Coatings of Metallic Bipolar Plates in PEM-Fuel Cells: Materials, Durability and Performance

EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476239 A1115 Page 1/9 A1115 Coatings of Metallic Bipolar Plates in PEM-Fuel Cells: Materials, Durability and Performance Jonathan Halter* (1), Isabel Vazquez-Fernandez (1), Stéphane Weissbaum (2), Samaneh Daviran (2), Jan Van Herle (2), Mardit Matian (1) (1) EH Group Engineering AG, Nyon/Switzerland; (2) Group of Energy Materials, Ecole Polytechnique Federal de Lausanne (EPFL)/Sion; *Contact corresponding authors: www.EFCF.com/ContactRequest Abstract Metallic bipolar plates (BBP) offer significant advantages over graphite BBP, such as significantly higher thermal and electrical conductivity, reduced thickness and weight, ease of manufacturing and finally increased volumetric power density of stacks. However, metals such as stainless steel or even titanium require protective coating layers to increase their capability to withstand the harsh corrosive environment within fuel cells. In this presentation, a range of different materials and approaches for such coatings are evaluated in terms of both their performance and their durability. The impact of the bipolar plate coating on performance, conductivity and internal contact resistance is evaluated and presented. To study their corrosion resistance, anodic and cathodic electrolyte conditions are simulated and potentiostatic and potentiodynamic measurements are performed. The results are supported by SEM/EDX measurements, allowing to map the elemental distribution of the samples, and through XPS, where the molecular contribution of the coating is derived. EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476239 A1115 Page 2/9 Introduction Metallic Bipolar plates are a possible candidate for PEM fuel cells, as they have significant advantages versus their composite or graphite counterparts. They have superior mechanical stability, a low gas permeability, are easy and cheap to produce and have a high electrical conductivity. Their major disadvantage is their corrosion resistance under fuel cell conditions [1]. Therefore, significant efforts in both academia and industry have been undertaken to develop coating materials to protect the base metal [2], [3], [4]. Typically, either a type of stainless steel or titanium is used as a base material, which is then coated using a range of coating methods and materials [5]. Stainless steel was chosen as a base material for this work as it is cheap and easy to manufacture which makes it an ideal candidate. The base metal has a low corrosion resistance, which amplifies the dependency on the to be evaluated coating. In this work, a range of possible coating materials and means of applying the coating were selected and evaluated for their use as a PEM fuel cell bipolar plate. 1. Scientific Approach The need for higher durability as well as both volumetric and gravimetric power density has driven the development of novel metal bipolar plates. The main role of the coating is to increase the corrosion resistance of the metallic plate, but it is to be expected to affect other crucial properties, such as conductive and contact resistance, thereby affecting not just durability but also performance. Investigating the influence of a range of state-of-the-art coatings and their suitability for using in PEM fuel cell bipolar plates is therefore needed to select the best candidate. Five possible metallic candidates were selected and in addition two graphite samples were used for comparison and to derive the advantages and disadvantages of each approach. As described in the experiment section, both electrochemical methods to assess the performance and durability of each candidate were employed and were combined with imaging to investigate elemental composition of the coatings. Both optical microscopy and SEM/EDX were used in this work. 2. Experiments In a first step, criteria to assess the candidates are to be defined. The US Department of Energy DoE, has defined targets for metallic bipolar plates (s. Table 1) and methods on how these candidates are to be evaluated [6]. Table 1: Targets for PEM fuel cell bipolar plates from the Department of Energy for 2025. DoE target 2025 Unit Plate weight 0.18 kg/kW Plate H2 permeation 2 x 10-6 cm3 s-1 cm-2 Pa-1 Corrosion anode <1, no active peak µA cm-2 Corrosion cathode <1 µA cm-2 Electrical conductivity > 100 S cm-1 Flexural strength > 40 MPa Forming elongation 40 % EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476239 A1115 Page 3/9 In order to assess if a coating of the base metal is suitable for PEM fuel cell application, the following criteria were selected. a. Interfacial Contact Resistance (ICR) To determine the internal contact resistance (ICR) between the bipolar plate candidate and the GDL, the set-up shown in Figure 1 was developed. The selection of the GDL with a MPL was chosen to minimize the contact resistance between the electrode and the GDL. The set-up therefore also closely represents a fuel cell (Figure 1. Left), with the bipolar plate sample in contact with the substrate side of the GDL. Figure 1: Set-up for measuring internal contact resistance. On the right side of Figure 1, the background measurement is shown, where the sample removed. The equations below show that by subtracting the background, all that is left is 2x the ICR (i.e. the resistance associated with the GDL/BP interface and the resistance of the sample material). However, as the intrinsic resistance of a material is typically orders of magnitudes smaller than the interfacial contact resistance, it can be neglected. For each candidate, a 2x2 cm sample was used, sandwiched between two GDLs (2.9x2.9 cm). A non-conductive gasket surrounding the sample was used to avoid electron conduction from GDL to GDL, which would bypass the sample. The entire assembly was compressed to 1.4 MPa. b. Resistivity measurement EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476239 A1115 Page 4/9 A 4-point probe device (Ossila T2001A) was used to derive the resistivity. A 2x2 cm sample of each candidate was used to measure the resistivity, from which the conductivity values are calculated. c. Corrosion measurement A 3-electrode set-up capable of heating an electrolyte solution to the required 80°C was developed following the DoE method of corrosion determination (s. Figure 2). The metallic sample was used as the working electrode, with an RHE as a reference electrode and a Pt wire as an auxiliary electrode. The electrolyte solution was created by acidifying DI water with concentrated sulfuric acid to a pH of 3. The addition of hydrofluoric acid was omitted. Both potentiostatic and potentiodynamic experiments were carried out. Figure 2: 3-electrode setup for corrosion measurements. For the potentiostatic measurements, a potential of 0.6 V vs Ag/AgCl was applied for a duration of 24 hours. For the potentiodynamic experiments, the voltage was increased between the starting value of -0.4 V to the final voltage of 0.6 V vs Ag/AgCl with a scan rate of 0.1 mV/s. Both experiments were carried out at 80 °C and a pH of 3. EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476239 A1115 Page 5/9 Figure 3: Scheme for deriving corrosion potential and current from a potentiodynamic measurement. Figure 3 shows how the corrosion potential and corrosion current are determined from the potentiodynamic sweep [7]. By extrapolating both the cathodic and anodic Tafel slopes, the intersection dictates both corrosion potential and current. In potentiostatic measurements, the current for the last hour was averaged to results in the corrosion current density. d. Imaging In addition to electrochemical testing, a set of different imaging techniques were applied. Optical microscopy using a Keyence VHX-7000 Digital Microscope was used for high resolution imaging and to derive surface roughness for the coated materials. SEM/EDX was carried out to determine the elemental composition of the coatings. To understand the elemental composition of bi-layer coatings, cross-section SEM/EDX was carried out. 3. Results In Table 2, the ICR and conductivity of all candidates are shown. As an overall trend, the metallic bipolar plates show a lower interfacial contact resistance with the GDL, as they tend to be below 5 mOhm cm2. All metallic samples fulfill the DoE target of less than 10 mOhm. Graphite samples tend to be closer to this target, with one of them not fulfilling it. Overall, metallic samples have lower ICR values than the 2 graphite samples. Table 2: Overview of thickness, internal contact resistance and conductivity for all samples. Sample Thickness [µm] ICR [mOhm cm2] Resistivity [Ohm m] Conductivity [S cm-1] Metallic # 1 103 2.65 ± 1.22 5.12E-07 ± 2.58E-07 19536 #2 105 3.59 ± 1.25 4.79E-07 ± 2.87E-07 20862 #3 102 0.58 ± 2.64 1.30E-07 ± 3.72E-08 76772 #4 207 1.42 ± 0.72 5.06E-07 ± 1.13E-07 19756 #5 102 8.86 ± 4.32 4.16E-07 ± 3.68E-08 24053 Graphite #6 1005 6.53 ± 2.27 8.9E-05 ± 1.24E-06 3358 #7 326 11.73 ± 1.45 2.98E-06 ± 3.24E-07 112 EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476239 A1115 Page 6/9 Bigger differences were observed in terms of conductivity, where metallic samples show higher values by 1-2 orders of magnitude. These measurements illustrate one of the key advantages of metallic bipolar plates. Although the graphite plates do also fulfill the DoE target of a conductivity higher than 100 S/cm, coated metallic plates are clearly superior in terms of conductivity. Furthermore, it should be noted that metallic samples were much thinner with a thickness between 100 µm to 200 µm, while the graphite samples had a thickness up to 1000 µm. To study the corrosion rates under anodic and cathodic fuel cell conditions, potentiodynamic (s. Figure 4) and potentiostatic (s. Figure 5) measurements were carried out. Figure 4: Potentiodynamic measurements for all sample. Furthermore, both potentiodynamic and potentiostatic measurements included an uncoated metal to better the understanding of the effect of the coating. The results of these tests are shown in Table 3. EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476239 A1115 Page 7/9 Figure 5: Potentiostatic Measurement for all samples. In the potentiodynamic test, the uncoated sample performs the worst, as expected. It shows the lowest corrosion potential and a high corrosion current. When analysing the different coatings for the corrosion current, all fulfill the DoE target of < 1 µA cm-2 easily. Their corrosion potential is between 400 mV and 550 mV vs SHE. One of the graphite samples shows a very high corrosion current of 10 µA cm-2, thereby missing the DoE target by a significant margin. Sample 7 showed a similar corrosion current as the coated metals, but did not outperform the coated metal samples. Table 3:Overview of corrosion rates from potentiostatic and potentiodynamic measurements. Potentiodynamic Potentiostatic Sample Corr. potential [mV vs SHE] Corr. current [µA cm-2] Corr. current [nA cm-2] Metallic #0 -20 1.2 60 ± 40 # 1 400 0.3 90 ±15 #2 540 0.2 140 ± 20 #3 510 0.2 70 ± 10 #4 470 0.5 50 ± 20 #5 500 0.6 70 ± 10 Graphite #6 280 10 280 ± 90 #7 340 0.3 140 ± 20 In potentiostatic measurement, the uncoated sample performed surprisingly well. The reason of this is unclear. As for the coated metallic samples, they all showed a corrosion current of less than 100 nA /cm2, except for sample 2 that exceeded that value slightly. All samples easily passed the DoE target of 1 µA cm-2. This is also true for both graphite samples, although as seen in the potentiodynamic measurements, they did not outperform the coated metals. In order to further understand the superior performance of the metallic samples, cross-section SEM/EDX was carried out to derive the elemental composition of the coatings (s. Figure 6). EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476239 A1115 Page 8/9 Figure 6: Cross section SEM (top) of sample 3; Titanium EDX map (bottom, left) and Carbon EDX map (bottom, right). Many of the samples showed a similar pattern of having a bi-layer coating of a bottom layer containing titanium and a top layer that is carbon coated (s. Table 4). The exception is sample 4, which is a monolayer of a chromium coating. Table 4: Overview of coating type and elemental composition. Sample Type of coating Bottom layer Top layer # 1 Bi-layer Titanium Carbon #2 Bi-layer Titanium Carbon #3 Bi-layer Titanium Carbon #4 Monolayer Chromium n/a #5 Bi-layer Titanium Carbon Interestingly, sample 4 outperformed all other samples in the potentiostatic measurement with a corrosion current of 50 nA cm-2. 4. Conclusion Overall, all the coated metals are well suited materials for PEM bipolar plates. Not only did they outperform their graphite counterparts in terms of interfacial contact resistance and conductivity, but they also show lower corrosion rates in potentiostatic and potentiodynamic measurements. Despite many samples showing a similar pattern of a bi-layer approach with a bottom layer of titanium coating and a top layer of carbon, significant differences in their performance and durability were observed. Acknowledgements (H2Marine) This work was supported by the European Union and the Swiss State Secretariat for Education, Research, and Innovation under the H2Marine project. The authors gratefully acknowledge the financial support that made this research possible. The authors acknowledge financial support from the Clean Hydrogen Partnership and its members under the grant agreement H2Marine 101137965. This work was supported by the Swiss State Secretariat for Education, Research and EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476239 A1115 Page 9/9 Innovation (SERI) under contract number 23.00624 –10113796 (EH Group) and 24.00109 – 101137965 (EPFL). References [1] X. Goa, J. Chen, R. Xu, Z. Zhen, X. Zeng, X. Chen und L. Cui, «Reserach progress and prospect of the materials of bipolar plates for proton exchange membrane fuel cells,» International Journal of Hydrogen Energy, pp. 711-743, 2024. [2] H.-S. Hea und S.-J. Kim, «Investigation of Electrochemical Characteristics and Interfacial Contact Resistance of TiN-Coated Titanium as Bipolar Plates in PEMFC,» Coatings, Bd. 13, Nr. 123, 2023. [3] R. Antues, M. Oliveira, G. Ett und V. Ell, «Corrosion of metal bipolar plates for PEM fuel cell: A review,» International Journal of Hydrogen Energy, pp. 3622-3647, 2010. [4] T. Liu, Y. Tao, Y. Wang, M. Wu, J. Zhang, Y. Yu, X. Wang und J. Shao, «Thermal Sprayed Protective Coatings for Bipolar Plates of Hydrogen Fuel Cells and Water Electrolysis Cells,» Coatings, Bd. 14, Nr. 307, 2024. [5] N. Asri, T. Husaini, A. Sulong, E. Majlan und w. Daud, «Coating of stainless steel and titanuim bipolar plates fro anticorrosion in PEMFC: A review,» International Journal of Hydrogen Energy, pp. 9135-9148, 2017. [6] Y. Wang, B. Seo, B. Wang, N. Zamel, K. Jiao und X. C. Adroher, «Fundamentals, materials, and machine learning of PEMFC technology,» Energy and AI, Bd. 1, 2020. [7] A. Hedayati, S. Asghari, A. H. Alinoori, M. Koosha und E. Vuorinen, «Effect of coating thickness on corrosion and contact resistance behavior of TiN coated ANSI 316L as bipolar plates for PEMFC,» Iranian Journal of Hydrogen & Fuel Cell 2, pp. 137-149, 2016. Keywords: EFCF2025, H2, Low-Temp. Fuel Cells & Electrolysers, Bipolar plates, Coatings, Corrosion rates, Interfacial contact resistance, Conductivity Remark: This work is licensed under Creative Commons Attribution 4.0 International