Available online at www.sciencedirect.

com
Available online at www.sciencedirect.com
ScienceDirect
ScienceDirect
Energy
Available
Available Procedia
online
online 00 (2018) 000–000
atatwww.sciencedirect.com
www.sciencedirect.com
Energy Procedia 00 (2018) 000–000 www.elsevier.com/locate/procedia
ScienceDirect
ScienceDirect
www.elsevier.com/locate/procedia

Energy Procedia
Energy Procedia 00
145(2017)
(2018)000–000
64–69
www.elsevier.com/locate/procedia
Applied Energy Symposium and Forum, Renewable Energy Integration with Mini/Microgrids,
Applied Energy Symposium and Forum,
REM 2017, 18–20 Renewable Energy
October 2017, Integration
Tianjin, China with Mini/Microgrids,
REM 2017, 18–20 October 2017, Tianjin, China
CFD modeling and simulation of PEM fuel cell using OpenFOAM
CFD modeling and
The 15th simulation
International of PEM
Symposium fuel Heating
on District cell using OpenFOAM
and Cooling
Jean-Paul Koneaa, Xinyu Zhanga,b,*, Yuying Yanb,c , Guilin Hudd, Goodarz Ahmadiee
Assessing
Jean-Paul Kone ,the feasibility
Xinyu of using
Zhang *, Yuying
a
a,b,
Yanthe, Guilin
b,c
heat Hudemand-outdoor
, Goodarz Ahmadi
International Doctoral Innovation Centre, University of Nottingham Ningbo China,
temperature function forTaikang
International Doctoral
a
199
Research 199
a long-term
Innovation
Taikang
Center
EastCentre,
for Fluids
b
Road, 315100
East and
Road,
district
University
Ningbo, China heat
of Nottingham
315100Engineering,
Thermal Ningbo, ChinaUniversity of
demand forecast
Ningbo China,

ResearchNingbo
Nottingham
b
CenterChina,
for Fluids
199 and Thermal
Taikang East Engineering,
Road; 315100University of
Ningbo, China
a,b,c c a a b c c
I. Andrić *, A. Pina , P. Ferrão , J. Fournier ., B. Lacarrière , O. Le Corre
Nottingham NingboEngineering
Fluids & Thermal China, 199 Taikang
ResearchEast Road; 315100 of
Group,University Ningbo, China
Nottingham,
c
Fluids &University
Thermal Engineering Research
Park, NG7 2RD Group,University
Nottingham, of Nottingham,
United Kingdom
a d
School
IN+ Center for Innovation, Technology University
of Mechanical andPark,
and Policy NG7 2RD
Automotive
Research Nottingham,
-Engineering,
Instituto United
Zhejiang
Superior Kingdom
University
Técnico, of Science
Av. Rovisco Pais and
1, 1049-001 Lisbon, Portugal
d b
School
Veolia of Mechanical
Recherche &and
Technology, Automotive
No 318, 291
Innovation, Engineering,
Liuhe Rd, 310023
Avenue Zhejiang University
Hangzhou,
Dreyfous Daniel, 78520 of
China Science
Limay, and
France
c e
Department
Département Systèmes Technology,
Énergétiques No 318,
of Mechanical Liuhe -Rd,
and Aeronautical
et Environnement IMT310023 Hangzhou,
Engineering,
Atlantique, China
Clarkson
4 rue Alfred University,
Kastler, 44300 Nantes, France
e
Department of MechanicalPotsdam,
and Aeronautical
New York, Engineering,
USA Clarkson University,
Potsdam, New York, USA

Abstract
Abstract
Abstract
ADistrict
proton heating
exchange membrane
networks are (PEM)
commonly fuel addressed
cell is an electrolytic
in the literaturecell that
as oneconverts
of thechemical energy solutions
most effective of hydrogen reacting with
for decreasing the
A protoninto
oxygen
greenhouse exchange
electricalmembrane
energy.
gas emissions from (PEM)
To meet
the fuel cell is anstringent
increasingly
building sector. electrolytic
These cell
application
systems thatneeds,
require converts chemical
highimproved
investments energyare
performance
which ofand
hydrogen
increased
returned reacting
through thewith
efficiency
heat
oxygen
are
sales. into electrical
paramount.
Due to the energy.climate
Computational
changed Tofluid
meetdynamics
increasingly
conditions andstringent
(CFD) is an application
building ideal meansneeds,
renovation for improved
achieving
policies, performance
heat these
demand andfuture
improvements.
in the increasedthisefficiency
Incould paper, a
decrease,
are paramount.
comprehensive
prolonging Computational
theCFD-based
investment tool
returnfluid
that dynamics
can
period. (CFD)
accurately is anthe
simulate ideal means
major for achieving
transport phenomena these improvements.
which take place withinIn this paper,
a PEM fuela
comprehensive
cell
Theismain
presented.
scope CFD-based
The toolpaper
of this toolisthat
is developed can accurately
using
to assess theOpenFOAMsimulate
feasibility and
of the major
it can
using transport
thebeheat
used phenomena
to rapidly
demand whichinto
gain insights
– outdoor take
temperature theplace within
for aheat
cell working
function PEM fuel
processes.
demand
cell
The isbase
presented.
forecast. case The tool
The results
district ofisAlvalade,
are developedlocated
compared usingprevious
with OpenFOAM model
in Lisbon andresults
it can be
(Portugal), and used
was to rapidly
experimental
used gain
as a casedata.insights into district
The present
study. The theI-V
cellcurve
working processes.
shows
is consisted better
of 665
The base case
agreement
buildings that results
with vary inare
bothcompared
the experimental trendwith
construction previous
at low andmodel
current
period results
densities.
typology. Theand
Three experimental
simulation
weather data.indicate
data also
scenarios The present
(low, that the
medium, I-V curve
chosen
high) shows
three better
andconcentration
district
agreement
constant
renovationhaswith
very the experimental
significant
scenarios impact
were developed trend at
on the low current
concentration
(shallow, densities. The simulation
overpotential.
intermediate, deep). To estimate datathe
also indicate
error, that the
obtained heatchosen
demand concentration
values were
constant
compared haswith
veryresults
significant
from impact
a dynamic on the
heatconcentration
demand model, overpotential.
previously developed and validated by the authors.
Copyright © 2018 Elsevier Ltd. All rights reserved.
The results showed that when only weather change is considered, the margin of error could be acceptable for some applications
Copyright © 2018 The Authors. Published by Elsevier Ltd.
Copyright
Selection ©in2018
and
(the errorand annualElsevier
peer-review
demand Ltd.
underAllresponsibility
was rights than
lower reserved.
20%of for
the scientific
all weather committee
scenarios ofApplied
the Applied
considered). Energy
However, Symposium
after introducing andrenovation
Forum,
Selection peer-review under responsibility of the scientific committee of the Energy Symposium and Forum,
Selection
Renewable
scenarios, and
the peer-review
EnergyerrorIntegration
value under
with
increased responsibility
Mini/Microgrids,
up to 59.5%
Renewable Energy Integration with Mini/Microgrids, REM 2017 of the
REM scientific
(depending 2017. on committee
the weather of
and the Applied
renovation Energy
scenarios Symposium
combination and Forum,
considered).
Renewable
The value Energy
of slope Integration
coefficient with Mini/Microgrids,
increased on average REM 2017.
within the range of 3.8% up to 8% per decade, that corresponds to the
Keywords:
decrease Computational
in the numberfluid of dynamics;
heating hours modelling; numerical;
of 22-139h OpenFOAM;
during proton
the heating exchange
season membraneon
(depending fuel
thecell; simulation of weather and
combination
Keywords:
renovation Computational fluid dynamics;
scenarios considered). Onmodelling;
the othernumerical; OpenFOAM;
hand, function proton
intercept exchangefor
increased membrane fuel cell;
7.8-12.7% per simulation
decade (depending on the
coupled scenarios). The values suggested could be used to modify the function parameters for the scenarios considered, and
improve the accuracy of heat demand estimations.

© 2017 The Authors. Published by Elsevier Ltd.

* Corresponding
Peer-review underauthor. Tel.: +86-18058261662;
responsibility fax:Committee
of the Scientific +86(0)574 8818 0175.
of The 15th International Symposium on District Heating and
*E-mail
Cooling. addresses:author.
Corresponding Tel.: +86-18058261662; fax:(X.
Xinyu.Zhang@nottingham.edu.cn +86(0)574
Zhang), 8818 0175.
E-mail addresses: Xinyu.Zhang@nottingham.edu.cn
Jean-Paul.KONE@nottingham.edu.cn (J. P. Kone). (X. Zhang),
Jean-Paul.KONE@nottingham.edu.cn (J. P. Kone).
Keywords: Heat demand; Forecast; Climate change
1876-6102 Copyright © 2018 Elsevier Ltd. All rights reserved.
1876-6102 Copyright
Selection and © 2018
peer-review Elsevier
under Ltd. All of
responsibility rights reserved. committee of the Applied Energy Symposium and Forum, Renewable Energy
the scientific
Selection
Integrationand peer-review
with under responsibility
Mini/Microgrids, REM 2017. of the scientific committee of the Applied Energy Symposium and Forum, Renewable Energy
Integration with Mini/Microgrids, REM 2017.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 Copyright © 2018 The Authors. Published by Elsevier Ltd.
Selection and peer-review under responsibility of the scientific committee of the Applied Energy Symposium and Forum, Renewable Energy
Integration with Mini/Microgrids, REM 2017
10.1016/j.egypro.2018.04.011
 Jean-Paul Kone et al. / Energy Procedia 145 (2018) 64–69 65
2 Jean-Paul Kone, Xinyu Zhang, Yuying Yan, Guilin Hu, Goodarz Ahmadi / Energy Procedia 00 (2018) 000–000

1. Introduction

Over the past several decades, fuel cells have become an important low or zero emission energy devices, and
thereby an attractive choice for succeeding some of the conventional power devices which depend on fossil fuels.
PEM fuel cells have high efficiency and operate at lower temperatures, which make them suitable for automotive
power supply, as well as power generation devices for portable electronics and stationary units [1]. The application in
the stationary sector covers a wide range of scales going from minigrid-connected combined heat and power systems
for residential use, to backup power systems (e.g. e.g. uninterruptible power systems (UPS) for buildings, etc.), to
main power supply for buildings.
The vast majority of current research work on PEM fuel cells is concern with improving the cell performance by
increasing its efficiency meanwhile decreasing the costs associated with manufacturing and test through the CFD
technique; and thereby, further promote their rapid integration with mini/microgrid systems. However, until now, truly
comprehensive and open-source code PEM fuel cell models are inexistent in the published literature. This work is an
attempt to fill this gap.
The model presented in this work is a single-phase flow and non-isothermal model. It is partly an adaptation of the
open-source computational model of a solid oxide fuel cell (SOFC) presented by Beale et al. [2] to a PEM fuel cell.
This model is different from that of Beale et al. [2] because: it is a different type of fuel cell, the geometry is different,
the boundary conditions are different, and the electrochemistry is different.

Nomenclature

cp specific heat capacity (J kg-1 K-1)

δ thickness (m)
F Faraday’s constant (96485 C mol-1 e-)
I current density (A m-2)
k thermal conductivity (W m-1 K-1)
K permeability (m2)
μ dynamic viscosity (Pa s)
ni number of electron of species i
p pressure (Pa)
ρ density (kg m-3)
∆S entropy of formation (J mol-1 K-1)
T temperature (K)
⃗U velocity (m s-1)
y mass fraction

2. Mathematical Model

2.1. Assumptions

The model assumptions are listed as follows:

• The fuel cell components are isotropic and homogeneous;

• The membrane is assumed to be fully humidified;
• The membrane is impermeable to reactant gases;
• Steady-state operating condition is assumed;
• The gas flow is laminar and incompressible due to low velocities;
• Ideal individual gases and gas mixture are assumed;
• The anode activation and mass transport overpotentials are neglected.
66 Jean-Paul Kone et al. / Energy Procedia 145 (2018) 64–69
Jean-Paul Kone, Xinyu Zhang, Yuying Yan, Guilin Hu, Goodarz Ahmadi / Energy Procedia 00 (2018) 000–000 3

2.2. Governing equations

The conservation equations are expressed by

for mass

⃗)=0
𝛻𝛻 ∙ (𝜌𝜌𝑈𝑈 (1)

for momentum

⃗ 𝑈𝑈
𝛻𝛻 ∙ (𝜌𝜌𝑈𝑈 ⃗ ) = −𝛻𝛻𝛻𝛻 + 𝛻𝛻 ∙ (𝜇𝜇𝜇𝜇𝑈𝑈
⃗ ) + 𝑆𝑆𝑀𝑀 (2)

where SM is the momentum source term. SM is equal to zero in the gas flow channels. In the porous media (gas diffusion
layers and catalyst layers), SM is equal to Darcy resistance given by

⃗
𝜇𝜇𝑈𝑈
𝑆𝑆𝑀𝑀 = − (3)
𝐾𝐾

for chemical species

⃗ 𝑦𝑦𝑖𝑖 ) = (𝛻𝛻 ∙ 𝜌𝜌𝜌𝜌𝜌𝜌𝑦𝑦𝑖𝑖 )

𝛻𝛻 ∙ (𝜌𝜌𝑈𝑈 (4)

where D is the effective diffusivity of species i in m2 s-1. The mass fraction of the inert species on each electrode is
𝑛𝑛−1
calculated from the solution of the other species, 𝑦𝑦𝑛𝑛 = 1 − ∑𝑖𝑖=1 𝑦𝑦𝑖𝑖

for energy

⃗ 𝑇𝑇) = 𝛻𝛻 ∙ (𝑘𝑘𝑘𝑘𝑘𝑘) + 𝑆𝑆𝐸𝐸

𝛻𝛻 ∙ (𝜌𝜌𝑐𝑐𝑝𝑝 𝑈𝑈 (5)

where SE is the energy source term. SE is essentially due to the heat released by the electrochemical reaction given by

𝐼𝐼 𝑇𝑇∆𝑆𝑆
𝑆𝑆𝐸𝐸 = (𝜂𝜂 − ) (6)
𝛿𝛿𝑀𝑀𝑀𝑀𝑀𝑀 𝑛𝑛𝑖𝑖 𝐹𝐹
The cell voltage is expressed by

𝑉𝑉𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 𝐸𝐸𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 − 𝜂𝜂 (7)

where E is Nernst potential and η is the combined voltage loss, also known as overpotential.

2.3. Boundary conditions

Dirichlet boundary conditions are applied at the fluid inlets. The outlet value is prescribed for pressure at the outlets
and Neumann boundary conditions are applied at the fluid outlets for all other variables. The membrane-electrode
interfaces are insulated against the flow of mass and chemical species. Impermeability, no-slip, and no-flux boundary
conditions are applied at all solid-fluid interfaces. A zero-gradient temperature is applied at all the external surfaces
of the cell.

3. Numerical Implementation

The present model has been developed using OpenFOAM, a highly flexible open-source CFD software package.
Almost every component of the software can be customized to suit the users’ needs. Nevertheless, one shortcoming
 Jean-Paul Kone et al. / Energy Procedia 145 (2018) 64–69 67
4 Jean-Paul Kone, Xinyu Zhang, Yuying Yan, Guilin Hu, Goodarz Ahmadi / Energy Procedia 00 (2018) 000–000

of the code is that it has no PEM fuel cell module. The pemfcModels-4.0 toolbox developed here is an attempt to
provide such a module.
The geometry and mesh of the fuel cell are shown in Fig. 1. A three-dimensional (3D) description of a complete
single cell is used. This consists of a membrane electrode assembly (MEA) placed between two bipolar plates (BP).
The MEA has two electrodes (i.e. anode and cathode) which are electrically insulated from each other by the
membrane. Each electrode has a gas diffusion layer (GDL) and a catalyst layer (CL). For feeding the cell with
hydrogen at the anode side and oxygen at the cathode side, gas flow channels (GFC) are cut into the bipolar plates.

Fig. 1. Geometry and mesh of a single cell PEM fuel cell.

Thus, the modeling domain has multiple regions with multi-physics phenomena existing. This requires different
conservation equations to be solved in the different fuel cell components. The solution algorithm follows a specific
procedure. After the initialization, several iterations are completed until convergence. For the potentiostatic case the
voltage is fixed, whereas for the galvanostatic case the voltage is adjusted until the current density is equal or very
close to the initial value. The base case operating conditions are given in Table 1.

Table 1. Base case operating conditions.

Quantity Symbol Value Unit Source
Cell voltage V 0.6 V -
Cell temperature Tcell 353 K -
Air outlet pressure pair 101325 Pa -
Fuel outlet pressure pfuel 101325 Pa -
Air inlet velocity Uair 3.082 ms -1
Calculated [3]
Fuel inlet velocity Ufuel 1.1055 m s-1 Calculated [3]
Mass fraction of oxygen yO2 0.2 - -
Mass fraction of water vapor in air yH2 O 0.15 - -
Mass fraction of nitrogen yN2 0.65 - -
Mass fraction of hydrogen yH2 0.2 - -
Fraction of water vapor in fuel yH2 O 0.8 - -
Stoichiometry ratio - cathode air ξair 2.0 - -
stoichiometry ratio - anode fuel ξfuel 1.5 - -
68 Jean-Paul Kone et al. / Energy Procedia 145 (2018) 64–69
Jean-Paul Kone, Xinyu Zhang, Yuying Yan, Guilin Hu, Goodarz Ahmadi / Energy Procedia 00 (2018) 000–000 5

4. Base Case Results

Fig. 2 (a) and 2 (b) show the distribution of hydrogen and oxygen mass fractions, respectively. The mass fractions
decrease from the cell inlet to the cell outlet and from the GFCs to the CLs as the electrodes are depleted of the
reactants by the electrochemical reaction.

Fig. 2. Species mass fraction distribution (a) Hydrogen; (b) Oxygen.

Fig. 3 (a) illustrates the distribution of the Nernst potential. This decreases from the cell inlet to the cell outlet as
the anode is depleted of the fuel due to the combined effects of the consumption of hydrogen and the production of
water vapor, which result in a decrease in the observed local current density as shown in Fig. 3 (b). Fig. 3 (c) shows
the distribution of the local temperature. Reduced heat sources such as Joule heating and the heat released by the
electrochemical reaction under the ribs are the origin of the non-uniform nature of the temperature field. The reduced
reaction heat under the ribs is due to lower current densities in these areas due to reduced mass transfer.

Fig. 3. Membrane cathode interface showing the distribution of (a) Nernst potential; (b) Local current density; (c) Temperature.

Fig. 4 plots the cell polarization or I-V curve. The results of the present work are compared to the results of the
numerical model and experimental data reported in Yuan et al. [4] since some of the parameters and properties used
in the present model were borrowed from Yuan et al. [4]’s multiphase flow model. The present I-V curve shows better
agreement with the experimental trend at low current density, which confirms the validity of the present model. The
difference between the present and experimental I-V curves can be attributed to, amongst other things, the combined
effects of concentration losses, the cell inlet flow rates, and the consideration of multiphase flow phenomenon. The
effective value of the concentration constant is often much larger than its theoretical value, leading to a much larger
drop in the current density. Therefore, this constant is often obtained empirically [5]. As for multiphase flow, the
 Jean-Paul Kone et al. / Energy Procedia 145 (2018) 64–69 69
6 Jean-Paul Kone, Xinyu Zhang, Yuying Yan, Guilin Hu, Goodarz Ahmadi / Energy Procedia 00 (2018) 000–000

existence of liquid water blocks gas diffusion passages, reduces the diffusion rate and the effective reaction surface
area, and hence leads to a deterioration in fuel cell performance.

Fig. 4. Polarization curve.

5. Conclusion and Outlook

A toolbox for the numerical simulation of single-phase flow in a PEM fuel cell has been developed using
OpenFOAM. The toolbox includes a main program, relevant library classes, and a constructed simulation case for a
co-flow gavanostatic simulation.
The base case results are as expected. The plotted polarization curve agrees better with the experimental trend at
low current densities compared to a known previous numerical model. Moreover, the simulation data indicate that the
value of the concentration constant has significant effects on the cell potential at high current densities.
The provided solver can serve as a basis to develop other features, such as improved catalyst and membrane models,
or multiphase flow. Arrangements are being made to make the toolbox available.

Acknowledgements

The authors acknowledge the financial support from the International Doctoral Innovation Centre, Ningbo
Education Bureau, Ningbo Science and Technology Bureau, and the University of Nottingham. This work was also
supported by the UK Engineering and Physical Sciences Research Council [grant numbers EP/G037345/1 and
EP/L016362/1], Faculty of Science and Engineering at University of Nottingham Ningbo China [SRG code
01.03.05.04.2015.02.001], Ningbo Natural Science Foundation Program [project code 2013A610107] and Ningbo
Science and Technology Bureau Technology Innovation Team [Grant No. 2016B10010].

References

[1] Kone JP, Zhang X, Yan Y, Hu G, Ahmadi G. Three-dimensional multiphase flow computational fluid dynamics models for proton exchange
membrane fuel cell: A theoretical development. The Journal of Computational Multiphase Flows. 2017;9:3-25.
[2] Beale SB, Choi H-W, Pharoah JG, Roth HK, Jasak H, Jeon DH. Open-source computational model of a solid oxide fuel cell. Computer
Physics Communications. 2016;200:15-26.
[3] Liu Z. Fuel cell performance. New York: Nova Science Publishers; 2012.
[4] Yuan W, Tang Y, Pan MQ, Li ZT, Tang B. Model prediction of effects of operating parameters on proton exchange membrane fuel cell
performance. Renewable Energy. 2010;35:656-66.
[5] O'Hayre RP, Cha S-W, Colella W, Prinz FB. Fuel Cell Fundamentals. New Jersey: John Wiley & Sons; 2006.