Observation of a robust and active catalyst for hydrogen evolution under high current densities

Efficient hydrogen evolution reaction (HER) requires catalysts to overcome activation barriers much larger than the thermodynamic minimum. While many catalysts show excellent activity at low current densities, durable performance at industrially relevant current densities (>500–1000 mA cm−2) remains challenging. Mechanisms governing HER differ between low and high current density regimes: beyond Sabatier-type adsorption energetics, interfacial charge-transfer resistance, intermediate coverage, mechanical stability, and bubble release kinetics become critical at high rates. Tafel slopes commonly reported at low current densities (~30 mV dec−1) increase substantially (>120 mV dec−1) at higher overpotentials, even for Pt, indicating emerging limitations from proton diffusion and hydrogen adsorption. Bubble dynamics can block active sites and induce local strain, impairing activity and stability. Conductivity also plays a crucial role; low-conductivity layered materials (e.g., TMDs) suffer from interlayer barriers and can yield misleading iR-corrected metrics that mask high applied potentials. Layered oxides offer platforms for surface reconstruction and catalyst loading, with in situ exsolution emerging as a route to socketed, stable metal nanostructures. Sr2RuO4 (SRO), a layered perovskite with strong correlations, surface magnetism, and metal-like conductivity, is a promising host, yet prior catalysis studies largely used polycrystalline samples at low current densities. This work addresses whether bulk SRO single crystals can serve as robust HER catalysts under high current density operation and why.

The study builds on insights that layered transition metal oxides can undergo surface reconstruction to expose active sites or host exsolved metallic species for enhanced electrocatalysis. Prior work on TMDs highlighted intrinsic conductivity limitations and the pitfalls of excessive iR correction at high currents. Exsolution in perovskites produces metal nanostructures rooted in the oxide support, improving dispersion, cohesion, electron transfer, and stability under harsh conditions. SRO is a highly conductive layered perovskite exhibiting correlated-electron phenomena (e.g., superconductivity, surface magnetism) that can influence catalysis. Previous HER studies on related ruthenates and perovskites often focused on polycrystalline forms and low current densities; here, bulk SRO single crystals enable direct probing of surface reconstruction, interfacial charge transfer, and durability at industrially relevant currents.

Materials synthesis: Millimeter-scale Sr2RuO4 single crystals were grown by floating-zone method. Precursors SrCO3 and RuO2 (molar ratio 1:0.6) were pelletized and sintered at 1150 °C for 24 h, reground, and formed into feed rods. Growth was conducted in O2/Ar at 0.35 MPa, with typical feed 28–30 mm/h, growth 45 mm/h, counter-rotation at 30 rpm. Plate-like crystals (~3 mm × 2 mm × 0.2 mm; geometric area ~5 mm2) were exfoliated for electrodes. Characterization: Crystal structure verified by Laue XRD; SEM (JEOL JSM 6700 F, 5 kV) for surface/cross-sections; TEM/HRTEM/EELS (TITAN 80/300, 200 kV) on FIB lamellae to observe surface amorphous layers; Raman spectroscopy for surface symmetry; XPS (UHV system with Scienta-200 analyzer, base pressure 2×10−10 mbar; C 1s at 284.6 eV reference) to analyze Sr 3d and Ru 3d states pre- and post-activation; magnetization for detecting ferromagnetic phases; ICP-OES for Sr and Ru leaching quantification post-stability tests; digital image correlation (DIC) for bubble dynamics; contact angle measurements (DCAT21) for wettability. Electrochemical testing: Three-electrode cell (100 mL electrolyte: 0.5 M H2SO4 or 1 M KOH). Reference: Ag/AgCl (3 M KCl); counter: graphite rod (Pt net for 70 °C tests). Working electrode: SRO single crystal attached to Cu wire with silver paint; controls confirmed negligible activity from Cu/Ag. LSV at 1 mV s−1 and CV at 50 mV s−1 recorded using Autolab PGSTAT302N; iR-compensation from EIS-derived solution resistance. Stability via chronoamperometry at fixed overpotentials/currents at 25 °C and 70 °C (water bath). EIS (10 kHz–0.1 Hz, 10 mV AC) with two-time-constant equivalent circuit to extract interfacial charge-transfer kinetics and dielectric interlayer response. Faradaic efficiency by gas collection via drainage at ~280 mV overpotential. Active surface area and TOF: Double-layer capacitance (no faradaic window) used to estimate ECSA; TOF computed from current density and estimated number of active Ru sites assuming top Ru atoms in Ru clusters are active (provided equations and constants). Mass activity estimated assuming HER activity arises from surface Ru clusters. Computations: DFT calculations assessed hydrogen adsorption free energies (ΔGH) on pristine SRO (001) and on Ru6 clusters (various geometries) supported on SRO, including charge-density difference analyses and Bader charge transfer to identify active sites and interfacial charge redistribution. Device-level test: Full electrolyzer assembled with activated SRO/Ru cathode and commercial 10% Ir/C anode; voltage recorded at 1000 mA cm−2 with and without iR compensation.

- Bulk SRO single crystals activate in situ to form a surface amorphous Ru-cluster layer (Ru0 signature at 279.8 eV in XPS), with Sr depletion; clusters exhibit room-temperature ferromagnetism, distinguishing them from bulk Ru and RuO2. - High intrinsic conductivity: ~1×10^6 S/m for SRO (decreasing to ~3×10^5 S/m after long-term testing), comparable to noble metals and far exceeding many chalcogenides/carbons. - Low-current-density performance: overpotential at 10 mA cm−2 is 18 mV (1 M KOH) and 28 mV (0.5 M H2SO4); Tafel slopes 22 mV dec−1 (KOH) and 29 mV dec−1 (H2SO4), comparable to Pt/C (30 mV dec−1). TOF at 100 mV overpotential is 121 s−1, among state-of-the-art. - High-current-density performance: to reach 1000 mA cm−2, overpotentials after iR correction are 182 mV (H2SO4) and 278 mV (KOH); without iR correction: 272 mV (H2SO4) and 354 mV (KOH). The iR difference (~90 mV) is far smaller than many reported high-rate catalysts (e.g., NiMoN, MoNi4, Se/Co–MoS2). - Tafel slopes at high current: ~120 mV dec−1 (acid) and ~200 mV dec−1 (alkaline) near −1000 mA cm−2, comparable to Pt under similar conditions, indicating fast adsorption/desorption kinetics at high coverage. - Stability: sustained −2000 mA cm−2 at −425 mV (no iR correction) for 5 days in acid; in 1 M KOH at 25 °C, ~1000 mA cm−2 stable for 35 days; at 70 °C, ~1300 mA cm−2 stable for 21 days with negligible LSV degradation; ICP-OES shows minor Sr (0.66 ppm) and Ru (0.02 ppm) in electrolyte after long-term tests. - Mass activity: assuming activity from Ru clusters, 16.6 A mg−1 at 50 mV in 1 M KOH, competitive among noble-metal-based HER catalysts. - Faradaic efficiency near 100% at 1000 mA cm−2 by gas collection. - EIS: charge relaxation time decreased from 0.05 s to 0.03 s post-activation for interfacial charge transfer; a fast high-frequency (~10−3 s) dielectric interlayer response observed. Post-activation kinetics faster than commercial Pt/C (0.62 s) and Ru single-atom catalysts (0.08 s). - Bubble dynamics and wettability: contact angle 69.2° (Ru/SRO) vs 80.9° (Pt foil); bubble sizes at 1000 mA cm−2 are smaller on Ru/SRO (~50 µm) than Pt (~100 µm), favoring rapid detachment and reduced polarization. - Mechanism (DFT): pristine SRO has unfavorable ΔGH (1.63–1.76 eV). Ru6 clusters on SRO (stable D4h configuration) exhibit charge redistribution (Bader: Ru1→SRO 0.29e, Ru2→SRO 0.19e). Calculated ΔGH: Ru2 −0.34 eV, Ru1 −0.12 eV (vs Pt ~−0.10 eV), identifying top Ru atoms as active HER sites with optimized H binding.

The work addresses the central challenge of sustaining high HER activity and durability at industrially relevant current densities. Upon electrochemical activation, SRO single crystals reconstruct to form socketed, amorphous Ru clusters anchored to a highly conductive oxide, minimizing interfacial charge-transfer resistance and mechanical degradation during vigorous bubble evolution. The in situ formed Ru clusters exhibit ferromagnetism and distinct electronic structure relative to bulk Ru, with interfacial charge redistribution that tunes H adsorption near the thermoneutral regime. DFT confirms that Ru6 clusters on SRO provide active Ru top sites with ΔGH comparable to Pt, while the SRO support acts as a charge reservoir facilitating rapid electron transfer. Experimentally, the catalysts demonstrate low overpotentials at 1000 mA cm−2 with minimal iR disparity, ultra-low Tafel slopes across regimes, and small, rapidly detaching bubbles due to improved wettability and a porous, hierarchical Ru layer with large ECSA. These factors collectively sustain high mass transport and charge transfer, leading to exceptional stability over weeks in both acidic and alkaline media at room and elevated temperatures. The findings highlight the importance of interface engineering and support conductivity in designing HER catalysts capable of operating under industrial-scale currents.

Bulk Sr2RuO4 single crystals, upon electrochemical activation, develop a robust surface layer of ferromagnetic Ru clusters that deliver outstanding HER activity and durability across acidic and alkaline electrolytes. The catalysts achieve 1000 mA cm−2 at low overpotentials (182 mV in acid; 278 mV in base, iR-corrected), maintain near-100% Faradaic efficiency, and exhibit long-term stability (up to 56 days combined tests) including at 70 °C. Mechanistically, interfacial charge redistribution and high support conductivity optimize hydrogen adsorption and minimize charge-transfer resistance, while improved wettability and a hierarchical Ru layer expedite bubble release. This work establishes interface-tuned, exsolved metal-on-conductive-oxide systems as a viable path for industrial-scale HER catalysts. Future studies could optimize cluster size/geometry and interlayer properties, quantify active-site densities more precisely, and extend the strategy to other conductive oxide hosts and electrolyzer architectures.

- The active catalytic phase arises from in situ reconstruction; performance metrics pertain to the activated Ru-cluster/SRO surface rather than pristine SRO, which has poor ΔGH for H adsorption. - Mass activity estimates assume HER activity originates solely from surface Ru clusters and that top Ru atoms are the active sites, introducing uncertainty in absolute mass-normalized metrics. - A minor Sr3Ru2O7 intergrowth was observed within the bulk crystal (though reported not to alter surface structure or conductivity), and Sr leaching was detected (0.66 ppm after long-term tests), indicating some compositional evolution. - Conductivity of SRO decreases after extended operation (~1×10^6 to ~3×10^5 S/m), which could impact long-term performance in different configurations. - The origin and properties of the fast high-frequency dielectric interlayer response remain under debate, limiting complete mechanistic assignment of EIS features.