Group Additivity and Modified Linear Scaling Relations for Estimating Surface Thermochemistry on Transition Metal Surfaces: Application to Furanics

We assess the applicability of DFT-parametrized group additivity (GA) and extended linear scaling relations (ELSRs) to the adsorption properties of biomass furanic derivatives on close-packed metal surfaces. We find that previous schemes are inadequate for larger furanic derivatives and even small unsaturated species, such as ethylene, and identify sources of error. We propose an interpolative GA-LSR scheme that accounts for (1) dispersion forces, (2) a modified valence model for unsaturated compounds, which depends on the number of metal atoms occupied by heteroatoms of an adsorbate, and (3) hydrogen-metal interactions, captured using a secondary atomic binding energy descriptor. This new model requires only the atomic carbon and oxygen adsorption energies on a metal surface to estimate the thermochemistry of a given furanic derivative. It predicts the preferred adsorption mode (di-sigma(CC) vs mu(CC)) of unsaturated species and the conformation (flat, tilted, vertical) of furanics and greatly reduces the error in the estimated species enthalpy of formation H-f,H-298.