Benchmark First-Principles Calculations of Adsorbate Free Energies

Abstract

Adsorbate free energies are fundamental quantities in the microkinetic modeling of catalytic reactions. In first-principles modeling, finite-temperature free energies are generally obtained by combining density functional theory energies with standard approximate models, such as the harmonic oscillator, the hindered translator, or the two-dimensional ideal gas. In this work, we calculate accurate free energies directly from first-principles potential energy surfaces combined with exact quantum mechanical solutions for the translational energy states to benchmark the reliability of common approximations. Through a series of case studies of monatomic adsorbates on metal surfaces, we show that no one free energy model performs satisfactorily in all cases. Moreover, even combinations of different approximations sometimes deviate significantly from the free energies calculated by our first-principles approach. Using observations from these case studies, we discuss how a full quantum mechanical approach can be extended to calculate accurate free energies for arbitrary adsorbate potential energy surfaces at computational cost similar to standard models.

NO Binding Energies to and Diffusion Barrier on Pd Obtained with Velocity-Resolved Kinetics

“… 29 , 30 This approach has been suggested earlier for applications of surface reactions, 28 and it is considered an accurate method to describe the partition function. 31 We define Q x as where Q x cHO is the partition function of the classical harmonic oscillator for hindered translation ( eq 9 ). As mentioned above, Q x qHO is a poor approximation of Q x at high temperatures, while Q x clas is not an appropriate approach at low temperatures.…”

Unraveling the Role of the Rh–ZrO2 Interface in the Water–Gas-Shift Reaction via a First-Principles Microkinetic Study

“…Even if the exact DFT functional is available, the evaluation of free-energies is not easy and may contribute to observed differences between experimental and computational results. 96 The construction of an atomistic model for the catalyst/support-interface introduces ad-…”

Progress in Computational and Machine‐Learning Methods for Heterogeneous Small‐Molecule Activation

“…[236] The recent publication by Bajpai et al shows that commonly used statistical mechanics methods such as harmonic oscillator, hindered translator, and the 2D ideal gas are not satisfactory in describing the energetics compared to the exact solution obtained by surveying the potential surface. [237] While the field has been addressing the effect of the solvation with implicit and explicit solvation method, obtaining the more accurate Gibbs free energy of solvation for energies are still difficult without using molecular dynamics ensemble. Continued improvement is critical to improving the predictability of the computational method.…”