Hydrogen and carbon nanotube production from CH4 decomposition on fe-based nanoparticles: atomistic insights from density functional theory and machine learning molecular dynamics

The dependence of the global energy system on fossil fuels is the primary driver of anthropogenic climate change. In this context, Hydrogen (H2) has emerged as a remarkably versatile energy carrier and chemical feedstock, yet the present supply of low-emission hydrogen remains very limited. Catalytic methane decomposition (CMD) offers a promising alternative in which methane (CH4) dissociates on tailored catalytic nanoparticles (NPs), and the carbon co-product reorganizes into nanostructured materials. In this way, CMD yields CO2-free H2 together with valuable nanocarbons that can improve process economics. Fe-based catalysts are particularly attractive owing to their low cost, abundance, and peculiar carbon affinity. Despite these prospects, several scientific and technological challenges still limit large scale deployment of Fe-based CMD. The identification of thermodynamically and kinetically optimal conditions for CH4 activation remains elusive, and a unified atomistic picture linking NP carburisation, CNT nucleation and growth, catalyst deactivation, and regeneration remains incomplete. To address these gaps, a multiscale computational strategy was employed by means of density functional theory (DFT) and machine learning molecular dynamics (ML-MD). The adsorption, dissociation, and activation energies of methane on pristine and promoted Fe and Fe3C were studied by static DFT calculations. The effectiveness of selected chemical scissors (e.g., H2, CO2, H2O) for CNT etching and detachment was investigated by DFT and ab initio molecular dynamics (MD). MD based on ML inter atomic potential trained on curated DFT data, extended time and length scales beyond ab initio MD and allowed us to observe in real time NP carburisation, cap nucleation, and CNT growth under controlled condition and NP size. The results provide a consistent, quantitative picture of CMD and early stage of CNT growth. Mechanistic insights and effect of catalyst phase, facet, and promoters allowed for informing catalyst design and reactor conditions optimisation for CO2-free H2 and CNT production.