Bisulfate and Sulfate Poisoning of PtCo Nanocatalysts: Effects on Oxygen Reduction Reaction Kinetics

Background

Polymer electrolyte membrane fuel cells rely on Pt‑based catalysts to reduce oxygen efficiently. Pt alloys, such as PtCo, have shown higher ORR activity than monometallic Pt [1–16]. Yet, the catalytic surface is prone to anion adsorption and surface‑oxide formation, both of which are potential dependent [17–19]. Even trace concentrations of sulfate, bisulfate, and halide species are inevitable in fuel‑cell systems due to ionomer degradation and trace impurities. Previous work has characterized anion adsorption on flat Pt surfaces and bulk Pt alloys [20–34], but systematic studies on carbon‑supported PtCo nanoparticles remain scarce.

Given the high sulfonate content of perfluorosulfonic ionomers, bisulfate and sulfate ions are especially problematic. Recent observations in single‑cell operation [38] revealed a linear relationship between Pt/C activity and log(anion concentration). Building on this, we investigate how bisulfate and sulfate ions quantitatively impact ORR kinetics on PtCo nanoparticles across multiple potentials, providing the first detailed assessment for practical carbon‑supported catalysts.

Methods

Materials

The catalyst employed was 30 wt.% PtCo dispersed on high‑surface‑area carbon (Tanaka Kikinzoku, Japan). A 15 mL ink comprising ultrapure water (Milli‑Q), 2‑propanol (HPLC grade), and 5.37 wt% Nafion solution (ethanol + Milli‑Q) was mixed in a 200:50:1 volume ratio and sonicated for 5 min. Ten microliters of this ink was deposited on a glassy‑carbon disk (0.196 cm²) and dried for 1 h before testing.

Electrochemical Evaluation

RDE experiments were performed in a standard three‑electrode cell using a potentiostat and a rotation controller (Pine Instrument Co). The electrolyte was 0.1 M HClO₄, with bisulfate or sulfate added via H₂SO₄ injections to achieve desired concentrations. Potentials were referenced to the reversible hydrogen electrode (RHE) via a Ag/AgCl reference electrode separated by a salt bridge. ORR curves were recorded at 5 mV s⁻¹ from 0.05 V to 1.0 V under O₂-saturated conditions at 1600 rpm. All measurements were conducted at ambient temperature.

Results and Discussion

Transmission electron microscopy confirmed a uniform particle size distribution of 3–7 nm on the carbon support (Fig. 1). Cyclic voltammetry exhibited symmetrical hydrogen adsorption/desorption peaks at ~0.15 V and Pt oxidation/reduction features at ~0.85 V/0.79 V vs. RHE (Fig. 2a), indicating reversible surface behavior. The measured H₂ adsorption charge (210 µC cm⁻²) corresponds to a BET surface area of 62 m² g⁻¹ for PtCo.

Linear sweep voltammetry (LSV) revealed that increasing bisulfate or sulfate concentration shifts the ORR polarization curves toward more negative potentials, reflecting activity loss (Fig. 2b). The half‑wave potential and limiting current both decline with anion concentration.

a TEM image of the nanocatalyst, scale bar = 50 nm. b Zoomed‑up TEM image, scale bar = 20 nm

a CV curve at 20 mV s⁻¹. b LSV curves at 0–100 mM bisulfate/sulfate

Figure 3a demonstrates a semilogarithmic linear dependence of ORR specific activity (I) on bisulfate concentration (C_HSO4), described by:

Here, G represents the activity at zero anion concentration, and D quantifies the slope of the decline. Across potentials from 0.88 to 0.95 V, the relationship holds robustly. The measured ORR activity at 0.9 V and 0 mM bisulfate (521 µA cm⁻² Pt) aligns with literature values [39].

Adopting a semilogarithmic adsorption isotherm for bisulfate on PtCo, the kinetic expression reduces to:

Thus, the exponential coverage term vanishes, indicating that bisulfate adsorption does not modify the activation free energy at a fixed potential. The rate law remains first order with respect to free Pt sites, supporting the conclusion that simultaneous occupation of two Pt sites by a single O₂ molecule is unlikely.

Illustration of ORR mechanism

Analysis of the slopes (G and D) versus potential yields a consistent transfer coefficient of α ≈ 0.8 (Fig. 3b), indicating an asymmetric activation barrier. Tafel plots (Fig. 5) show slopes of 77–89 mV dec⁻¹ that are invariant with bisulfate concentration, further confirming that the rate‑determining step remains unchanged by anion adsorption.

Tafel slope of 30 wt.% PtCo at various bisulfate concentrations

Conclusions

Bisulfate and sulfate ions linearly suppress ORR activity on PtCo nanoparticles, with a transfer coefficient of ~0.8. Importantly, anion adsorption does not alter the activation free energy at a given potential, and the ORR mechanism remains first order in free Pt sites. These findings suggest that surface blockage, rather than kinetic pathway alteration, underlies the observed poisoning.

Abbreviations

CV:

Cyclic voltammetry

LSV:

Linear sweep voltammetry

ORR:

Oxygen reduction reaction

PEMFC:

Polymer electrolyte membrane fuel cell

RDE:

Rotating disk electrode

RHE:

Reversible hydrogen electrode

RRDE:

Rotating ring‑disk electrode

TEM:

Transmission electron microscopy

XAS:

X‑ray adsorption spectroscopy