Generalized adsorption isotherms for molecular and dissociative adsorption of a polar molecular species on two polar surface geometries: Perovskite (100) (Pm-3m) and fluorite (111) (Fm-3m)

Abstract

Lattice based kinetic Monte Carlo simulations have been used to determine a functional form for the second order adsorption isotherms on two commonly investigated crystal surfaces: the (111) fluorite surface and the (100) perovskite surface which has the same geometric symmetry as the NaCl (100) surface. The functional form is generalized to be applicable to all values of the equilibrium constant by a shift along the pressure axis. Functions have been determined for estimating the pressure at which a desired coverage would be achieved and, conversely, for estimating the coverage at a certain pressure. The generalized form has been calculated by investigating the surface adsorbate coverage across a range of thermodynamic equilibrium constants that span the range 10−26 to 1013. The equations have been shown to be general for any value of the adsorption equilibrium constant.

Ethanol Conversion over La0.7Sr0.3MnO3–x(100): Autocatalysis, Adjacent O-Vacancies, Disproportionation, and Dehydrogenation

“…Kinetic Monte Carlo (KMC) simulations were performed to compare to the experimentally observed selectivities and assess the viability of the chemical mechanism proposed. KMC simulations were performed using kmos software, and the symmetry of a perovskite lattice as established in a previous publication. , Additional details on the KMC simulations are included in the Supporting Information.…”

Progress in Accurate Chemical Kinetic Modeling, Simulations, and Parameter Estimation for Heterogeneous Catalysis

“…This can lead to simulated catalytic rate differences of up to an order of magnitude . The more accurate calculated free energies can then be incorporated into either kinetic Monte Carlo models ,− or mean field equations. ,− …”

Machine-Learned Corrections to Mean-Field Microkinetic Models at the Fast Diffusion Limit

“…Other functional forms have been employed to model the encounter probability from kMC simulations. 54,55 While sampling around specific values of coverages may improve the neural network performance in that region, we assume no prior information about the system here; therefore, we use a uniform prior to sample the coverage values for our lattice Monte Carlo simulations. Monte Carlo simulations were run for 50 randomly generated sets of coverage values (uniform distribution between 0 and 1) to generate the data for training the neural networks.…”

“…The basic requirement is that the model should be continuous and differentiable in the whole space so that the microkinetic model is solvable using existing solvers. Other functional forms have been employed to model the encounter probability from kMC simulations. , …”