Harnessing structural disorder: unraveling hydrogen evolution in monolayer amorphous carbon via first-principles simulations and machine-learned potentials

Disorder and defect coordination in the catalytic plane are crucial in enhancing hydrogen production via Hydrogen Evolution Reaction (HER) on two-dimensional catalysts. Amorphous materials are disordered, making them catalytically adaptive for many reactions. In this work, we demonstrate for the first time the HER efficacy of monolayer amorphous carbon (MAC), identify the governing structural features, and compare its performance with crystalline carbon derivatives such as pristine graphene (GE) and various graphyne derivatives. MAC generated from melt-quench simulations revealed a diverse framework of predominantly sp 2 and sp 3 carbons with numerous 5-, 6-, and 7-membered rings. Density Functional Theory (DFT) calculations were performed to investigate free-energy variations in hydrogen adsorption for each material. Crystalline carbon materials possess a limited number of active sites, in which β -graphyne showed the best Δ G H value of + 0.34 eV. DFT adsorption calculations were performed for MAC across 30 distinct isolated local environments to establish a reference understanding of site-dependent hydrogen adsorption behavior. Core structural properties of the environment were analyzed against varying environment radius. Calculations showed Δ G H distribution of MAC in a range of − 0.02 eV to + 1.35 eV. An extended approach employing a MACE MLIP was adopted to evaluate the activity across the entire MAC surface. The finetuned foundation model showed an optimal fit for energy and force corresponding to 1.65 meV/atom and 29.15 meV/Å respectively. The MLIP was benchmarked against DFT on independently sampled adsorption sites, achieving a mean absolute error of 0.161 eV in Δ G H and a classification accuracy of 95% in distinguishing active from inactive sites. The MLIP-predicted Δ G H values ranged from − 0.91 eV to + 1.70 eV for a larger MAC system, with approximately 15% of the sites exhibiting values below + 0.25 eV. Local feature analysis of individual adsorption sites in MAC revealed that 7-membered rings, local bond distortion, and ripple height in the structure enhance HER activity. Our findings suggest that, with careful optimization of the local features and their corresponding synthesis routes, MAC can be tuned to achieve Δ G H values approaching those of noble-metal catalysts, positioning it as a promising candidate for further kinetic and experimental validation.