Trace Platinum Turns Cheap Sponge Into Durable Seawater Splitting Electrode

Hydrogen made by splitting seawater could become far cheaper if the electrodes doing the splitting did not rely on thick coatings of precious metal. A team reporting in the journal Carbon Neutrality has now built a working electrode that contains just 0.02 atomic percent platinum, yet drives an overall water-splitting current of 200 milliamperes per square centimetre at only 1.846 volts and keeps running for 720 hours at an industrial current density of 0.5 amperes per square centimetre. The trick is not a new catalyst composition so much as a new way of putting one together: a one-step, mild electroless plating process that grows an amorphous nickel boride alloy, decorated with trace platinum, directly onto an ordinary melamine sponge.

The electrode, named Pt-NiB/MS, was made by Xinke Huang, Liugang Wu and colleagues, who immersed the sponge in a plating solution containing platinum ions, nickel ions and dimethylamine borane, a reducing agent. The redox reaction deposits the catalytic layer at room-temperature-friendly, mild conditions, without the high-temperature steps above 120 degrees Celsius that most noble-metal doping protocols demand. Because the process is chemical rather than electrochemical, no external energy input is needed, and the authors report that electrodes ranging from 1 to 25 square centimetres could be fabricated rapidly at 298 kelvin.

Structural characterisation explains why the design works. X-ray diffraction showed peaks typical of metallic nickel alongside broad features from the sponge, while high-resolution transmission electron microscopy revealed nanoparticles wrapped in wrinkled nanosheets, with no distinct diffraction spots in the selected-area electron diffraction pattern. The catalyst is therefore a hybrid of amorphous and crystalline phases: the short-range order of the amorphous regions increases the density of active sites and structural defects, while crystalline nickel domains contribute mechanical stability. Energy-dispersive X-ray spectroscopy and ICP-OES analysis both confirmed the platinum content at roughly 0.02 atomic percent, with molar ratios of about 71.7 percent nickel, 11.3 percent boron and 17.5 percent oxygen.

The electronic consequences of that whisper of platinum are substantial. X-ray photoelectron spectroscopy showed that the doped electrode retains a distinct zero-valent boron component and stronger nickel-boron bonding signals, whereas the undoped NiB/MS sample showed stronger oxidised nickel features and only boron-oxygen peaks. The authors attribute this to platinum’s high work function and electronegativity, which pull electron density toward neighbouring nickel and boron atoms, raising local electron density, suppressing boron oxidation and introducing additional surface disorder. In practical terms, the doped electrode’s charge-transfer resistance fell to about 0.38 ohms for the hydrogen evolution reaction and 0.408 ohms for the oxygen evolution reaction, well below the undoped control.

Performance figures back the mechanism up. For hydrogen evolution in simulated seawater, 1.0 molar potassium hydroxide plus 0.5 molar sodium chloride, Pt-NiB/MS needed an overpotential of just 122 millivolts to reach 100 milliamperes per square centimetre, with a Tafel slope of 57.5 millivolts per decade, compared with 254 millivolts and 115.7 millivolts per decade for the undoped electrode. For oxygen evolution it delivered an overpotential of 287 millivolts at the same current density with a Tafel slope of 42.9 millivolts per decade. Trace platinum doping reduced the overall water-splitting overpotential by 40 millivolts at 200 milliamperes per square centimetre, and control tests confirmed that the bare sponge substrate itself is essentially inactive, so the catalysis comes from the deposited NiB phase and its platinum-modulated surface.

Durability is where sponge electrodes usually fail, and this is where the new results are most striking. In the simulated seawater electrolyte, the electrode maintained its performance over 720 hours of continuous operation at 0.5 amperes per square centimetre, a current density relevant to industrial electrolysis, with only a 7.1 percent decrease in current density. In alkaline real seawater collected from Fuxing Island, Shanghai, it retained 92 percent of its initial performance. Faradaic efficiency measurements, using a gas collection setup that recorded hydrogen and oxygen every 30 seconds, yielded 10 millilitres of hydrogen and 4.95 millilitres of oxygen in 270 seconds, a ratio of actual to theoretical gas volume of 0.99 to 1, indicating efficiency close to unity once potential system leakage is considered.

The corrosion resistance that makes this longevity possible appears to come from surface reconstruction rather than the sponge alone. After oxygen evolution, zeta potential measurements showed the reconstructed surface carried a more negative charge, at minus 9 millivolts versus minus 1.5 millivolts initially, which electrostatically repels chloride ions and shields the electrode from salt attack. Corrosion measurements reinforced the point: the doped electrode showed higher corrosion potentials, 0.218 and 0.196 volts versus a saturated calomel electrode, and lower corrosion current densities, 4.70 and 3.84 microamperes per square centimetre, in simulated and alkaline natural seawater respectively. The authors are careful to note that ex situ FTIR and XPS measurements reveal oxidised boron species on the surface but cannot confirm formation of a specific borate-rich layer in operation, so they describe the mechanism conservatively as overall surface reconstruction and oxidised surface species.

In situ Raman spectroscopy during oxygen evolution traced how the active layer forms. As the potential rose from 1.23 to 1.63 volts, the doped electrode showed nickel-oxygen vibrational bands of nickel oxyhydroxide, NiOOH, emerging at 1.43 volts versus reversible hydrogen electrode, 50 millivolts earlier than the 1.48 volt onset of the undoped sample. Post-electrolysis XPS showed partial dissociation of nickel-boron bonding after hydrogen evolution and a nickel valence transition from the divalent to trivalent state during oxygen evolution, consistent with formation of a surface-reconstructed NiOOH active layer. The authors conclude that trace platinum lowers the energy barrier for this reconstruction from nickel boride to active nickel hydroxide and NiOOH phases, maximising the density of intrinsic active sites while smoothing the redox transitions that preserve long-term efficiency.

The mild plating strategy also proved general. Substituting ruthenium or iridium salts for platinum in the same one-step bath produced Ru-NiB/MS and Ir-NiB/MS electrodes, again with only 0.02 atomic percent noble metal by ICP analysis. These delivered hydrogen evolution overpotentials of 159 and 142 millivolts and oxygen evolution overpotentials of 428 and 425 millivolts at 100 milliamperes per square centimetre, confirming that the electronic-modulation approach is not platinum-specific. The authors also report stable operation of the platinum-doped electrode in wastewater and simulated urine, and successful integration into an anion exchange membrane water electrolysis system, where the self-supporting sponge architecture mitigated gas bubble accumulation.

Cost is the quiet argument underlying the whole study. Nickel and iron foams, the standard three-dimensional substrates for such electrodes, cost roughly 0.008 dollars per square centimetre and suffer corrosion or dissolution under alkaline conditions, while the melamine sponge substrate costs about 0.0003 dollars per square centimetre. The sponge-based electrode can be tuned between flexible and rigid states by adjusting plating time, supporting loads of up to 1.0 kilogram while still bending, and the authors position it as a potential alternative to industrial-grade Raney nickel. Caveats remain: the seawater tested was simulated or from a single Shanghai location, the corrosion mechanism is only partially resolved, and scaling from laboratory cells to industrial electrolysers will demand further validation. But as a demonstration that 0.02 percent of a precious metal, deposited at mild conditions on a cheap sponge, can sustain industrial current densities for a month in salty alkaline electrolyte, the work offers electrolysis developers a notably economical starting point.

Subject of Research: Trace noble-metal-doped amorphous nickel boride sponge electrodes for water electrolysis in seawater

Article Title: Pt (Ru, Ir)-doped amorphous electrocatalysts enabling practical sponge electrodes for industrial-scale seawater splitting

Article References: Huang, X., Wu, L., Wang, Y., Rao, Y., Wang, Y., Weng, S., Qu, S., Chen, Z., & Hao, W. (2026). Pt (Ru, Ir)-doped amorphous electrocatalysts enabling practical sponge electrodes for industrial-scale seawater splitting. Carbon Neutrality, 5(1), Article 14. https://doi.org/10.1007/s43979-026-00170-1

Image Credits: AI Generated

DOI: 10.1007/s43979-026-00170-1

Keywords: water electrolysis, hydrogen production, seawater splitting, electrocatalysis, nickel boride, platinum doping, electroless plating, melamine sponge, oxygen evolution reaction, hydrogen evolution reaction, anion exchange membrane, corrosion resistance