CC BY
24
-□ □-
Одним з факторiв, яК можуть тдви-щити властивкть поверхневого перенесен-ня для дифузп газу з матерiалу мембрани в ПЕПОМ, е пористьсть матерiалу газоди-фузшного шару, що впливае нарозподш час-ток на електродах. В даному до^джент проводиться моделювання продуктивность ПЕПОМ шляхом чисельного дослгдження впливу пористость ГДШв деяких комерцш-них ELAT-TEK-1200W (г=0,31) i SIGRACET 25ВА (г=0,63), а також оргашчних матен рiалах, таких як кокосова койра (е=0,88). Для цього використовували COMSOL Multiphysics 5.3а у виглядi масових концентраций частинок, нанесених на профшь поверхш i точок розрiзу на електродах за минулий час. Потiм отримаш результаты були використан для визна-чення продуктивность ПЕПОМ шляхом розрахунку деяких втрат; активацшних, омiчних i масовог концентрацшног поля-ризацп. Результати показали, що на про-дуктивтсть ПЕПОМ впливае тьльки масо-ва поляризация. Це означае, що питома потужтсть сильно залежить вiд концен-трацг частинок в анод i катод. Масова концентрация сильно залежить вiд розпо-дЫу часток; в ходi реакцг утворюють-ся Н2, О2 i Н2О. Найбшьша концентрация Н2 на анодi спостер^аеться в ГДШ з використанням ELAT-TEK-1200W, що мае найменшу пористьсть, що забезпечуе най-вищу питому потужтсть в порiвняннi з тшими матерiалами ГДШ. Це полегшуе процес дифузп мiж частинками Н2 i 02. Однак кокосова койра як орга^чний мате-рiал може в майбутньому стати перспек-тивним ГДШ завдяки свогй ефективность в порiвняннi з тшими
Ключовi слова: чисельне дослгджен-ня, ПЕПОМ, продуктивтсть, пористьсть, газодифузтний шар, матерiал, COMSOL
Multiphysics -□ □-
UDC612.822
I DOI: 10.15587/1729-4061.2020.181442]
A NUMERICAL STUDY OF PROTON EXCHANGE MEMBRANE FUEL CELL PERFORMANCES AFFECTED BY VARIOUS POROSITIES OF GAS DIFUSSION LAYER MATERIALS
Widya Wijayanti
Associate Professor* Е-mail : widya_dinata@ub.ac.id Rizky Kusumastuti
Master student** E-mail: rizkykusumastuti94@gmail.com Sasmoko PhD student** E-mail: sasmoko@ymail.com Mega Nur Sasongko Associate Professor* Е-mail : megasasongko@ub.ac.id *Department of Mechanical Engineering Brawijaya University Jl. Mayjen Haryono, 167, Malang, Indonesia, 65145 **Department of Mechanical Engineering National Central University No. 300, Zhongda Road, Zhongli District, Taoyuan City, Taiwan, 320
Received date 23.10.2019 Copyright © 2020, Widya Wijayanti, Rizky Kusumastuti, Sasmoko, Mega Nur Sasongko
Accepted date 14.02.2020 This is an open access article under the CC BY license
Published date 24.02.2020 (http://creativecommons.org/licenses/by/4.0)
1. Introduction
One of the energy sources that promise in the future is a fuel cell. It is an electro-chemical device that produces direct electrical current from the chemical energy stored in a fuel [1]. It is the most promising energy conversion device because it has the cleanest technology, relatively low operating temperature (around 80 °C), almost no noise, and has a high efficiency to generate electricity (around 70 %). The exhaust emissions are water and heat, which are used as "fuel cell by-products". At present, it is used in various applications including electricity for households and trans-
portation [2]. From several types of fuel cells, the Proton Exchange Membrane Fuel Cell (PEMFC) is the type that is very suitable for power plants because it uses a membrane on the stack. It is a very practical certainty compared to after fuel cell types.
The PEMFC basic elements are bipolar plates, diffusion layers, electrodes and an electrolyte [3]. In the components, the membrane electrode assembly (MEA) is composed by a proton exchange membrane, catalyst layers, and gas diffusion layers (GDL) sandwiched in between flow field plates. The GDL is one of PEMFC parts, a vital electrode part located at Membrane-Electrode Assembly (MEA) [4-6]
that functions to provide a good pathway for reactants to catalysts [6]. Another function of the GDL is as gas diffuser and provides additional mechanical strength (mechanical support) for the membrane [6]. Consequently, the main requirements in selecting materials for GDL are:
1) having high electrical conductivity;
2) having high permeability;
3) having corrosion resistance;
4) facilitating water and CO2 disposal;
5) having good mechanical properties, good porosity for the inlet and outlet of the reactant and products;
6) a cheap price [7].
Various improvements and developments continue to be made to expand the efficiency of PEMFC because one of the weaknesses is the high water content in each electrode due to the trapping of water from the reaction between H2 and O2. Therefore, there have been many efforts focused on developing conductor membranes, including the GDL material with high conductivity and can operate at higher temperatures to expand the ability of H2 diffusivity can cause further surface transfer properties.
To meet the requirements of the GDL material, it is necessary to choose materials that meet the requirements that the availability of material that is easy and abundant, high permeability, and also has a cheap price. An organic matter that meets these criteria seems to be a material composed of compounds containing carbon. As we know, carbon has allotropic properties so that it can be used as a catalyst or material that is easy to absorb the water content. The GDL material must also transfer the generated heat on the catalyst away from the MEA. Therefore, the use of organic materials by the GDL must be done. Recently, the most widely used materials for GDLs is carbon in the form of commercial carbon [8] and organic carbon.
2. Literature review and problem statement
The commercial GDL used is generally produced using carbon-based polyacrylonitrile materials, which are derived from non-renewable energy materials having high production costs [9]. This problem makes some researchers develop the GDL by using carbon materials derived from biomass which is very abundant and is a renewable material. One of the promising biomass as the GDL materials which has been studied is coconut fiber [10]. It was observed that carbon coconut coir can be useful in distributing reactant gas. It has many parallel hollow tubes and functions as porous media material. It has a certain size porosity which is capable of distribution of the reactants throughout the catalyst layer. Coconut coir is also able to eliminate water and CO2 out of the cell. Besides, it is able to transfer the generated heat on the catalyst away from the MEA.
One of the factors that can increase the surface transfer property for gas diffusivity apart from the membrane material itself is the H2 mass distribution of the catalyst [11] which states that the reduction in membrane porosity size can increase the electronic conductivity of GDL. The size of porosity will also significantly affect the uniformity of reactant gas distribution, so that it will affect the generated electric current which shows the PEMFC performance. Besides the porosity is one of significant factors in determining gas and water transport through/in the GDL, the
distribution and variation in porosity give the effect on the capillary pressure driving the water out of the PEM-FC [12]. Therefore, in the present study, various porosities of the used organic materials as GDL will be considered. Some effects of the reactant transports in the PEMFC electrodes will be compared.
The porosity can be defined as the ratio of the volume of empty space in a material [13]. It can also be defined as the ratio of non-fiber volume to the total volume of material [14]. In the recent study, the effect of porosity that greatly affects the performance of GDL in 3 types of organic matter will be observed using COMSOL numerical simulation. The numerical simulation will investigate the diffusion of reactants that greatly affect the mass and mass transport that occurs in the electrodes (Anode and PEMFC cathode). Therefore, the study will compare the porosity effects of coconut carbon fiber (e=0.74) [10] with the two of commercial GDL materials; SIGRACET 25BA (e=0.88) [7] and ELAT-TEK-1200W (e=0.31) [12] using COMSOL Multiphysics 5.3a software. In the simulation results, the species mass concentration at the anode and cathode in the Gas Diffusion Layer (GDL) will be presented. If the previous studies [15] observed the gas velocity and the mass conservation in the steady conditions, however, in this study, the mass concentration will be simulated under transient conditions until the reactants reach the equilibrium state.
3. The aim and objectives of the study
The aims of the study are to determine the PEMFC performance due to different porosities among three GDL materials affecting the surface transport properties of gas diffu-sivity causing some losses or polarizations or overpotentials.
On the other hand, the transport properties make some losses in the PEMFC, so that it will give an impact on the PEMFC performance.
To achieve this aim, the following objectives are accomplished:
- by calculating the performance of PEMFC by investigating the effect of GDL porosities in some commercial ELAT-TEK-1200W (e=0.31), and SIGRACET 25BA (£=0.88), also an organic material coconut coir (£=0.74) numerically using COMSOL Multiphysics 5.3a;
- by determining the species mass distribution represented by the gas flow velocity and total mass conservation at the anode and cathode of the Gas Diffusion Layer (GDL);
- by calculating the electricity voltage to get the power density of the fuel cell. It explains that the PEMFC performance curve was influenced by three polarizations:
1) an activation polarization;
2) an ohmic polarization;
3) a concentration polarization.
4. Numerical simulation for GDL species transport
There are some assumptions for analysis conditions in the present study considering that the PEMFC (Fig. 1) is operated in unsteady (transient) and isothermal conditions. Then, pressure drops are ignored. It works in adiabatic conditions. It is only H2 that reacts electrochemically, while all gases and water obey ideal gas law.
©-
Table 1
Anode
Cathode
Dimensions
©-
Gh -©
©-
10
10
Anode Outlet
Anode Current Controller
Anode Inlet
Anode
Cathode inlet
Cathode Current Collector
Cathode Outlet
Cathode
Membrane
0.25 mm 0.1 mm 0.25 mm
0.25 mm
b
Fig. 1. Geometry model of PEMFC: a — size, domain, and boundary layers; b — cut point method measuring mass concentration
Parameters Value Unit
Cell height 2 mm
Cell width 0.25 mm
Membrane thickness 0.1 mm
Current collector height 1 mm
Table 2 Operating conditions
Parameters Anode Cathode
Reactant H2 O2
Pressure, Pa/Pc 1 atm 1 atm
Relative moisture, RHa/RHc 100 % 100 %
Operating temperature Ta/Tc 333 K 333 K
4. 1. Modeling approach
The physical properties of various organic matters for GDL materials are assumed to have the same mathematical parameters. However, the usage of different materials causes a difference in terms of porosity. Modeling of porosity effects of PEMFC has an essential material for the optimization of parameters affecting the performance of the fuel cell. The present study predicts the mass fraction effects and the transport phenomena in the transient conditions. The model is developed as having a single flow channel presented in Fig. 1, a and using model parameters given in Table 1, 2. In addition, some parameters and physical properties are listed in Table 3, using recent work assumptions and the other previous research.
The recent study uses a two-dimensional model, two-phase flow, ideal gas, non-isothermal PEM Fuel Cells and uses the agglomeration model simulated with COMSOL Multiphy-sics 5.3a. The model is divided into 3 domains; an anode, a membrane, and a cathode. Each GDL material is integrated directly with the gas distributor that has an inlet channel, a current collector, and a channel outlet. The same notation is used for the cathode side (Table 1).
Furthermore, the anode side is divided into 10 cut points to investigate the H2 mass concentration that occurs during PEMFC operation as shown in Fig. 1, b (Table 2).
Table 3
List of parameters for governing equations
Parameter Value Unity
Conductivity, solid phase, ks 5,000 S/m
Conductivity, membrane, km 7.2 S/m
Cell voltage, Vceu 0.6 V
Fluid viscosity, ^ 2.10x10-5 Pa s
Reference pressure, Pref 1 atm
Anode inlet pressure, Pa, in 1 atm
Cathode inlet pressure, Pc, out 1 atm
Equilibrium potential, anode 0 V
Equilibrium potential, cathode 1 V
Proton-water drag coefficient through membrane, D 3 -
Exchange current density, anode, i0, a 9.23x108 A/m2
Exchange current density, cathode, i0, c 1.05x106 A/m2
Specific surface area, S 1.04x107 m2/m3
Aggregate radius, Ragg 0.1 ^m
Faraday constant, F 96,487 C/mol
Microscopic porosity inside agglomerates, zmic - ELAT-TEK-1200 W=0.31; - coconut coir=0.74; - SIGRACET 25 BA=0.88 -
Macroscopic porosity between agglomerates, zmac - ELAT-TEK-1200 W=0.31; - coconut coir=0.74; - SIGRACET 25 BA=0.88 -
Anode inlet weight fraction H2, wH^ 0.2 -
Cathode inlet weight fraction O2, wO 0.168 -
Cathode inlet weight fraction, H2O, wh2o,in 0.2 -
Molar mass, H2, M„ 2 g/mo1
Molar mass, O2, M„ O2 32 g/mo1
Molar mass, H2O, MHO 18 g/mo1
Molar mass, N2, M„ 28 g/mo1
Henry's law constant, H2 in agglomerate, HH 4.56x103 Pa m3/mol
Henry's law constant, O2 in agglomerate, HO 0.101x1012 Pa m3/mol
Active layer thickness, Iact 10 ^m
1
1
2
2
3
3
4
4
5
5
6
6
7
7
9
8
8
9
9
a
4. 2. Model equations at porous media fluid flow of GDL
To model the gas flow, the reactant gas velocity can be calculated using Darcy's law based on the law of continuity determined by:
V. (pu) = 0 at the anode and cathode,
(1)
where p is the mixture density of the gas phase (kg/m3) and u is the velocity of the gas (m/s). Then, the ideal gas law states the density in the gas phase mixture:
p = — YM-x-.
(2)
At the electrode boundary aimed at the anode and cathode, the gas velocity was calculated from the total mass flow agreed by the stoichiometric coefficients using Faraday's law, resulting:
_T 7, _ 3anode
— 77
M„
nUchatode _
.^cathode
M„
1 ^H
Mu
(3)
(4)
concentration influenced by the use of three GDL materials with different porosity. Then, the PEMFC performance will be represented by the power density versus the resulting current. Before getting the result, several parameters need to be calculated.
Initially, the PEMFC operating voltage can be determined by the result of ideal voltage reduced by all three types of polarizations [16]:
Vcell ~ E ^aa nac nr ^ma nm
(5)
where E is open-circuit voltage (OCV), na, nr nm, respectively, are the activation polarization, the ohmic polarization and the mass concentration polarization.
E, the open-circuit voltage, is the ideal potential known as the theoretical potential which is not possible to be obtained in actual applications due to its irreversibility. Later, it is referred to as the polarization or the losses or the overpotentials.
5. 1. Activation overpotential
This polarization is the amount of voltage needed to generate the activation energy of an electrochemical reaction expressed with the ButtlerVolmer equation:
By combining all the boundary conditions, Darcy's law calculates the gas flow velocity and preserves the total mass conservation in the anode and cathode.
Further, in the anode, there are two gases; H2 and H2O while in the cathode, it presents three gases; O2 and N2 (air), as well as H2O. The model used one instance of the transport of concentrated species interface for each electrode side, then each electrode was focused in some cut points illustrated by Fig. 1, b.
4. 3. 2D Comsol Multiphysics model simulation procedures
In the recent study, Fuel Cell and Battery module were selected for the PEMFC system. The PEMFC model was modeled by adding Electrochemical in Secondary Current Distribution (CD) to model electrochemical currents using Ohm's law and the Butler-Vormer equation. Using the equations, some polarization plots and distributions of current density will be obtained. Then, determining the velocity of fluid flow in the GDL, the physics was selected in Fluid Flow in Porous Media and Subsurface Flow in Darcy's Law (dl). Meanwhile, finding out mass transportation that occurs at PEMFC, Transport Species Materials in Transport of Concentrates Species (tcs) was added. The physics was added twice at the anode and cathode sides because the mass transport at the anode is H2 and H2O, while at the cathode there was the mass transport of O2 and N2 (air). The Maxwell-Stefan equation was used to solve the mass transfer in the porosity of GDL. The Navier Stokes and the Brinkman equations are also used in the GDL and electrodes. The equations were also used to obtain the mass concentration profiles of substances at the anode and cathode in the physics.
5. Performance assessment of PEMFC using various organic materials
In addition, the study also calculates the PEMFC performance considering numerical simulation results of the mass
J = J0
exp
-exp
azF n*
RT
(1 - *) zF n*
RT
(11)
where a is known as the transfer coefficient (or symmetry factor) and the value changes in the range 0<a<1. Experimentally, the value often used for the value of a is around 0.5. Meanwhile, Jo,; is exchange current density and represents the readiness of the electrodes to form electrochemical reactions. Then, the activation overpotential for the anode and cathode can be calculated as follows:
nact,i
2RT
zF
sinh
J
2J
, i=a, c.
(12)
5. 2. Ohmic overpotential
Charge transport regulates the charge of the electrons and the ions from the production to the consumption. The resistance to the charge transport results in a lost voltage (the ohmic loss) [16]. For excess PEMFC resistance by considering the contribution of each cell component; anode, cathode, electrolyte, and interconnector, Ohm's law can be determined as follows [17]
nohm ,J(daPa+dePe+dcPc+dmtPmt),
(13)
in which, d and p are thickness and resistivity for anodes, cathodes, electrolytes and GDL.
5. 3. Concentration overpotential
The mass transfer process of uncharged species of both H2 and O2 is an important matter because the reactants and products mixing in the catalyst layer ensure the performance of the fuel cell. In order to minimize overpotential (concentration loss), it is able to be optimized at the electrode. The mass transport in a porous solid is eventually studied as a complex phenomenon. Knudsen diffusion re-
P
F
gards the diffusion of gas molecules over some very small capillary pores. An effective Knudsen diffusion takes into account the porosity and tortuosity of the electrodes as follows:
Deff = l l 8RT
l'k 3 r V nMi
p'
(14)
where £, r and rp represent porosity, tortuosity, and pore average radius of porous media. M is the molecular mass of species.
The difference in concentration of the two gas species between the electrode interface and the bulk becomes the concentration overpotential written by the equation as follows:
RT
JF'
In
Ph
RT
JF'
In
PA.
P'
(15)
(16)
P = JVcel,
Afterward, the partial pressure of the electrode-electrolyte of the gas species is explained by the Fickian diffusion [16].
Finally, after finding the appropriate parameters in some of the equations above, we will come at the purpose of the research, which is to get the PEMFC performance which can be formulated as follows :
(17)
where P stands for power density, J is commonly used as current density and VceU is the fuel cell operating voltage.
Finally, from some equations, the performance of PEMFC based on the effect of the porosity from the 3 different organic materials will be found.
6. Results
6. 1. Effect of porosity on the mass concentration profiles in elapsed time
In the concentration profiles, the surface contour and mass concentration of species at the anode and cathode for H2, O2, and H2O will be presented. The surface contour illustrates the mass concentration distribution during the reaction process, then it will present at the mass concentration profiles at each cut point measuring the mass concentration as illustrated in Fig. 1.
6. 2. Concentration profiles of H2
Due to H2 inlet through the anode, the H2 mass concentration is shown their surface contour and mass concentration as illustrated in Fig. 2. It is represented by one of the organic material; the coconut coir.
Then, the H2 mass concentrations of three GDL materi als are plotted in Fig. 3 from each cut point method measur ing illustrated in Fig. 1, b.
c
Fig. 2. Hydrogen mass concentration at anode of Coconut Coir: a - t=0.02 s; b - t=0.05 s; c - t=0.09 s
Afterward, the mass concentration of H2 in each material having different porosities can be compared as shown in Fig. 4.
n
am, u
a
n
con, c
b
0.04 0.06
time (s)
c
Fig. 3. H2 mass concentration at anode: a - ELAT-TEK-1200W; b - Coconut Coir; c - SIGRACET 25BA
Fig. 3 explains that the H2 mass concentration increases in elapsed time. It increases with the time, even it is close to the anode outlet, the H2 concentration tends to be constant. It is due to H2 has reacted on the electrode to release the protons in the electrolyte membrane toward the cathode. The conditions are different from the area
that is far from the inlet. In the area, the H2 concentration increases significantly because the unreacted H2 flows upward, causing the increase of the H2 mass concentration at the anode outlet. Correspondingly, if we look at the surface contour in Fig. 2, it can be understood that there is an increasing H2 concentration in elapsed time and pile up in the anode outlet. Automatically, in the steady state conditions (after t=0.09 s), the anode outlet has the highest mass concentration of hydrogen.
I S
0.07 0.06 0.05 0.04 0.03 0.02 0.01 0
-ELAT-TEK-1200W -Coconut Coir -SIGRACET 25BA
23456789 Cut Point in Steady State
10 11
Fig. 4. The effect of different porosities of Hydrogen mass concentrations
Initially, at the anode inlet (point 10), the H2 concentration is the highest if it is compared to other points. Afterward, after point 7, it begins to decrease because H2 has reacted with electrodes to break down the protons and the electrons. At the points 10 to 8, it tends to be constant because the fuel cell is a reversible system in which the reactants (H2) will continue to be supplied through the anode inlet.
Afterward, Fig. 4 compares the H2 mass concentration among three GDL materials. It shows that the highest hydrogen concentration was owned by ELAT-TEK-1200W which had a porosity of 0.31 followed by coconut coir and SIGRACET 25BA.
6. 3. Concentration profiles of O2
At the cathode, the flow rate of O2 can be observed in the form of the O2 mass surface contour in Fig. 5 and also the mass concentration profiles for each cut point in Fig. 6. Then, the effect of each type of material with different porosities can be compared in Fig. 7.
Fig. 5 illustrates the surface contour changes of the O2 mass concentration in surface contour to calculate the mass concentration plotting in Fig. 6. The figure represents the O2 mass concentration in elapsed time for the 3 GDL materials. From the cathode inlet (point 1), the O2 mass concentration decreases significantly over time approaching 0 (runs out).
In all porosities, based on the surface contour, the highest mass concentration is at the cathode inlet. Then, they were plotted in Fig. 7 to describe the comparison of the O2 mass concentration in the difference porosities. The result shows that the highest O2 mass concentration is SIGRACET 25BA, then coconut coir, and finally ELAT-TEK-1200W.
a
b
x 10
0.15 0.14 0.13 0.12 0.11 0.1 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01
-A- [02].1
~¥r- [02J.2
-*- [02].3
-e- [02].4
[02].5
—i— [02].6
-m- [02].7
-B- [02].a
- [02].9
- [02].10
0.02
0.04 0.06
Time (s)
r r 0, "k
0.18 Ol
r 0.16 C o
IS
■ 0.14
£= o u EZ o
0.12
0.1 1/1 l/l (TJ
0.08 £ <z 0)
0.06 rjl * X
0.04 O
0.02
0.12 _ -A- [02].1
0 11 [02].2
[02].3
0.1 -e- [02].4
0.09 -6- [02].5
0.08 [02].6
0.07 - —[02].7
0.06 _ -B- [02].8
0 ns -[02].9
-[02].10
0 04 -
0.02 0.04 0.06 0.08 0.1
Time <s)
b
Fig. 5. Oxygen mass concentration at anode Coconut Coir: a — ¿=0.02 s; b — i=0.05 s; c
(kg/m3) of - ¿=0.09 s
a ^
¡5 m
o ^ a
S
o
0,16 0,14 0,12 0,1 0,08 0,06 0,04 0,02 0
Fig. 6. Oxygen mass concentration at anode: a — ELAT-TEK-1200W; b — Coconut Coir; c — SIGRACET 25BA
-ELAT-TEK-1200W -Coconut Coir -SIGRACET 25BA
0 123456789 10 11
_Cut Point in Steady State_
Fig. 7. The effect of different porosities of Oxygen mass concentrations
a
a
c
c
6. 4. Concentration profiles of H2O
The insufficient PEMFC that cannot be avoided is the water formation in the electrode; cathode and anode. Like H2 and O2 species, the surface contour and the water mass concentration at each electrode point are also plotted in the comparison graph among the GDL materials shown in Fig. 8.
10 11
—•— ELAT-TEK- 1200W (anode)
■ Coconut Coir (anode) -*- SIGRACET 25BA (anode) --•— ELAT-TEK- 1200W (cathode) --■-- Coconut Coir (cathode)
SIGRACET 25BA (cathode)
Cut Point in Steady State
Fig. 9. Activation overpotential in the different GDL porosities 6. 6. Activation overpotential
The activation overpotential represents the activation polarization among three materials having the same values shown in Fig. 9, the activation polarization for both the anode and cathode. In fact, the three different materials show the same changes from 0 to 0.295 V.
Fig. 8. The effect of different porosities of water mass concentrations at the anode
From Fig. 8, it can be seen that the porosity of the GDL also influences the water mass concentration distribution at each point.
At the anode, the water concentration at each point decreases in elapsed time. Meanwhile, the highest water content in each GDL material is located at point 10, in which it is the anode inlet. If the GDL that is a part of the electrode has high porosity, the displacement of protons containing water molecules will be inhibited, so that the diffusion velocity will slow down. It will cause the water in the GDL to be retained [17]. It is illustrated in Fig. 8 that the lowest water mass concentration is located at the anode outlet.
The water mass concentrations for the three different GDL materials at the cathode also are illustrated in Fig. 8 using the dash line in which it tends to increase. It opposites to the H2O mass concentration at the anode due to the different inlet/outlet of the anode and the cathode.
6. 5. Performance of PEMFC affected by various GDL porosities
After finding the numerical simulation results in the mass concentration of the three different GDL porosities, both in the surface contour and the cut point of mass concentration at the anode and cathode, then the PEM-FC performance influenced by the GDL materials with different porosities will be calculated based on the data written on the physical parameters in Table 1-3 by using equations (5)-(17).
Initially, before calculating the PEMFC performance determined using eq. (17), the losses (overpotentials) had been calculated. The overpotentials of the PEMFC operation will reduce the ideal cell potential stated in eq. (5). After considering the three overpotentials of PE FC operation, it can be illustrated by the following graphs as shown in Fig. 9-11.
0,08 0,07 0,06 0,05 0,04 0,03 0,02 0,01 0
-ELAT-TEK-1200W -Coconut Coir -SIGRACET 25BA
0 2500 5000 7500 10000 12500 15000 Current Density (A/m2)
Fig. 10. Ohmic overpotential in the different GDL thickness
6. 7. Ohmic overpotential
Fig. 10 presents the ohmic overpotential that depends on the thickness and the resistivity for anodes, cathodes, electrolytes and GDL. The porosity of GDL materials does not affect the ohmic overpotential, however, each material has a different thickness and resistivity, which greatly affects the ohmic overpotential value.
6. 8. Concentration overpotential
After finding the mass concentration values plotted in Fig. 2-8, the overpotential activation is illustrated in Fig. 9.
-ELAT-TEK- 1200W -Coconut Coir -SIGRACET 25BA
0 2500 5000 7500 10000 12500 15000 Current Density (A/m2)
Fig. 11. Concentration overpotential in the different GDL porosities
Meanwhile, the excess concentration of the anode concentration can be neglected, but the excess concentration of the cathode concentration is considered at high current densities as shown in Fig. 11.
Table 4
Thickness and resistivity of GDL materials
Unit ELAT-LT-1200W COCONUT COIR SIGRACET 25 BA
da m 0.0003 0.00025 0.0002
Pa Q-m 8,000 5,000 8,000
de m 0.00001 0.00001 0.00001
Pe Q-m 9 9 9
dc m 0.0003 0.00025 0.0002
Pc Q-m 8,000 5,000 8,000
In Fig. 11, it looks that porosity greatly influences the concentration losses which are very significant results. The biggest losses occur in SIGRACET 25BA which has the biggest pores.
6. 9. Power density (PEMFC performance)
After calculating the amount of losses/polarization occurring in the fuel cell stack using eq. (17), finally the PEMFC performance can be illustrated in Fig. 12. At different porosities, the concentration polarization has great losses affecting the PEMFC performance.
450 400 350 300 250 200 150 100 50 0
ELAT-TEK-1200W -Coconut Coir -SIGRACET 25BA
0 500 1000 1500 2000 2500 3000 3500 4000 Current Density (A/m2)
Fig. 12. PEMFC performances of the various porosities GDL materials
7. Discussion of PEMFC performance influenced by the porosities of the GDL materials affecting various activations
In the concentration profiles depicted in Fig. 2-8, the surface contour and mass concentration of species at the anode and cathode for H2, O2, and H2O were presented. The surface contours illustrate the mass concentration distribution during the reaction process. The H2 mass concentration increases in elapsed time. However, the H2 concentration close to the anode outlet tends to be constant due to H2 has reacted on the electrode to release the protons in the electrolyte membrane toward the cathode. The conditions are far different from the outlet area because the unreacted H2 flows upward, causing the increase of the H2 mass concentration at the anode outlet. Correspondingly, if we look at the surface contour, the anode outlet has the highest mass
concentration of hydrogen. Then, if to compare the H2 mass concentration among three GDL materials, it shows that the highest hydrogen concentration was owned by ELAT-TEK-1200W which had a porosity of 0.31 followed by coconut coir and SIGRACET 25BA. It means that the porosity factor certainly plays an important role to influence it. The higher the porosity, the more difficult for H2 to release protons and even the anode outlet function to release the protons. As a result, the less proton production, H2 is more difficult to flow into the membrane then it will bond with O2.
The surface contour of the O2 mass concentration represents the O2 mass concentration that decreases significantly over time in the cathode inlet. It is because O2 reacts with protons and electrons to form H2O. Among the three GDL materials, they have the same trend, which decreases in elapsed time. In all porosities, based on the surface contour, the highest mass concentration is at the cathode inlet. Then, after regarding comparison among them, the result shows that the highest O2 mass concentration is SIGRACET 25BA, then coconut coir, and finally ELAT-TEK-1200W. At the cathode, the porosity affects the distribution of O2 significantly. The greater the porosity, the smaller the O2 diffusivity ability.
Afterwards, the insufficient PEMFC that cannot be avoided is the water formation in the electrode; cathode and anode. Like H2 and O2 species, the surface contour and the water mass concentration at each electrode were compared. It can be seen that the porosity of the GDL also influences the water mass concentration distributions. At the anode, the highest water concentration is the anode inlet. If the GDL that is a part of the electrode has high porosity, the displacement of protons containing water molecules would be inhibited, so that the diffusion velocity will slow down. It will cause the water in the GDL to be retained [17]. On the contrary, the lowest water mass concentration is located at the anode outlet. It means that the mass of water has not accumulated at the anode outlet, then it makes easier to flow H2. Afterward, beside the H2 mass is enough to initiate the reaction, some of them move to the cathode through the membrane.
Then, the water mass concentrations for the three different GDL materials at the cathode opposites to the H2O mass concentration at the anode due to the different inlet/outlet of the anode and the cathode. In fact, the water mass concentration at the cathode is much greater than at the anode one. This is the result of the effects of "electro-osmotic drag" transport, in which together with the electrochemical water production resulting in an accumulation of water at the cathode [18].
After finding the numerical simulation results in the mass concentration of the three different GDL porosities, both in the surface contour and the cut point of mass concentration at the anode and cathode, the PEMFC performance influenced by the GDL materials with different porosities was calculated based on the data written on the physical parameters in Table 1-3 by using equations (5)-(17). Initially, before calculating the PEMFC performance determined using eq. (17), the losses (overpotentials) had been calculated. The overpotentials of the PEMFC operation reduce the ideal cell potential stated in equation (5). Firstly, the activation overpotential represents the required energy to initiate the reaction at PEMFC. It is also illustrated as a barrier that the reactants must overcome. It will influence the electrochemical reaction involving the absorption process among
species on the electrode surface. However, the results show that the activation polarization among three materials has the same values shown in Fig. 9. It is the fact that, based on equation (11), (12), the activation polarization only depends on the operating temperature, thus the porosity does not affect the activation losses. In the present study, the activation polarization for both anode and cathode and also from the three different materials show the same changes from 0 to 0.295 V.
The other overpotential, the ohmic overpotential depends on the thickness and the resistivity for anodes, cathodes, electrolytes and GDL. Here, the ion transport is more difficult and complex than electricity transmission. In the electrolytes, it is a dominant loss and is caused by an excess of resistance (ohmic overpotential). Therefore, it depends on some parameters, such as the proton conductivity, the electrolyte thickness and the current density, as presented in eq. (13), (14). The porosity of GDL materials does not affect the ohmic overpotential, however, each material has a different thickness and resistivity, which greatly affects the ohmic overpotential value. In addition to the size of the thickness and resistivity of each material calculated in eq. (13), it can be seen in Table 4 giving some different properties of materials.
After finding the mass concentration values, the overpotential activation was illustrated in Fig. 9. The mass transportation is simultaneous activity in the fuel cell in which the reactants sent to the electrochemical reaction area producing the product must be quickly removed from the reaction to get a stable electricity production. Some species appear to be overreacting, such as unexpected water formation. Thus, one of the main reasons for excess concentration is the convective mass transfer from the gas channel to the porous electrodes. The anode consists of H2 and O2 for the cathode side. However, there are H2, O2, and H2O flowing in porous electrodes. H2 can pass to the interface of the electrodes quickly on the anode side due to its high effective diffusion coefficient. On the other hand, O2 and H2O, which have a high molecular weight, diffuse at the cathode side with a lower diffusion coefficient. Meanwhile, the excess concentration of the anode can be neglected, but the excess concentration of the cathode is considered at high current densities as shown in Fig. 11. It looks that the effect of the porosity greatly influences the concentration losses which are very significant results. The biggest losses occur in SIGRACET 25BA which has the biggest pores. It is due to the big porosity will produce the overreacting process, such as unexpected water formation, it gives an effect in an accumulation of species concentrations at the electrode.
After calculating the amount of losses/polarization that occurs in the fuel cell stack using eq. (17), finally the PEMFC performance can be illustrated in Fig. 12. At different porosities, the concentration polarization has great losses
affecting the PEMFC performance. It has a significant loss besides the ohmic and activation losses. As a result, the ideal power density of the maximum power density cannot be achieved due to the polarizations. The highest power density is achieved by the smallest porosity material (ELAT-TEK-1200W). If it is look at the mass concentration species in Fig. 2-8, ELAT-TEK-1200W is capable of producing very high-power densities because the H2 concentration in the anode generates the highest concentration compared to other GDL materials. As a result, the supply of fuel is very large to react with O2. This is evidenced by the lowest O2 concentration at the cathode side. It means the best H2 and O2 reactions compared to coconut coir and SIGRACET 25BA. On the other hand, the H2O formation at the anode for ELAT-TEK-1200W is the lowest one. It makes an easier way for the diffusion between H2 and O2 species to work properly. Finally, ELAT-TEK-1200W produces a high generated power density for PEMFC. In the study, coconut coir as an organic material nevertheless has the potential GDL material to replace other commercial GDL such as ELAT-TEK-1200W and SIGRACET because it has a quite promising performance. Coconut coir can be a promising GDL material in the future because of its performance and environmentally friendly materials.
The future work needs to investigate the performance of PEMFC comparing GDL materials among other some derived organic materials, coconut coir, and commercial materials. The other physical properties of the GDL need to be considered in the numerical simulation properties to predict the PEMFC performance more accurately due to this research only examines the porosity effects. Hence, further research will be needed to calculate all the physical properties of organic materials having some potencies to become GDL. Afterward, the effect of temperature and pressure that is currently under STP conditions will be able to be varied in real terms according to experimental conditions.
8. Conclusions
1. The highest H2 mass concentration was owned by ELAT-TEK-1200W which had a porosity of 0.31, however, the highest O2 mass concentration is SIGRACET 25BA.
2. The greater the porosity, the smaller the O2 diffusiv-ity. The porosity does not affect the activation and also the ohmic overpotential. It only influences the concentration polarization that gives an impact on the PEMFC performance.
3. The biggest losses occur at SIGRACET 25BA, so that ELAT-TEK-1200W produces the highest generated power density for PEMFC. Coconut coir can be a promising GDL material in the future because of its performance and environmentally friendly materials.
References
1. Wang, Y., Chen, K. S., Mishler, J., Cho, S. C., Adroher, X. C. (2011). A review of polymer electrolyte membrane fuel cells: Technology, applications, and needs on fundamental research. Applied Energy, 88 (4), 981-1007. doi: https://doi.org/10.1016/ j.apenergy.2010.09.030
2. Qiu, D., JanBen, H., Peng, L., Irmscher, P., Lai, X., Lehnert, W. (2018). Electrical resistance and microstructure of typical gas diffusion layers for proton exchange membrane fuel cell under compression. Applied Energy, 231, 127-137. doi: https:// doi.org/10.1016/j.apenergy.2018.09.117
3. Mohammed, H., Al-Othman, A., Nancarrow, P., Tawalbeh, M., El Haj Assad, M. (2019). Direct hydrocarbon fuel cells: A promising technology for improving energy efficiency. Energy, 172, 207-219. doi: https://doi.org/10.1016/j.energy.2019.01.105
4. Fadzillah, D. M., Rosli, M. I., Talib, M. Z. M., Kamarudin, S. K., Daud, W. R. W. (2017). Review on microstructure modelling of a gas diffusion layer for proton exchange membrane fuel cells. Renewable and Sustainable Energy Reviews, 77, 1001-1009. doi: https:// doi.org/10.1016/j.rser.2016.11.235
5. Tseng, C.-J., Lo, S.-K. (2010). Effects of microstructure characteristics of gas diffusion layer and microporous layer on the performance of PEMFC. Energy Conversion and Management, 51 (4), 677-684. doi: https://doi.org/10.1016/j.enconman.2009.11.011
6. Park, S., Popov, B. N. (2011). Effect of a GDL based on carbon paper or carbon cloth on PEM fuel cell performance. Fuel, 90 (1), 436-440. doi: https://doi.org/10.1016/jj.fuel.2010.09.003
7. O'Hayre, R., Cha, S.-W., Colella, W., Prinz, F. B. (2016). Fuel Cell Fundamentals. John Wiley & Sons. doi: https://doi.org/ 10.1002/9781119191766
8. Bruno, M. M., Viva, F. A. (2013). Carbon Materials for Fuel Cells. Direct Alcohol Fuel Cells, 231-270. doi: https:// doi.org/10.1007/978-94-007-7708-8_7
9. Destyorini, F., Irmawati, Y., Widodo, H., Khaerudini, D. S., Indayaningsih, N. (2018). Properties and Performance of Gas Diffusion Layer PEMFC Derived from Coconut Coir. Journal of Engineering and Technological Sciences, 50 (3), 409-419. doi: https:// doi.org/10.5614/j.eng.technol.sci.2018.50.3.7
10. Indayaningsih, N., Irmawati, Y., Destyorini, F. (2016). Performance of gas diffusion layer derived from carbon powder of coconut coir for PEMFC application. ARPN Journal of Engineering and Applied Sciences, 11 (6), 4040-4044.
11. Jeon, D. H., Kim, K. N., Baek, S. M., Nam, J. H. (2011). The effect of relative humidity of the cathode on the performance and the uniformity of PEM fuel cells. International Journal of Hydrogen Energy, 36 (19), 12499-12511. doi: https://doi.org/10.1016/ j.ijhydene.2011.06.136
12. El-kharouf, A., Mason, T. J., Brett, D. J. L., Pollet, B. G. (2012). Ex-situ characterisation of gas diffusion layers for proton exchange membrane fuel cells. Journal of Power Sources, 218, 393-404. doi: https://doi.org/10.1016/j.jpowsour.2012.06.099
13. Chen, F., Chang, M., Hsieh, P. (2008). Two-phase transport in the cathode gas diffusion layer of PEM fuel cell with a gradient in porosity. International Journal of Hydrogen Energy, 33 (10), 2525-2529. doi: https://doi.org/10.1016/j.ijhydene.2008.02.077
14. Hinebaugh, J., Bazylak, A. (2011). PEM Fuel Cell Gas Diffusion Layer Modelling of Pore Structure and Predicted Liquid Water Saturation. ASME 2011 9th International Conference on Fuel Cell Science, Engineering and Technology. doi: https:// doi.org/10.1115/fuelcell2011-54422
15. Sezgin, B., Caglayan, D. G., Devrim, Y., Steenberg, T., Eroglu, I. (2016). Modeling and sensitivity analysis of high temperature PEM fuel cells by using Comsol Multiphysics. International Journal of Hydrogen Energy, 41 (23), 10001-10009. doi: https:// doi.org/10.1016/j.ijhydene.2016.03.142
16. Kalinci, Y., Dincer, I. (2018). Analysis and performance assessment of NH3 and H2 fed SOFC with proton-conducting electrolyte. International Journal of Hydrogen Energy, 43 (11), 5795-5807. doi: https://doi.org/10.1016/j.ijhydene.2017.07.234
17. Jeon, D. H., Kim, K. N., Baek, S. M., Nam, J. H. (2011). The effect of relative humidity of the cathode on the performance and the uniformity of PEM fuel cells. International Journal of Hydrogen Energy, 36 (19), 12499-12511. doi: https://doi.org/10.1016/ j.ijhydene.2011.06.136
18. Ji, M., Wei, Z. (2009). A Review of Water Management in Polymer Electrolyte Membrane Fuel Cells. Energies, 2 (4), 1057-1106. doi: https://doi.org/10.3390/en20401057
