The effect of pulsating pressure on the performance of a PEM fuel cell with a wavy cathode surface

Highlights

• The active approach pore-scale LBM was used for PEMFC simulation.

• The GDL was modeled by a stochastic arrangement of circular solid obstacles.

• The waveform channel with pulsatile pressure improves the fuel cell performance around 7%.

Abstract

In the context of attempts to improve the performance of Proton Exchange Membrane (PEM) fuel cells with a heterogeneous porous gas diffusion layer (GDL) consisting of carbon paper, we investigated whether - and to which degree - pulsating the pressure in a single waveform cathode channel affects the flow field in the channel and the performance of the fuel cell. In this 2-D study, the GDL was modeled by a stochastic arrangement of circular solid obstacles the macroscopic transport properties of which, such as permeability and tortuosity, were numerically simulated and found to compare favorably with experimental data. The focus of this paper is on the effects of varying amplitude and frequency of the pressure pulsations on cell performance. The results obtained show that a pulsating pressure enhances the convective species transport to the reaction sites and thereby increases cell performance. We found that in a waveform channel a pulsatile pressure with an amplitude as high as 0.7 times the pressure drop over the cathode channel improves the fuel cell performance by around 7%, while the effect of pulsation frequency on output power is marginally small only.

Introduction

In a fuel cell, the chemical energy stored in fuels is directly converted into electricity power; the efficiency of this process is relatively high, e.g. with respect to heat engines [1]. Due to its low operating temperature, very low emissions and – in principle - high power density, the Proton Exchange Membrane Fuel Cell (PEMFC) looks very attractive as a power source for future automotive and portable applications. The size of the fuel cell is a decisive issue for portable applications: the smaller the more attractive but also the more difficult to remedy certain drawbacks.

Particularly in many of the smaller devices, and depending on the ambient operating conditions, the output power density of this type of fuel cell suffers from a low diffusion flux of oxygen (air) and from flooding by the water produced [6]. Imposing a forced convective flow with the help of a blower or compressor may cure the issue of mass transfer limitation. Several investigators tried to improve the performance of a PEMFC even further. Kuo and Chen [2], [3], [4] proposed a waveform cathode surface and claimed that this novelty improved the power density delivered by the fuel cell by almost 40% [4]. Recent experimental data reported by Han et al. [5], however, did not confirm this claim. Han et al. [5] investigated the effect of a wavy channel wall in a PEMFC both numerically and experimentally, and found the fuel cell's performance improved by just some 5%. Their results further indicated that the optimum value for the Expansion Ratio (ER) is two, where ER is defined as the ratio of channel depth to the wave's amplitude. Another option to reduce serious power loss while keeping system size small is by means of a pulsating or oscillating air flow [7].

This paper investigates the combined effect of an oscillating air flow and a wavy cathode surface on the performance of a PEMFC. The geometry investigated is shown in Fig. 1.

There are several studies about the effect of oscillations on mass transfer in straight and waveform channels. Nishimura and Kojima [8] experimentally studied the effects of pulsating flow on the mass-transfer in a symmetrical sinusoidal wavy-walled channel. They showed that the combination of flow separation and oscillation leads to a significant increase in mass transfer rates under laminar-flow conditions. In another study, Nishimura et al. [9] experimentally investigated the influence of an oscillation frequency on mass transfer in grooved channels at a high Schmidt number. They found that intermediate Strouhal numbers resulted in substantial transport enhancement in laminar flows.

There are just a few reports, however, on applying oscillating air flow in fuel cells. Hwang et al. [7] experimentally studied the oxygen diffusion rate at the cathode of PEMFC in a pure oscillating flow (the mean flow rate being zero) and measured the effects of oscillation frequency f and sweep distance (oscillation amplitude). The effect of the frequency was expressed in terms of the non-dimensional Womersley number α defined by

$$\alpha =H_c\sqrt{\pi f/2\upsilon }$$

in which H_c stands for the channel height and υ is the kinematic viscosity. According to Hwang et al. [7], the reaction rate of the oxygen increases linearly with sweep distance, while the non-dimensional effective diffusivity varies linearly with α for α » 1. Recently, Ramiar et al. [10] numerically studied the effect of pulsation of the air flow at the cathode side of a PEMFC with an interdigitated flow field by using a transient 2-D isothermal two-phase multi-component transport model. They found that the pulsation amplitude increased the fuel cell performance while the pulsation frequency didn't have a significant effect. To the best of the authors' knowledge, the effect of air pulsation in a waveform cathode channel in PEMFC has never been studied before.

In all of these numerical studies, the porous medium of the cathode electrode, i.e. the gas diffusion layer (GDL), is considered homogeneous and isotropic with uniform morphological properties such as porosity, tortuosity and permeability. In reality, however, GDLs are inhomogeneous and anisotropic [11], since in general they are fabricated from carbon fiber based paper or cloth. The microscopic structure of a GDL may significantly influence the performance of a PEMFC. The real microstructure of the cathode electrode porous medium may be mimicked by means of pore-scale numerical tools among which the Lattice Boltzmann (LB) methods seem very attractive [12].

In fact, LB has superior properties as to dealing with a domain with a complicated morphology, parallelization of the algorithm, and modeling multi-phase fluid flow with a dynamic interface in a porous medium [13]. Classical chemical engineering models fall short due to the use of empirical correlations incapable of distinguishing between the transport of the various species. Although several studies modeled specific phenomena in a fuel cell such as the water management [14], [15], [16], [17], the structure of the GDL [18], [19], [20] and the reactive flow in a porous medium [21], [22] by means of LBM, just a few studies adopted LB methods to investigate the flow in all different layers of a fuel cell and its effect on the cell performance [23], [24].

In the present study, the Pseudo-Potential LB method originally proposed by Shan and Chen [25] has been used for simulating multicomponent and transport processes in a 2-D PEMFC. Rather than using an approach in which the mixture behavior is the starting point, we used the so-called ‘active approach’ in which the coupled velocity and concentration fields of each individual species are solved. In this approach, the mutual interactions of all components are much more accurately taken into account. Recently, this model was also used by Molaeimanesh et al. [26], [27], [28] for simulating a PEMFC with interdigitated flow field. Another advantage of this approach is that the modified bounce back rule proposed by Kamali et al. [29] for the reaction surface can easily be incorporated.

The purpose of the present study is to investigate the effect of a pulsating air pressure in a waveform cathode channel of PEMFC while considering a pore-scale model for the GDL and to explore the effects of both pulsation amplitude and pulsating frequency on the fuel cell performance. The focus of this paper is on the cathode side as its lay-out seems to have the biggest impact on the performance of a fuel cell, the hydrogen conversion at the anode being less critical.