Molecular-dynamics simulations of nickel oxide surfaces

Abstract

Molecular-dynamics-simulation techniques have been used to investigate the effect of temperature on the three low index surfaces of nickel oxide. The surface energies for the [100},[110},and Ni [111} surfaces have been calculated as a function of temperature for NiO. The surface energies for all three surfaces were essentially constant up to 2000 K, when surface melting was observed for the less densely packed surfaces such as the [110}and Ni [111}surfaces. The [110}surface reconstructed to give [100} microfacets that are densely packed, and hence the surface disorder was less than the Ni [111 } surface.The defective (Ni'+) Ni [111}surface showed a similar tendency but was less marked, partly due to the increased charge at the surface. The oxidation energy was negative, indicating that thermodynamically the surface would oxidize.

Interpretation of the Ni K-edge EXAFS in nanocrystalline nickel oxide using molecular dynamics simulations

“…The formal ion charges Z (Z=+2 for Ni and Z=-2 for O) were used in the Coulomb potential [23,24]. We found that the Fisher's FF model [23] gives slightly better agreement with the experimental Ni K-edge EXAFS signal for c-NiO than the Oliver's one [24], therefore only the former was used in the simulation of nano-NiO.…”

“…The MD simulations were performed using two FF potential models, developed in the past by Fisher [23] and Oliver [24]. The two models include the same pair interactions, but differ slightly in the numerical values of the FF parameters (Table 2).…”

“…The two force-field (FF) models by Fisher [23] and Oliver [24] were applied to c-NiO (Fig. 5 and Table 2).…”

“…The Buckingham potential parameters (A, ρ, and C in Table 2) were taken from available data [23,24] for c-NiO. The formal ion charges Z (Z=+2 for Ni and Z=-2 for O) were used in the Coulomb potential [23,24].…”

“…The formal ion charges Z (Z=+2 for Ni and Z=-2 for O) were used in the Coulomb potential [23,24]. We found that the Fisher's FF model [23] gives slightly better agreement with the experimental Ni K-edge EXAFS signal for c-NiO than the Oliver's one [24], therefore only the former was used in the simulation of nano-NiO. However, the original value of the Buckingham potential parameter ρ in the Fisher's FF model [23] was adjusted ( Table 2) for nano-NiO to reflect the change in the frequency of the experimental EXAFS signal being related to the lattice expansion ( Table 1).…”