Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2024 Feb 12;9(8):9383–9395. doi: 10.1021/acsomega.3c08756

Effects of Inhomogeneous Gas Diffusion Layer Properties on the Transportation Phenomenon and Performances of Proton-Exchange Membrane Fuel Cells

Hangwei Lei 1, Yuzhen Xia 1, Guilin Hu 1,*
PMCID: PMC10905583  PMID: 38434892

Abstract

graphic file with name ao3c08756_0013.jpg

The encapsulation of the proton-exchange membrane fuel cell (PEMFC) is an essential step of fuel cell stack assembly. The selection of the assembly pressure is very important to the stack performance and life. Based on that, this paper presents a method to describe the internal physical deformation of the gas diffusion layer (GDL) after inhomogeneous pressure by using user-defined functions (UDFs). The results show that the transmission difference caused by the nonuniform GDL deformation can be clearly seen by the UDFs method, and there is an obvious transition of GDL at the interface of the channel and the rib. A three-dimensional single-channel PEMFC model is established, and an optimal assembly pressure range is obtained, between 1.5 and 2.0 MPa. The maximum thermal stress inside the cell occurs in the middle of the membrane electrode assembly and decreases as the assembly pressure increases. Furthermore, the influence of rib–channel ratios is discussed. Compared to the fuel cells with ratios of 2:1, 2:3, and 1:2, the one with 1:1 exhibits the maximum current density and the highest power density.

1. Introduction

With the development of industry, the issues of resource depletion and environmental problems have become more and more serious. Clean and efficient energy utilization methods are gradually becoming the mainstream direction for future energy development and utilization. A proton-exchange membrane fuel cell (PEMFC) has been considered a very strong competitor for the power sources of future vehicles for its advantages, such as high energy utilization, low or zero emission, and short startup time.1−4

In order to further understand the mechanisms and improve fuel cell systems, many researchers have engaged in modeling and computer simulation studies of PEMFC. In terms of its developmental journey, the models have evolved from simple one-dimensional single-channel mathematical models to complex three-dimensional cell stack models.5−7 Furthermore, the characteristics being studied by the models have become increasingly complex, ranging from single-phase flow to multiphase flow,8,9 from steady-state simulations to transient simulations,10,11 and from macroscopic cell models to the establishment of mathematical models within the gas diffusion layer (GDL)12,13 and the catalyst layer (CL)14,15 at the micro and mesoscale. In particular, pore-scale models based on the Lattice Boltzmann method gained significant attention for their capability to elucidate the transport processes of water within the GDL at the micro and mesoscales.16,17 However, the aforementioned mathematical models established through numerical simulations have not considered the deformation of the GDL caused by assembly pressure, as well as the resulting changes in the GDL’s physical properties, internal temperature, and thermal stress distribution and their impact on cell performance.

Therefore, the selection of fuel cell assembly pressure and detailed theoretical research have become increasingly important. The choice of assembly load is crucial for the performance and lifespan design of fuel cell stacks.18 Since GDL has the worst mechanical properties among all components, it is the component that undergoes major deformation during assembly. Therefore, some researchers have studied the performance of the GDL after deformation. Zhou et al.19 studied the effects of compression and different parameters (carbon fiber diameter, porosity) on two-phase flow in GDL using a fluid volume method. The results showed that the larger the fiber diameter and porosity, the more favorable the water transport. Zhang et al.20 utilized the finite element method (FEM) to investigate the compressive deformation of GDL from a microscopic perspective. The results demonstrated that different porosities, fiber diameters, and thicknesses lead to varied microstructure effects on GDL deformation. With increasing pressure, the porosity decreases. Chen et al.21 examined the impact of assembly pressure on the mechanical properties of commercial carbon paper-based GDL. The study revealed that excessive assembly stress can lead to carbon fiber breakage, affecting its transport performance. It was also found that the spatial arrangement of pores with a diameter below 0.175 nm was largely unaffected, while it did have an influence on the spatial arrangement of larger pores with a diameter exceeding 0.2 nm. These are studies on the physical properties and mechanical strength of GDL after deformation, and the influence of GDL deformation on internal transmission and cell performance is not studied. Because the deformation of the cell model is complex under various assembly pressures, many researchers have studied it by simplifying the model. Chi et al.22 assumed that the physical characteristics of GDL in the compressed flow channel were uniform, and the deformation of GDL was not considered. By taking the average value of the physical parameters of the compressed GDL, a numerical simulation was carried out. Su et al.23 established a nonuniform deformation GDL, but the model ignored the deformation at the junction of the channel and the rib and treated the GDL inside the channel as the thickness. It was found that this model is superior to the uniform GDL model. Chippar et al.24 subdivided the uniform and nonuniform GDL models into three working conditions: uncompressed GDL, uninvaded GDL, and invaded GDL. The results showed that nonuniform GDL compression and intrusion had a significant effect on PEMFC internal transport properties. All of the above studies simplify the GDL deformation model and do not consider the transition of the junction position of the rib and the channel, as well as the nonuniform distribution of the physical parameters. Some studies focus on the influence of different working conditions and materials on the performance of GDL cells after compression deformation. Shrivastava et al.25 conducted experimental research on the influence of different gas humidity levels and compression ratios on the performance of PEMFC. The experiments revealed that the performance initially improves and then deteriorates with increasing pressure. In their study, the best performance of the cell was achieved at a compression ratio of 14.2 and 25.7% at 100 and 50% gas humidity, respectively. Movahedi et al.26 studied the effect of assembly pressure on the performance of cells with GDL thicknesses. The study found that the optimal assembly pressures for GDL thicknesses of 110, 254, and 370 μm were determined to be 0.395, 0.696, and 1.101 MPa, respectively. Kang et al.27 developed a two-dimensional FEM of the stacked metal bipolar plate (BP) to examine the phenomenon of stress concentration in the membrane electrode assembly (MEA) under assembly pressure. The consequences indicated that stress significantly concentrates in the membrane region beneath the rib edges of BP, potentially causing plastic deformation of the membrane. The aforementioned works mainly focused on mass transport and cell performance due to GDL deformation but did not consider the thermal stress and temperature distribution within the compressed cell. The internal temperature and thermal stress distribution during fuel cell operation play crucial roles in its performance and lifespan.

Therefore, this paper considers the nonuniform GDL deformation and proposes a method using UDFs to set the nonuniform porosity and permeability of GDL after deformation. At the same time, this paper explores mass transport, temperature distribution, thermal stress, mechanical stress, and cell performance under different assembly pressures and observes that the optimal assembly pressure varies between low-voltage and intermediate-voltage conditions. Moreover, the impact of rib–channel ratios on the distribution of current density and cell performances at 1.5 and 2.0 MPa are studied.

2. Model Development

2.1. Model Assumptions

The following assumptions are considered to develop the simplified mathematical model for PEMFC:

  • (1)

    The fuel cell is in the steady operation state.

  • (2)

    Except for GDL, other materials are isotropic and homogeneous porous media.

  • (3)

    Except for GDL, other components of PEMFC do not deform under various assembly pressures.

  • (4)

    All gases are ideal gases, and the fluid flow is laminar.

  • (5)

    The contact thermal resistance of the interface between GDL and BP is not taken into account

2.2. Geometric Model

Figure 1a shows a single PEMFC model. One single channel shown in Figure 1b is selected as the computation domain for saving the computation time. The fuel cell model is composed of BP, GDL, CL, and PEM, as illustrated in Figure 1c, whose dimensions are given in Table 1. The fuel cell model and its meshes with deformed GDL for the compression are illustrated in Figure 1d.

Figure 1.

Figure 1

Open in a new tab

PEMFC model and computation domain, (a) single cell model, (b) single-channel cell model, (c) assembly pressure and cell parts schematic, and (d) grid diagram of the XY cross section after compression deformation.

Table 1. Some Basic Parameters and Working Conditions.

parameters values
channel width/height (mm) 1/1
rib width (mm) 0.5
BP width/length/height (mm) 2/50/1.5
primary GDL thickness (mm) 0.28
membrane thickness (mm) 0.05
CL thickness (mm) 0.01
working temperature (K) 353.13
working pressure (Pa) 101 325
anode stoichiometry ratio 1.5
cathode stoichiometry ratio 2
open circuit voltage (V) 1.1
Faraday constant (C·mol–1) 96 485
primary contact electric resistance (mΩ·cm2) 14.328
Open in a new tab

2.3. Stress–Strain Model for GDL

To calculate the deformation of GDL, the equations representing the connection between stress, strain, and external load are described as follows, which are similar to our previous work.29

The force balance for the GDL is written as eq 1.

2.3. 1

where Fx, Fy, and Fz are the body forces in the x-, y-, and z-directions, respectively; σx, σy, and σz are the positive stresses in the corresponding directions (Pa); and τxy, τyz, and τzx are the shear stresses at the xy-, yz-, and zx-planes, respectively (Pa).

The deformation and strains of GDL relate to the displacements as in eq 2.

2.3. 2

where ψx, ψy, and ψz are positive strains in x-, y-, and z-directions; u, v, and w are displacements in the corresponding directions; and ψxy, ψyz, and ψzx are shear strains at the xy-, yz-, and zx-planes, respectively.

The strain or deformation is physically related to stress by eq 3.

2.3. 3

where E is the elastic modulus (Pa) and μ is the porosity of GDL.

2.4. PEMFC Model Based on CFD

The numerical simulation of the transportation phenomenon in PEMFC is based on a 3D CFD model considering geometric deformation. The conservation equations of the PEMFC mathematical model include heat/mass transfer, momentum, and charge transport. The conservation equations of the simplified model for mass, momentum, temperature, species, and charges are written as eqs 4–8.

2.4. 4
2.4. 5
2.4. 6
2.4. 7
2.4. 8

where ε is the porosity of different components, which equals zero for BP and unit for channel; u represents the intrinsic velocity (m·s–1), which is zero in BP; ρ is the fluid density (kg·m–3); μ is the dynamic viscosity of fluid (kg·m–1·s–1); p is fluid pressure (Pa); cp is the specific heat capacity of the fluid (J·kg–1·K–1); λ stands for thermal conductivity (W·m–1·K–1); ci is the species concentration; Deffisi is the diffusion coefficient of species i (m2·s–1); γ represents the electric conductivity (S·cm–1); φ is the potential (V), subscript j = s or m represents the solid phase and membrane phase, respectively; superscript eff indicates effective value; and S means source terms for mass, energy, momentum, species, and charges.

Sm is the mass source term, which indicates the total mass change for the electrochemical reaction as eq 9.

2.4. 9

The source term Su of the momentum equation is zero in other components except in GDL and CL for the interaction force between fluid and solid base in the porous media as eq 10.

2.4. 10

where K is the permeability (m2) and β2 represents the inertial resistance factor. The velocity in GDL and CL is normally low; the second-term inertia force is much lower than the first-term viscous force.

Considering the ohm heat of current for the membrane with high resistance, reversible, and irreversible reaction heat for CL, the source term SQ for the energy conservation equation can be described as eq 11.

2.4. 11

where I is the current density (A·cm–2), Acv represents the specific surface area (m–1), ΔS stands for the entropy change of electrochemical reaction, η means the activation over potential (V), and F is the Faraday constant (C·mol–1).

The gas species source terms Si are not zero in CL for the electrochemical reaction, which can be described as following via Faraday’s law as eq 12.

2.4. 12

where M represents the molecular weight and R stands for electrochemical reaction rate. The electrochemical reaction rate can be defined by the Butler–Volmer equation (13).31

2.4. 13

where d stands for a or c for anode or cathode, respectively; irefj is the reference exchange current density (A·cm–2), Crefj is the reference species concentration, and α represents the charge transfer coefficient (W·m2·K–1).

According to the electric neutrality, the source terms for electric potential and membrane potential in anode and cathode CL are equal with opposite symbols as in eqs 14 and 15

2.4. 14
2.4. 15

2.5. Numerical Procedures and Boundary Conditions

The stress–strain model for the three-dimensional GDL geometry is solved using the STATIC STRUCTURE module of commercial software ANSYS. Then, the geometry with GDL deformation is obtained and rebuilt by SolidWorks and imported into ANSYS to remesh, providing the computation domain for the CFD simulation of fuel cells. The characteristic parameters, such as porosity and permeability of GDL after deformation, are computed by empirical formula, and the obtained data are imported into FLUENT through UDFs. The continuity and momentum equations as well as the heat transfer and charge transfer equations of the PEMFC model are solved by FLUENT. The current density is obtained for a given voltage from 1.0 to 0.3 V with steps of 0.1 V. Some of the physical parameters and basic operation case conditions used in the simulation are listed in Tables 1, 2, and 3.

Table 2. Mechanical Parameters of the PEMFC Components.

parameters PEM GDL CL BP
density (kg·m–3) 1980 1000 1000 1000
Young’s modulus (MPa) 23229 6.338 249 13 000
poisson ratio 0.253 0.25638 0.3 0.26
Open in a new tab

Table 3. Physical Properties of the PEMFC Components.

parameter PEM GDL CL BP
primary electrical conductivity (S·m–1) 9.825 300 300 20 000
primary porosity 0.4 0.78 0.3 0
thermal conductivity (W·m–1·K–1) 0.95 1 1 20
specific heat capacity (J·kg–1·K–1) 83338 56830 3300 158038
thermal expansion coefficient (K) 1.23 × 10–4 2.5 × 10–6 3.7 × 10–5 0.9 × 10–6
Open in a new tab

For solid-mechanics calculations, the assembly load is applied to BP and the other outer contour surfaces are set as frictionless supports. Therefore, the cell can only generate displacement in the Y-direction, while those in the X- and Z-directions are fixed to zero.

3. Results and Discussion

3.1. Grid Reliability Verification

The hexahedral grid is used to divide the single-channel model, and the reliability of grid numbers is studied. The obtained current densities at 0.4 V under assembly pressure of 1.0, 2.0, and 2.5 MPa are plotted in Figure 2, with the grid numbers from 10 080 to 308 000. The current densities increase with the grid number. After the grid number reaches 200 736, the variation of current density with the increasing grid number is below 1%. Therefore, 200 736 grids are used in this study, considering the balance between calculation time and accuracy.

Figure 2.

Figure 2

Open in a new tab

Numerical experiments on grid number independence.

3.2. Experimental Verification

In order to verify the reliability of the PEMFC model and the calculation results, the polarization curve under 1.5 MPa is compared with the experimental results from the reference,32 as shown in Figure 3. It can be clearly seen that the numerical calculation results agree well with the experimental data, which proves the reliability of the numerical simulation.

Figure 3.

Figure 3

Open in a new tab

Comparison between the simulated and experimental performance

3.3. Effects of Assembly Pressures of Deformation and Characteristic Parameters of GDL

Based on the assumption that the channels and other components do not deform after compression, the structure deformation only occurs in GDL, as shown in Figure 4. The channel–rib flow field causes the nonuniform deformation of GDL. In the direction of the X-axis, the thickness of GDL is different in the area under the channel and rib, especially for the large assembly pressures. Then, the physical parameters such as permeability, porosity, and conductivity of deformed GDL can be calculated because they are the functions of the thickness.

Figure 4.

Figure 4

Open in a new tab

Deformation and physical parameters of GDL under different assembly pressures: (a) compression deformation of GDL, (b) porosity distribution after deformation, (c) permeability distribution after deformation, and (d) contact resistance and conductivity under different assembly pressures.

The GDL deformation for various assembly pressures from 0 to 2.5 MPa is shown in Figure 4a. Comparatively, the deformation in the area under the rib is larger than that under the channel, while the deformation in the middle of the channel region is very small and neglected in this work. The thickness of GDL under the ribs decreases from primary 0.28–0.09 mm as the assembly pressure increases from 0 to 2.5 MPa. The compression ratio (CR) is defined as eq 16.33

3.3. 16

where δprimary indicates the primary GDL thickness and δ indicates the thickness after compression. It can be seen from the figure that the maximum CR reaches 68% under the assembly pressure of 2.5 MPa.

The porosity and permeability of GDL also correspondingly vary with compression deformation.31 The porosity, ε, in Figure 4b is obtained by34

3.3. 17

where εprimary indicates the primary porosity, which is equal to 0.78 in this paper. The porosity under the rib is lower than that under the channel and decreases with the increase of assembly pressure. Meanwhile, a larger difference in porosity variation is observed under a higher assembly pressure. The difference in the porosity value varies about 4% as assembly pressure increases from 0 to 0.5 MPa, while the value reaches 17% as assembly pressure increases from 2.0 to 2.5 MPa.

The permeability, K, in Figure 4c is obtained by35

3.3. 18

where df is the diameter of carbon fiber and is equal to 9.75 μm. With the increase of pressure, the permeability decreases continuously. The permeability value under the rib greatly changes from 4.19 × 10–5 to 3.75 × 10–7 mm2 when the assembly pressure increases from 0 to 2.5 MPa.

The changes in contact electric resistance and conductivity under different pressures are plotted in Figure 4d. The contact electric resistance, Rcontact, is obtained.36

3.3. 19

where Acontact is the contact area of BP and GDL and Passembly and Aassembly are the assembly pressure and the assembly area, respectively. The change in Acontact is ignored in this work, which is equal to 0.5 cm2. It can be seen that the contact resistance reduces as assembly pressure increases from 0 to 1.0 MPa. Then, the slope of the curve becomes small, with an electric resistance value of 2.922 mΩ·cm2 at 2.5 MPa. The obtained contact electric resistance values are compared with the experimental results from Mishra et al.36 It is shown that the two curves fit very well, indicating the high reliability of our simulation results.

The effective conductivity γeff of GDL in Figure 4d is obtained by37

3.3. 20

where γprimary indicates the primary conductivity, which is equal to 300 S·m–1.

The conductivity increases from 516 to 1509 S·m–1 with an increase in assembly pressure. Different from the contact electric resistance curve, the slope of the conductivity curve increases with the value of assembly pressure.

The obtained porosity, permeability, and conductivity of GDL are imported into FLUENT by UDFs for CFD calculations for the fuel cell performance discussion.

3.4. Distribution of Water Concentration

Figure 5 shows the distribution of water concentration in the cathode GDL at the (a) XZ-plane and (b) XY-plane under different assembly pressures. It is observed that the molar concentration of water in the region below the rib is higher than that below the channel for each assembly pressure. This is because the water produced in the cathode CL is more easily carried away by the airflow in the channel. In addition, the lower porosity and permeability of GDL under the ribs after compression makes the removal of water away from the rib more difficult. Thus, the GDL region under the rib is more likely to be flooded.38 These results are in good agreement with previous experimental results.39−41 The average molar concentration of water in the cathode GDL increases from 7.79 to 9.63 mol·m–3 as the assembly pressure increases from 0 to 2.5 MPa.

Figure 5.

Figure 5

Open in a new tab

Distribution of molar concentration of water in the middle of cathode GDL at 0.5 V, (a) XZ section, and (b) XY section.

3.5. Distribution of Oxygen and Hydrogen

Figure 6 shows the distribution of oxygen and hydrogen in the GDL of the cathode and anode, respectively. It can be seen that the gas concentration decreases as the assembly pressure increases; the average concentrations of oxygen and hydrogen decreased from 4.21 to 2.05 mol·m–3 and from 30.91 to 28.74 mol·m–3, respectively; the concentrations of reactants oxygen and hydrogen in the region under the channel are higher than that under the rib. The reason is that the mass transfer of compressed GDL under the rib is greatly reduced compared to that under the channel. At the same time, the volume of the channel decreases as GDL intrudes into the channel, the velocity of the gas flow in the flow passage increases, and the convection mass transfer is enhanced.42

Figure 6.

Figure 6

Open in a new tab

Oxygen distribution at (a) XZ-plane and (b) XY-plane, and hydrogen distribution at (c) XZ-plane and (d) XY-plane in the GDL under 0.5 V.

3.6. Distribution of Temperature

Temperature has a significant effect on the PEMFC performance, especially on the proton electric resistance, gas distribution, and water content in the membrane.43Figure 7 illustrates the distribution of temperature at the XZ-plane inside the PEM under 0–2.5 MPa. The temperature of the membrane below the channel is higher than that under the rib for better mass transfer and faster reaction. With the increase of assembly pressure, the maximum temperature decreases from 355.15 to 353.48 K. This is because a higher electrical conductivity of GDL under higher assembly pressure subsequently reduces the contact electrical resistance for the neighboring components, thereby reducing the Ohmic heat.44

Figure 7.

Figure 7

Open in a new tab

Temperature distribution in the middle of the proton-exchange membrane under 0.5 V.

3.7. Distribution of Mechanical Stress and Thermal Stress

Figure 8 shows the distribution of mechanical stress in MEA. It indicates that the maximum stress appears in the membrane during the assembling process. The maximum value increases from 1.372 to 6.859 MPa as the assembly pressure increases from 0.5 to 2.5 MPa. The mechanical stress under the rib of PEM is obviously greater than that under the channel for the compression force coming from the rib vertically. For the GDL, the higher mechanical stress exists at the junction point of the channel and the rib because a sudden change of the assembly force leads to stress concentration.

Figure 8.

Figure 8

Open in a new tab

Mechanical stress distribution of MEA under different assembly pressures.

Thermal stress is mainly generated during the operation of cells, and it affects the durability of PEMFC.30 Thus, it is very important to study the magnitude and distribution of thermal stress in cells under different pressures. Figure 9 shows the thermal stress distribution of MEA in the PEMFC, which is obtained by45

3.7. 21

where E indicates Young’s modulus (Pa), T0 indicates the operating temperature (K), T indicates the actual temperature (K), and α indicates the thermal expansion coefficient. It can be seen that the thermal stress appears mainly in PEM under the middle part of the channel region, while the influence is smaller in CL and GDL. This is because the thermal expansion coefficient of PEM is larger than those of CL and GDL.30 The maximum value of the thermal stress in PEM decreases from 60912.6 to 14641.3 Pa with the increase of the assembly pressure. A more uniform distribution of thermal stress is observed under a higher assembly pressure. It is consistent with the temperature distribution as depicted in Figure 9. Higher stress values, both mechanical and thermal stress, were found in the membrane. While the maximum mechanical stress is 6.8599 MPa, obtained at 25 MPa, the maximum thermal value is only 61523.6 Pa at 0.5 MPa.

Figure 9.

Figure 9

Open in a new tab

Thermal stress distribution of MEA under 0.5 V.

3.8. Performance of PEMFC

Figure 10 plots the performance of PEMFC under different assembly pressures. Results show that the current density at 0.3 V increases from 0 to 1.5 MPa and then decreases from 20 to 25 MPa. The highest values, 2.397 and 2.271 A·cm–2, are obtained at 1.5 and 2.0 MPa. The performance increases with the pressure increase and then decreases as the assembly continuously increases. There is an obvious mass transfer limit around voltage 0.5 V for the high assembly pressure due to the big compression effect. On the one hand, lower contact electric resistance and higher conductivity of GDL obtained under high pressure can reduce the internal electric resistance and contact electric resistance. On the other hand, mass transfer reduces the porosity and permeability of GDL with decreasing assembly pressure. Therefore, the optimized cell performance at 1.5–2.0 MPa is a result of the balance of both the effects.

Figure 10.

Figure 10

Open in a new tab

Performance of PEMFC under different assembly pressures.

3.9. Rib–Channel Ratio Effect

The rib–channel ratios affect the transportation phenomenon and cell performance of the PEMFC by changing the passages of fluid flow and charge transfer. The rib width is fixed at 0.5 mm, and the channel width varies from 0.25 to 1 mm to obtain rib–channel ratios of 2:1, 1:1, 2:3, and 1:2 and study the effect of rib–channel ratios on cell performance.

Figure 11 shows the current density distribution at the z = 25 mm plane for different rib–channel ratios with an assembly pressure of 1.5 MPa and an operating voltage of 0.5 V. It can be observed from Figure 11a that under different assembly pressures, the distribution of current density is generally similar. The highest current density is found at the corners of the current collector, while the lowest current density is observed in the middle position at the top and bottom of the current collector. The reason for this phenomenon is that the electrons in the GDL below the channel move to the closest point of the current collector and lots of electrons aggregate over there; channels do not conduct electric current, and the current collector at the top of the channel is larger than that in other places. Furthermore, the maximum current density increases with increasing assembly pressure, which means the higher aggregation effect of electrons. In Figure 11b, it is found that for the 2:1 and 2:3 models, the maximum current density appears below the junction position of the ribs and channels and close to the PEM. For the 1:1 and 1:2 models, the maximum current density occurs near the connection position of GDL, rib, and channel.

Figure 11.

Figure 11

Open in a new tab

Current density distribution at z = 25 mm plane for different rib–channel ratios with assembly pressure of 1.5 MPa and operating voltage of 0.5 V (a) in BP and (b) GDL.

The effects of rib–channel ratios (2:1, 1:1, 2:3, 1:2) on the cell performances are shown in Figure 12. In Figure 12a, it is observed that the flow field with a 1:1 rib–channel ratio has the highest current density at 0.3 V under 1.5 MPa. In addition, a higher current density at 0.3 V is obtained by the flow field with a ratio of 1:1 under 20 MPa, as shown in Figure 12b. While an obvious concentration polarization is found for the flow field with a ratio of 2:3 and 1:2, the current density shows a downward trend at a high current density. Therefore, the optimum rib–channel ratio is 1:1 under both the assembly pressures of 1.5 and 2.0 MPa.

Figure 12.

Figure 12

Open in a new tab

Polarization performance of PEMFC with different rib–channel ratios at (a) 1.5 and (b) 2.0 MPa.

4. Conclusions

The effect of assembly pressure on GDL deformation is investigated using FEM. Based on the deformed GDL and its variable physical properties, the 3D PEMFC model is developed and computed to elucidate the transportation phenomena and cell performance by commercial software Ansys Fluent with UFDs. The following are some main research results:

  • (1)

    The porosity and permeability of GDL decrease after compression, in particular in the region under the rib, which hinders the mass transfer performances, including the accumulation of more water and a lower gas concentration.

  • (2)

    GDL compression results in higher conductivity and lower contact electric resistance between GDL and BP, which reduces the ohmic electric resistance and decreases the internal temperature of the cell.

  • (3)

    The mechanical stress increases with the increase of assembly pressure, and the thermal stress decreases with the increase of assembly pressure.

  • (4)

    The maximum current density at 0.3 V is obtained under 1.5 and 2.0 MPa due to the trade-off between mass transfer resistance and ohmic resistance.

  • (5)

    Under the assembly pressures of 1.5 and 2.0 MPa, the flow field with the rib–channel ratio of 1:1 has the highest current density at 0.3 V.

Acknowledgments

This work was supported by Zhejiang Provincial Outstanding Youth Fund (R1100065) and Hangzhou Lingfei Technology Co., Ltd.

The authors declare no competing financial interest.

References

  1. Lu J.; Xia Y.; Hu Y.; Wang Z.; Lei H.; Hu G. Study on transport phenomena and performance of proton exchange membrane fuel cell with radial flow fields. Eng. Appl. Comput. Fluid Mech. 2023, 17 (1), 2156925 10.1080/19942060.2022.2156925. [DOI] [Google Scholar]
  2. Jiao K.; Xuan J.; Du Q.; Bao Z.; Xie B.; Wang B.; Zhao Y.; Fan L.; Wang H.; Hou Z.; Huo S.; Nigel P.; Yin Y.; Michael D. Designing the next generation of proton-exchange membrane fuel cells. Nature 2021, 595 (7867), 361–369. 10.1038/s41586-021-03482-7. [DOI] [PubMed] [Google Scholar]
  3. Niu Z.; Jiao K.; Wang Y.; Du Q.; Yin Y. Numerical simulation of two-phase cross flow in the gas diffusion layer microstructure of proton exchange membrane fuel cells. Int. J. Energy Res. 2018, 42 (2), 802–816. 10.1002/er.3867. [DOI] [Google Scholar]
  4. Bao N.; Zhou Y.; Zhou Y.; Yin Y.; Du Q.; Chen J. Effect of gas diffusion layer deformation on liquid water transport in proton exchange membrane fuel cell. Eng. Appl. Comput. Fluid Mech. 2014, 8 (1), 26–43. 10.1080/19942060.2014.11015495. [DOI] [Google Scholar]
  5. Bernardi D. M.; Verbrugge M. W. A mathematical model of the solid-polymer-electrolyte fuel cell. J. Electrochem. Soc. 1992, 139, 2477–2491. 10.1149/1.2221251. [DOI] [Google Scholar]
  6. Chen H.; Guo H.; Ye F.; Ma C. F. A numerical study of baffle height and location effects on mass transfer of proton exchange membrane fuel cells with orientated-type flow channels. Int. J. Hydrogen Energy 2021, 46 (1–2), 7528–7545. 10.1016/j.ijhydene.2020.11.226. [DOI] [Google Scholar]
  7. Yu Y.; Zhang Z. G.; He L. Y.; Yang X. X.; Wan X. B.; Sui P. C.; Pan M. Effects of distribution zone design on flow uniformity and pressure drop in PEMFC. J. Electrochem. Soc. 2021, 168 (9), 094505 10.1149/1945-7111/ac2656. [DOI] [Google Scholar]
  8. Zhang Z. Q.; Liu W.; Wang Y. L. Three dimensional two-phase and non-isothermal numerical simulation of multi-channels PEMFC. Int. J. Hydrogen Energy 2019, 44, 379–388. 10.1016/j.ijhydene.2018.05.149. [DOI] [Google Scholar]
  9. Ding Y. J.; Xu L. F.; Zheng W. B.; Hu Z. Y.; Shao Y. B.; Li J. Q.; Ouyang M. G. Characterizing the two-phase flow effect in gas channel of proton exchange membrane fuel cell with dimensionless number. Int. J. Hydrogen Energy 2023, 48 (13), 5250–5265. 10.1016/j.ijhydene.2022.09.288. [DOI] [Google Scholar]
  10. Franz T.; Papakonstantinou G.; Sundmacher K. Transient hydrogen crossover in dynamically operated PEM water electrolysis cells - A model-based analysis. J. Power Sources 2023, 559, 232582 10.1016/j.jpowsour.2022.232582. [DOI] [Google Scholar]
  11. Huang Z. Y.; Shen J.; Chan S. H.; Tu Z. K. Transient response of performance in a proton exchange membrane fuel cell under dynamic loading. Energy Convers. Manage. 2020, 226, 113492 10.1016/j.enconman.2020.113492. [DOI] [Google Scholar]
  12. Liu H.; Chen J.; Hissel D.; Hou M.; Shao Z. G. A multi-scale hybrid degradation index for proton exchange membrane fuel cells. J. Power Sources 2019, 437, 226916 10.1016/j.jpowsour.2019.226916. [DOI] [Google Scholar]
  13. Hou Y. Z.; Li X.; Du Q.; Jiao K.; Zamel N. Pore-scale investigation of the effect of micro-porous layer on water transport in proton exchange membrane fuel cell. J. Electrochem. Soc. 2020, 167 (14), 144504 10.1149/1945-7111/abc30a. [DOI] [Google Scholar]
  14. Huang T. M.; Huang J.; Feng M. C.; Yi D. X.; Chen X.; Ou C. J.; Zhang J.; Zhang Y.; Wan Z. M. Optimization of the thickness of catalytic layer for HT-PEMFCs based on genetic algorithm. Energy Rep. 2022, 8, 12905–12915. 10.1016/j.egyr.2022.09.111. [DOI] [Google Scholar]
  15. KP V. B.; Varghese G.; Joseph T. V.; Chippar P. Optimization of graded catalyst layer to enhance uniformity of current density and performance of high temperature-polymer electrolyte membrane fuel cell. Int. J. Hydrogen Energy 2022, 47 (6), 4018–4032. 10.1016/j.ijhydene.2021.11.006. [DOI] [Google Scholar]
  16. Zhu L. J.; Zhnag H.; Xiao L. S.; Bazylak A.; Gao X.; Sui P. C. Pore-scale modeling of gas diffusion layers: Effects of compression on transport properties. J. Power Sources 2021, 496, 229822 10.1016/j.jpowsour.2021.229822. [DOI] [Google Scholar]
  17. Bao Y. Y.; Wang Z. Z.; Gan Y. X. Numerical investigation of compressive stress and wettability effects on fluid transport in polymer electrolyte membrane fuel cell porous layers. Energy Technol. 2022, 10 (12), 2201028 10.1002/ente.202201028. [DOI] [Google Scholar]
  18. Song K.; Wang Y. M.; Ding Y. H.; Xu H. J.; Mueller-Welt P.; Stuermlinger T.; Bause K.; Ehrmann C.; Weinmann H. W.; Schaefer J.; Fleischer J.; Zhu K.; Weihard F.; Trostmann M.; Schwartze M.; Albers A. Assembly techniques for proton exchange membrane fuel cell stack: A literature review. Renewable Sustainable Energy Rev. 2022, 153, 111777 10.1016/j.rser.2021.111777. [DOI] [Google Scholar]
  19. Zhou X.; Niu Z. A.; Li Y. N.; Sun X. Y.; Du Q.; Xuan J.; Jiao K. Investigation of two-phase flow in the compressed gas diffusion layer microstructures. Int. J. Hydrogen Energy 2019, 44 (48), 26498–26516. 10.1016/j.ijhydene.2019.08.108. [DOI] [Google Scholar]
  20. Zhang Z.; He P.; Dai Y. J.; Jin P. H.; Tao W. Q. Study of the mechanical behavior of paper-type GDL in PEMFC based on microstructure morphology. Int. J. Hydrogen Energy 2020, 45 (53), 29379–29394. 10.1016/j.ijhydene.2020.07.240. [DOI] [Google Scholar]
  21. Chen Y. Q.; Zhao J. H.; Jin C. H.; Ke Y. C.; Li D. C.; Wang Z. X. Effect of clamping compression on the mechanical performance of a carbon paper gas diffusion layer in polymer electrolyte membrane fuel cells. Membranes 2022, 12 (7), 645. 10.3390/membranes12070645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Chi P. H.; Chan S. H.; Weng F. B.; Su A. Y.; Sui P. C.; Djilali N. On the effects of non-uniform property distribution due to compression in the gas diffusion layer of a PEMFC. Int. J. Hydrogen Energy 2010, 35 (7), 2936–2948. 10.1016/j.ijhydene.2009.05.066. [DOI] [Google Scholar]
  23. Su Z. Y.; Liu C. T.; Chang H. P.; Li C. H.; Huang K. J.; Sui P. C. A numerical investigation of the effects of compression force on PEM fuel cell performance. J. Power Sources 2008, 183 (1), 182–192. 10.1016/j.jpowsour.2008.04.060. [DOI] [Google Scholar]
  24. Chippar P.; Kyeongmin O.; Kang K.; Ju H. A numerical investigation of the effects of GDL compression and intrusion in polymer electrolyte fuel cells (PEFCs). Int. J. Hydrogen Energy 2012, 37 (7), 6326–6338. 10.1016/j.ijhydene.2011.04.154. [DOI] [Google Scholar]
  25. Shrivastava N. K.; Chatterjee A.; Harris T. A. L. Effect of cell compression on the performance of a non–hot-pressed MEA for PEMFC. Int. J. Energy Res. 2020, 44 (1), 370–387. 10.1002/er.4933. [DOI] [Google Scholar]
  26. Movahedi M.; Ramiar A.; Ranjber A. A. 3D numerical investigation of clamping pressure effect on the performance of Proton Exchange Membrane fuel cell with interdigitated flow field. Energy 2018, 142, 617–632. 10.1016/j.energy.2017.10.020. [DOI] [Google Scholar]
  27. Kang H.; Zhan Z. G.; Yang X. F.; Zhang Z. B.; Shi J. R.; Jiang P. X.; Pan M. Numerical Study on the Stress Concentration Phenomenon in the Membranes of PEMFCs in an Assembled State. Int. J. Hydrogen Energy 2020, 146 (5), 04020053 10.1061/(ASCE)EY.1943-7897.0000696. [DOI] [Google Scholar]
  28. Li W.; Yang W.; Zhang W.; Qu Z.; He Y. Three-dimensional modeling of a PEMFC with serpentine flow field incorporating the impacts of electrode inhomogeneous compression deformation. Int. J. Hydrogen Energy 2019, 44 (39), 22194–22209. 10.1016/j.ijhydene.2019.06.187. [DOI] [Google Scholar]
  29. Hu G.; Ji C.; Xia Y.; Suo Y.; Wu X.; Zhang Z. Assembly mechanics and its effect on performance of proton exchange membrane fuel cell. Int. J. Electrochem. Sci. 2019, 14 (2), 1358–1371. 10.20964/2019.02.51. [DOI] [Google Scholar]
  30. Oh K.; Chippar P.; Ju H. Numerical study of thermal stresses in high temperature proton exchange membrane fuel cell (HT-PEMFC). Int. J. Hydrogen Energy 2014, 39 (6), 2785–2794. 10.1016/j.ijhydene.2013.01.201. [DOI] [Google Scholar]
  31. Xia Y.; Hu Y.; Hu G.; Lei H.; Lu J.; Wang Z.; Wang Q. Numerical analysis on the effects of manifold design on flow uniformity in a large proton exchange membrane fuel cell stack. Int. J. Hydrogen Energy 2023, 48, 5643–5655. 10.1016/j.ijhydene.2022.11.131. [DOI] [Google Scholar]
  32. Corda G.; Fontanesi S.; d’Adamo A. Methodology for PEMFC CFD Simulation Including the Effect of Porous Parts Compression. Int. J. Hydrogen Energy 2022, 47 (32), 14658–14673. 10.1016/j.ijhydene.2022.02.201. [DOI] [Google Scholar]
  33. Zhang H.; Xiao L.; Chuang P.; Djilali N.; Sui P. Coupled stress-strain and transport in proton exchange membrane fuel cell with metallic bipolar plates. Appl. Energy 2019, 251, 113316 10.1016/j.apenergy.2019.113316. [DOI] [Google Scholar]
  34. Jiao K.; Park J.; Li X. Experimental investigations on liquid water removal from the gas diffusion layer by reactant flow in a PEM fuel cell. Appl. Energy 2010, 87 (9), 2770–2777. 10.1016/j.apenergy.2009.04.041. [DOI] [Google Scholar]
  35. Tomadakis M. M.; Robertson T. J. Viscous permeability of random fiber structures: comparison of electrical and diffusional estimates with experimental and analytical results. J. Compos. Mater. 2005, 39 (2), 163–188. 10.1177/0021998305046438. [DOI] [Google Scholar]
  36. Mishra V.; Yang F.; Pitchumani R. Measurement and prediction of electrical contact resistance between gas diffusion layers and bipolar plate for applications to PEM fuel cells. J. Fuel Cell Sci. Technol. 2004, 1 (1), 2–9. 10.1115/1.1782917. [DOI] [Google Scholar]
  37. Das P. K.; Li X.; Liu Z. S. Effective transport coefficients in PEM fuel cell catalyst and gas diffusion layers: Beyond Bruggeman approximation. Appl. Energy 2010, 87 (9), 2785–2796. 10.1016/j.apenergy.2009.05.006. [DOI] [Google Scholar]
  38. Zhou Y.; Jiao K.; Du Q.; Yin Y.; Li X. Gas diffusion layer deformation and its effect on the transport characteristics and performance of proton exchange membrane fuel cell. Int. J. Hydrogen Energy 2013, 38 (29), 12891–12903. 10.1016/j.ijhydene.2013.05.150. [DOI] [Google Scholar]
  39. Iranzo A.; Boillat P. Liquid water distribution patterns featuring back-diffusion transport in a PEM fuel cell with neutron imaging. Int. J. Hydrogen Energy 2014, 39 (30), 17240–17245. 10.1016/j.ijhydene.2014.08.042. [DOI] [Google Scholar]
  40. Wang Y.; Chen K. S. Effect of spatially-varying GDL properties and land compression on water distribution in PEM fuel cells. J. Electrochem. Soc. 2011, 158 (11), B1292. 10.1149/2.015111jes. [DOI] [Google Scholar]
  41. Owejan J.; Trabold T.; Jacobson D.; Baker D.; Hussey D.; Arif M. In situ investigation of water transport in an operating PEM fuel cell using neutron radiography: Part 2 – Transient water accumulation in an interdigitated cathode flow field. Int. J. Heat Mass Transfer 2006, 49 (25), 4712–4720. 10.1016/j.ijheatmasstransfer.2006.07.004. [DOI] [Google Scholar]
  42. Taymaz I.; Benli M. Numerical study of assembly pressure effect on the performance of proton exchange membrane fuel cell. Energy 2010, 35 (5), 2134–2140. 10.1016/j.energy.2010.01.032. [DOI] [Google Scholar]
  43. Toghyani S.; Nafchi F.; Afshari E.; Hasanpour K.; Baniasadi E.; Atyabi S. Thermal and electrochemical performance analysis of a proton exchange membrane fuel cell under assembly pressure on gas diffusion layer. Int. J. Hydrogen Energy 2018, 43 (9), 4534–4545. 10.1016/j.ijhydene.2018.01.068. [DOI] [Google Scholar]
  44. Kulkarni N.; Cho J.; Jervis R.; Roberts E.; Francesco I.; Kok M.; Shearing P.; Brett D. The effect of non-uniform compression on the performance of polymer electrolyte fuel cells. J. Power Sources 2022, 521, 230973 10.1016/j.jpowsour.2021.230973. [DOI] [Google Scholar]
  45. Tang Y.; Santare M.; Karlsson A.; et al. Stresses in proton exchange membranes due to hygro-thermal loading. J. Fuel Cell Sci. Technol. 2006, 3 (2), 119–124. 10.1115/1.2173666. [DOI] [Google Scholar]

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES