Water splitting on the Basal Plane of Graphene Grown on Nickel

Although graphene (Gr) has significant potential in hydrogen production and storage, its chemical neutrality poses limitations to the practical realization of such applications. When there is a strong interaction with the underlying metal support, a large charge transfer may occur at the film-metal interface, likely leading to the formation of a dipole layer capable of catalyzing chemical reactions. In this context, Gr/Ni(111) has often been identified as one of the most reactive Gr/metal systems.  Notably, it has been recently demonstrated that when Ni single crystals covered by Gr monolayers are exposed to water vapor doses sufficient to enable intercalation underneath Gr, a fraction of the intercalated water molecules can undergo dissociation, facilitated by the presence of Stone–Wales defects in the Gr lattice. This result, although of great interest, does not address the key question, namely whether the Gr/Ni(111) interface itself possesses intrinsic catalytic activity for water splitting.

To answer this question, we focused on the early phase of the interaction between water and Gr/Ni(111) by performing X-ray photoelectron spectroscopy (XPS) combined with temperature programmed desorption (TPD) at the SuperESCA beamline of Elettra. Synchrotron experiments were complemented by laboratory scanning tunneling microscopy (STM), surface X-ray diffraction (XRD) and were substantiated by density functional theory (DFT) calculations.  Figure 1a shows the C 1s spectrum recorded for Gr/Ni(111) after exposure to 2×10⁴ L (1L =1.33×10^-6 mbar s) of H₂O molecules at room temperature. Notably, the spectrum exhibits the same components observed for hydrogenated Gr/Ni(111) after exposure to a low dose of atomic hydrogen, suggesting that water splitting has occurred. Gr hydrogenation was confirmed by studying the evolution of the C1s core level photoemission during sample annealing (Fig. 1b), which shows the disappearance of the C-H bonds around 600 K - the same temperature at which desorption of molecular hydrogen is detected by TPD. Importantly, surface XRD measurements ruled out the intercalation of water molecules, demonstrating that water dissociation occurs on top of Gr, not beneath it. STM imaging of the clean Gr/Ni(111) surface (Fig. 1c) reveals the presence of some dispersed adatoms on the Gr lattice, likely corresponding to Ni atoms trapped at C vacancies. After a water dose of 4.5×10³ L the Gr layer appears covered by a uniform distribution of H atoms, which remains stable even near the step edge, strongly supporting the conclusion that water dissociates on the basal plane of Gr (Fig. 1d).

Figure 1: a) C 1s core level spectrum measured for the Gr/Ni(111) surface exposed to 2×10⁴ L of water at RT.  The C₀ component represents non-hydrogenated C atoms; C atoms directly bonded to H contribute to A2  (H monomers and dimers) and to A₁ (H trimers or larger clusters), whereas graphene sites neighboring C–H bonds originate the B1  (neighbors of one and two C–H bonds) and the B₂ (neighbors of three C–H bonds) components.  b) Integrated areas of the C 1s components vs. temperature.  (c-d) STM images of the Gr/Ni(111) surface before and after the exposure water: (c) image of the Gr honeycomb lattice with a few nanostructures, likely Ni adatoms trapped at C vacancies; (d) images of the Gr layer evenly covered by H atoms also in the vicinity of a step edge after the exposure to 4.5×10³ L of water. Adapted from Carbon 243, 120422 (2025); DOI: 10.1016/j.carbon.2025.120422, under a Creative Commons CC-BY license.

To understand the mechanism driving the splitting of a single water molecule, we evaluated the energetics of the reaction by performing DFT calculations. On as grown Gr/Ni(111) the dissociative water chemisorption turns out to be an endothermic process, incompatible with the prominent dissociation yield observed experimentally. The presence of both a pre-adsorbed H atom and a C monovacancy, renders water dissociation exothermic. However, in both cases the energy barrier to reach the final state indicates that the reaction would require the input of external energy. A somewhat stronger substrate reactivity is observed when the C monovacancy is saturated by a Ni adatom (see Fig. 1c). In this configuration, the final state is reached by overcoming a barrier of 0.8 eV (Fig. 2), a value consistent with room-temperature dissociation at a reasonable rate. The energy gained in the reaction can be used by the H and OH fragments to diffuse away from the defect site, which then remains free to catalyze new dissociation reactions. Therefore, the dissociation of isolated water molecules can occur on the basal plane of graphene at sites where Ni atoms are trapped at C vacancies. The limited surface density of such structural inhomogeneities may not fully account for the H coverage observed by XPS and may instead suggest the occurrence of collective adsorption processes. Cooperative effects that can lower the dissociation energy barrier of water molecules adsorbing as dimers or larger clusters are likely to occur at the pressures used in our experiments. Nonetheless, the role of trapped Ni atoms in initiating the splitting reaction locally may be crucial in triggering an autocatalytic process, which is facilitated by adsorbed water fragments. 

Figure 2: Configurations for a dissociated water molecule on the plain Gr/Ni(111) surface and in the presence of an extra H atom pre-adsorbed on Gr, a C vacancy and a C vacancy site filled by a trapped Ni atom. The right panel shows the calculated energy barriers for the three exothermic dissociation routes. Adapted from Carbon 243, 120422 (2025); DOI: 10.1016/j.carbon.2025.120422, under a Creative Commons CC-BY license.

Further experiments could help clarify the complex mechanism of water dissociation, thereby enabling the optimization of experimental parameters that maximize hydrogen production in this system.

This research was conducted by the following research team:

M. Pozzo^1,2,3, P. Lacovig⁴, M. Bianchi⁴, M. Schied⁴, L. Bignardi⁵, F. Zarotti⁶, R. Felici⁷, D. Alfè^3,8,9, S. Lizzit⁴ and R. Larciprete⁶

¹ Faculty of Technological & Innovation Sciences, Universitas Mercatorum, Roma, Italy 
² Institute for Materials Discovery, UCL East, Stratford, London, UK  
³ Department of Earth Sciences, Thomas Young Center, University College London, London, UK  
⁴ Elettra-Sincrotrone Trieste, Trieste, Italy  
⁵ Department of Physics, University of Trieste, Trieste, Italy  
⁶ CNR- Institute for Complex Systems, Rome, Italy  
⁷ CNR-ISM, Roma, Italy  
⁸ Dip. di Fisica ‘‘Ettore Pancini’’, Università Federico II, Napoli, Italy  
⁹ London Centre for Nanotechnology, Thomas Young Centre, University College London, London, UK

Contact person emails: 

Local contact email: