IUST International Journal of Engineering Science,

Vol. 19, No.5-2, 2008, Page 73-81

A PARAMETRIC STUDY OF THE PEM FUEL CELL

CATHODE
S. Rowshanzamir, M. H. Eikani, M. Khoshnoodi & T. Eshagh Nimvar

Abstract: Most of the overpotential in the solid polymer electrolyte fuel cell is
related to the cathode electrode, particularly the cathode catalyst layer. In this
research work, a mathematical model for the cathode electrode catalyst layer has
been developed. The model is one dimensional, isothermal, pseudo-homogeneous
and uses an adjustable parameter. The mathematical model has four nonlinear
couple differential equations that have been solved using MATLAB software. The
results obtained by comparing the existing data and model predictions are
acceptable. In an optimized fuel cell electrode, it is desirable to achieve higher

D
current densities, lower overpotential and lower catalyst consumption. In this
regard, some of design parameters such as overpotential, proton conductivity,
catalyst layer porosity and catalyst surface area on the performance of fuel cell
have been studied.

SI
Keywords: Solid polymer electrolyte fuel cell, Modeling, Catalyst layer, Gas
diffusion electrode
of
1. Introduction1 fuel cell. The mathematical models devoted to the
The proton exchange membrane fuel cell catalyst layer of a fuel cell electrode include the
(PEMFC), is one of the most attractive energy pseudo-homogeneous [4-11], the single pore [12], the
conversion devices. It is regarded as a power source for thin film [13], and agglomerate models [7, 13]. Among
ive

stationary and mobile application. PEM fuel cell is these models the pseudo-homogeneous model, which
particularly suited to electric vehicle primarily because assumes that the catalyst layer has a uniform thickness
of its relatively low temperature (70-90oC), efficiency and uniform distribution of catalyst particles, ionomer,
and high power density, quick start-up capability, zero and gas pores throughout the catalyst layer, is the one
emissions is comparable to existing internal most often used, mainly because its compares well
combustion technology. The application to electric with experimental results.
ch

vehicle requires a PEMFC offering a high power The pseudo-homogeneous model has been adopted the
density and a low catalyst loading to reduce weight, following studies in regard to the PEM fuel cell
volume and financial costs. Because the thin cathode cathode. Bernardi and Verbrugge [4,5] presented a
catalyst layer of PEM fuel cell is the region where the complete model for a gas diffusion cathode bonded to a
Ar

electric power is generated and the largest irreversible polymer electrolyte. They incorporated the electro-
losses in cell voltage occur in it, the cell performance osmotic and pressure-driven water transport consonant
depends largely on the optimization of the various of the PEM fuel cell to investigate the overpotential
processes occurring there[1-3]. characteristics, water transport, and catalyst utilization.
Mathematical models are useful for analysis and Springer et al. [6] developed a model considering, in
optimization of the performance of fuel cell and detail, the voltage losses caused by interfacial kinetics
especially for cathode catalyst layer in the case of PEM at the Pt/ionomer interface, gas transport limitations
and proton conduction limitations, in the cathode
backing.
Paper first received May.20, 2007, and in revised form Broka and Ekdung [7] studied the discrepancies in
April. 13. 2009.
S. Rowshanzami, Fuel Cell Laboratory, Green Research Center
polarization curves given by the pseudo-homogeneous
(GRC), Iran University of Science and Technology (IUST), Tehran, model and the agglomerate model under various values
Iran, rowshanzamir@iust.ac.ir of oxygen permeability, conductivities and thickness of
Mohammad H. Eikani, Department of Chemical Industries, Iranian the cathode catalyst layer. Srinivasan et al. [8] studied
Research Organization for Science & Technology (IROST), Tehran,
Iran, eikani@irost.org
the effects of ohmic, diffusion and activation
M. Khoshnoodi, Department of Chemical Engineering, University of overpotentials on electrode kinetics using both pseudo-
Sistan and Baluchestan, Zahedan, Iran, khosh@hamoon.usb.ac.ir homogeneous and agglomerate models. They also
T. Eshagh Nimvar, M.S. student at the Department of Chemical compared the kinetics of oxygen reduction reaction on
Engineering, University of Sistan and Baluchestan, Zahedan, Iran,
tooba_eshagh@yahoo.com
platinum/carbon powders in Nafion film using the

www.SID.ir
74 A Parametric Study of the PEM Fuel Cell Cathode

rotating disk electrode and gas diffusion electrode. 2. Model Description

Marr and Li [9] used a mathematical model to study Fig. 1 presented a schematic of the cathode bonded
the performance of the cathode catalyst layer in a PEM to a proton exchange membrane in a PEM fuel cell.
fuel cell by including both electrochemical reaction, The system can be divided into three regions: a
and mass transport processes. The composition and membrane region of hydrated solid polymer
performance optimization of catalyst platinum has also electrolyte, an active catalyst layer that provides
been investigated. Sui et al. [10] compared both catalytic site for reduction of oxygen, and a gas
pseudo-homogeneous and agglomerate catalyst diffusion region (gas diffusion layer) composed of
models. highly porous and conductive materials.
Wang et al. considered a spherical flooded- The gas diffusion layer is located adjacent to the gas
agglomerate model for the cathode catalyst layer of a chamber (air/ pure oxygen) and the current collector.
proton exchange membrane fuel cell, which includes Its open pores and electronically conductive materials
the kinetics of oxygen reduction, at the transport O2 and electrons to the catalyst layer,
catalyst/electrolyte interface, proton transport through respectively.
the polymer electrolyte network, the oxygen diffusion Membrane is an electronic insulator, but an excellent
through gas pore, and the dissolved oxygen diffusion conductor of protons. The protons produced at the
through electrolyte [14]. The optimization of anode go through membrane to the catalyst layer of the

D
distributions of Nafion content and Pt loading in order cathode, where they react with electrons and O2 to
to get a best performance was presented by Sond et al. form H2O.
[15].

SI
Berg et al., developed a macroscopic model for a non-
O2 4 H 4e 2 H 2O (1)
isothermal cathode catalyst layer (CL) in a proton
exchange membrane (PEM) fuel cell is presented, in
which liquid water in the CL pores is neglected. The The catalyst layer is formed as a thin layer of proton
model couples three phases: an electrically conductive conductive ionomer (e.g. Nafion), a carbon supported
carbon/platinum phase, gas pores, and a proton- catalyst (e.g. Pt/C), and a void space.
of
conducting Nafion phase [16]. A steady-state The ionomer portion of the catalyst layer forms the
mathematical model for the ordered cathode of proton transport avenue for protons, the matrix portion
exchange membrane fuel cells is developed to consisting of carbon and catalyst plays the role of
investigate the dependence of the cathode performance electronic conductor and the void space portion
on the structural parameters of the catalyst layer by Du transport O2 to active site [6, 7].
ive

et al. [17] In the present study we focus on the catalyst layer of

Wang et al. presented a pore-scale model for oxygen, the cathode, which is in the region of 0 z and
water vapor and proton transport through a three- describe a macro-pseudo-homogeneous model with the
dimensional, random catalyst layer microstructure [18]. following assumptions:
A model-based control structure is proposed that 1. The fuel cell is operating under a steady state.
ch

comprises a multivariable control of the cathode 2. The distribution of oxygen concentration, current
pressure and the excess ratio of oxygen using the mass density, oxygen transport flux, and the overpotetial
flow controller and the outlet throttle as actuators by in the catalyst layer is considered to be one
Danzer et al.[19] A three-dimensional, steady-state, dimensional.
multi-agglomerate model of cathode catalyst layer in 3. The catalyst layer is isothermal.
Ar

polymer electrolyte membrane (PEM) fuel cells has 4. The ionomer is considered as being completely
been developed to assess the activation polarization saturated of water.
and the current densities in the cathode catalyst layer 5. The oxygen permeation through the PEM is
by Das et al. [20] negligible.
Liu and Eikerling accounted for the full coupling of 6. The diffusion coefficient is constant.
random porous morphology, transport properties, and 7. The electron conductivity of carbon matrix is
electrochemical conversion in cathode catalyst layers. substantially greater than the proton conductivity of
It relates spatial distributions of water, oxygen, ionomer, and the potential drop among solid matrix
electrostatic potential, and reaction rates to the carbon is negligible compared to the potential drop
effectiveness of catalyst utilization, water handling in the electrolyte phase [4,6,9].
capabilities, and voltage efficiency [21]. In this paper, 8. The anodic overpotential is disregarded with
a pseudo-homogeneous model similar to You et al. respect to the cathode ones [4,5]. Consequently, the
[11] is used to study the cathode catalyst layer. The voltage across the anode is considered to be
model is derived from a basic mass current balance by negligible.
the control volume approach. The two-point problem is
solved using shooting method with Matlab software
(version. 6.5) and the model results are compared with
open literature data [11].

www.SID.ir
S. Rowshanzamir, M. H. Eikani, M. Khoshnoodi & T. Eshagh Nimvar 75

concentration is constant at the catalyst layer.

Therefore, the exchange current density is function of
dissolved oxygen concentration:

c c
C Of
2
i 0 i 0 , ref ( ) (7)
C Of
2 , ref

The change in current is proportional to reaction

current density and catalyst surface area per unit
volume, a .
dI z
i.a (8)
dz
Fig. 1. Schematic of the cathode bonded to a proton or
exchange membrane in a PEM fuel cell
dI z COf 2 F F
ai c0 , ref ( f
) exp c exp a
(9)
dz CO RT RT

D
We consider the catalyst layer in Fig.1 as a 2 , ref

homogeneous medium. Consider a small control

volume as shown in Fig.2. When water velocity is neglected, the equation of
Ohms law for ionomer phase expressed as:

SI
CO2, Z,
d I z
(10)
Nz Nz+dz
dz k eff

Where, the proton current is related to the ionomer

Iz Iz+dz
phase potential difference.
of
From Eqs. (5), (10) and assuming the matrix of solid
Fig.2. Control volume for oxygen mass flux and phase to be equipotential (as assumption 7), changes in
current balance. overpotential can be expressed as:
The decrease of oxygen molar flux should be balanced d ( ) I
by the increase of proton current density. From Fig. 2, z (11)
ive

the oxygen consumption rate in the control volume is: dz keff

. dN The oxygen diffusion in the z direction occurs in
ms N z N z dz dz (2) both the gas phase and ionomer, fluxes in the gas phase
dz
are greater than ionomer phase. Thus, diffusion in the
The balance of current density is:
ionomer is a controller parameter, because the oxygen
ch

. dN
I z dz I z m s 4 F I z 4 F dz reduction reaction takes places in the ionomer. Finally,
(3)
dz according to Ficks law, the oxygen flux in the ionomer
or is related to dissolve oxygen concentration in the
ionomer by:
dI z dN
d (C O ) f
Ar

4 F (4) eff (12)

dz dz N O 2 D O2 , f
2

dz
When the electrochemical reaction (1) take place at eff
cathode, the electrode potential deviates from Where D is the effective oxygen diffusion
O2 , f
equilibrium potential, and it is defined as: coefficient, and can be determined from:
1.5
E E eq solid (5) DOeff2 , f DO 2 m (13)

The electrochemical reaction rate (oxygen reduction There are four dependent variables, I , , N , C
z O2 O2
per unit volume) in the catalyst layer can be described involved in four first-order differential equations, Eqs.
using the Butler-Volmer rate expression with the (4), (9), (11) and (12). The appropriate boundary
assumption that reduction current is positive [4]. conditions are listed below. At interface between the
c F F cathode gas diffuser and catalyst layer (z = 0):
i i0 exp c exp a (6)
RT RT Iz 0 (14)

Where, c I
i0 is the exchange current density at the NO2 (15)
4F
equilibrium potential. Since, the catalyst layer is
assumed to be grain supported, expansion of ionomer CO COz 2 ,0gas (16)
2
with hydration is neglected and assumes that the proton

www.SID.ir
76 A Parametric Study of the PEM Fuel Cell Cathode

At interface between the membrane and catalyst layer At 1

(z = ):
I* I * (31)
I 1 I (17) 1

The governing equations, 24-27 represent the four

dependent variables: I * , N * , x and . Once their values
3. Solution Technique
The governing equations and the corresponding at 0 are designated, the nonlinear first-order initial-
boundary conditions together pose a boundary-value value problem can be solved to yield I * , N * , x and
problem. This boundary-value problem can be reduced
using following relations: profiles using numerical methods (e.g. the Runge-Kutta
z method at Matlab). Before integration is begun, the
(18) current density, I * for the electrode is assumed. We

C f have the initial values of I * , N * and x from Eqs. (28),

O2
x (19) (29) and (30). However, the value of overpotential of
COz 2 ,0gas the cathode gas diffuser layer /active catalyst layer

D
Because of the resistance of porous electrode, the interface must be chosen to ensure that the solution
oxygen transport flux across the catalyst layer is profiles satisfy Eqs. (31).
limited. The maximum oxygen flux can be defined as:
COz 2 ,0gas

SI
eff (20) 4. Results and Discussions
N max D O 2 , gas After the solution of equations 24-27 and with

neglecting the overpotential loss in the anode side, the
NO (21)
N* 2 cell output voltage, is calculated as:
N max
U U oc IAR m c (32)
Z
The corresponding current density is defined as the
of
limiting current density Fig. 3 compares the calculated fuel call potential as a
4 FDOeff , gas COz 2 ,0gas function of current density with open literature data for
I lim 4 FN max 2 (22) the base-case conditions (Tab.1). The model parameter
was adjusted to yield model results that suitably mimic
ive

I the open literature data. With this one adjustable

I* (23) parameter, the agreement between model and open
I lim
literature data[11] is quite good.
By using the dimensionless variables, I * , N * , x, , the The dissolved oxygen concentration is calculated by
governing equations take the following form equations:
Pwsat
ch

dN * dI * x wsat (33)
(24)
d d Pc
ai c COz ,0gas 1 1
dI * K O 2 , 2 K O 2 , 1 exp G (34)
0 , ref ( f 2 ) S x R T T 1
2
Ar

d I lim CO
2 , re f
(25) The constant Henry in Nafion at 50 0C is 0.265
c F F atm.cm3/mol [22].
exp exp a
RT P c
RT C Osat2 (1 x wsat ) or ,
d ( ) I lim * K O2 ,2
I (26) (35)
d k eff f P gas O 2
C O2
D Oeff , gas K O 2 ,2
dx
2
eff
N* (27) The effective oxygen diffusion coefficients in gas
d SD O , f phase, effective proton conductivity, and open circuit
2

The corresponding dimensionless boundary conditions potential are calculated by:

become:
At 0
DOeff , gas DO 1.5 (36)
2 2

I* 0 (28)
(29)
k eff m k (37)
N* I

RT
x 1 (30) Uoc 1.23 0.9 103 (T 298) 2.3 log(PH2 PO ) (38)
4F 2 2

www.SID.ir
S. Rowshanzamir, M. H. Eikani, M. Khoshnoodi & T. Eshagh Nimvar 77

4.1. The Effects of Overpotential so absolute overpotential increases across the catalyst
Fig. 4 show the effects of overpotential (at catalyst layer. While is close to 0.35, the current density
layer/membrane interface) on the proton current, reaches the limiting current density, and oxygen
oxygen concentration and overpotential distribution concentration becomes zero at catalyst layer/membrane
across the catalyst layer, respectively. Other parameters interface. With further increase of overpotential, the
are given in Table 1. The electrochemical reaction rate oxygen reduction rate increase, most oxygen is
depends on the local oxygen concentration and consumed in a thin region of the catalyst layer, the
overpotential. Across the catalyst layer, from the gas active catalyst layer and because of the oxygen mass-
diffuser/catalyst layer interface (=0) to the catalyst transport limitation; the current density cannot
layer/membrane interface (=1), the dimensionless increase.
current density increase; dimensionless oxygen
concentration decrease, and the absolute overpotential 4.2. The Effect of Proton Conductivity
increases due to ohmic losses across the catalyst layer. Fig. 5 show effects of proton conductivity on the
When overpotential at catalyst layer/membrane proton current, oxygen concentration and overpotential
distribution across the catalyst layer, respectively. The
interface is small, (0.1), the oxygen reduction
comparison is based on the same overpotential at the
rate is very low, little oxygen is consumed, and the
catalyst layer/membrane interface (total activation and

D
oxygen concentration remains almost unchanged.
Thus, only a small current density can be produced concentration overpotential is 0.3) and in this case the
(across the catalyst layer, the dimensionless current highest possible current is desired to improve fuel cell
density increase). Since the gradient of overpotential is performance. When proton conductivity at catalyst

SI
inversely proportional to the current density when layer decrease, ohmic resistance increases and the
corresponding potential loss in the electrolyte phase
proton conductivity is assumed as constant, the
potential loss in the electrolyte is small, and the from catalyst layer/membrane interface to the catalyst
overpotential across the width catalyst layer remains layer/gas diffuser interface increases. At low proton
almost unchanged. conductivity (= 0.005 (cm)-1), the reaction is proton
conduction controlled and oxygen consumption,
of
Tab. 1.Cathode catalyst layer parameters for base current generation rate are low in the region 0.5,
case [11]. and most of the current generation in a thin region
closest the catalyst layer/membrane interface. While
Parameter value when proton conductivity is high, oxygen reduction is
Catalyst layer porosity, 0.25 [-]
diffusion controlled, and most of the reaction occurs in
ive

Catalyst layer thickness , c 0.00129 cm

Reference kinetic parameter/ Oxygen 5.0174107 A/mol
a thin region closest to the catalyst layer/gas diffuser
reference concentration, interface. High proton conductivity can be achieved by
ai0 , ref / C Of increasing water content and ionmer fraction in the
2 , re f catalyst layer. While water content is mainly related to
(model parameter) operating conditions, ionomer fraction in the catalyst
ch

Cathodic transfer coefficient, c .52 [-] layer is a design parameter. As shown in Fig. 5, when
Anodic transfer coefficient, a .54 [-]
Oxygen diffusion coefficient in plain 0.0056 cm2/s
= 0.5(cm)-1, the overall current density approaches
medium, Do2 the limiting current density at about =1, which means
proton conductivity, 0.05 1/. cm an optimum ionomer fraction exists in the catalyst
O2 concentration parameter for i0, 1 [-] layer, because enough carbon particle, catalyst particle
Ar

Cell temperature, T 350 K and porosity for oxygen transport should also be
O2 Henrys constant, KO2 3.078105 atm-
provided in the catalyst layer.
cm3/mol
Air-side pressure , P 3 bar
Fuel-side pressure , P 1 bar
Ralative humidity of inlet air and fuel 100% [-]
Inlet fuel and air temperature, To 333 k
Thermodynamic open-circuit potential, 1.174 V
U
Volume fraction membrane in active 0.78 mol/cm3
layer, m
Ratio of oxygen concentration in 0.96 mol/cm3
ionomer to gas phase,CfO2 /CgO2
Activation energy, G 5.21 kJ/ mol
Overpotential at catalyst/membrane 0.3 V
interface

With the increase of overpotential, current density

increase, more oxygen is consumed, and the
corresponding of high current density, the electrolyte Fig. 3. Model and open literature data for fuel
phase potential loss across the catalyst layer is high and cell potential as a function of operating current
density at base-conditions (tab. 1)

www.SID.ir
78 A Parametric Study of the PEM Fuel Cell Cathode

and enough polymer electrolyte for proton transport,

thus pores in the catalyst layer for oxygen transport is
limited. The oxygen transport resistance depends on
the effective porosity. As seen in Fig.6, when the
effective porosity of the catalyst layer decrease from
0.35 to 0.15, the oxygen transport resistance increase
and oxygen is quickly consumed at catalyst layer,
limiting current density decrease.

D
SI
of
ive
ch
Ar

Fig. 4. The variation of current density, oxygen

concentration with overpotential and overpotential
across the catalyst layer.

4.3. The Effects of Effective Porosity of the Catalyst

Layer
Fig. 6 shows the influences of effective porosity
on the performance of the catalyst layer. Oxygen can
reach to surface catalyst through the hydrophobic pores
of solid matrix. Therefore, the catalyst layer must have Fig. 5. The variation of current density, oxygen
enough carbon matrix for good electron conductivity, concentration and overpotential with proton
enough Pt particle surface for electrochemical reaction, conductivity across the catalyst layer.

www.SID.ir
S. Rowshanzamir, M. H. Eikani, M. Khoshnoodi & T. Eshagh Nimvar 79

related to the electrochemical reaction and can be

improved by controlling electrochemical reaction
conditions, such as temperature, concentration, and
favored electrochemical reaction paths.

D
SI (a)
of
ive

Fig. 6. The variation of current density, oxygen

concentration with porosity across the catalyst
layer.
ch

4.4. The Effects of the Active Catalyst Surface Area

Fig. 7 shows the influences of the active catalyst (b)
surface area per unit volume on the performance of the
catalyst layer. The active catalyst surface area increases
Ar

by using finer catalyst particles (such as Pt) and/or by

increasing catalyst loading, which usually comes
higher capital. As shown in Fig. 7(a), when active
catalyst surface area is increased, the current density
increased, though the overall current density is still
controlled by the limiting current density. Fig. 7(b)
shows the variations of oxygen concentration with
surface area. When a =base case/2.5, the
electrochemical reaction will not be strong enough, and
the overall current density will be lower than a = base
case. Thus, there also exists an optimum catalyst
surface area. Fig. 7(c) provides the variations of
overpotential with catalyst surface areas. There are (c)
other parameters that influence the performance of the
Fig. 7. (a) The variation of current density with
catalyst layer, the exchange current density, cathodic
catalyst surface area, (b) the variation of oxygen
transfer coefficient and anodic transfer coefficient.
concentration with catalyst surface area, (c) the
Increasing exchange current density and cathodic
variation of overpotential with catalyst surface area
transfer coefficient can increase the current density
across the catalyst layer.
when the overpotential is given. These parameters are

www.SID.ir
80 A Parametric Study of the PEM Fuel Cell Cathode

5. Conclusions : Dimensionless oxygen concentration

A pseudo-homogeneous model of proton exchange : Dimensionless length through catalyst layer
membrane fuel cell for cathode catalyst layer has been : O2 concentration parameter for i0
proposed and described in details. This model has been : catalyst layer thickness, cm
solved by Matlab software. Although the model is : potential, V
relatively straightforward, fuel cell performance : Porosity
predicted by this model compared well with the open : ionic conductivity, (. Cm)-1
literature data [11]. The parametric effects on the : overpotential, V
performance of the cathode catalyst layer in a proton
exchange membrane fuel cell are systematically
studied. Based on the modeling results are as: Subscripts and superscripts
Before reaching to limiting current density, c: cathode catalyst layer
increasing the overpotential leads to increasing the eff: effective property which accounts for porosity
current density. eq: equilibrium
Increasing of the proton conductivity increases the f: ionomer film
current density and decreases the variation of lim: limiting current density
m: membrane

D
overpotential in the catalyst layer and also decreases
O2: oxygen
the catalyst layer effective thickness.
oc: open circuit
By decreasing the catalyst layer porosity, the oxygen
Solid: electrode surface
mass transfer, the limiting current density, the ratio

SI
sat: saturation
of transfer to consumption of oxygen and the catalyst
w: water
layer effective thickness are decreased.
z: dimension through catalyst layer
By increasing the catalyst surface area, the current
density, the rate of oxygen consumption, the ratio of
consumption to transfer of oxygen and the variation References
of
of overpotential are increased and the catalyst layer [1] Ceraolo, M., Miulli, C., Pozio, A., Modeling tatic and
effective thickness is decreased. Dynamic Behaviourl of Proton Exchange Membrane
Fuel Cells on the Basis of Electro-Chemical
It should be mentioned that in the low limitting
Description, J. Power Sour. Vol. 113, 2003, pp.131-
current densities (low oxygen transfer), increasing of 144.
catalyst surface area has no influence on the fuel cell
ive

performance. [2] Antoine, O., Bultel, Y., Ozil, P., Durand, R., Catalyst
Based on the results, the design parameters for the Gradient for Cathode Active Layer of Proton Exchange
Membrane Fuel ell, J. Electrochemica. Acta., Vol. 45,
cathode catalyst layer can be optimized for different
2000,4493-4500.
working conditions.
[3] Song, D., Wang, Q., Liu, Z., Navessin, T., Eikerling, M.,
Holdcroft, S., Numerical Optimization Study of the
ch

6. Nomenclature
Catalyst Layer of PEM Fuel Cell Cathode, J. Power
a: surface area per unit volume of catalyst layer, Sour. Vol. 126 , 2004, pp. 104-111.
cm2/cm3
ARm: membrane rsistance, . cm2 [4] Bernardi, D.M., Verbrugge, M.W., Mathematical Model
C: concentration, mol/cm3 of a Gas Diffusion Electrode Bonded to a Polymer
Ar

D: diffusion coefficient, m2/s Eectrolyte, AIChE J. Vol. 37 (8), 1991, pp.1151-1163.

E: potential, V [5] Bernardi, D.M., Verbrugge, M.W., Mathematical Model
F: Faraday constant, 96487 C/ mol of the Solid-Polymer-Electrolyte Fuel Cell, J.
G: activation energy, kJ/mol Electrochem. Soc. Vol. 139 (9), 1992, pp. 2477-2490.
i0: catalyst exchange current density, A/cm2
I: current density, A/cm2 [6] Springer, T.E., Wilson, M.S., Gottesfeld, S., Modeling
m s and Experimental Fiagnostics in Polymer Fuel Cells, J.
: Oxygen consumption rate, mol/s Electrochem. Soc. Vol. 140 (12), 1993, pp. 3513-3526.
N: molar flux, mol/cm2.s
R: universal gas constant, 8.314 J/mol.K [7] Broka, K., Ekdunge, P., Modeling the PEM Fuel Cell
S: ratio of oxygen concentration in ionomer to gas Cathode, J. Appl. Electrochem., Vol. 27, 1997, pp.
phase, 281-9.
CfO2 /CgO2 [8] Gloaguen, F., Convert, P., Gamburzer, S., Velve, O.A.,
T: absolute temperature, K Sirinivasan, S., An Evaluation of the Macro-
U: voltage, V Homogeneous Model for Oxygen Reduction in PEM
z: dimension through the catalyst layer, cm Fuel Cell, J. Electrochemica. Acta., Vol. 43 (9), 1998,
pp. 1077-1086.
Greek symbols [9] Marr, C., Li, X., Composition and Performance
c: Cathodic charge transfer coefficient Modeling of Catalyst Layer in a Proton Exchange

www.SID.ir
S. Rowshanzamir, M. H. Eikani, M. Khoshnoodi & T. Eshagh Nimvar 81

Membrane Fuel Cell, J. Power Sour., Vol. 77 , 1999, Interface - Electrode-Kinetics and Mass-Transport," J.
pp.17-27. Electrochem. Soc., Vol, 139, 1992, pp. 2856-2862.

[10] Sui, P.C., Chen, L.D., Seaba, J.P., Wariishi, Y.,

Modeling and Optimization of a PEMFC Catalyst
Layer, SAE Congress, 1999-01-0539, 1999, pp. 61-70.

[11] You, L., Liu, H., A Parametric Study of the Cathode

Catalyst Layer of PEM Fuel Cells Using a Pseudo-
Homogeneous Model, Int. J. Hydrogen Energy, Vol.
26, 2001, pp. 991-999.

[12] Srinivasan, S., Hurwitz, H.D., Bockris, O.M., Chem.

Phys., J., "Fundamental Equations of Electrochemical
Kinetics at Porous gas-Diffusion Electrodes," Vol. 46
(8), 1967, pp. 3108.

[13] Bultel, Y., Ozil, P., Durand, R., Modified Thin Film

D
and Agglomerate Models for Active Layers of P.E.M
Fuel Cells, Electrochim. Acta, Vol. 43 (9), 1998,
pp.1077.

[14] Wang, Q., et al., A Mathematical Model and

Optimization of the Cathode Catalyst Layer Structure
in PEM Fuel Cells, Electrochimica Acta, Vol. 50,
2004, pp. 725730.

[15] Sond, D., et al., A Method for Optimizing Distributions

SI
of
of Nafion and Pt in Cathode Catalyst Layers of PEM
Fuel Cells, Electrochimica Acta, Vol. 50, 2005, pp.
33473358.

[16] Berg, P., et al., Analysis of a Cathode Catalyst Layer

ive

Model for a Polymer Electrolyte Fuel Cell, Chemical

Engineering Science, Vol. 61, 2006, pp. 4316 4331.

[17] Du, C.Y., et al., Parametric study of a novel Cathode

Catalyst Layer in Proton Exchange Membrane Fuel
Cells, Journal of Power Sources, Vol. 160, 2006, pp.
ch

224231.

[18] Wang, G., et al., Optimization of Polymer Electrolyte

Fuel Cell Cathode Catalyst Layers Via Direct
numerical Simulation Modeling, Electrochimica Acta,
Vol. 52, 2007, pp. 63676377.
Ar

[19] Danzer, M.A., et al., Model-Based Control of Cathode

Pressure and Oxygen Excess Ratio of a PEM Fuel Cell
System, Journal of Power Sources, Vol. 176, 2008, pp.
515522.

[20] Das, P.K., et al., A Three-Dimensional Agglomerate

Model for the Cathode Catalyst Layer of PEM Fuel
Cells, Journal of Power Sources, Vol. 179, 2008, pp.
186199.

[21] Liu, J., Eikerling, M., Model of Cathode Catalyst

Layers for Polymer Electrolyte Fuel Cells: The Role of
Porous Structure and Water Accumulation,
Electrochimica Acta, Vol. 53, 2008, pp. 44354446.

[22] Parthasarathy, A., Srinivasan, S.A.J., Appleby, C.R.,

Martin, "Pressure-Dependence of the Oxygen Reduction
Reaction at the Platinum Microelectrode Nafion

www.SID.ir