Experimental and mechanistic studies of gradient pore polymer electrolyte fuel cells

Min Li; Chongcai Xu; Tianya Li; Guangyi Lin

doi:10.1515/htmp-2025-0090

Article Open Access

Experimental and mechanistic studies of gradient pore polymer electrolyte fuel cells

Min Li , Chongcai Xu , Tianya Li and Guangyi Lin

Published/Copyright: September 8, 2025

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information

From the journal High Temperature Materials and Processes Volume 44 Issue 1

Abstract

This article uses carbon black as a conductive material, polytetrafluoroethylene as a hydrophobic material, and calcium carbonate (CaCO₃) as a pore-forming agent. Meanwhile, a gas diffusion layer (GDL) with a gradient pore size was prepared using the spray coating method. The results show that a three-layer gradient pore size GDL has good water vapor transmission ability. When humidity is 40%, the current density and power density of gradient pore size GDL are 127.9 mA·cm⁻² and 63.31 mW·cm⁻² higher than commercial GDL29BC, respectively. When humidity is 60%, the current density and power density of gradient pore size GDL are 250.3 mA·cm⁻² and 127.3 mW·cm⁻² higher than commercial GDL29BC, respectively. When humidity is 100%, the current density and power density of the gradient pore size GDL are 255.69 mA·cm⁻² and 127 mW·cm⁻² higher than commercial GDL29BC, respectively.

Keywords: proton exchange membrane fuel cell; gas diffusion layer; graded pore size; pore-forming agent

Nomenclature

P C: capillary pressure
P l: water pressure
P g: gas pressure
r c: the critical radius of liquid water entering the pores
σ: surface tension of liquid water
θ: water contact angle
D i: The diffusion coefficient in porous media
ε τ: porosity
s: the saturation of liquid water
D i 0: The diffusion coefficient of i at temperature
T 0 and pressure p 0 . γ: stress factor
q 1: flow rate of liquid water on the section

1 Introduction

Proton exchange membrane fuel cells (PEMFCs) use oxygen as oxidant, hydrogen as fuel. With the cooperation of the proton exchange membrane, the catalyst layer (CL), and the gas diffusion layer (GDL), they can directly generate electricity [1,2,3,4,5]. Lower operating temperature, high efficiency, and pollution-free characteristics have received widespread attention [6,7,8,9,10]. GDL is one of the key components in a fuel cell, mainly playing a role in diffusion and transport of hydrogen, oxygen, and water [11,12,13,14].

Although significant progress has been made in the study of GDL [7,8,15,16], there are still some shortcomings and areas for improvement [17–20]. First, although pore size distribution has a significant impact on water transport and thermal conductivity of GDL [21–23], current research is mostly focused on performance optimization under specific conditions [24–26], and the adaptability research for complex and changing working environments(different humidity) is not sufficient [27–29]. Second, precise control and uniform distribution of pore size remain a technical challenge, especially in the preparation of GDL with a gradient pore structure [30,31]. Gradient pore size GDL often focuses on changes in pore size and ignores the effects of capillary pressure and capillary pressure difference on water transport. In addition, gradient pore size studies often overlook the synergistic effects with other parameters such as thickness, hydrophilicity, and hydrophobicity, which collectively affect the overall performance of GDL [32–36]. Therefore, in future research, it is necessary to comprehensively consider the interactions between various parameters and their performance under different working conditions to promote the further development of GDL technology.

Based on the above research, we chose multi-layer gradient pore size GDL as the research focus and analyzed the water–gas transport mechanism of gradient pore size GDL. Through study, it can be found that capillary pressure is the main factor affecting water transport, and the larger the capillary pressure difference, the higher the water transport efficiency. A gradual decrease in pore size from the support layer (carbon paper) to the microporous layer (MPL) direction can achieve better fuel cell performance.

2 Experimental materials and equipment

Main materials and testing instruments used in sample preparation are shown in Tables 1 and 2.

Table 1

Experimental materials

Material	Company	Model
Carbon paper	Japan Toray Group	TGP-H-060
Anhydrous ethanol	Beijing Tongguang Fine Chemical Co., Ltd.	Purity ≥99.7%
Carbon black	Cabot Corporation	Vulcan XC-72
Polytetrafluoroethylene (PTFE)	China Biochemical Technology Co., Ltd.	60 wt% by mass

Table 2

Experimental equipment

Equipment	Company	Model
Scanning electron microscope	Japan, Hitachi	S-4700
Tube furnace	China Nanjing Boyuntong Instrument Technology Co., Ltd.	TL1200
Water contact angle measuring instrument	DATA Physics Instrument Company	TBU 90E
Four-probe conductivity tester	Guangzhou Four-probe Technology Co.	RTS-4
Spray gun	Taizhou Huangyan Ronghao Tools Co., Ltd.	K3
Roughness tester	Olympus Investment Co., Ltd.	OLS4000
PEMFC testing equipment		model 850e

3 Preparation

Sample 1: Add 7 g carbon black and 100 mL anhydrous ethanol to a beaker, stir and sonicate for 30 min, and repeat the above steps until carbon black is evenly dispersed in anhydrous ethanol. In this experiment, stirring and ultrasonication were performed four times each to obtain a carbon black solution with a good dispersion effect. Then, add 5 g PTFE solution to the evenly mixed carbon black solution and stir. The stirring time of 30 min can obtain a good dispersion effect. Spray the above solution onto carbon paper with a thickness of 60 μm. Place the sample in the tube furnace for sintering. Heat up at 5°C·min⁻¹, hold for 30 min after reaching 250°C, then heat up to 350°C, and hold for 30 min. Perform physical and chemical performance tests after the sample has cooled down.

Sample 2: Add 7 g carbon black and 100 mL anhydrous ethanol into a beaker and stir and sonicate for 30 min. Repeat the above steps until carbon black is completely dispersed in anhydrous ethanol. Stirring and sonication for four times can achieve a good dispersion effect. Add 5 g PTFE solution and take it to the carbon black solution, stirring for 30 min. Prepare two sets of solutions with the same ratio according to the above steps, labeled as (1) and (2), respectively. Add 5 g calcium carbonate (CaCO₃) to (2) and stir with ultrasound for 30 min. First, spray (2) onto carbon paper, with a thickness of 30 μm, and then spray (1) onto (2), with a thickness of 30 μm. Place the sample in the tube furnace for sintering. Heat up at 5°C·min⁻¹ and hold for 30 min after reaching 250°C; then, heat up to 350°C and hold for 30 min. After cooling the sample, soak it in dilute hydrochloric acid for 2 h, then clean it with distilled water, and dry it to obtain the sample to be tested.

Sample 3: Pour 7 g carbon black and 100 mL anhydrous ethanol into a beaker, and stir and sonicate for 30 min each. Repeat this process four times to obtain a carbon black solution with a good separation effect. Add 5 g PTFE solution and stir for 30 min to obtain a uniformly dispersed mixed solution. Prepare three identical mixed solutions according to the above steps, denoted as (1), (2), and (3). Add 5 g CaCO₃ to (3) and sonicate for 30 min. Add 5 g CaCO₃ to (2) and sonicate for 60 min. The particle size of CaCO₃ decreases with increasing ultrasonic time, resulting in CaCO₃ mixed solutions with different particle sizes. Spray mixed solutions (3), (2), and (1) onto carbon paper in sequence, with a thicknesses of 20 μm each. Place the sample in the tube furnace for sintering. Heat up at 5°C·min⁻¹ and hold for 30 min after reaching 250°C; then heat up to 350°C and hold for 30 min. After cooling the sample, soak it in dilute hydrochloric acid for 2 h, then clean it with distilled water, and dry it to obtain the sample. The similarities and differences of the above three samples are shown in Table 3.

Table 3

Differences between GDLs

	Same	Difference
Sample 1	Ultrasound time, carbon black and PTFE content sintering time, the total thickness of MPL is 60 µm	As shown in Figure 3a, no CaCO₃ added, the total thickness of MPL is 60 µm
Sample 2		As shown in Figure 3b, MPL2 contains CaCO₃, and ultrasound time is 30 min, MPL1 does not contain CaCO₃, the thickness of each layer MPL is 30 µm
Sample 3		As shown in Figure 3c, MPL3 contains CaCO₃ and sonicated for 30 min, MPL2 contains CaCO₃ and sonicated for 60 min, MPL1 does not contain CaCO₃, the thickness of each layer MPL is 20 µm

Sample 1 is traditional MPL, sample 2 is gradient pore size GDL with double-layer MPL structure, and sample 3 is gradient pore size GDL with three-layer MPL structure. Adding CaCO₃ for long-term ultrasound can reduce particle size, and the longer the ultrasound time, the greater the change in particle size. Taking the three-layer gradient pore size MPL as an example, the ultrasonic time for CaCO₃ in the MPL near the carbon paper side is 30 min, the ultrasonic time for CaCO₃ in the second layer of MPL is 60 min, and the third layer is MPL without added CaCO₃. By using the above ultrasonic method, the pore size distribution in GDL treated with dilute hydrochloric acid gradually decreases along the direction from carbon paper to MPL. Figure 1 is the preparation process of the three-layer gradient pore size GDL. The preparation methods and processes of samples 1 and 2 refer to sample 3.

Figure 1

Preparation process of sample 3.

4 Results and discussion

4.1 Polarization curve and power density

Figure 2 shows polarization curves and power density test results under the humidity of 40, 60, and 100%. It can be seen that sample 3 exhibits good performance regardless of low humidity, medium humidity, or high humidity. When humidity is 40%, maximum current density and power density have increased by 127.9 mA·cm⁻² and 63.31 mW·cm⁻² compared to sample 1. When humidity is 60%, maximum current density and power density have increased by 250.3 mA·cm⁻² and 127.3 mW·cm⁻² compared to sample 1. When humidity is 100%, maximum current density and power density have increased by 255.69 mA·cm⁻² and 127 mW·cm⁻² compared to sample 1. From the test results, it can also be seen that during the process of increasing the humidity from 40 to 60%, the difference in maximum current density and power density between sample 3 and sample 1 gradually increases.

Figure 2

Polarization curve and power density test results (1: traditional GDL; 2: double layer gradient pore size GDL; 3: three layer gradient pore size GDL).

When PEMFC is running, air needs to enter fuel cell to participate in reaction, and internal water needs to be discharged to ensure the normal operation. Only by improving the efficiency of bidirectional flow of water and gas, the performance can be further improved. Transmission of air comes from diffusion, while the transmission of water comes from capillary pressure. When capillary pressure is relatively high, water transport efficiency will increase, and when capillary pressure is relatively low, the water transport efficiency will decrease. As shown in Figure 2, regardless of the humidity being 40, 60, or 100%, gradient pore size GDL has the best limit current density and power density. From test results, it can be seen that the GDL with a gradient pore size can generate a larger capillary pressure to help fuel cell discharge water.

From equation (1), it can be seen that capillary pressure is the pressure difference between water and air. During fuel cell operation, air will enter the fuel cell at a fixed flow rate to participate in the reaction. As the generated water continues to increase, the water pressure inside the fuel cell will also continue to increase. According to equation (1), it can be concluded that capillary pressure increases, which helps water to be discharged from the fuel cell through the pores. Due to the fact that water is a liquid and air is a gas, the transport of water requires a larger pore size compared to the diffusion of air during the bidirectional flow process. By designing, the pore size of GDL has changed. According to statistical results in Table 4 obtained from mercury intrusion experiment, the GDL with three-layer gradient pore size has higher content of large (50 µm∼) and medium (7 µm∼) pores, which proves that this type of GDL can not only improve water transport efficiency but also accelerate free diffusion of air.

From equation (2), it can be seen that when surface tension and hydrophobicity of GDL remain unchanged, capillary pressure and critical radius ( r c ) of liquid water entering the pore are inversely proportional. When capillary pressure is low, water can pass through large pores. When capillary pressure is high, it can transmit through large and small pores. Specifically, the reason for low capillary pressure comes from two aspects. On the one hand, at the beginning of operation, fuel cells generate relatively less water, resulting in lower liquid water pressure, and the constant flow rate of air; after calculation (equation (2)), the capillary pressure is relatively small. According to inverse proportionality between capillary pressure and critical radius, water can be transported in the larger pore size of GDL.

On the other hand, after long-term operation, due to the unreasonable design of the GDL, external air cannot smoothly enter the interior, causing air to accumulate outside the fuel cell. According to formula (1), capillary pressure will decrease. According to formula (2), water can only be transported through large pores. Long-term exposure to such an environment will make water transportation increasingly difficult, untreated water will cover the MPL and cause “flooding,” ultimately fuel cell to stop working. Based on the test results, it can be concluded that the three-layer gradient pore size GDL designed achieves continuous variation of pore size in the thickness direction. It not only maintains good cell performance at low current densities but also ensures high capillary pressure in the thickness direction at high current densities to improve water transport capacity.

P C is the capillary pressure, P l is the water pressure, P g is the gas pressure, r c is the critical radius of liquid water entering the pores, σ is the surface tension of liquid water, and θ is the water contact angle.

(1) P C = P l − P g ,

(2) r c = − 2 σ cos θ P C .

4.2 Capillary pressure

Water transport in GDL is accomplished through capillary pressure gradients, with higher capillary pressure resulting in higher water transport efficiency. Calculation of capillary pressure is based on (1) and (3). When analyzing the relationship between pore and capillary pressure, it is believed that fuel cell temperature, hydrogen oxygen flow rate, and other factors are all the same. From equation (3), it can be seen that as porosity increases, the capillary pressure shows an upward. Combined with the statistical results of pore size and porosity from mercury intrusion, it can be seen that sample 3 has the highest porosity, which is 45.7%. Meanwhile, based on the calculation results of water contact angle, GDL (sample 3) with a gradient pore size can achieve the maximum capillary pressure.

(3) P c = σ cos θ c ( K / ε ) 0.5 ( 1.417 S − 2.12 S 2 + 1.26 S 3 ) , θ C > 90 ° .

For hydrophobic GDLs, the capillary pressure is negative. K is absolute permeability, ε is the porosity, θ c is the water contact angle, and S is the saturation.

According to equation (3), capillary pressure shows that the GDL with a three-layer MPL gradient structure has the best capillary pressure. The effect of capillary pressure on water flow velocity is shown by Darcy’s law in equation (4). Thus, it can be concluded that the higher the capillary pressure, the faster the water flow speed. q ₁ is the water flow velocity

(4) q l = − ρ K k r l υ ∇ p l .

Diffusion of gas has a significant impact on fuel cell performance. Although GDL is a porous structure, the transport of internal water affects the diffusion of gas. Diffusion in porous media is described by Fick’s law (equation (5)). By equation (5), it can be concluded that the higher the water saturation (s), the greater the diffusion coefficient, which is more conducive to the transport of water and gas:

(5) D i = ε τ ( 1 − s ) b D i 0 p 0 p γ T T 0 1.5 .

D i is the diffusion coefficient in porous media, ε τ is the porosity, s is the saturation of liquid water, D i 0 is the diffusion coefficient of i at temperature T 0 and pressure p 0 , and γ is the stress factor.

The flow rate of liquid water on the cross-section is shown in equation (6). From equation (6), it can be concluded that the flow rate of liquid water on the cross-section is proportional to the porosity. Comparison of porosity test results for samples 1, 2, and 3, it can be concluded that sample 3 has the fastest flow rate, followed by 2, and the GDL without gradient treatment has the worst flow rate.

(6) q l = − σ cos θ c υ ε 2 4 ( 1 − ε ) K K S 3 ( 1.417 − 4.24 S + 3.789 S 2 ) d s d x ,

where q 1 is the flow rate of liquid water on the section.

According to the above analysis, sample 3 has the best water transmission efficiency. Free diffusion is the best way for oxygen to be transported in pores; moreover, free diffusion requires a pore size of greater than 7 μm. According to Table 4, it can be seen that sample 3 has the most pores larger than 7 μm. Combining the pore size and porosity distribution of sample 3, it can be concluded that sample 3 not only improves the water transport capacity of the fuel cell but also enhances the gas transport capacity.

Table 4

Test results of the pore size distribution interval

Pore content (µm)	>100	50–100	20–50	7–20	0–7	ε (%)
1	1.3	1.7	18.3	2.6	3.9	36.5
2	1.1	2.9	19.7	3.8	3.5	39.2
3	0.8	9.2	23.8	5.9	1.3	45.7

4.3 Capillary pressure difference

In Figure 3, a represents traditional MPL, b represents two-layer MPL, and c represents three-layer MPL processed by gradient pore. CaCO₃ in MPL3 was sonicated for 30 min, CaCO₃ in MPL2 was sonicated for 60 min, and there is no CaCO₃ in MPL1. According to ultrasound time of CaCO₃, as shown in Table 4, the shorter the ultrasound time, the larger the pore size left after treatment with dilute hydrochloric acid. Thus, the pore size in Figure 3 gradually decreases from carbon paper to the CL direction. When the cell is running, humidity, flow rate of hydrogen and air, temperature, and content of CL side catalyst are all the same, so that the water content generated at the contact between MPL and cathode CL is the same, thus obtaining P a = P b = P c . According to the calculation of formula (2), P a 1 > P b 2 > P c 3 , and δ a = δ b = δ c . We can obtain

(7) P a − P a 1 δ a < P b − P b 2 δ b < P c − P C 3 δ c .

Figure 3

Schematic diagram of the capillary pressure interface at cathode GDL ((a) traditional GDL; (b) GDL with double-layer gradient pore; and (c) GDL with three-layer gradient pore).

We define the capillary pressure per unit thickness as the capillary pressure difference. The higher the capillary pressure difference, the higher the water transmission efficiency. Thus, it can be concluded that the water transfer efficiency of sample 3 is higher than that of sample 2, while the water transfer efficiency of sample 1 is the lowest. Comparing the polarization curve and power density test results, it can be seen that sample 3 has the best performance, followed by sample 2, and sample 1 has the worst performance. The excellence of sample 3 is mainly manifested in the reasonable pore size distribution and the maximum capillary pressure difference per unit thickness.

4.4 Electrochemical impedance spectroscopy (EIS)

Figure 4 shows EIS test results. It can be concluded that GDL with a three-layer gradient pore has the lowest material transmission impedance when the humidity is 40, 60, and 100%, respectively. The reason is that the three-layer gradient pore size GDL can effectively transport water and gas. From the pore size distribution range in Table 4, it can be seen that the pore size distribution is widest in 7–20 and 20–100 μm. The more pore sizes between 7 and 20 μm, the better efficiency of oxygen transport, and the more pore sizes between 20 and 100 μm, the more favorable for water transport. Based on the analysis, GDL with a three-layer MPL structure is not only beneficial for gas diffusion but also for water transport.

Figure 4

EIS test results of three samples at different humidity levels.

According to the equivalent circuit in Figure 4, ohmic impedance plays a major role when the frequency is high, and the material transmission impedance plays a major role when the frequency is low. The test results show that the material transmission impedance of sample 3 is lowest, consistent with the polarization curve and pore size distribution.

4.5 Water permeability and residual rate

As shown in Figure 5, the permeability of sample 3 was 40.2 mL·min⁻¹, and the permeability of sample 1 was 25.6 mL·min⁻¹. Water residual rate of sample 3 is 2.2 mL·min⁻¹, and sample 1 is 5.6 mL·min⁻¹. Water permeability and residual rate represent the transfer efficiency of GDL for water and oxygen, respectively. When water permeability is relatively high, the transmission efficiency of water will increase. When the water residual rate is relatively high, it will cause some water to block the pores in GDL, thereby affecting the transmission of oxygen. In the polarization curve test, it can be concluded that the GDL with a three-layer gradient pore has the best performance under the three humidity conditions, mainly due to its ability to improve water transport efficiency and reduce water residue rate. GDL with a three-layer gradient pore structure can quickly discharge water, not only improving the efficiency of water and gas transmission but also increasing the stability of fuel cell operation.

Figure 5

Test results of water permeability and water residual rate.

4.6 SEM and resistance

As shown in Figure 6, it can be seen that sample 3 has the highest distribution of pores on its surface, followed by sample 1 and sample 2. Sample 1 has almost no obvious pores on its surface. Sample 3 has the lowest resistivity, while sample 1 has the highest resistivity. It can explain why sample 3 has the lowest ohmic impedance at a humidity of 40%. The pore size distribution of GDL can be improved by CaCO₃, which is not only beneficial for the transmission of water and gas but also ensures the stability of the fuel cell during operation.

Figure 6

Comparison of SEM images and resistance for the three samples.

4.7 Water contact angle test results

From Figure 7, it can be seen that there is no significant difference in water contact angle among the samples. Therefore, in theoretical and experimental analyses, the influence of hydrophobicity can be ignored. The hydrophilicity of MPL directly affects the water transfer efficiency, ultimately leading to changes in fuel cell performance. In the process of studying three different structures of GDL, due to the small difference in hydrophobicity among the three samples, we can better explore the influence of structure and pore size on GDL water management.

Figure 7

Water contact angle test results.

4.8 Comparison between GDL29BC and the gradient pore size GDL

We compared the performance differences between GDL with a three-layer gradient pore structure and commercial GDL29BC. Test results are shown in Figure 8. It can be seen from the test results that the GDL with a three-layer gradient pore structure has good water and gas transmission ability when the humidity is 40, 60, and 100%, respectively.

Figure 8

Polarization curve comparison between GDL29BC and gradient pore size GDL.

5 Conclusion

This article prepared three types of GDLs; test results show that the three-layer gradient pore size GDL can exhibit good performance at different humidity levels. Specifically, when humidity is 40%, power density increases by 63.31 mW·cm⁻². Humidity is 60%, power density increases by 127.3 mW·cm⁻². Humidity is 100%, power density increases by 127 mW·cm⁻².

Capillary pressure is positively correlated with porosity. Through calculation and analysis, a three-layer gradient pore size GDL can obtain optimal capillary pressure, thereby providing good power for water transport. Mercury intrusion experiment showed that the pore size distribution of the three-layer gradient pore size GDL was the widest at 7 μm and above, providing assistance for the free diffusion of oxygen.

Three-layer gradient pore size GDL has the largest capillary pressure difference in the thickness direction, which proves that this structure can improve water transport efficiency. Water permeability of GDL with a three-layer gradient pore size is highest, while the residual amount of water is the smallest, which not only proves the high transmission efficiency of this structure for liquid water; moreover, smaller residual water can provide more pores for the diffusion of air. Three-layer gradient pore size GDL can exhibit good performance under low, medium, and high humidity conditions, thanks to its reasonable pore size distribution and capillary pressure difference. Improved the stability and efficiency of water and gas transmission and is easy to prepare, which can meet the needs of large-scale production.

Acknowledgments

This work was supported by Suqian Sci&Tech Program under Grant K202205 and Grant S202218 and Shandong Province Key R&D Program (Major Scientific and Technological Innovation Project) Project 2020CXGC010312.

Funding information: This work was supported by Suqian Sci&Tech Program under Grant K202205 and Grant S202218 and Shandong Province Key R&D Program (Major Scientific and Technological Innovation Project) Project 2020CXGC010312.
Author contributions: Min Li completed the execution of the experiment and the writing of the paper. Chongcai Xu completed the data collection. Tianya Li completed the literature research and organization. Guangyi Lin provided financial support and article revisions.
Conflict of interest: The authors state no conflict of interest.
Ethical approval: This article is not applicable for both human and/or animal studies. This declaration is not applicable.
Data availability statement: The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

[1] Wei, J., F. Ning, C. Bai, T. Zhang, G. Lu, H. Wang, et al. An ultra-thin, flexible, low-cost and scalable gas diffusion layer composed of carbon nanotubes for high-performance fuel cells. Journal of Materials Chemistry A, Vol. 8, No. 12, 2020, pp. 5986–5994.10.1039/C9TA13944CSearch in Google Scholar

[2] Nanadegani, F. S., E. N. Lay, and B. Sunden. Computational analysis of the impact of a micro porous layer (MPL) on the characteristics of a high temperature PEMFC. Electrochimica Acta, Vol. 333, 2020, id. 135552.10.1016/j.electacta.2019.135552Search in Google Scholar

[3] Chen, L., R. Lin, M. Lou, and K. Lu. Structural design of microporous layer to mitigate carbon corrosion in proton exchange membrane fuel cells. Carbon, Vol. 199, 2022, pp. 189–199.10.1016/j.carbon.2022.07.066Search in Google Scholar

[4] Lin, R., M. Dong, S. Lan, and M. Lou. Numerical simulation of liquid water transport in perforated cracks of microporous layer. Energy, Vol. 262, 2023, id. 125372.10.1016/j.energy.2022.125372Search in Google Scholar

[5] Zhou, C., R. Wenqing, W. Yucheng, Z. Jiawei, L. Xiaoteng, and X. Chenxi. A highly CO-tolerant anode Pt/Ti3C2Tx-CNT hybrid catalysts for PEMFC. Electrocatalysis, Vol. 14, 2023, pp. 1–8.10.1007/s12678-022-00768-wSearch in Google Scholar

[6] Kai-Qi, Z., D. Quan, X. Jiang-Hai, Y. Chen, Z. Jing, Z. Yan, et al. Dynamic performance for a kW-grade air-cooled proton exchange membrane fuel cell stack. International Journal of Hydrogen Energy, Vol. 47, 2022, pp. 35389–35411.10.1016/j.ijhydene.2022.08.015Search in Google Scholar

[7] Yun, Z., Z. Jian, L. Shuai, P. Chao, and X. Yucen. Nonlinear model predictive control of PEMFC anode hydrogen circulation system based on dynamic coupling analysis. International Journal of Hydrogen Energy, Vol. 48, 2023, pp. 2385–2400.10.1016/j.ijhydene.2022.09.306Search in Google Scholar

[8] Zheng, L., W. Yameng, M. Yongbiao, W. Buke, J. Yuting, Z. Lin, et al. Recent advances in the anode catalyst layer for proton exchange membrane fuel cells. Renewable and Sustainable Energy Reviews, Vol. 176, 2023, id. 113182.10.1016/j.rser.2023.113182Search in Google Scholar

[9] Zhang, Z., C. Xiao, H. Huang, M. Ul Haq, Z. Li, and L. Zeng. Aiming the water transport issues of anion exchange membrane fuel cells by introducing hydrophobic carbon nanotube diffusion layer. Journal of Power Sources, Vol. 616, 2024, id. 235123.10.1016/j.jpowsour.2024.235123Search in Google Scholar

[10] Tang, H., L. Wang, P. He, Q. Huang, and X. Wang. Bulk hydrophobic gas diffusion layer with interpenetrating network for high-performance fuel cells. Chemical Engineering Journal, Vol. 495, 2024, id. 152968.10.1016/j.cej.2024.152968Search in Google Scholar

[11] Johnson, M. S., R. C. Buck, I. T. Cousins, C. P. Weis, and S. E. Fenton. Estimating environmental hazard and risks from exposure to per-and polyfluoroalkyl Substances (PFASs): outcome of a SETAC focused topic meeting. Environmental Toxicology and Chemistry, Vol. 40, No. 3, 2021, pp. 543–549.10.1002/etc.4784Search in Google Scholar PubMed PubMed Central

[12] Yang, T. F., C. Y. Hsueh, B. L. Chen, W. K. Li, and W. M. Yan. Effects of fluorinated ethylene propylene contents in a novel gas diffusion layer on cell performance of a proton exchange membrane fuel cell. International Journal of Energy Research, Vol. 46, No. 2, 2022, pp. 1553–1564.10.1002/er.7269Search in Google Scholar

[13] Guo, Q. Y. and Y. Z. Qin. Numerical investigation of water droplet removal characteristics in novel block channels of PEMFC using dynamic wettability model. International Journal of Hydrogen Energy, Vol. 46, No. 74, 2021, pp. 36890–36902.10.1016/j.ijhydene.2021.08.221Search in Google Scholar

[14] Shigemasa, K., H. Sato, Y. Otsuki, M. Kurosu, and T. Araki. Investigating the vapour transport phenomena inside the cathode gas diffusion layer media by controlling local temperature gradient inside an operating proton exchange membrane fuel cell. Electrochimica Acta, Vol. 365, 2021, id. 137383.10.1016/j.electacta.2020.137383Search in Google Scholar

[15] Teuku, H., I. Alshami, J. Goh, M. S. Masdar, and K. S. Loh. Review on bipolar plates for low-temperature polymer electrolyte membrane water electrolyzer. International Journal of Energy Research, Vol. 45, No. 15, 2021, pp. 20583–20600.10.1002/er.7182Search in Google Scholar

[16] Li, C. Z., Z. T. Xu, L. F. Peng, and X. Lai. An electric-pulse-assisted stamping process towards springback suppression and precision fabrication of micro channels. International Journal of Mechanical Sciences, Vol. 218, 2022, id. 107081.10.1016/j.ijmecsci.2022.107081Search in Google Scholar

[17] Wang, X. R., Y. Ma, J. Gao, T. Li, G. Z. Jiang, and Z. Y. Sun. Review on water management methods for proton exchange membrane fuel cells. International Journal of Hydrogen Energy, Vol. 46, No. 22, 2021, pp. 12206–12229.10.1016/j.ijhydene.2020.06.211Search in Google Scholar

[18] Wu, S. D., W. M. Yang, H. Yan, X. Zuo, Z. Cao, H. Li, et al. A review of modified metal bipolar plates for proton exchange membrane fuel cells. International Journal of Hydrogen Energy, Vol. 46, No. 12, 2021, pp. 8672–8701.10.1016/j.ijhydene.2020.12.074Search in Google Scholar

[19] Shen, J. and Z. K. Tu. Flow channel design in a proton exchange membrane fuel cell: From 2D to 3D. International Journal of Hydrogen Energy, Vol. 47, No. 5, 2021, pp. 3087–3098.10.1016/j.ijhydene.2021.10.252Search in Google Scholar

[20] Shen, J., Z. Tu, and S. H. Chan. Enhancement of mass transfer in a proton exchange membrane fuel cell with blockage in the flow channel. Applied Thermal Engineering, Vol. 12, No. 138, 2018, pp. 1–51.Search in Google Scholar

[21] Shi, X., D. K. Jiao, Z. M. Bao, K. Jiao, W. Chen, and Z. Liu. Liquid transport in gas diffusion layer of proton exchange membrane fuel cells: Effects of microporous layer cracks. International Journal of Hydrogen Energy, Vol. 47, No. 9, 2022, pp. 6247–6258.10.1016/j.ijhydene.2021.11.248Search in Google Scholar

[22] Tian, C., F. Yuan, T. Deng, Q. He, C. Hu, Y. Chen, et al. Coupled optimization of auxiliary channels and porosity gradient of GDL for PEMFC. Energy, Vol. 301, 2024, id. 131734.10.1016/j.energy.2024.131734Search in Google Scholar

[23] Tao, X., K. Sun, R. Chen, Q. Li, H. Liu, W. Zhang, et al. Effect of gas diffusion layer parameters on cold start of PEMFCs with metal foam flow field. Applied Energy, Vol. 364, 2024, id. 123186.10.1016/j.apenergy.2024.123186Search in Google Scholar

[24] Guo, Y., F. Cheng, X. Liang, F. Yue, Y. Yan, Y. Li, et al. Effect of gas diffusion layer surface property on platinum nanowire array electrode performance in proton exchange membrane fuel cells. International Journal of Hydrogen Energy, Vol. 74, 2024, pp. 322–328.10.1016/j.ijhydene.2024.06.169Search in Google Scholar

[25] Etzi, M., J. Dangbegnon, A. Chiodoni, and C. F. Pirri. Impact of gas diffusion layer architecture on the performances of Cu-based electrodes for CO2 electroreduction to ethylene in flow reactors. Journal of CO2 Utilization, Vol. 83, 2024, id. 102772.10.1016/j.jcou.2024.102772Search in Google Scholar

[26] Calili-Cankir, F., E. M. Can, D. B. Ingham, K. J. Hughes, L. Ma, M. Pourkashanian, et al. Patterned hydrophobic gas diffusion layers for enhanced water management in polymer electrolyte fuel cells. Chemical Engineering Journal, Vol. 484, 2024, id. 149711.10.1016/j.cej.2024.149711Search in Google Scholar

[27] Ma, T., H. Guo, Z. Gu, W. Lin, J. Qi, C. Yu, et al. Performance analysis and experimental study of titanium GDL in proton exchange membrane fuel cell. International Journal of Hydrogen Energy, Vol. 83, 2024, pp. 604–613.10.1016/j.ijhydene.2024.08.012Search in Google Scholar

[28] Li, M., J. Liu, P. Nachtigal, and D. Mimic. Pore-scale investigation of water-gas transport in reconstructed gas diffusion layers with binder and polytetrafluoroethylene coating. Journal of Power Sources, Vol. 612, 2024, id. 234819.10.1016/j.jpowsour.2024.234819Search in Google Scholar

[29] Liu, W., J. Lee, T. J. Schmidt, and P. Boillat. Quantitative observation of water phase transition in gas diffusion layers utilizing advanced in situ differential scanning calorimetry. Journal of Power Sources, Vol. 602, 2024, id. 234268.10.1016/j.jpowsour.2024.234268Search in Google Scholar

[30] Zhang, T., L. Meng, C. Chen, L. Du, N. Wang, L. Xing, et al. Similarities and differences between gas diffusion layers used in proton exchange membrane fuel cell and water electrolysis for material and mass transport. Advanced Science News, Vol. 11, 2024, id. 2309440.10.1002/advs.202309440Search in Google Scholar PubMed PubMed Central

[31] Sim, J., J. Kim, H. Chun, J. Chang, Y. Kim, B. Kim, et al. Variation in performance and characteristics of long-term operated gas diffusion layer in a fuel cell electric vehicle. Journal of Power Sources, Vol. 595, 2024, id. 234026.10.1016/j.jpowsour.2023.234026Search in Google Scholar

[32] Li, Y. and T. Van Nguyen. A one-dimensional model of a PEM fuel cell with the cathode catalyst layer hydrophobically treated for water management. Journal of The Electrochemical Society, Vol. 169, 2022, id. 114505.10.1149/1945-7111/ac9bdfSearch in Google Scholar

[33] Wang, X. and T. Van Nguyen. Modeling the effects of the microporous layer on the net water transport rate across the membrane in a PEM fuel cell. Journal of The Electrochemical Society, Vol. 157, No. 4, 2010, pp. B496–B505.10.1149/1.3292287Search in Google Scholar

[34] Gaiselmann, G., R. Thiedmann, I. Manke, W. Lehnert, and V. Schmidt. Stochastic 3D modeling of fiber-based materials. Computational Materials Science, Vol. 59, 2012, pp. 75–86.10.1016/j.commatsci.2012.02.038Search in Google Scholar

[35] Shen, J., Z. K. Tu, and S. H. Chan. Performance enhancement in a proton exchange membrane fuel cell with a novel 3D flow field. Applied Thermal Engineering, Vol. 164, 2020, pp. 1–8.10.1016/j.applthermaleng.2019.114464Search in Google Scholar

[36] Mukherjee, P. P., C. Y. Wang, and Q. Kang. Mesoscopic modeling of two-phase behavior and flooding phenomena in polymer electrolyte fuel cells. Electrochimica Acta, Vol. 54, No. 27, 2009, pp. 61–75.10.1016/j.electacta.2009.06.066Search in Google Scholar

Received: 2025-04-13

Revised: 2025-06-27

Accepted: 2025-07-06

Published Online: 2025-09-08

This work is licensed under the Creative Commons Attribution 4.0 International License.

Articles in the same Issue

https://doi.org/10.1515/htmp-2025-0090

Keywords for this article

proton exchange membrane fuel cell; gas diffusion layer; graded pore size; pore-forming agent

Creative Commons

BY 4.0