In-plane and through-plane water distribution in a passive PEMFC observed with neutron radiography

EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476288 B1009 Page 1/6 B1009 In-plane and through-plane water distribution in a passive PEMFC observed with neutron radiography L. Duque* (1), E. Diaz-Alvarez (1), N. Weber (2), M. Sarma (2), T. Lappan (2), J. Lee (3), P. Boillat (3), M.A. Folgado (1), A.M. Chaparro (1) (1) CIEMAT Madrid/Spain; (2) Helmholtz-Zentrum Dresden – Rossendorf, Dresden/Germany; (3) Paul Scherrer Institut, Villigen PSI/Switzerland; *Contact corresponding authors: www.EFCF.com/ContactRequest Abstract Passive polymer-electrolyte membrane fuel cells (PEMFCs) may open new fundamental and technological possibilities for low temperature fuel cell technologies. Since they operate undisturbed from external convective gas flow, they are ideal platforms for the study of transport properties in porous electrodes. At the same time, passive PEMFC systems overcome conventional fuel cell systems in power and energy densities in many small and portable hydrogen applications. The principal power limitation of passive PEMFCs is the accumulation of liquid water generated by the cathodic reaction in porous electrodes (O2 + 4H+ + 4e- → 2H2O). The generated water saturates the pores and hinders the access of gases to the catalyst sites. In this presentation, the liquid water distribution in a passive PEMFC is observed in-operando using neutron radiography. A PEMFC configuration adapted for passive operation was imaged in in-plane and through-plane (see Figure) [1] [2]. Neutron images show distinct profiles of liquid water inside the cells as a result of the combination of the different passive forces that govern its transport. In particular, liquid water accumulates towards the upper part of the cell surface when in vertical position. In general, water distribution is found very dependent on ambient conditions (temperature, humidity), and cell orientation (vertical vs. horizontal), which may hinder the application of these cells in some portable devices. A second presentation at this conference shows means to diminish such dependence of cell response based on superhydrophobic catalyst layers. Top: Frontal and lateral photographs of the passive PEMFC; Bottom: neutron images of the water generated after some minutes of passive operation, showing in plane (bottom, left) and cross-sectional distributions (bottom, right). EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476288 B1009 Page 2/6 Introduction Hydrogen fuel cells have a high potential as power source for the small and portable applications [3,4]. They improve the autonomy, durability, and safety of the applications, either alone or in hybrid systems with batteries or supercapacitors. A key inherent characteristic of fuel cells is the decoupling of power generation and energy storage functions, that can be dimensioned independently, which makes more flexible configurations and safer power systems. The same characteristic limits the power and energy densities and makes less compact power systems, which hinders in many cases their use in portable applications. In order to overcome these penalties, new cell designs are proposed, based on different electrode configurations [5,6], all having in common the introduction of passive elements (dead end anode, air breathing cathode). Our research group has also patented and published [2,7] an air breathing passive fuel cell for small, portable applications. These passive configurations have few but important drawbacks, such as their higher dependence on ambient conditions (temperature and humidity) and on the cell/stack orientation. All of these variables mainly affect the water distribution inside the cell, which is difficult to study directly in operando without substantially modifying the cell´s design. Neutron imaging can be used without severely influencing the cell´s assembly, since the most important requirement is that the materials employed exhibit low neutron interaction. This can be achieved by fabricating the cell plates using low absorption metals, like aluminum, or polymers that have very low amount of hydrogen atoms, like PTFE. In any case these materials are easily available and cheap, which facilitates their implementation to the cell. 1. Scientific Approach Cell and support hardware has been prepared to make neutron radiography on passive PEMFCs. In operando neutron imaging was carried-out under two different neutron beam incidences, in-plane or lateral incidence (MEA parallel to beam), and through-plane or frontal incidence (MEA perpendicular to beam). The in-plane incidence imaging was carried out with cell in vertical position and in horizontal position to see the effect of the gravity field. Different MEA configurations were tested to analyze their influence on water dynamics. 2. Experiments Cell assembly Following the cell design described in [2] (fig 1), the passive fuel cells were assembled with slight modifications: some silicone gaskets were exchanged with Teflon ones, and the central ring and columnar plate were made of aluminum. Fig. 1 Passive PEMFC developed at CIEMAT [2]. These modifications were directed to reduce the “background” absorption of the cell, thus improving the contrast of the water signal. EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476288 B1009 Page 3/6 MEA preparation Four different MEAs were analyzed, one with commercial electrodes on both sides (FuelcellStore, Pt/C on carbon cloth with microporous layer, 0.3mgPt/cm2), and three with electrosprayed (ES) catalyst layers in anode, in cathode, and in both. For preparation of the electrosprayed catalyt layers, the electrospray deposition technique was used by imposing a high potential difference (5kV – 10kV) between an ejector needle and a substrate (Nafion 212 membrane from Ion Power, Dupont). In all cases the active area of the cell was 12.5 cm2. The electric field generated between the needle and the substrate accelerates charged catalyst and ionomer particles towards the substrate, where, upon discharging, deposit forming highly porous and superhydrophobic layers. Figure 2: Schematic representation of the electrospray deposition setup used at CIEMAT [3] The process also maximizes catalyst utilization thanks to the electric attraction of the particles to the substrate. Electrospray deposited catalyst layers have demonstrated a positive effect on water management and performance of PEMFCs [3], [4]. Experimental setup and cell protocol for neutron radiography A dedicated support was manufactured in order to hold the cells and allow easier cell and neutron beam orientation changes (fig. 2). Figure 2. 1.- Support design. 2.- Experimental setup. 3.- Cell orientations: a) Vertical, through-plane; b) Vertical in-plane; c) Horizontal in-plane, EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476288 B1009 Page 4/6 Figure 3. Current profile during Neutron imaging This support along with the cell were positioned at the beamline, with three thermocouples to measure ambient and cell temperatures (top and bottom of cell). Electrochemical measurements were carried out with a potentiostat (Biologic) following the current load profile depicted in figure 3. The ramp up was performed until reaching maximum 4000 mA, or until the cell voltage was close to 0.3V to preserve the cells. Upon reaching this point, an electrochemical impedance was performed. The cells were operated under ambient conditions (average 23ºC and 40% RH) and hydrogen was fed to the cells at 0.8 barg from small metal hydride containers (Hydrostik Pro, Horizon) with 1g hydrogen weight capacity. 3. Results Electrochemical characterization Electrochemical characterization with polarization curves was performed just after cell assembly. Figure 4 shows polarization curves of the cells with the four different MEAs configurations, showing the stabilized, but dynamic, response of the cells. A significant improvement of the response is observed when having the electrosprayed electrode in the anode. Such improvement of the performance with electrosprayed CLs is supported by the imaging results that are shown further below, although neutron imaging corresponds to a static current demand profile as shown in Fig. 3. Figure 4: Comparison of the four different MEAs tested under the neutron beam. Curves taken under ambient conditions (23 oC, 30%RH) feeding the cathode with static ambient air, and the anode with a static dry hydrogen pressure of 0.8bar. EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476288 B1009 Page 5/6 Neutron imaging Neutron radiography was carried out at the NEUTRA facility of the SINQ spallation source (Paul Scherrer Institute, Switzerland) [5]. The neutron radiographies were used to obtain images of the water thickness inside the cells. In order to eliminate experimental effects caused by changes in the neutron beam intensity, the detector homogeneity, and absorption of the components of the cell, corrections were applied to the images (filtering, subtraction of background…), ending with substraction of the “dry cell” image (without water) [6]. Figure 5 shows the raw neutron radiography obtained of the dry and in-operando cell, and the result of the corrections and analysis performed to obtain a grey scale of water thickness. Figure 5: Raw neutron radiography of the dry cell (a), the in-operando cell (b), and the calculated water thickness image (c). Figure 6 shows water thickness images of the 4 different MEAs in vertical through-plane and vertical in-plane orientations at the same operation time. Figure 6: Water thickness images obtained from neutron radiographies of two passive PEMFCs with different MEAs configurations at the same operation time. The yellow circle marks the active electrode areas. Electrospray catalyst layers have a great impact on the cell water distribution as it can be seen in figure 5. Thanks to its properties (high porosity and superhydrophobicity [2,3]) the water distribution over the cell is more homogeneous and covers all the electrode area; in contrast, the cell with commercial electrodes shows accumulation of water on the top side of the cell. This more homogeneous distribution is attributed to better water transport properties within the electrosprayed electrodes and gives rise to better electrochemical response in the polarization curves. Commercial electrodes ES anode / Com cathode Top Top a b c EFCF 2025: Low-Temp. Fuel Cells, Electrolysers & H2 Processing 1 – 4 July 2025, Lucerne Switzerland https://doi.org/10.5281/zenodo.17476288 B1009 Page 6/6 Horizontal orientation had a great impact on the cell performance, reducing its operation time and maximum current by half. This is due to a faster water accumulation leading to an earlier flooding of the cell. A more in-depth analysis of the water distribution observed will be provided in another communication at this conference: “Neutron radiography analysis of water management in a passive PEMFC with superhydrophobic electrosprayed catalyst layers”, B1701. 4. Conclusion Neutron radiography, although not being an easily accessible technique, is really useful to elucidate water dynamics in a PEMFC. This imaging technique helps understand how different layers and their properties affect the water management in fuel cells while needing minimal changes to their design and assembly. In our case, it gives a visual confirmation of the enhance performance given by electrosprayed catalyst layers. Acknowledgements: Funded by Ministerio de Ciencia, Educación y Universidades of Spain, Project Dualcell (PID2023-151637OB-I00), and European Union, Project DECODE Grant agreement ID: 101135537. Neutron imaging was performed at the NEUTRA beamline of SINQ, PSI (P20240329) References [1] A. M. Chaparro, P. Ferreira-Aparicio, M.A. Folgado, R. Hübscher, C. Lange, N. Weber, J. Power Sources 480 (2020) 228668. https://doi.org/10.1016/j.jpowsour.2020.228668. [2] M.A. Folgado, L. Duque, A.M. Chaparro, Int.J.Hydrogen Energy. 52 (2024) 1315-1324. https://doi.org/10.1016/j.ijhydene.2023.08.291. [3] J. J. Conde, P. Ferreira-Aparicio, A. M. Chaparro, ACS Applied Energy Materials 2021 4 (8), 7394-7404. https://doi.org/10.1021/acsaem.1c01445 [4] P. A. García-Salaberri, L. Duque, M. Antonia Folgado, E. Diaz-Alvarez, and A. M. Chaparro, Fuel, vol. 380, p. 133175, Jan. 2025, doi: 10.1016/J.FUEL.2024.133175. [5] E.H. Lehmann, P. Vontobel, L. Wiezel. Nondestruct. Test. Eval. 16 (2001) 191–202, doi:http://dx.doi.org/10.1080/10589750108953075. [6] Q. Meyer et al. Electrochimica Acta, vol. 203, pp. 198–205, Jun. 2016, doi: 10.1016/j.electacta.2016.04.018. [7] A.M. Chaparro, M.A. Folgado, L. Duque. Dispositivo electroquímico para generación de electricidad y/o hidrógeno. ES 2932991. Keywords: EFCF2025, H2, Low-Temp. Fuel Cells & Electrolysers, Passive PEMFC, Neutron Imaging, Water dynamics, electrospray Remark: This work is licensed under Creative Commons Attribution 4.0 International