Understanding Hydrogen Crossover and Oxidation Reactions in High-Pressure PEM Water Electrolyzers

Introduction

This study evaluates hydrogen crossover in standard PEM Membrane Electrode Assemblies (MEAs) under high differential pressure using electrochemical characterization and online gas chromatography (GC).

(1) Under high differential pressure, part of the permeated hydrogen is oxidized at the anode catalyst layer. The HOR current was characterized using Linear Sweep Voltammetry (LSV) at 1 mV/s under differential pressures of 0-30 bar. The observed HOR current is diffusion-limited, proportional to the cathode pressure, and independent of the iridium oxide type in the anode catalyst layer or the PGM coating on the Porous Transport Layer (PTL).

(2) Online GC was used to quantify hydrogen concentration at the anode outlet. Since some permeated hydrogen is oxidized at the anode catalyst layer, GC measurements at the outlet underestimate the actual permeation rate under high pressure. Combining these measurements with HOR current data allows accurate calculation of the hydrogen crossover rate under various operating current densities and differential pressures (0-30 bar).

(3) The hydrogen crossover rate increases with both differential pressure and operating current density, suggesting two distinct gas permeation mechanisms: a diffusion/pressure-driven pathway and an oversaturation/concentration-driven pathway.

Schematic of the experimental setup showing PEMWE equipped with manual backpressure at the cathode and online gas chromatography to measure hydrogen content at the anode. (https://pubs.rsc.org/)

Experimental Methods

MEA Preparation & Single-Cell Setup: MEAs were prepared using Nafion N115 membrane (127 μm). The cathode catalyst was Pt/C (50 wt%, loading 0.1 mg/cm²), and anode catalysts included various IrO₂ types (loading 0.4 mg/cm²). The single cell had an active area of 5 cm². A custom-built pressure setup allowed cathode back-pressure control from 0 to 30 bar gauge pressure, while the anode operated at ambient pressure. Tests were conducted at 80°C.

Electrochemical Characterization: Polarization curves were recorded from 5 mA/cm² to 4 A/cm². High-Frequency Resistance (HFR) was measured for iR-correction. The reversible cell voltage was calculated considering increased hydrogen partial pressure at the cathode under high back-pressure. HOR currents were measured via LSV (1 mV/s) from OCV to the oxygen evolution reaction (OER) onset potential under different cathode pressures.

Online Gas Chromatography: A GC with a Thermal Conductivity Detector (TCD) quantitatively measured the hydrogen-in-oxygen volume fraction (%H₂/O₂) at the anode outlet after water vapor removal. These values were converted to hydrogen flux at the anode.

Data Combination for Crossover Calculation: The actual hydrogen crossover rate was calculated by summing the HOR current density (representing hydrogen consumed at the anode) and the hydrogen flux derived from GC measurements (representing hydrogen exiting the anode).

Results and Discussion

HOR Current under Pressure: Significant HOR currents were observed only under high differential pressure (>0 bar). The HOR polarization curves showed three regions: facile kinetics up to ~0.1 V, a pronounced diffusion-limited current plateau from ~0.1 V to ~1.4 V, and OER dominance above ~1.4 V. The magnitude of the diffusion-limited HOR current was directly proportional to the cathode pressure. This current was independent of the specific IrO₂ anode catalyst type and the PGM coating (Pt, Ir, Au) on the PTL, confirming that the IrO₂ surface itself catalyzes the HOR under these conditions.

Underestimation by GC Measurements Alone: GC measurements showed that the hydrogen volume fraction at the anode outlet increased with differential pressure but decreased with increasing current density due to oxygen dilution from OER. However, the hydrogen flux calculated from GC data at operating currents was significantly lower than the hydrogen permeation rate measured at zero current (i.e., the diffusion-limited HOR current). This discrepancy confirms that a substantial portion of the permeated hydrogen is oxidized at the anode catalyst layer before reaching the outlet and is not detected by GC.

Accurate Crossover Quantification: By adding the HOR current density to the GC-derived hydrogen flux, the true hydrogen crossover rate was determined (Figure 4c). This combined analysis reveals that the crossover rate depends on both pressure and current density.

Pressure-Driven Crossover: At zero operating current, the crossover rate (represented by the diffusion-limited HOR current) is proportional to the differential pressure, indicating a classic diffusion-driven mechanism.

Current-Driven Crossover: When the electrolyzer operates at high current densities, the crossover rate increases significantly beyond the pressure-driven baseline. This suggests an additional mechanism, likely related to hydrogen supersaturation within the ionomer or water phase, driven by the high hydrogen generation rate at the cathode.

The electrolyzer functions similarly to a hydrogen pump at applied voltages below ~1.4 V when the operating current is less than the diffusion-limited HOR current, confirming the consumption of permeated hydrogen at the anode.

Conclusion

This study demonstrates a critical methodological advance for quantifying hydrogen crossover in PEM electrolyzers under high differential pressure.

The HOR reaction occurring at the IrO₂-based anode catalyst layer under high differential pressure consumes a significant fraction of the permeated hydrogen. This HOR current is diffusion-limited and proportional to the cathode pressure.

Relying solely on anode outlet gas composition (e.g., via GC) severely underestimates the true crossover rate because it misses the hydrogen consumed by HOR. The actual permeation rate must be calculated by combining the GC-measured hydrogen flux with the electrochemically measured HOR current density.

Hydrogen crossover is driven by two mechanisms: a pressure-driven diffusion process dominant at low/zero current, and a current-dependent process (likely concentration-driven via supersaturation) that becomes significant at high operating current densities.

These findings are essential for accurate safety assessment, modeling, and design of mitigation strategies (like recombination layers) for high-pressure PEM electrolyzers. The combined electrochemical/GC approach provides a robust tool for quantifying crossover under realistic operating conditions.

References

[1] Gawas R., Kushner D.I., Peng X. Importance of hydrogen oxidation reaction current in quantifying hydrogen crossover in PEM water electrolyzers at high differential pressure. Energy Environ. Sci., 2025, (10):18.