Guangrong Deng, Liang Liang, Zhao Jin, Chenyang Li

Transport Phenomena and Fluid Mechanics AIChE Journal
DOI 10.1002/aic.16193
Enhancing Mass Transport in Direct Methanol Fuel Cell by Optimizing
the Microstructure of Anode Micro-Porous Layer
Guangrong Deng1,2,3 , Liang Liang1,3 , Zhao Jin1,3 , Chenyang Li1,3 ,
Changpeng Liu*1,3 , Junjie Ge*1,3 , Wei Xing*1,3,4
1.Laboratory of Advanced Power Sources, Changchun Institute of Applied Chemistry,
Chinese Academy of Sciences, Changchun, Jilin, 130022, PR China
2.University of Chinese Academy of Sciences, Beijing, 100039, PR China
3.Jilin Province Key Laboratory of Low Carbon Chemical Power Sources, Changchun, Jilin,
130022, PR China
4.State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied
Chemistry, Chinese Academy of Sciences, Changchun, Jilin, 130022, PR China
Corresponding authors:
Changpeng Liu (liuchp@ciac.ac.cn)
Junjie Ge (gejj@ciac.ac.cn)
Wei Xing (xingwei@ciac.ac.cn)
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process which may lead to
differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/aic.16193
© 2018 American Institute of Chemical Engineers (AIChE)
Received: Oct 10, 2017; Revised: Apr 16, 2018; Accepted: Apr 25, 2018
This article is protected by copyright. All rights reserved.

AIChE Journal Page 2 of 40
Abstract
The microstructural characteristics of the anode micro-porous layer (MPL) can significantly
affect the mass transport in direct methanol fuel cells (DMFCs) by influencing the methanol
delivery and CO2 removal processes. The hydrophilic-hydrophobic balance and pore
structure of the flow path were established by optimizing the content of
polytetrafluoroethylene (PTFE) in the anode MPL. An empirical model was developed to
design and optimize the anode MPL to achieve better mass transport and cell performance.
From the simulated and experimental results, increasing the content of PTFE enhanced the
CO2 removal ability in the anode MPL, thereby alleviating CO2 blockage in the anode
catalyst layer (CL), whereas the narrowed flow path hindered methanol delivery in the anode
MPL. A good balance between methanol delivery and CO2 removal in terms of mass
transport was achieved when the PTFE content was adjusted to 15 wt.%, leading to the best
cell performance.
Topical Heading: Transport Phenomena and Fluid Mechanics
Keywords: direct methanol fuel cell, anode micro-porous layer, microstructure, methanol
delivery, CO2 removal
AIChE Journal
Page 3 of 40 AIChE Journal
Introduction
The complex two-phase mass transport process in direct methanol fuel cells (DMFCs) is
one challenge that must be addressed to achieve improved cell performance.1-4 At the anode
side of DMFCs, the delivery of methanol and removal of CO2 affect the cell performance, as
well as the cell durability. The micro-porous layer (MPL), which is a key component of the
membrane electrode assembly (MEA), plays an important role in the mass transport
process.5,6 Thus far, several researchers have investigated the effect of the cathode MPL
structure on water management in proton exchange membrane fuel cells (PEMFCs).7-9 In
DMFCs, gas-liquid flow and methanol crossover make the situation more complicated.10
Hence, systematic investigation of the effects of the components and structure of the anode
MPL on methanol delivery and CO2 removal is highly desirable. The major factors
influencing the mass transport are the hydrophilic-hydrophobic balance and pore structure of
the MPL, both of which are strongly determined by polytetrafluoroethylene (PTFE), the most
frequently used binder. Some efforts have focused on optimizing the content of PTFE in the
anode MPL. Mullai et al.11 increased the PTFE loading of the anode MPL, leading to a 20%
decrease in the methanol crossover. The peak power density improved from 13 to 24 mW
cm-2. Oedegaard et al.12 reported that adding PTFE to the diffusing layer facilitated gas
transport and led to more stable power output of the cell. Li et al.13 developed a new filtration
method for preparation of the anode MPL and optimized the carbon and PTFE content based
on the method used for passive DMFCs. Shao et al.14 optimized the carbon type and then
studied the effect of the PTFE loading of MPL on the performance of a DMFC at 90℃. Under
the test conditions, the CO2 repelling effect of PTFE was weakened and the best cell
performance was achieved with a PTFE content of 10 wt.%. Other studies focused mainly on
employing new materials for the MPL. For example, RuO2-based and silicon-based MPLs
AIChE Journal
that effectively enhanced the mass transport process and consequently led to good cell
performance were reported by Park et al.15 and Lu et al.,16 respectively.
In all the mentioned studies, the optimum component of the MPL was determined from
cell tests, where the cell preparation process is time-intensive and expensive. If the
relationship between the microstructure of the anode MPL and mass transport process can be
comprehensively understood, it is possible to employ computer-aided design to predict the
mass transport behavior and cell performance. In this study, PTFE and XC-72R, commonly
used MPL materials, are selected for study. The hydrophilic-hydrophobic nature and pore
structure of MPLs with different PTFE contents are characterized, and an empirical model is
proposed to describe the gas-liquid flow and methanol concentration gradient of the cell.
Finally, prediction of the cell performance from the model is demonstrated to be
well-matched with the experimental data.
Experimental
Preparation and evaluation of MEAs
The MPL consisted of carbon black and PTFE binder. A PTFE suspension (10 wt.%,
Dupont Inc.) was added to the suspension of Vulcan XC-72R (Cabot, USA) by mechanical
mixing to form a homogeneous ink, which was then sprayed onto the backing layer (carbon
paper with 20% proof, Toray Inc.) and heat-treated at 320℃ for 0.5 h. After the heat treatment
process, the remaining solvent was evaporated and PTFE was distributed uniformly
throughout the MPL. The loading of carbon black in the anode MPL was 1.5 mg cm-2 and the
PTFE content in the series of anodes prepared for Cell 1, Cell 2, and Cell 3 was 5, 15, and 25
wt.%, respectively. The catalysts used for the anode and cathode were 60 wt.% PtRu/C (atom
radio 2:1) and 60 wt.% Pt/C from Johnson Matthey Company, respectively. The noble metal
loading was 2 mg cm-2. The catalyst ink suspension was sprayed onto the MPL. Nafion 115
AIChE Journal
(Dupont Inc.) was employed in the cells as the electrolyte. The anode and cathode electrodes
were used to sandwich the membrane by heat pressing under a pressure of 50 kg cm-2 at
135℃ for 90 s. The MEAs with a geometric area of 5.0 × 5.0 cm2 were clamped in a qCf FC
25 (Baltic Fuel Cells, Germany) device at 2 N mm-2.
The polarization tests were performed on a fuel cell test station (Arbin Instrument) and
the electrochemical impedance was measured by using an FC impedance meter (KIKUSUI,
Japan) with oxygen as the oxidant at 60℃. The ohmic resistance of the cells was obtained
from the electrochemical impedance spectra, where the plots intersected the horizontal line,
Im (z) = 0, at the high-frequency side.
Characterization of surface morphology and porosity
Water was dropped onto the micro-porous layers and the contact angle data were
recorded by a contact angle tester (KRUSS DSA 30, German). For analysis of the surface
morphology, the MPL was sprayed and treated on a smooth and flat slide glass instead of
carbon paper, where the uneven landscape of the latter would interfere with the measurement
and cause errors in the plot and roughness. Thus, the structural influence of the carbon paper
was eliminated. The 3D surface profile of MPL was obtained by using a non-contact optical
instrument (ST400; NANOVEA, USA). The scanning area was 100 × 100 µm2 and the step
length was set to 30 nm.
The porosity of the MPLs was measured by employing an automated gas sorption
analyzer (Quantachrome Autosorb iQ2, USA) and the results were calculated from the
nitrogen adsorption/desorption data by the Barrett-Joyner-Halenda (BJH) method.
CO2 concentration at cathode
AIChE Journal
CO2 emerging in the cathode was mainly derived from the diffusion of CO2 from the
anode to the cathode and methanol oxidation due to crossover.17,18 The concentration of CO2
at the cathode outlet was measured by a CO2 sensor in the constant-current mode.11 For the
CO2 diffusion measurement, the anode was fed with 1 mol methanol, the cathode was
supplied with pure nitrogen, and an external voltage was applied. Hence, methanol was
oxidized on the PtRu electrode and only hydrogen evolution occurred on the Pt electrode.
The CO2 detected at the cathode was completely derived from the anode and diffused through
the membrane. The test started when the CO2 concentration approached 0 ppm at open circuit.
For measurement of the total CO2 concentration, oxygen replaced nitrogen in the cathode and
the crossover methanol was oxidized, which produced a large amount of CO2.
Model development and analysis
Basic equations and model development
Modification of the microstructure of the MEA can enhance mass transport and further
improve the cell performance.19 An empirical model was developed to reveal the relationship
between the hydrophilic-hydrophobic character, pore structure, and mass transport process in
the anode MPL. However, methanol delivery and CO2 removal in the DMFC are complicated
processes. For instance, the anisotropy of the backing layer (BL) causes the mass transfer not
to be strictly one-dimensional.20 Methanol will diffuse through the porous media in the
in-plane direction of MEA. Additionally, temperature gradients exist because of the uneven
reaction at the catalyst layer. To develop a simplified model that captures the essence of the
mass transport behavior, some assumptions were made as follows:
(1) Steady state
(2) Isothermal conditions
AIChE Journal
(3) An ideal gas mixture and uniform gas pressure over the entire cell
(4) CO2 only exists in the gas phase
(5) The crossover methanol is completely oxidized in the cathode catalyst layer
(6) Transport of the species is approximated as one-dimensional
The model mainly analyzes the mass transport process on the anode side of the cell.
Methanol transfer in the anode is determined by the sum of the methanol flux, NMeOH, in
the liquid and gas phases. Fick’s law is used to describe methanol transport in a liquid and its
conveyance in a gas. The Maxwell-Stefan multi-component diffusion equation was
applied.21,22
N MeOH = N lMeOH + N gMeOH (1)
dClMeOH
N lMeOH = − DlMeOH
, eff ⋅ (2)
dx
C gMeOH N gCO2 − Cg ,t DgMeOH

, eff ∇C g
MeOH
N MeOH
= (3)
C g ,t − C gMeOH
g
DMeOH
l,eff and DMeOH
g,eff indicate the effective diffusion coefficient of aqueous and gaseous methanol
corrected by the Bruggeman correlation for porous media, respectively; l indicates the liquid
state and g indicates the gas state. CMeOH is the methanol concentration and NgCO is the flux of
2
CO2. The effective diffusion coefficient can be calculated from the porosity, ε, and liquid
saturation, s, of the porous media.
= DlMeOH ( ε s )
1.5
DlMeOH
, eff
= DgMeOH ε (1 − s ) 
1.5
DgMeOH
, eff (4)
AIChE Journal
In the gas phase, the concentration of H2O vapor, which is in equilibrium with the solution,
remains unchanged. Therefore, Eq. 3 only describes the binary diffusion system that includes
methanol vapor and CO2. The concentration of methanol vapor, CMeOH

g , is determined from
Henry’s law and the total gas concentration, Cg,t, is obtained from the ideal gas law. The
transport behavior of CO2 gas can also be evaluated from the Maxwell-Stefan Equation.
The methanol flux through the anode catalyst layer (CL) is consumed by two paths: the
methanol oxidation reaction (MOR) occurs at the anode CL|PEM interface with crossover
through the PEM to the cathode side. Methanol consumption by the MOR obeys Faraday’s
Law and the equation for methanol conservation at the anode CL can be determined. The first
and second terms on the right side of Eq. 5 represent the electro-oxidation consumption and
methanol crossover flux, respectively.
i
N MeOH = + N cross
MeOH
(5)
6F
By combining Eq. 5 with Eq. 1, we evaluate the balance between the methanol delivery and
consumption.
i
N lMeOH + N gMeOH = + N cross
MeOH
(6)
6F
To solve Eq. 6, the two-phase flow in porous media was studied. The capillary pressure,
which is defined as pc = pg – pl, drives the gas-liquid flow and is a function of the contact
angle, θ, porosity, ε, and liquid saturation, s, as described by Eq. 7. This semi-empirical
correlation was first proposed by Leverett23 and then modified by Rose and Bruce.24
0.5
ε 
pc = σ cos θ  
K
(1.417s − 2.120s 2
+ 1.263s 3 ) (7)
AIChE Journal
To determine the liquid pressure, Darcy’s law was employed, where K and kr,l denote the
permeability of the porous media and the relative permeability of the liquid phase. With the
capillary pressure known, the liquid saturation can be calculated if the contact angle and
porosity are obtained. Furthermore, the effective diffusion coefficient and concentration
distribution of methanol on the anode side can be determined.
Kkr ,l
vl = − ∇pl (8)
µl
The gas-liquid flow is more complex in the anode catalyst layer due to the accumulation
of CO2 gas. The flux of CO2 through the membrane is employed as shown in Eq. 9.25 The
CO
CO2 concentration at the anode CL, Cg,aCL , is assumed as constant because of the relatively
2
low methanol vapor concentration. For membranes of similar thickness, the gas fraction in
the anode catalyst layer, sg,aCL, is obtained from the CO2 diffusion through the membrane.
C gCO,aCL
2
CO2
N mem = Dmem
CO2
s g ,aCL (9)
δ mem
The anode overpotential, ηa, of the DMFC can be calculated from Eq. 10, which was
proposed by Meyers and Newman to describe the four-step mechanism of the MOR. The
reaction constant kr was introduced to capture the methanol oxidation reaction from the 0th
order to 1st order;26-28 The value is 1.5 in this study.
α F 
MeOH
CaCL exp  a ηa 
i MOR = i0MOR  RT  (10)
 αa F 
CaCL + kr exp 
MeOH
ηa 
 RT 
In the equation, iMOR equals the cell current density and iMOR
0 is the exchange current density.
CMeOH
aCL and αa are the methanol concentration and the transfer coefficient in the anode CL,
respectively. A detailed description of these parameters is presented in Table 1.
AIChE Journal
Model analysis
The model focuses on enhancing the mass transfer of the DMFC by optimizing the
hydrophilic-hydrophobic character and pore structure of the anode MPL. Equation 7 is used
to establish the relationship between the mass transport and structural variation of the anode
MPL. Noting that the contact angle and porosity change suddenly at the interface of the
anode BL and MPL, a saturation jump model was used to analyze the transformation during
mass transport. Across the interface, the capillary pressure remains uniform.
pc ,aBL = pc , aMPL (11)
By combining Eq. 11 with Eq. 7, the influence of the contact angle and porosity of the anode
MPL can be obtained as follows:
0.5 0.5
ε  ε 
σ cos θ BL  BL  J ( sint, BL ) = σ cos θ MPL  MPL  J ( sint, MPL ) (12)
 K BL   K MPL 
J(s) represents the Leverett function described in Equation 7 and KMPL is the gas permeability
of the MPL, which follows an exponential relationship with the porosity.29 At the anode BL,
the capillary pressure remains the same for all the cells due to the identical structure of the
carbon paper. The contact angle and porosity of the MPL will control the liquid saturation in
this layer, which affects the effective diffusion coefficient as well as CO2 removal in the
anode catalyst layer. To optimize the microstructure of the anode MPL by adjusting the PTFE
content, the correlation of the contact angle, porosity, and PTFE loading must first be
clarified.
Moreover, it is believed that the addition of PTFE to the anode MPL will increase the
gas path and alleviate CO2 blockage in the anode CL, i.e., the gas fraction of the anode CL is
also associated with the PTFE content. Specifically, the anode MPL with more PTFE
AIChE Journal
undergoes higher gas saturation, which increases the effective diffusion coefficient of the gas
phase. Thus, the CO2 removal ability is enhanced and less CO2 accumulates in the anode CL.
By fitting the experimental data, the gas fraction in the anode CL is solved in the next
section.
The methanol crossover flux, NMeOH

cross , involves two parts: the electro-osmotic drag effect
originating from proton migration and methanol diffusion from the anode to the cathode due
to the concentration gradient, where the following equation is obtained:
MeOH
i MeOH CaCL
MeOH
N cross = N EOD
MeOH
+ N diff
MeOH
= n MeOH + Dmem (13)
F δ mem
Here, nMeOH and DMeOH

mem are the electro-osmotic drag coefficient of methanol and the methanol
diffusivity in the PEM, respectively; δmem is the thickness of the membrane. At the cathode,
the crossover methanol is completely oxidized and the methanol concentration is zero.
Obviously, the amount of crossover methanol can be calculated from the methanol
concentration at the anode CL, which is affected by the hydrophilic-hydrophobic character
and pore structure of the anode MPL.
The cathode overpotential, ηc, can be described as a modified Bulter-Volmer equation.30
α F 
O2
CcCL
i ORR
=i
ORR
0 O2
exp  c ηc  (14)
Cref  RT 
The cathode current density, iORR, for the oxygen reduction reaction (ORR) consists of two
parts: the cell current density, i, and parasitic current density, ip, which originates from the
oxidation of crossover methanol. Thus, we relate the cathode overpotential, ηc, to the
microstructure of the anode MPL.
The cell voltage, E, of the DMFC can then be calculated as:
AIChE Journal
E = E ORR − E MOR − ηa − ηc − iRm (15)
where EORR and EMOR are the thermodynamic potential of the ORR and MOR, respectively.
Rm is the resistance of the cell with different PTFE contents in the MPL, which is determined
by the electrochemical impedance, as presented in Table 2. Combined with the overpotential
equation, the cell performance with a certain PTFE content can be obtained.
In summary, a relationship between the mass transport and PTFE content, which alters
the hydrophilic-hydrophobic character and pore structure of the anode MPL, was defined. A
plot of the methanol concentration and PTFE content can be obtained at a certain current
density. Moreover, the anode and cathode overpotential can be derived by determination of
the contact angle and porosity. The voltage-current density curves or polarization curves of
the DMFC can also be obtained by setting the PTFE content to a constant value.
Results and discussion
Hydrophilic-hydrophobic character and pore structure of MPLs
The empirical model is sensitive to the contact angle value of the MPL. As shown in Eq.
12, the saturation, s, is directly affected by the contact angle. Thus, the effective methanol
diffusion coefficient is also influenced. Generally speaking, the methanol delivery and CO2
removal are both controlled by this parameter. Therefore, the contact angles of the MPLs
with PTFE contents ranging from 5 to 35 wt.% (with intervals of 5 wt.%) were measured.
The repeatability was confirmed by depositing and analyzing seven drops on random parts of
the surfaces. Figure 1 shows the arithmetic average of the contact angles within ±3° deviation.
The increase in the contact angle with increasing PTFE content in the MPL is less than linear.
A sharp increase occurs when the content is less than 15%. Once the PTFE content exceeds
15%, the values change minimally. The model focuses on the mass transport process inside
AIChE Journal
MPL and takes the contact angle of ideal surface into consideration. Since the measured
surfaces of MPLs are nonideal and the measured contact angles are higher due to the inflating
effect of surface roughness: macroscopic roughness created by accumulation of the XC-72R
particles and microscopic roughness of the individual particles. Thus, plugging the measured
contact angles directly into the model would lead to a lower saturation value. The correction
of measured contact angle is necessary to exclude the influence of roughness and evaluate the
hydrophilic-hydrophobic character of MPL which is controlled by the heterogeneity of
carbon black and PTFE on the surfaces. The surfaces, including the profile and roughness,
were characterized by a non-contact optical instrument that would not damage the MPLs.
The 3D profiles of the MPLs and the corresponding Abbott-Firestone curves are shown in
Figure 2. The surface of the MPL with 5 wt.% PTFE is smooth, whereas deeper valleys
appeared in the 15 and 25 wt.% samples. The depth of the surfaces, which follows a Gaussian
distribution, becomes greater with increasing PTFE content. We conclude that the addition of
PTFE enlarges the aggregates of XC-72R particles in the ink. During the spraying process,
the accumulation of bigger aggregates leads to greater depth and results in an increase in the
mean surface roughness, as shown in Table 3. The difference in the roughness is less than 0.4
µm. This small increment in the roughness is negligible when considering its contribution to
the measured contact angle. In terms of the microscopic roughness, the morphology of each
XC-72R particle is assumed to be identical. Thus, the microscopic roughness of the surfaces
is the same. Overall, the inflating effect of the surface roughness on the contact angle is
similar for all samples and the heterogeneity of carbon black and PTFE on the surfaces leads
to differences in the contact angle.
XC-72R particles and PTFE are materials of completely different wettability, i.e., θ ≈
55−64°31,32 for water on XC-72R and θ ≈ 110−120°33,34 for water on PTFE. Thus, the actual
contact angle of the MPLs must be between these two boundary values. During the
AIChE Journal
heat-treatment process, the molten PTFE tends to be evenly distributed on the surface of the
carbon particles and the hydrophobicity of the surfaces increases. In the basis of the measured
contact angle, the hydrophobicity of the surface with 15 wt.% PTFE is close to the maximum
value and there is little increase when the content exceeds 15 wt.%. In other words, full
coverage of PTFE on carbon black is probably achieved and the material behaves as if it is
PTFE. For simulation, the contact angles are fitted to the maximum value of 120° for the
MPLs with 15 wt.% and 25 wt.% PTFE and 100° for that with 5 wt.% PTFE (Table 3).
The pores with diameters ranging from 2 nm to 200 nm were investigated. The pore
volume of the MPLs decreases with the addition of PTFE, and this tendency is more
pronounced in moving from 15 wt.% to 25 wt.% PTFE (Table 3). As discussed above, the
coverage of PTFE on XC-72R gradually reaches a maximum value in moving from 5 wt.% to
15 wt.%. Further addition of PTFE mainly thickens the covering and enhances the
pore-filling effect, blocking the path for mass transport in the MPLs. The fitted porosity
values also follows a similar decreasing trend, as shown in Table 3. The changes in the pore
structure will affect the mass transport of the cells. At open circuit, the reduced pore volume
hinders methanol diffusion, leading to less methanol crossover and resulting in an increase in
the open circuit voltage (OCV).
Two-phase distribution in anode catalyst layer
According to Eq. 9, the gas fraction, sg,aCL, at the anode CL is proportional to the flux of
CO2 diffusing through the PEM. The two-phase distribution in the anode CL can be
calculated from the CO2 diffusion data measured at the outlet of the cathode (see Figure 3a).
The gas fraction expands rapidly at low current density and then gradually increases and
finally fluctuates around a fixed value. When the current density is low, the CO2 bubbles are
small and separated, making them difficult to remove from the anode catalyst layer. The
AIChE Journal
bubbles accumulates and becomes bigger and continuous with an increase in the current
density. Concomitantly, the CO2 removal ability is improved, accompanied by a change in
the appearance of the bubbles. Eventually, a balance is achieved between the production and
removal of CO2. A similar two-phase flow and current density were observed in the anode
flow channel.35-37
The rate of production of CO2 in the anode CL obeys Faraday’s law and the removal rate
is limited by the gas path of each cell. A modified Logistic Equation is used to describe the
relation between the gas fraction, sg,aCL, and current density, i. By fitting the experimental
data, the influence of the PTFE content, z, on the gas fraction is simplified as a polynomial
function as follows:
 
− 1 ( 2.23z 2 − 1.62 z + 0.75 )
2
sg ,aCL =  (16)
 1 + exp ( −0.0035i ) 
As depicted in Figure 3b, the sg,aCL values calculated from the CO2 diffusion data fit well with
the simulated results. The gas fraction of the anode CL declines with the addition of PTFE in
the MPL due to an increase in the gas path, which enhances the CO2 removal ability of MPL.
The maximum gas fraction of Cell 1 is 0.67 at the anode CL, and this value declines to 0.56
for Cell 2 and 0.48 for Cell 3. Accordingly, Cell 1 most likely suffers from CO2 blockage in
the anode CL with an increase in the current density, whereas this situation is alleviated for
Cell 2 and Cell 3 at the same current density.
Distribution of liquid saturation and methanol concentration
Figure 4 shows the simulated results for liquid saturation at the anode side. The curves for
liquid saturation in the anode BLs almost overlap at 40 and 240 mA cm-2 due to the high
porosity of the BLs, which makes it easy for liquid and gas to pass through. The
liquid-saturation jump, as shown in Table 4, is observed at the interface of the anode BL and
AIChE Journal
MPL caused by the sudden change in the porosity and contact angle. The addition of PTFE
increases the hydrophobicity and reduces the porosity of the anode MPL, which decrease the
liquid saturation from 0.565 in Cell 1 to 0.225 in Cell 3 at 40 mA cm-2. The reduction in the
liquid saturation of the anode MPL is conducive to gas movement, and consequently, less
CO2 accumulates in the anode CL. Cell 1 with the anode MPL having the weakest CO2
removal ability exhibites the highest gas saturation in the CL, followed by Cell 2 and Cell 3
successively. At 40 mA cm-2, the difference in the liquid saturation at the anode CL for Cell 1
and Cell 3 is less than 0.12 because of the low CO2 production rate. When the current density
reaches 240 mA cm-2, a large amount of CO2 is generated by the reaction and the liquid
saturation difference increases to 0.19 owing to the difference in the CO2 removal ability of
the anode MPL. The effect of liquid saturation on the methanol delivery could be clarified
from the methanol concentration distribution.
The distribution of the methanol concentration at 40 and 240 mA cm-2 along the anode
side is shown in Figure 5. The difference in the methanol concentration is minor in the BLs as
the pore structure and liquid saturation of the cells is similar. In the MPL, the concentration
of methanol in Cell 3 decreases dramatically as this MPL has the lowest porosity, which
reduces the effective diffusion coefficient of methanol, making methanol delivery more
difficult. At 40 mA cm-2, a limited amount of CO2 accumulate and the concentration gap
across the anode CL is about 100 mol m-3 for all cells. When CO2 blockage becomes serious
in the anode CL at 240 mA cm-2, the difference in the methanol concentration becomes
obvious. The methanol concentration in the anode CL of Cell 1 declines rapidly, resulting
from the weak CO2 removal ability of the MPL. Among the cells, Cell 2 provides the highest
methanol concentration of 202 mol m-3 owing to the appropriate porosity and hydrophobic
character of the anode MPL. In summary, the porosity of the anode MPL is the dominant
factor influencing the methanol delivery at low current density, and as the current density
AIChE Journal
continues to increase, the accumulated CO2 will hinder methanol delivery and the CO2
removal ability plays an increasingly important role. With the addition of PTFE, the
improved CO2 removal ability is accompanied by reduced porosity. A good balance between
the methanol delivery and CO2 removal is achieved when the PTFE content of the anode
MPL is adjusted to 15 wt.%, and the optimal hydrophilic-hydrophobic balance and pore
structure are achieved.
The methanol crossover rate is linearly related to the methanol concentration at the anode
CL. As shown in Figure 6, the crossover current density is high at low current density and
declines sharply with increasing current density for Cell 1. We ascribe this sharp decrease to
massive CO2 accumulation in the anode CL. The crossover current density of Cell 3 remains
low at low current density due to the poor pore volume of the anode MPL. At high current
density, the methanol reaching the anode CL of Cell 3 is more than that in Cell 1 because of
the better CO2 removal ability of the former. Evidently, the crossover current density of Cell
2 is the highest when the current density exceeded 20 mA cm-2. In other words, Cell 2 with
15 wt.% PTFE in the anode MPL provided the highest methanol concentration for the MOR
at the anode CL. The calculated results are consistent with the experimental data shown in the
same figure. The experimental data points are slightly lower than the simulated results
because CO2 that is partially dissolved in the water could not be detected by the CO2 sensor.
As the current density increases, the effect of the crossover current density on the cathode
overpotential is weakened. Thus, the methanol concentration at the anode CL, CMeOH
aCL , plays a
key role in influencing the cell performance, especially at high current density.
Cell performance
After determining the methanol concentration at the anode CL and the crossover current
density, the anode and cathode overpotential are calculated as shown in Figure 7. Cell 2 with
AIChE Journal
15 wt.% PTFE in the anode MPL has the highest CMeOH

aCL value when the current density
exceeds 20 mA cm-2. The rate of methanol adsorption on the PtRu electrode is also higher
and CO is more linearly bonded on this electrode (where CO is a long-term and predominant
electrode poison during the electrooxidation of methanol) in the anode CL of Cell 2.38 Thus,
the anode overpotential of Cell 2 is higher than that of Cell 1 and Cell 3 in the low current
density region. When the current density exceeds 220 mA cm-2, the methanol concentration at
the anode CL is very low and methanol transport to the surface of the electrode becomes the
rate-determining step of the reaction, resulting in the lowest anode overpotential for Cell 2.
Based on Equation 14, the cathode overpotential of the cells exhibits a positive correlation
with the crossover current density, as illustrated in Figure 7b. Given that the difference in the
cathode overpotential of the cells is less than 4 mV at the same current density, it is plausible
that the anode overpotential and ohmic loss dominate the cell performance.
The simulated voltage-current density plot is compared with the experimental data in
Figure 8a. Some deviation is observed at low current density, which may be caused by the
use of the kinetic expression and the inaccurate liquid-saturation boundary condition,
although the model does capture the ohmic loss and mass transport behavior of the cells. In
the ohm-dominated region of the polarization curves, Cell 1 shows better performance due to
less adsorption of poisons on the anode and the lowest cell resistance (Table 2). When the
current density exceeds 220 mA cm-2, the voltage of Cell 1 and Cell 3 declines rapidly due to
poor methanol delivery. As shown in Figure 8b, Cell 2 exhibites better and steadier cell
performance than Cell 1 and Cell 3. Operation of Cell 2 at 240 mA cm-2 is adequate as this is
close to the peak power density; the power density is improved by 7.7% and 17.3% when
compared with that for Cell 1 and Cell 3, respectively. Overall, the model clarifies the
optimization of the microstructure based on the PTFE content and illustrates its effect on the
mass transport process.
AIChE Journal
Generally, this model uses structural parameters as its basis; if the porosity, contact
angles, and other parameters of MPL with other PTFE contents are obtained, the model could
be applied to predict the mass transport behavior and cell performance as well. Moreover, the
model could be extended to the rational design of MPL structures other than the PTFE
content (e.g. carbon black types or heterogeneous application) where the structural
parameters are known. The structures of the BL and CL are also involved in this model as
they affect the results of the conservation equation and two-phase equation. Certainly, based
on structural characterization, the optimal PTFE loading could be obtained even if it is
changed due to the use of a different BL or CL.
Conclusions
This study successfully reveals the effect of the hydrophilic-hydrophobic character and
pore structure of the anode MPL on the mass transport process. The microstructure of MPLs
with different PTFE contents and the accompanying methanol delivery and CO2 removal
behavior were investigated. The addition of PTFE enhances the CO2 removal ability of the
anode MPL, which alleviates CO2 blockage in the anode catalyst layer, whereas the narrowed
flow path hinders methanol delivery in the anode MPL. Excellent mass transport balance
between the methanol delivery and CO2 removal is achieved when the PTFE content in the
anode MPL is adjusted to 15 wt.%. The calculated polarization data for the DMFCs correlate
well with the experimental data, which confirms the validity of the empirical model. This
model can be further used in the design of DMFCs to achieve enhanced mass transport.
Acknowledgment
The work is supported by the National Natural Science Foundation of China (21673221,
21433003), National Science and Technology Major Project (2016YFB0101202 and
2017YFB0102905), Jilin Province Science and Technology Development Program
AIChE Journal
(20150101066JC, 20160622037JC and 20170203003SF), Hundred Talents Program of
Chinese Academy of Sciences and the Recruitment Program of Foreign Experts
(WQ20122200077).
Notation
αa = transfer coefficient of the anode CL
αc = transfer coefficient of the cathode CL
δmem = thickness of membrane, m
ε = porosity
ηa = anode overpotential, V
ηc = cathode overpotential, V
µl = dynamic viscosity of liquid phase, kg m-1 s-1
ρ = density, kg m-3
σ = surface tension, N m-1
CMeOH
l = mole concentration of aqueous methanol, mol m-3
DMeOH
l,eff = effective diffusion coefficient of aqueous methanol, m2 s-1
E = cell voltage, V
F = Faraday’s constant, C mol-1
i = operating current density of DMFC, A m-2
iMOR = anode current density, A m-2
AIChE Journal
iORR = cathode current density, A m-2
ip = parasitic current density, A m-2
K = gas permeability, m2
NMeOH = mole flux of methanol, mol m-2 s-1
NMeOH
cross = crossover methanol flux, mol m-2 s-1
n = electro-osmotic drag coefficient
pc = capillary pressure, Pa
R = universal gas constant, J mol-1K-1
Rm = cell resistance, Ω cm2
s = liquid saturation
T = cell temperature, K
vl = velocity of liquid phase, m s-1
z = PTFE content in anode MPL
Subscripts
aBL = anode backing layer
aMPL = anode micro-porous layer
aCL = anode catalyst layer
cCL = cathode catalyst layer
eff = effective value
AIChE Journal
EOD = electro-osmotic drag
g = gas phase
l = liquid phase
Literature Cited
1. St-Pierre J, Wilkinson D. Fuel cells: a new, efficient and cleaner power source. AIChE
Journal. 2001;47(7):1482-1486.
2. Zhao TS, Xu C, Chen R, Yang WW. Mass transport phenomena in direct methanol fuel
cells. Prog. Energy Combust. Sci. 2009;35(3):275-292.
3. Basri S, Kamarudin SK. Process system engineering in direct methanol fuel cell.
International Journal of Hydrogen Energy. 2011;36(10):6219-6236.
4. Li X, Faghri A. Review and advances of direct methanol fuel cells (DMFCs) part I:
Design, fabrication, and testing with high concentration methanol solutions. J. Power
Sources. 2013;226:223-240.
5. Krishnamurthy B, Deepalochani S. Effect of PTFE content on the performance of a Direct
methanol fuel cell. International Journal of Hydrogen Energy. 2009;34(1):446-452.
6. Sudaroli BM, Kolar AK. Heat and mass transfer characteristics of direct methanol fuel
cell: Experiments and model. Energy Procedia. 2014;54:359-366.
7. Park S, Lee JW, Popov BN. Effect of PTFE content in microporous layer on water
management in PEM fuel cells. J. Power Sources. 2008;177(2):457-463.
8. Gallo SP, Cristiani C, Dotelli G, Omati L, Zampori L, Pelosato R, Guilizzoni M. Effect of
different substrates, inks composition and rheology on coating deposition of microporous
layer (MPL) for PEM-FCs. Catalysis Today. 2009;147, Supplement:S30-S35.
9. Gostick JT, Ioannidis MA, Fowler MW, Pritzker MD. On the role of the microporous layer
in PEMFC operation. Electrochemistry Communications. 2009;11(3):576-579.
AIChE Journal
10. Schonewill PP, Leighton DT Jr. Mass transport enhancement in the DMFC using an
externally applied oscillatory flow. AIChE Journal. 2008;54(6):1410-1423.
11. Sudaroli BM, Kolar AK. An experimental study on the effect of membrane thickness and
PTFE (polytetrafluoroethylene) loading on methanol crossover in direct methanol fuel cell.
Energy. 2016;98:204-214.
12. Oedegaard A, Hebling C, Schmitz A, Møller-Holst S, Tunold R. Influence of diffusion
layer properties on low temperature DMFC. J. Power Sources. 2004;127(1–2):187-196.
13. Li J, Ye DD, Zhu X, Liao QA, Ieee. Effects of microporous layer preparation on the
performance of an air-breathing direct methanol fuel cell. International Conference on
Sustainable Power Generation and Supply 2009, April 6-7.
14. Shao ZG, Hsing IM, Zhang H, Yi B. Influence of anode diffusion layer on the
performance of a liquid feed direct methanol fuel cell by AC impedance spectroscopy.
International Journal of Energy Research. 2006;30(14):1216-1227.
15. Park KW, Kwon BK, Choi JH, Park IS, Kim YM, Sung YE. New RuO2 and
carbon–RuO2 composite diffusion layer for use in direct methanol fuel cells. J. Power
Sources. 2002;109(2):439-445.
16. Lu GQ, Wang CY, Yen TJ, Zhang X. Development and characterization of a
silicon-based micro direct methanol fuel cell. Electrochim. Acta. 2004;49(5):821-828.
17. Nakagawa N, Abdelkareem MA, Sekimoto K. Control of methanol transport and
separation in a DMFC with a porous support. J. Power Sources. 2006;160(1):105-115.
18. Eccarius S, Garcia BL, Hebling C, Weidner JW. Experimental validation of a methanol
crossover model in DMFC applications. J. Power Sources. 2008;179(2):723-733.
19. Cai W, Yan L, Liang L, Xing W, Liu C. Model-based design and optimization of the
microscale mass transfer structure in the anode catalyst layer for direct methanol fuel cell.
AIChE Journal. 2013;59(3):780-786.
AIChE Journal
20. Gostick JT, Fowler MW, Pritzker MD, Ioannidis MA, Behra LM. In-plane and
through-plane gas permeability of carbon fiber electrode backing layers. J. Power Sources.
2006;162(1):228-238.
21. Spiegel C. Modeling the gas diffusion layers. PEM fuel cell modeling and simulation
using Matlab. Burlington: Academic Press; 2008:197-241.
22. Ko J, Chippar P, Ju H. A one-dimensional, two-phase model for direct methanol fuel cells
- Part I: Model development and parametric study. Energy. 2010;35(5):2149-2159.
23. Leverett MC. Capillary behavior in porous solids. Transactions of the American Institute
of Mining and Metallurgical Engineers. 1941;142:152-169.
24. Rose W, Bruce WA. Evaluation of capillary character in petroleum reservior rock.
Transactions of the American Institute of Mining and Metallurgical Engineers.
1949;186(5):127-142.
25. Ren X, Myles TD, Grew KN, Chiu WKS. Carbon dioxide transport in Nafion 1100 EW
membrane and in a direct methanol fuel cell. J. Electrochem. Soc.
2015;162(10):F1221-F1230.
26. Meyers JP, Newman J. Simulation of the direct methanol fuel cell - I. Thermodynamic
framework for a multicomponent membrane. J. Electrochem. Soc. 2002;149(6):A710-A717.
27. Meyers JP, Newman J. Simulation of the direct methanol fuel cell - II. Modeling and data
analysis of transport and kinetic phenomena. J. Electrochem. Soc. 2002;149(6):A718-A728.
28. Meyers JP, Newman J. Simulation of the direct methanol fuel cell - III. Design and
optimization. J. Electrochem. Soc. 2002;149(6):A729-A735.
29. Garcia-Salaberri PA, Vera M, Iglesias I. Modeling of the anode of a liquid-feed DMFC:
Inhomogeneous compression effects and two-phase transport phenomena. J. Power Sources.
2014;246:239-252.
AIChE Journal
30. Wang ZH, Wang CY. Mathematical modeling of liquid-feed direct methanol fuel cells. J.
Electrochem. Soc. 2003;150(4):A508-A519.
31. De Ridder DJ, Verliefde ARD, Heijman BGJ, Gelin S, Pereira MF, Rocha RP, Figueiredo
JL, Amy GL, Van DH. A thermodynamic approach to assess organic solute adsorption onto
activated carbon in water. Carbon. 2012;50(10):3774-3781.
32. Vol'fkovich YM, Sosenkin VE, Nikol'skaya NF. Hydrophilic-hydrophobic and sorption
properties of the catalyst layers of electrodes in a proton-exchange membrane fuel cell: A
stage-by-stage study. Russian Journal of Electrochemistry. 2010;46(4):438-449.
33. Goswami S, Klaus S, Benziger J. Wetting and absorption of water drops on nafion films.
Langmuir. 2008;24(16):8627-8633.
34. He C, Mighri F, Guiver MD, Kaliaguine S. Sorption of water/methanol on Teflon and
hydrocarbon proton exchange membranes. ACS Applied Materials & Interfaces.
2016;8(19):12541-12551.
35. Argyropoulos P, Scott K, Taama WM. Carbon dioxide evolution patterns in direct
methanol fuel cells. Electrochim. Acta. 1999;44(20):3575-3584.
36. Yang H, Zhao TS, Ye Q. In situ visualization study of CO2 gas bubble behavior in DMFC
anode flow fields. J. Power Sources. 2005;139(1-2):79-90.
37. Liao Q, Zhu X, Zheng X, Ding Y. Visualization study on the dynamics of CO2 bubbles in
anode channels and performance of a DMFC. J. Power Sources. 2007;171(2):644-651.
38. Gasteiger HA, Markovic N, Ross PN, Cairns EJ. Methanol electrooxidation on
well-characterized Pt-Ru alloys. Journal of Physical Chemistry. 1993;97(46):12020-12029.
AIChE Journal
List of Figure Captions
Figure 1. Apparent contact angle data for the investigated surfaces.
Figure 2. 3D profiles and corresponding Abbott-Firestone curves for the MPL surfaces:
(a,d) 5%, (b, e) 15%, (c, f) 25%.
Figure 3. (a) CO2 diffusion through the membrane; (b) calculated and experimental gas
saturation in anode catalyst layer.
Figure 4. Liquid saturation distribution in the anode side of the cells at
(a) 40 mA cm-2 and (b) 240 mA cm-2.
AIChE Journal
Figure 5. Methanol concentration distribution in anode side of the cells at
(a) 40 mA cm-2 and (b) 240 mA cm-2.
Figure 6. Comparison between simulated and experimental crossover current density of the
cells.
Figure 7. Simulated (a) anode and (b) cathode overpotential data for the cells.
Figure 8. Comparison between experimental and simulated data for the cells: (a)
polarization curve and (b) power density versus current density curve.
List of Table Captions
Table 1. Parameter Values
Table 2. Characteristics of the Cells
Table 3. Surface Morphology and Pore Structure of Anode MPLs
Table 4. Saturation Jump at the Interfaces of Anode Side
AIChE Journal
Figure 1. Apparent contact angle data for the investigated surfaces.
288x200mm (300 x 300 DPI)
AIChE Journal
Figure 2. 3D profiles and corresponding Abbott-Firestone curves for the MPL surfaces: (a,d) 5%, (b, e) 15%,
(c, f) 25%.
73x35mm (300 x 300 DPI)
AIChE Journal
Figure 3. (a) CO2 diffusion through the membrane; (b) calculated and experimental gas saturation in anode
catalyst layer.
47x17mm (300 x 300 DPI)
AIChE Journal
Figure 4. Liquid saturation distribution in the anode side of the cells at (a) 40 mA cm-2 and (b) 240 mA cm-
2.
47x18mm (300 x 300 DPI)
AIChE Journal
Figure 5. Methanol concentration distribution in anode side of the cells at (a) 40 mA cm-2 and (b) 240 mA
cm-2.
49x19mm (300 x 300 DPI)
AIChE Journal
Figure 6. Comparison between simulated and experimental crossover current density of the cells.
181x126mm (300 x 300 DPI)
AIChE Journal
Figure 7. Simulated (a) anode and (b) cathode overpotential data for the cells.
49x19mm (300 x 300 DPI)
AIChE Journal
Figure 8. Comparison between experimental and simulated data for the cells: (a) polarization curve and (b)
power density versus current density curve.
61x25mm (300 x 300 DPI)
AIChE Journal
Table 1. Parameter Values
Symbol and value Description

T = 333.15 K Cell temperature
αa =0.45 Anodic transfer coefficient
αc =0.80 Cathodic transfer coefficient
σ = 0.0625 N m-1 Surface tension
 35570  1 1 
i0MOR = 94.25 × exp  × −   A m-2 Exchange current density of anode
 R  353 T 
 73200  1 1 
i0ORR = 0.0422 × exp  × −   A m-2 Exchange current density of cathode
 R  353 T 
  1 1  Electro-osmotic drag coefficient of
n H 2 O = 2.9 × exp 1029 ×  − 
  333 T  water
Electro-osmotic drag coefficient of
n MeOH = n H 2 O ClMeOH M H 2 O ρ H 2 O
methanol
  1 1  Diffusion coefficient of methanol in
MeOH
Dmem = 4.9 × 10−10 exp  2436 ×  −   m2 s-1
  333 T   membrane
Diffusion coefficient of methanol in
DlMeOH = 10−5.416 − 999.778 T m2 s-1
water
DgMeOH = −6.954 × 10 −6 + 4.5986 × 10−8 T + 9.4979 × 10 −11T 2 m2 s-1 Diffusion coefficient of methanol in
gas
Diffusion coefficient of carbon
CO2
Dmem = 4.61 × 10 −10 m2 s-1
dioxide in membrane
K BL = 10−12 m2 Gas permeability of backing layer
K MPL = 2.31 × 10 −14 exp ( 6.8 × ε MPL ) m2 Gas permeability of micro-porous

layer
εaBL= 0.7 Porosity of anode backing layer
εaCL= 0.3 Porosity of anode catalyst layer
δaBL= 0.00019 m Thickness of anode backing layer

Thickness of anode micro-porous
δaMPL= 0.00003 m
layer
δaCL= 0.00001 m Thickness of anode catalyst layer
δmem= 0.00014 m Thickness of the Nafion membrane
AIChE Journal
Table 2. Characteristics of the Cells
Cell 1 Cell 2 Cell 3

PTFE Content 5% 15% 25%
Resistance / Ω cm2 0.219 0.276 0.303
OCV / V 0.693 0.698 0.709
Power density @ 240 mA cm-2/ mW cm-2 75.11 80.90 68.94
AIChE Journal
Table 3. Surface Morphology and Pore Structure of Anode MPLs
PTFE content 5% 15% 25%

Apparent contact angel θ* / ˚ 143 161 168
Mean roughness /µm 1.50 1.61 1.89
Fitted contact angel θ / ˚ 100 120 120
Pore volume of anode MPL/ cm3/g 1.110 0.989 0.787
Porosity of anode MPL 0.35 0.27 0.13
AIChE Journal
Table 4. Saturation jump at the interfaces of anode side
Interfaces of anode side Cell 1 Cell 2 Cell 3

BL/MPL @40 mA/cm2 0.799/0.565 0.799/0.347 0.799/0.225
MPL/CL @40 mA/cm2 0.564/0.592 0.346/0.663 0.224/0.706
BL/MPL @240 mA/cm2 0.798/0.564 0.798/0.346 0.798/0.225
MPL/CL @240 mA/cm2 0.563/0.325 0.345/0.442 0.223/0.515
AIChE Journal

Guangrong Deng, Liang Liang, Zhao Jin, Chenyang Li

Uploaded by

Copyright:

Available Formats

Guangrong Deng, Liang Liang, Zhao Jin, Chenyang Li

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Guangrong Deng, Liang Liang, Zhao Jin, Chenyang Li

Uploaded by

Copyright:

Available Formats

Transport Phenomena and Fluid Mechanics AIChE Journal

Enhancing Mass Transport in Direct Methanol Fuel Cell by Optimizing

the Microstructure of Anode Micro-Porous Layer

Guangrong Deng1,2,3 , Liang Liang1,3 , Zhao Jin1,3 , Chenyang Li1,3 ,

Changpeng Liu*1,3 , Junjie Ge*1,3 , Wei Xing*1,3,4

1.Laboratory of Advanced Power Sources, Changchun Institute of Applied Chemistry,

Chinese Academy of Sciences, Changchun, Jilin, 130022, PR China

2.University of Chinese Academy of Sciences, Beijing, 100039, PR China

4.State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied

Chemistry, Chinese Academy of Sciences, Changchun, Jilin, 130022, PR China

Changpeng Liu (liuchp@ciac.ac.cn)

Wei Xing (xingwei@ciac.ac.cn)

This article is protected by copyright. All rights reserved.

structure of the flow path were established by optimizing the content of

polytetrafluoroethylene (PTFE) in the anode MPL. An empirical model was developed to

Topical Heading: Transport Phenomena and Fluid Mechanics

delivery, CO2 removal

structure on water management in proton exchange membrane fuel cells (PEMFCs).7-9 In

performance were reported by Park et al.15 and Lu et al.,16 respectively.

comprehensively understood, it is possible to employ computer-aided design to predict the

Finally, prediction of the cell performance from the model is demonstrated to be

well-matched with the experimental data.

Preparation and evaluation of MEAs

25 (Baltic Fuel Cells, Germany) device at 2 N mm-2.

the electrochemical impedance was measured by using an FC impedance meter (KIKUSUI,

Im (z) = 0, at the high-frequency side.

Characterization of surface morphology and porosity

length was set to 30 nm.

nitrogen adsorption/desorption data by the Barrett-Joyner-Halenda (BJH) method.

CO2 concentration at cathode

Model development and analysis

Basic equations and model development

mass transport behavior, some assumptions were made as follows:

(1) Steady state

(2) Isothermal conditions

(4) CO2 only exists in the gas phase

(6) Transport of the species is approximated as one-dimensional

conveyance in a gas. The Maxwell-Stefan multi-component diffusion equation was

N MeOH = N lMeOH + N gMeOH (1)

C gMeOH N gCO2 − Cg ,t DgMeOH

saturation, s, of the porous media.

methanol vapor and CO2. The concentration of methanol vapor, CMeOH

methanol crossover flux, respectively.

angle, θ, porosity, ε, and liquid saturation, s, as described by Eq. 7. This semi-empirical

distribution of methanol on the anode side can be determined.

order to 1st order;26-28 The value is 1.5 in this study.

respectively. A detailed description of these parameters is presented in Table 1.

pc ,aBL = pc , aMPL (11)

MPL can be obtained as follows:

The methanol crossover flux, NMeOH

to the concentration gradient, where the following equation is obtained:

Here, nMeOH and DMeOH

concentration at the anode CL, which is affected by the hydrophilic-hydrophobic character

and pore structure of the anode MPL.

The cathode overpotential, ηc, can be described as a modified Bulter-Volmer equation.30

microstructure of the anode MPL.

The cell voltage, E, of the DMFC can then be calculated as:

E = E ORR − E MOR − ηa − ηc − iRm (15)

by the electrochemical impedance, as presented in Table 2. Combined with the overpotential

Changpeng Liu1,3 , Junjie Ge1,3 , Wei Xing*1,3,4