Numerical simulation of droplet dynamics in a proton exchange membrane (PEMFC) fuel cell micro-channel

https://doi.org/10.1016/j.ijhydene.2014.09.077Get rights and content

Highlights

  • LB simulations were used to understand the droplet dynamics in gas micro-channels.

  • Effect of different parameters on the droplet behavior is examined.

  • Droplet evacuation is faster for a hydrophobic micro-channel than a hydrophilic one.

Abstract

Water management is emerging as one of the problems related to Proton Exchange Membrane Fuel Cell. In fact, the humidity in a PEMFC plays a key role on its performance. The membrane must be sufficiently moistened to ensure the transport of protons. However, liquid water may form and block the transport of gas to the electrodes. This can generate a sharp decrease in current produced by the cell.

In this paper, the droplet behavior in a proton exchange membrane (PEM) fuel cell micro-channel was simulated by using the lattice Boltzmann method (LBM) based on the Shan-Chen Pseudo-potential model. A three-dimensional case was considered and a D3Q19 scheme was utilized to keep track of the deformation of the liquid–gas interface. Visualization of droplet shape is obtained for different capillary numbers and the hysteresis between the advancing and receding contact angle is clearly observed. Also flow structures in the micro-channel were illustrated. The effect of wettability on droplet displacement behavior is also explored. It was found that hydrophobic micro-channel is better than the hydrophilic micro-channel for droplets evacuation. This work presents a basic understanding for the droplet behavior in a fuel cell micro-channel and the effect of important parameters on its dynamics.

Introduction

A fuel cell is a device that generates electricity by a chemical reaction. It is the object of intensive research by the scientific community and industry [2]. There are different types of fuel cells. A fuel cell proton exchange membrane (PEM) is composed of an anode, a cathode, two bipolar plates, two catalysts and a membrane. The electricity generation is done through oxidation on an electrode of a fuel in this case hydrogen (H2 → 2H+ + 2e) coupled with the reduction on the other electrode of an oxidant, such as the oxygen which produces water (1/2O2 + 2H+ + 2e → H2O)

The hydrogen and oxygen are fed to the anode and cathode by the flow channels and the gas-diffusion-layers (GDL) which is an essential element for the proper functioning of the fuel cell; it should allow the gas transfer on the electrode's entire surface, evacuate water from the active layers, evacuate the heat generated by the electrochemical reactions, improve the strength of the membrane electrode assembly and spreading electrons injected from the electrode in the plane of the active layer.

The diffusion layer is a porous medium made of carbon fiber and usually treated with a hydrophobic agent to improve its water removal properties. It must have sufficient pores to provide reactant gas access from flow-field channels to catalyst layers and provide passage for removal of product water from catalyst-layer area to flow-field in channels which must have a well-defined shape.

The water produced in the porous membrane is shunted out to the exterior of the fuel cell [17]. The polymer membrane must be hydrated to ensure high conductivity of protons. The bipolar plate containing the channels must have an adequate geometry for improving the performance of a fuel cell. PEM operating at reduced temperature and pressure levels, which triggers the development of two-phase flows inside the bipolar plates channels. There are many types of bipolar plates: the original GlobeTech geometry, the serpentine geometry, the spiral geometry and the discontinuous channels geometry as cited by Yuan et al. [20].

The water management [10] in fuel cells is currently one of the critical issues. Thus, it is important to understand how water behaves inside the fuel cell. Removing the water produced by the electrochemical reactions at the cathode from anode side GDL is also essential to achieve continuous operation.

The removal of the water produced by the fuel cell is carried out by mechanical entrainment at the same time as the evacuation of unconsumed reactant gas.

The water present in the fuel cell can come from two sources: a portion of this water is fed through the wet gas stream entering the cell across the catalyst layer to the electrode-membrane interface, this water is then absorbed by the electrolyte and contributing to its hydration, when the second portion of water is produced by the chemical reactions of the fuels, in fact, after migrating through the membrane, the protons react with the electrons and oxygen gas, introduced on the cathode side, producing water.

Water vapor condensation is responsible for the growth of water droplets in the feed channels of the fuel cell that are discharged by the gas flow. This phenomenon is important to analyze, several studies address the mechanisms of gas dynamics and its effects on the mobility of water droplets in the micro-channels. However, there are few direct observations of two-phase flow during fuel cell operation; some authors employ transparent fuel cell, magnetic resonance imaging, or X-rays to visualize the water. Nevertheless, these diagnostic tools require complex and expensive equipment.

It is imperative to prevent fuel cell flooding, for this it is possible to optimize the stack design and the operating conditions. For example varying the shape of the bipolar plate channel can be a good solution for flooding problem [14].

Proton exchange membrane fuel cells micro-channels are generally rectangular in shape but there are other shapes such as trapezoidal, triangular, and circular shapes. The change of the micro-channel’s geometry may affect the water removal ability; for circular micro-channels, the condensed water forms a liquid film at the bottom of the channel and for tapered micro-channels the water forms small droplets.

Liu et al. [15] studied the reactant gas transport and the cell performance of a proton exchange membrane fuel cell (PEMFC) with a tapered flow channel design, a two dimensional numerical modeling was done with the reduction in the channel depth along the streamwise direction. The authors concluded that the application of tapered micro-channels improve water management and fuel cell performance. The effects may be increased with decreasing the taper ratio of the fuel channel.

Metz [18] proposes a flooding passive management using capillary forces to move the excess water. This requires a special structure of gas distribution micro-channels. He shows that the use of tapered channels may allow permanent gas intake even when large quantities of water are produced. The channels are designed so that to force water to rise along the hydrophilic walls then extended in a second micro-channel located at the top of the conical section. For such a system to be effective it is essential to add an absorbent material to the upper end of each channel.

Buie [4] chooses an active water management by incorporating an electroosmotic pump to remove product water from the cathode area.

Minor [19] has studied and modeled the droplet dynamic behavior in air flow field channels using experimental studies. Micro digital particle image velocimetry techniques were used to provide quantitative visualizations of the flow inside the liquid phase for the case of air flow around a droplet adhered to the wall of a rectangular gas channel model. The results of the study yield very interesting results such as the observation of a variety of rotational secondary flow patterns within the droplet.

Using a transparent fuel cell, Chen [6] shows that several water flow regimes inside the microchannels can be observed. When the air flow increases, we can go from a regime with droplets stalling to film flow before reaching a chaotic flow regime. Hussaini et al. [13] also show in their work the existence of these different flow regimes.

Anderson [1] also reviewed numerical simulations on the two-phase flow in gas flow channels of proton exchange membrane fuel cells. In his review, he exhibits in situ and ex situ experimental setups utilized to visualize and quantify two-phase flow phenomena in terms of flow regime maps, flow maldistribution, and pressure drop measurements. He summarizes the current literature on CFD simulations of gas–liquid two-phase flow in PEM fuel cells including the multi-fluid model, mixture model, volume of fraction method (VOF), and Lattice Boltzmann method (LBM).

In his work, Cho et al. [8] did both an experimental and numerical study of droplet deformation and detachment in micro gas flow channels. The volume of fluid (VOF) scheme was used to simulate the water droplet dynamics and a high resolution CCD camera is employed to capture the droplet shape-change and detachment. They found that the local gas pressure is increased at the stagnation point in the front of the droplet and small droplets are more sensitive to viscous force.

The volume-of-fluid model (VOF) was also applied by Zhu et al. [21] to investigate the effect of micro-channel geometry on droplet dynamics. Many geometric configurations were tested and it was shown that for the wettable configuration, the movement of water changes drastically when a droplet surface is in contact with the hydrophilic walls.

In this work, a three-dimensional, two-phase model has been developed to simulate droplet emerging from pores of a porous fuel cell electrode into the gas flow channel as sketched in Fig. 1.

Numerical algorithm based on lattice Boltzmann method is developed to solve the Boltzmann equation with the two-phase Shan–Chen model and appropriate boundary conditions. The model is capable of predicting the droplet dynamics in a micro-channel of a proton exchange membrane fuel cell.

Section snippets

Lattice Boltzmann method

The Numerical Approach used in this work is based on the lattice Boltzmann method which is an alternative to the traditional approaches.

Recently the lattice Boltzmann method has met with significant success for the numerical simulation of many industrial and scientific problems [3], [5], [7], [9], [11], [12], [16]. Traditional numerical methods solve the macroscopic transport equations of fluid flow, mass and heat transfer by directly discrediting them. Contrary to classical methods (difference

Shan and Chen-type lattice Boltzmann

Shan and Chen proposed a multiple phases LBM model by introducing an interparticle potential between fluid components and based on the BGK collision model. In this model, one distribution function is introduced for each of the fluid components.

In the Shan-Chen model, a force, between the two fluids is introduced that effectively perturbs the equilibrium velocity for each fluid [16].

In D3Q19 model, this force is given by:F(x,t)=Gψ(x,t)iwiψ(x+eiΔt,t)eiwhere G is the interaction strength, wi is

Results and discussion

In this section, some simulation results are repeated to examine the validity of this model and code.

First, a simulation of a spinodal phase separation has been conducted in a 51 × 51 × 51 lu3 domain initialized by a random distribution of two phases; the system will automatically separates to the liquid phase and the vapor phase due to the instability of van der Waals equation of state. Fig. 5 shows the Phase field morphology at different times. The variation of the density is shown in color

Conclusion

We have developed an LBM algorithm with an external force for two-phase droplet dynamics in three-dimensional micro-channel of a proton exchange membrane fuel, in this model the two phase fluid is modeled by Shan–Chen scheme. The density is simulated inside the micro-channel. We remark that the droplet is highly deformed into a shape with a tip. It is observed that the droplet deformation increases with increase in capillary number.

It is clear that the average contact angle hysteresis, which is

References (21)

There are more references available in the full text version of this article.

Cited by (45)

View all citing articles on Scopus
View full text