasted since hydrogen molar fraction all along the channel has been restored to one. If the purge is sufficiently long that all of the accumulated water and nitrogen are removed then the power output in the subsequent cycle (galvanostatic operation) would be highest, compared with incomplete purges which do not fully restore hydrogen concentration in the anode. Such purge schedule, however, is associated with certain amount of hydrogen loss. Therefore, there is a trade-off between hydrogen loss and power output, and a corresponding purge interval that produces the largest efficiency. The optimum purge intervals for different cycle durations are identified. The calculated DEA efficiencies are compared with flow-through (FT) operation. The analysis and model-based optimization methodology presented in this paper can be used for optimizing DEA operation of PEMFC with minimum experimentation and development time. ∗Address all correspondence to this author. NOMENCLATURE ∆t Cycle duration, s δ t Purge interval, ms t0 Time reference, the end of a DEA cycle and the start of the purge t1 Purge interval to place the hydrogen starvation front at the channel end, referenced to t0, ms t2 Purge interval to restore the hydrogen in the whole channel with unity molar fraction, referenced to t0, ms i Current density A cm−2 n Molar fraction r Source term in transport equation, mol cm−3 s−1 s Liquid volume fraction E Voltage, V M Molar mass, Kg mol−1 N Convective flux, mol cm−2 s−1 J Diffusive flux, mol cm−2 s−1 Q Energy, J W Purge flow rate, m3 s−