• First-principles investigation of TDMAH adsorption on pristine and oxygen-functionalized graphene surfaces. • DFT analysis shows that COOH, OH, and epoxide groups can enhance surface reactivity. • Proton-donating COOH/OH sites promote strong chemisorption, while epoxide groups act as anchoring sites during the initial half-cycle of ALD. • Oxygen functionalities increase charge transfer, strengthen interfacial Hf–O bonding and can provide nucleation centres. • Reactivity contrast revealed by DFT explains the origin of uniform and selective ALD nucleation on modified graphene. Hafnium oxide (HfO 2 ) is a leading high-κ dielectric for graphene-based field-effect transistors yet achieving conformal and defect-free HfO 2 films on graphene remains challenging. The chemical inertness of pristine graphene suppresses precursor chemisorption during atomic layer deposition (ALD), resulting in discontinuous coverage and increased gate-leakage currents. Two-photon laser oxidation (2PLO) is a promising approach to introduce oxygen-containing groups that enhance surface reactivity, but a systematic atomistic comparison of TDMAH adsorption on pristine graphene and oxygen-functionalized graphene has not yet been reported. This study employs first-principles density functional theory (DFT) to investigate the adsorption behaviour of tetrakis(dimethylamido)hafnium (TDMAH) on graphene during the initial half-cycle of ALD. Adsorption energy, charge density difference (CDD), projected density of states (PDOS) and Bader charge analyses reveal that pristine graphene exhibits weak physisorption, hydroxyl and carboxyl groups promote strong chemisorption, whereas epoxide groups show intermediate behaviour. A higher density of oxygen functionalities is suggested to provide more nucleation sites and uniform HfO 2 growth. These findings explain the nucleation delay on inert region and define a practical selectivity window during early growth stages. The results provide an atomistic framework for interpreting selective ALD behaviour on graphene and support process optimisation for dielectric integration in advanced nanoelectronics and quantum applications.
Khosravi et al. (Sun,) studied this question.