Skip to main content

Tracking platinum catalyst oxidation in fuel cells

11-04-2025

Using stroboscopic operando high-energy X-ray scattering at beamline ID31, researchers have directly observed platinum oxidation in a proton exchange membrane fuel cell (PEMFC) occurring at lower voltages than previously thought. These findings challenge existing degradation models and call for new pathways to improve fuel cell durability.

Share

As the world shifts towards decarbonized energy, PEMFCs represent a promising zero-emission power generation technology, converting hydrogen into electricity with only water and heat as by-products. However, the long-term stability of the required platinum-based catalysts remains a critical challenge in improving their cost-efficiency and reliability.

Platinum oxidation is widely recognized as a key factor in catalyst degradation, leading to reduced performance and a shorter lifespan. Until now, most studies have focused on oxidation occurring above 1.0 V versus the reversible hydrogen electrode (RHE), conditions not typically encountered during practical PEMFC operation. This research provides new evidence that platinum oxidation can occur even at lower potentials, fundamentally challenging the established understanding of catalyst degradation mechanisms.

To investigate platinum oxidation under real operating conditions, researchers employed stroboscopic operando high-energy wide-angle X-ray scattering (WAXS) at beamline ID31. Experiments were conducted on a commercial platinum catalyst in a ‘model’ cell configuration (liquid electrolyte at room temperature) and on a commercial membrane electrode assembly (MEA) in an X-ray-transparent fuel cell setup, under realistic operating conditions at 80°C (Figure 1a).

The stroboscopic strategy involved merging together signals collected from consecutive platinum oxidation and reduction cycles, thereby accumulating beam exposure time for different cell voltages. This approach enabled the capture of fast structural changes in platinum nanoparticles with high spatial and temporal resolution. X-rays tracked changes in platinum surface crystallinity, serving as a direct indicator of oxidation (as confirmed by density functional theory (DFT) calculations in Figure 1b).

Fig.1_chattot.jpg

 

Click figure to enlarge

Fig. 1: Methodology of the study, integrating (a) structural inputs from operando WAXS and (b) DFT calculations.


In parallel, WAXS data were complemented by electrochemical online inductively coupled plasma mass spectrometry (ICP-MS) to correlate oxidation events with platinum dissolution. The combination of these advanced techniques provided unprecedented insights into the oxidation dynamics of platinum catalysts at the nanometre scale.

The main result of the study was the observation of a metal-oxide phase transition in platinum catalysts occurring at potentials as low as 0.80 V vs. RHE, well below the previously reported range of 1.05–1.10 V. This transition is characterized by a loss of surface crystallinity and the formation of an amorphous oxide layer, which was clearly resolved as a function of electrode potential using WAXS (Figure 2a).

Pair distribution function (PDF) calculations were performed to analyse the local structure of the amorphous phase formed. While a platinum-platinum (Pt-Pt) distance of 3.3 Å was observed – indicating a structure distinct from metallic platinum – the exact composition of the platinum oxide did not correspond to classical PtO2, indicating the need for further investigations (Figure 2b).

Complementary measurements of platinum dissolution rates using online ICP-MS confirmed that this oxidation process significantly contributes to platinum loss (Figure 2c-d). Interestingly, PEMFC conditions (high temperature, low pH) were found to enhance platinum dissolution compared to model laboratory conditions. Furthermore, platinum dissolution in PEMFC conditions was higher at 1.0 V vs. RHE than at higher potentials.

Fig.2_chattot.jpg

Click figure to enlarge

Fig. 2: Platinum oxidation in PEMFC and its impact on metal dissolution. a) Lattice constant and scale factor variations (orange curves) as a function of cell voltage. The blue curve indicates the emergence of a non-metallic Pt-Pt distance revealed by PDF analysis of WAXS patterns. b) Waterfall plot of the differential PDFs showing the evolution of the oxide phase over time. The observed phase does not correspond to classical PtO2 oxide. c) Pt dissolution rate under different conditions following a potential step up to 1.23 V vs. RHE. d) Pt dissolution rate under different conditions following a potential step up to 1.0 V vs. RHE.


These findings have profound implications for fuel cell durability. Since typical PEMFC operation occurs within a potential range of 0.6–1.0 V, the study demonstrates that platinum oxidation is not limited to extreme operating conditions but is an ongoing process during normal fuel cell use, explaining the gradual platinum loss over time.

The final part of the study focused on platinum oxidation and reduction kinetics. The high temporal resolution provided by the stroboscopic strategy enabled researchers to demonstrate that while platinum reduction is fast, platinum oxidation at 1.0 V vs. RHE is relatively slow, with a first-order time constant of approximately three seconds. This finding reveals a fundamental limitation of using ‘accelerated’ degradation studies to predict platinum catalyst lifetime.

Future research will focus on optimizing mitigation techniques and exploring alternative materials with enhanced oxidation resistance. By leveraging the capabilities of the ID31 beamline, further studies can refine our understanding of catalyst behaviour under real-world conditions, ultimately accelerating the development of more robust and efficient fuel cell technologies.


Principal publication and authors
Metal-oxide phase transition of platinum nanocatalyst below fuel cell open-circuit voltage, C.A. Campos-Roldán (a), A. Gasmi (a), M. Ennaji (a), M. Stodel (b), I. Martens (c), J.-S. Filhol (a), P.-Y. Blanchard (a), S. Cavaliere (a), D. Jones (a), J. Drnec (c), R. Chattot (a), Nat. Commun. 16, 936 (2025); https://doi.org/10.1038/s41467-024-55299-3
(a) ICGM, Univ. Montpellier, CNRS, ENSCM, Montpellier (France)
(b) CIRIMAT, Université Toulouse 3 Paul Sabatier, CNRS, Toulouse (France)
(c) ESRF


References
[1] K. Sasaki et al., ACS Catal. 6(1), 69-76 (2016).
[2] T. Fuchs et al., Nat. Catal. 3(8), 754-761 (2023).

 

About the beamline: ID31

Beamline ID31 is dedicated to the study of interfaces and materials processing using high-energy X-rays and flux provided by the EBS. It offers a range of hard X-ray characterization techniques, including reflectivity (XRR), wide- and small-angle X-ray scattering (WAXS and SAXS), and diffraction in both transmission and grazing incidence geometries, with the option to integrate imaging methods. The beamline’s optical design enables rapid adjustments of beam properties, such as photon energy, bandwidth, and focus, accommodating complex and heavy experimental setups.

Equipped with a state-of-the-art Pilatus CdTe 2M detector, along with CMOS and other imaging detectors, ID31 supports multimodal characterization and simultaneous operando experiments, including electrocatalysis, batteries, and additive manufacturing. Specialized sample environments include grazing incidence electrochemical cells, fuel cells, electrolyser cells, microfluidic systems, and hydrogen storage reactors. These capabilities enable the investigation of energy materials under real operating conditions, bridging the gap between fundamental and applied research.

Through the combination of diffraction, scattering, and imaging, ID31 provides a comprehensive approach to understanding the relationship between microscopic material properties and macroscopic device performance, driving the optimization of next-generation energy technologies.