Two-Dimensional Simulations Of A Proton Exchange Membrane Fuel Cell With Dead-End Anode

Proton exchange membrane fuel cells (PEMFC) with dead-end anodehave many advantages. People do not need to consider any excesshydrogen that will exit from the anode outlet. Moreover, the system doesnot need the hydrogen circulation system as an additional auxiliaryequipment. The operation of PEMFC with dead-end anode requires carefulwater and nitrogen management to achieve optimal performance. Theamount of water and nitrogen accumulated in the anode are particularlyimportant factors.Appropriate water management is necessary with the purpose ofkeeping the membrane sufficiently humidified while ensuring that theanode avoid water flooding. Flooding can lead the performancedeterioration with time that has been reported frequently in the literature.Visualization experiments show that there are considerable amount ofliquid water in the channel. Large amounts of liquid water often gather inthe channel and block it; this phenomenon is more serious for fuel cellwith a dead-end. Once that happens, hydrogen is difficult to spread to theend of the anode, and cell performance will decrease. Most of the modeldoesn’t consider the flooding problems within the anode channel, however,this problem is a very important factor that can affect the cell performance.At present, there is a problem about the prediction of liquid water in themost models. The calculated liquid water saturations are much lower thanthat measured in experiments. For the majority of two-phase model, itoften use a mist flow model in the channel (assuming the liquid water tospray the form of droplets of uniform distribution in the flow channel and maintain the same flow rate of gas). This assumption is appropriate for thecase of high gas flow rate. But this assumption can seriously underestimatethe liquid water saturation in the channel for the case of low gas flow rate.The gas flow rate is very low in the PEMFC with a dead-end anode. Somist flow model is no longer valid for that kind of fuel cell. It is necessaryto adopt a new two-phase model.In this paper, the extended capillary pressure model is used to explainthe water flooding in the channel. Firstly, assuming channel is a specialkind of porous media; secondly, using two-phase flow model simulatetwo-phase transport in the channel. Compared with previous two-phasemodel, this model is able to reflect the flooding phenomena in the anodechannel. The results show that the liquid water saturation is increasingalong the direction of the flow channel; liquid water saturation reaches itsmaximum in the end of the flow channel; liquid water saturation in the endof the anode side is higher than that in cathode side.In this paper, nitrogen penetration model is used to explain thenitrogen accumulation. One set of nitrogen control equation for the wholearea of the cell are established. In solving the problem of nitrogenaccumulation, most previous models are one-dimensional and few modelsare two-dimensional. The one-dimensional models use different controlequations for different area of the cell. And there is not a set of unifiedcontrol equations for the whole area of the cell. Therefore, we establish atwo-dimensional CFD model with the nitrogen accumulation effectconsidered. Comparing to the previous one-dimensional model, this modelhas been able to reflect the accumulation of nitrogen and the significantimpact on hydrogen transport. The results show that along the direction ofthe flow channel, the hydrogen molar concentration decreases, the fractionis reduced to0in the middle of the cell. It is identified that quite amountnitrogen accumulates in the anode channel near the dead end, whichseverely reduces the performance of the cell.