Modeling High Pressurized Three Dimensional Polymer Electrolyte Membrane Fuel Cell

Transcription

1 HEFAT007 5 th International Conference on Heat Trsfer, Fluid echics, d Thermodynamics 1-4 July 007, Sun City, South Africa, AK odeling High Pressurized Three Dimensional Polymer Electrolyte embre Fuel Cell Khaled Alhuss 1 1 Deputy Director of Space esearch Institute, King Abdulaziz City for Science & Technology, P.O. Box 6086, iyadh 1144, Tel: , Fax , Saudi Arabia. alhuss@kacst.edu.sa ABSTACT A fuel cell is energy conversion device that converts the chemical energy of fuel into electrical energy. Fuel cells operate continuously if they are provided with the reactt gases, not like batteries. Fuel cells c provide power in wide rge. Fuel cells are environmentally friendly; the by-product of hydrogen/oxygen fuel cell is water d heat. This paper will show a numerical modeling of high pressurized Polymer Electrolyte embre fuel cell. Numerical modeling requires understding the physical principles of fuel cells, fluid flow, heat trsfer, mass trsfer in porous media, electrochemical reactions, multiphase flow with phase chge, trsport of current d potential field in porous media d id conducting regions, d water trsport across the polymer bre; d this will result in optimal design process. This paper will show fuel cell models that are used in this alysis. Such as; electrochemical model: predicts local current density, voltage distributions. Potential field model: predicts current d voltage in porous d id conducting regions. ultiphase mixture model: predicts liquid water d gas flow in the porous diffusion layers. Thin film multiphase model: tracks liquid water flow in gas flow passages. The numerical results of the theoretical modelling of high pressurized fuel cells d trsport effects of the PE fuel cells are shown in this paper. esults including contour plot of mass d mole fraction of H, O, d H O across the alyst layers d the bre are shown in this paper. As shown in this paper the concentration of the H, O, d H O are clear. INTODUCTION A fuel cell is energy conversion device that converts the chemical energy of fuel into electrical energy. Fuel cells operate continuously if they are provided with the reactt gases, not like batteries. Fuel cells c provide power in wide rge. Fuel cells are environmentally friendly; the by-product of hydrogen/oxygen fuel cell is water d heat. This paper will show a numerical modeling for high pressurized of Polymer Electrolyte embre fuel cell. Numerical modeling requires understding the physical principles of fuel cells, fluid flow, heat trsfer, mass trsfer in porous media, electrochemical reactions, multiphase flow with phase chge, trsport of current d potential field in porous media d id conducting regions, d water trsport across the polymer bre; d this will result in optimal design process. This paper will show fuel cell models that are used in this alysis. Such as electrochemical model which predicts the local current density, voltage distributions. Potential field model which predicts current d voltage in porous d id conducting regions. ultiphase mixture model which predicts liquid water d gas flow in the porous diffusion layers. Thin film multiphase model which tracks liquid water flow in gas flow passages. DISCUSSION A fuel cell is electrochemical energy conversion device that combines hydrogen d oxygen to produce electricity, with water d heat as its by-product. In principle, fuel cells operate like batteries. Unlike batteries, fuel cells do not run down or require recharging. It will produce energy in the form of electricity d heat as long as there is a flow of chemicals into the cell. A Polymer Electrolyte embre (PE) fuel cell, also called Proton Exchge embre, is one of the most promising technologies. A PE fuel cell uses a simple chemical reaction to combine hydrogen d oxygen into water, producing electric current in the process. The PE fuel cell uses one of the simplest reactions of y fuel cell. Polymer electrolyte bre (PE) fuel cells deliver high power density d offer the advtages of low weight d volume, compared to other fuel cells. PE fuel cells use a id polymer as electrolyte d porous carbon electrodes containing a platinum alyst. They need only hydrogen, oxygen from the air, d water to operate d do not require corrosive fluids like some fuel cells. They are typically fueled with pure hydrogen supplied from storage tks or onboard ormers. Polymer electrolyte bre fuel cells operate at relatively low temperatures, around 100 C. PE fuel cells are used primarily for trsportation appliions d some stationary appliions. Due to their fast startup time, low sensitivity to orientation, d favorable power-to-weight ratio, PE fuel cells are particularly suitable for use in passenger vehicles. However, it requires that a platinum alyst be used

2 to separate the hydrogen's electrons d protons. The platinum alyst is extremely sensitive to CO poisoning, making it necessary to employ additional reactor to reduce CO in the fuel gas if the hydrogen is derived from alcohol or hydrocarbon fuel. Pressurized hydrogen gas is fed to the ode. This gas is forced through the alyst by the pressure. When hydrogen atoms come in contact with the alyst, it splits into two H + ions d two electrons e -. The electrons are forced to travel in external circuit because the bre is electrically insulating. The protons migrate through the electrolyte to the hode, where they reunite with oxygen d the electrons to produce water d heat. On the hode alyst of the fuel cell, oxygen molecules are fed to the hode d react with the electrons d protons, where it forms two oxygen atoms. Each of these atoms has a strong negative charge. This negative charge attracts the two H + ions through the bre, where they combine with oxygen atom d two of the electrons from the external circuit to form a water molecule (H O) [1-5]. ost fuel cells designed for use in vehicles produce less th 0.9 volts of electricity-far from enough to power a vehicle. Theore, multiple cells must be assembled into a fuel cell stack. The potential power generated by a fuel cell stack depends on the number d size of the individual fuel cells that comprise the stack d the surface area of the PE. Fuel cells are more energy-efficient th internal combustion engines in terms of the amount of energy used per weight of fuel d the amount of fuel used vs. the amount wasted. However, hydrogen gas is very diffuse, d only a small amount in terms of weight c be stored in onboard fuel tks of a reasonable size. This c be overcome by increasing the pressure under which the hydrogen is stored or through the development of chemical or metal hydride storage options. esearchers are developing high-pressure tks d hydride systems that will store hydrogen more effectively d safely. Fluent is used to model the flow in PE fuel cells [6]. Two electric potential fields are ved. One potential is ved in the bre d alyst layer. The other is ved in the alyst layers, the diffusion layers, d the current collectors. Surface reactions on the porous alyst region are ved d the reaction diffusion balce is applied to compute the rates. Based on the cell prescribed voltage, the current density value is computed. Alternatively, a cell voltage c be computed based on a prescribed average current density. The electrochemical processes of the rate of the hydrogen oxidation d the rate of oxygen reduction are treated as heterogeneous reactions that take place on the alyst surfaces inside the two alyst layers on both sides of the bre. The driving difference between the phase potential of the id d the phase potential of the electrolyte/bre. Theore, two potential equations are ved for in the PE model: one potential equation accounts for the electron trsport through the id conductive materials the other potential equation represents the ionic trsport of H +. There are two types of external boundaries. Those through which there passes electrical current d those through which there passes no current. As no protonic currents leave the fuel cell through y external boundary, there is a zero flux boundary condition for the bre phase potential on all outside boundaries. For the id phase potential there are external boundaries on the ode d the hode sides that are in contact with the external electric circuit d only through these boundaries pass the electrical current generated in the fuel cell. On all other external boundaries there is a zero flux boundary condition. The Butler-Volmer function is used to model the trsfer currents inside the alyst layers. The driving force for the kinetics is the local surface over-potential also known as the activation loss. It is generally the difference between the id d bre potentials. The gain in electrical potential from crossing from the ode to the hode side c then be taken into account by subtracting the opencircuit voltage on the hode side [6]. y challenges must be overcome before fuel cell vehicles will be a successful competitive alternative for users. Since the conversion of the fuel to energy takes place via electrochemical process, not combustion, the process is cle, quiet d highly efficient about two to three times more efficient th combustion. THEOETIL ODELING OF PE FUEL CELL The electrochemistry modeling concerns with calculation of the rate of the hydrogen oxidation d the redaction rate of oxygen. Heterogeneous reactions are assumed for these electrochemical processes that take place on the alyst surfaces on both sides of the bre [6-9]. The driving difference between the phase potential of the id d the phase potential of the electrolyte/bre. Theore, two potential equations are ved. The first potential equation accounts for the electron trsport e - through the id conductive materials, as shown in equation 1. The other potential equation is shown in equation represents the ionic trsport of H + [6]. ( Φ ) + = 0 ( Φ ) + = 0 σ (1) σ () Where: σ = electrical conductivity (1/ohm-m), Φ = electric potential (volts), d = volumetric trsfer current (A=m 3 ). There are two types of external boundaries; first type through which there passes electrical current d second type those through which there passes no current. Note that there is no protonic currents leave the fuel cell through y external boundary; there is a zero flux boundary condition for the bre phase potential, Φ, on all outside boundaries. Also note that for the id phase potential, Φ, there are external boundaries on the ode d the hode sides that are in contact with the external electric circuit d only through

3 these boundaries pass the electrical current generated in the fuel cell. On all other external boundaries there is a zero flux boundary condition for Φ. The source terms in equations 1 d are non-zero only inside the alyst layers. For the bre phase: on the ode side = + (> 0) d on the hode side =- (< 0). For the id phase: on the ode side = - (< 0) d on the hode side = + (> 0). The source terms in equations 1 d are also known as the exchge current density (A/m 3 ), d are shown in equations 3 d 4 [6]. (3) = J γ α Fη / T αnf / T ( e e ) η γ α Fη / T αnfη / T ( e e ) (4) Where: J is the volumetric erence exchge current density (A/m 3 ),. is the local species concentration, erence value (kgmol/m 3 ), γ is the dimensionless concentration dependence, α is dimensionless trsfer coefficient, d F is Faraday constt (9.65 x10 7 C/kgmol). Equations 3 d 4 are the general formulation of the Butler-Volmer function. Equations 5 d 6 are the simplified form of the Butler-Volmer function which is known as the Tafel formulation [6]. γ α Fη / T ( e ) (5) γ α Fη / T ( e ) (6) The driving force for the kinetics is the local surface overpotential, η. It is the difference between the id d bre potentials, Φ d Φ. The gain in electrical potential from crossing from the ode to the hode side c be taken into account by subtracting the open-circuit voltage V oc on the hode side as shown in equations 7 d 8 [6-9]. η = Φ Φ (7) η = Φ Φ V (8) oc When hydrogen atoms come in contact with the alyst, it splits into two H + ions d two electrons e -. The electrons are forced to travel in external circuit because the bre is electrically insulating. The protons migrate through the electrolyte to the hode, where they reunite with oxygen d the electrons to produce water d heat. On the hode alyst of the fuel cell, oxygen molecules are fed to the hode d react with the electrons d protons, where it forms two oxygen atoms. Each of these atoms has a strong negative charge. This negative charge attracts the two H + ions through the bre, where they combine with oxygen atom d two of the electrons from the external circuit to form a water molecule (H O). The following reactions occur, respectively, at the ode d the hode, as shown in equations 10 d 11: H H + + e + + 4H + 4e (9) H O (10) O The volumetric source terms for the species equations d energy equation include ohmic heating, heat of formation of water, electric work d latent heat of water are given in equations 11 to 14 [6]. ω,h S H = F (11) ω, O SO = 4F (1) ω, H O S H O = F (13) S = I + h η + h (14) h ohm reaction +, phase Heterogeneous reactions that take place on the alyst surfaces in the porous media for the electrochemical reaction of the alyst layers are assumed. Theore, the species concentrations of hydrogen d oxygen in the rate calculation, equations 3 d 6 are the surface values. The reactions are treated as surface reactions in the two alyst layers, d it is assumed that the diffusive flux of y reacting species is balced by its rate of production [6]. ρ δ nf ω i ( y i, surf yi, cent ) η =, D i, (15) Where: D i is the mass diffusivity of species i (m /s), η is the specific reacting surface area of the alyst layer, or surfaceto-volume ratio (1/m), y i;surf is the mass fraction of species i at the reacting surface, y i;cent is the mass fraction of species i at the cell center, d δ is the average distce between the reaction surfaces d the cell center (m). The left hd side of equation 15 represents the diffusive flux at the reacting surface d the right hd side represents the rate of mass generation. The average distce from the center of the cell to the reacting surface is assumed as δ = 1/s. Equation 15 is used to obtain the surface values of H d O concentrations, applying a Newtoni ution procedure. These surface values are then used to compute the rates in equations 3 to 6 [6-13].

4 The numerical results of the theoretical modeling of pressurized PE fuel cells are shown in figures 1 to. Figure 1 shows contour plot of mass fraction of H, O, d H O across the alyst layers d the bre. As shown in this figure the species concentrations of the H, O, d H O are clear. Figure 3 shows the contour plots of the relative humidity across the alyst layers d the bre. Fig. Contour plots of elative Humidity Fig. 1(a) Contour plots of mass fraction of H Fig. 1(b) Contour plot of mass fraction of O Fig. 1(c) Contour plots of mass fraction of H O CONCLUSION Two electric potential fields are ved. One potential is ved in the bre d alyst layer. The other is ved in the alyst layers, the diffusion layers, d the current collectors. Surface reactions on the porous alyst region are ved d the reaction diffusion balce is applied to compute the rates. Based on the cell prescribed voltage, the current density value is computed. Alternatively, a cell voltage c be computed based on a prescribed average current density. The electrochemical processes of the rate of the hydrogen oxidation d the rate of oxygen reduction are treated as heterogeneous reactions that take place on the alyst surfaces inside the two alyst layers on both sides of the bre. The driving difference between the phase potential of the id d the phase potential of the electrolyte/bre. Theore, two potential equations are ved for in the PE model: one potential equation accounts for the electron trsport through the id conductive materials the other potential equation represents the ionic trsport of H +. The numerical results of the theoretical modeling of high pressurized PE fuel cells are shown in figures 1 to. Figure 1 shows contour plot of mass fraction of H, O, d H O across the alyst layers d the bre. As shown in this figure the species concentration of the H, O, d H O are clear. Figure 3 shows the contour plots of the relative humidity across the alyst layers d the bre. Acknowledgements The author gratefully acknowledges sponsorship of this research from the Space esearch Institute of the King Abdulaziz City for Science d Technology. EFEENCES 1. Alhuss, K. "A Spiral Design in High Pressurized Polymer Electrolyte embre Fuel Cell", The nd International Battery, Hybrid d Fuel Cell Electric Vehicle Symposium & Exposition,Paper no. EVS10371 Pacifico Yokohama, Yokohama, Jap October 3-8, 006.

5 . Alhuss, K. "ethod of Energy Trsfer", International Symposium on Physics of Fluids Hugsh, China, 9-1 June, Alhuss, K. Appliion of Computational Fluid Dynamics in Discontinuous Unsteady Flow with Large Amplitude Chges; The shock Tube Problem IASE Trsaction Issue 1 Volume, pp , Juary Alhuss, K. Supersonic Flow over Blunt Body with a Decelerator IASE Trsaction Issue 3 Volume 1, , August Alhuss, K. Supersonic Flow over Blunt Body with a Decelerator IASE Trsaction Issue 3 Volume 1, , August Fluent Inc. 6. Copyrighted by Fluent Inc. 7. A.A. Kulikovsky J. Divisek d A.A. Kornyshev. odeling the Cathode Compartment of Polymer Electrolyte Fuel J. Electrochemical Society, 146(11): , J.V. Cole d S. azumder. igorous 3-D athematical odeling of PE Fuel Cells J. Electrochemical Soc., 150:A1510, Sukkee Um C.Y. Wg d K.S. Chen Computational Fluid Dynamics odeling of PE Fuel Cells J. Electrochemical Society, 147(1): , J.H. Nam d. Karviy Effective Diffusivity d Water-Saturation Distribution in Single- d Twolayer PEFC Diffusion edium Int. J. Heat ass Trsfer, T. V. Nguyen. odeling two-phase flow in the porous electrodes of proton exchge bre fuel cells using the interdigitated flow fields Electrochemical Soc. Proceedings, 99(14):-41, T.E. Springer T.A. Zawodzinski d S. Gottesfeld Polymer Electrolyte Fuel Cell odel J. Electrochemical Soc., 138:334, Brdon. Eaton. One dimensional, Trsient odel of Heat, ass, d Charge Trsfer in a Proton Exchge embre..s. Thesis, 001.