Green hydrogen made by splitting water could become far cheaper if the anode reaction stops wasting energy, and a new study reports a catalyst that does exactly that by carefully balancing two design knobs that researchers usually treat separately. Chemists engineered a five-metal layered double hydroxide in which configurational entropy and atomic-scale defects were tuned in a controlled sequence, revealing that the two factors are coupled rather than simply additive. The optimized catalyst, labeled HE-D-Mn-LDH, needed an overpotential of just 256 millivolts to drive the oxygen evolution reaction at a current density of 10 milliamperes per square centimeter, and it ran steadily for 450 hours at 100 milliamperes per square centimeter without obvious degradation.
The work, published as an open-access original article in the journal Carbon Neutrality, tackles the oxygen evolution reaction, the sluggish four-electron half reaction at the anode that limits the overall efficiency and cost of electrolyzers. Benchmark catalysts such as iridium dioxide and ruthenium dioxide are highly active, but their scarcity, expense and instability under operating conditions make them impractical for large-scale deployment. That has pushed intense interest toward earth-abundant transition metal hydroxides and oxides, which balance reasonable activity with scalability.
Two strategies dominate modern electrocatalyst design for this reaction. High-entropy materials mix many elements in near-equal proportions, and their elevated configurational entropy brings enhanced structural stability, tunable electronic structures and synergistic interactions among the constituent elements. Defect engineering, meanwhile, deliberately introduces atomic-scale imperfections, particularly cation vacancies, which create coordinatively unsaturated sites that favor the adsorption and transformation of reaction intermediates. Both approaches have shown promise independently, but their interplay has remained poorly understood, because in most synthetic routes increasing defect density inadvertently alters composition and therefore entropy, making it impossible to isolate the individual contributions.
The research team, led by Dongliang Zhang, Fei Wei and Xia Long with colleagues, solved this problem with a pentanary FeAlNiCoZn high-entropy layered double hydroxide used as a tunable model system. The inherent stability of high-entropy systems allows them to accommodate a high density of defects without phase separation or structural degradation. The key trick was selective alkaline leaching of zinc, an amphoteric element that dissolves under mild basic conditions. Removing zinc simultaneously creates cation vacancies and lowers configurational entropy, so the material slides along a well-defined entropy-defect axis as etching proceeds.
The measurements traced that axis precisely. The pristine material showed a configurational entropy of 1.53 R, calculated from experimentally determined elemental compositions with uncertainty intervals of roughly 0.009 to 0.022 R from triplicate measurements. After half an hour of alkaline treatment, partial zinc leaching reduced entropy to 1.45 R; two hours of etching removed nearly all the zinc, dropping entropy to 1.37 R and marking a transition toward a medium-entropy regime; and fifteen hours drove it down to 1.08 R. Across the same series, the estimated vacancy concentration climbed from about 7.5 percent to about 52.6 percent, and electron paramagnetic resonance showed a monotonic rise in defect density with etching time.
Electrochemical testing in 1 molar potassium hydroxide revealed a volcano-shaped relationship. As entropy fell from 1.53 R to 1.45 R, the overpotential at 10 milliamperes per square centimeter dropped from 337 to 314 millivolts and the Tafel slope improved from 75 to 55 millivolts per decade. Pushing entropy further down to 1.37 R reversed the trend and degraded performance. Excessive etching continued to raise the electrochemically active surface area, with double-layer capacitance increasing from 0.43 to 0.54 millifarads per square centimeter, yet the turnover frequency collapsed from 0.72 to 0.092 per second, showing that severe entropy loss disrupts the cooperative electronic interactions among the metals and suppresses intrinsic activity even as more sites appear.
Spectroscopy explained what the defects were doing. X-ray photoelectron spectroscopy showed positive binding-energy shifts for nickel, iron and cobalt after etching, with partial oxidation of nickel to the trivalent state and iron averaging slightly above plus three, states widely recognized as active centers for oxygen evolution. Synchrotron X-ray absorption spectroscopy confirmed these surface findings in the bulk, with the nickel and iron K-edge absorption edges shifting to higher energy, while extended X-ray absorption fine structure showed that the local metal-oxygen and metal-metal coordination shells remained intact, meaning the electronic modulation happened within the original framework without phase decomposition.
With the optimum etched host in hand, the team applied an etch-and-fill strategy, impregnating manganese ions into the vacancy-rich framework. Manganese was chosen for its multiple accessible oxidation states, its ionic radius comparable to the host cations, and its role in the natural oxygen-evolving cluster of photosystem II. The dopant restored and even raised configurational entropy to 1.607 R, above both the etched sample at about 1.445 R and the pristine material at 1.525 R. Elemental mapping showed manganese distributed uniformly without clusters, and control experiments in which pristine, unetched material received the same manganese treatment showed no obvious improvement, supporting the idea that the zinc-leaching defects are what enable incorporation.
The finished catalyst outperformed every intermediate and, notably, a directly synthesized five-metal hydroxide of identical overall composition prepared by conventional hydrothermal methods, which managed only 268 millivolts. That comparison underscores that elemental composition alone does not guarantee activity; the spatial arrangement of defects and dopants is decisive. The optimized sample also showed the smallest Tafel slope at 43.7 millivolts per decade, the lowest charge-transfer resistance at 2.1 ohms, and a Faradaic efficiency of about 99.3 percent, indicating the measured current came from water splitting rather than side reactions or catalyst degradation.
Evidence points to a lattice oxygen participation mechanism underlying the gains. In a bulky trimethylammonium hydroxide electrolyte, which suppresses lattice-oxygen pathways, the new catalyst showed much stronger activity decay than the pristine material, and in situ Raman spectroscopy revealed a growing feature near 1065 wavenumbers assigned to peroxo-like oxygen species, alongside enhanced nickel-oxygen vibrations indicating formation of a thin reconstructed nickel oxyhydroxide surface layer. The authors caution that the volcano-shaped relationship applies specifically to the alkaline etching series and that rigorous theoretical modeling of these disordered high-entropy systems remains an open challenge. Even so, the demonstration that entropy and defects must be jointly optimized, rather than independently maximized, offers a generalizable design principle for multicomponent electrocatalysts for green hydrogen and beyond.
Subject of Research: Entropy-defect co-engineering of high-entropy layered double hydroxide electrocatalysts for the oxygen evolution reaction in water electrolysis
Article Title: Balancing high entropy complexity and defects for efficient green hydrogen production
Article References: Zhang, D., Wei, F., Zhan, Y., Lu, H., Shen, G., Qian, X., & Long, X. (2026). Balancing high entropy complexity and defects for efficient green hydrogen production. Carbon Neutrality, 5(1), Article 22. https://doi.org/10.1007/s43979-026-00176-9
Image Credits: AI Generated
DOI: 10.1007/s43979-026-00176-9
Keywords: green hydrogen, water electrolysis, oxygen evolution reaction, high-entropy materials, layered double hydroxides, defect engineering, configurational entropy, cation vacancies, manganese doping, electrocatalysis, lattice oxygen mechanism, overpotential
