Pynta─An Automated Workflow for Calculation of Surface and Gas–Surface Kinetics

Abstract

Many important industrial processes rely on heterogeneous catalytic systems. However, given all possible catalysts and conditions of interest, it is impractical to optimize most systems experimentally. Automatically generated microkinetic models can be used to efficiently consider many catalysts and conditions. However, these microkinetic models require accurate estimation of many thermochemical and kinetic parameters. Manually calculating these parameters is tedious and error prone, involving many interconnected computations. We present Pynta, a workflow software for automating the calculation of surface and gas–surface reactions. Pynta takes the reactants, products, and atom maps for the reactions of interest, generates sets of initial guesses for all species and saddle points, runs all optimizations, frequency, and IRC calculations, and computes the associated thermochemistry and rate coefficients. It is able to consider all unique adsorption configurations for both adsorbates and saddle points, allowing it to handle high index surfaces and bidentate species. Pynta implements a new saddle point guess generation method called harmonically forced saddle point searching (HFSP). HFSP defines harmonic potentials based on the optimized adsorbate geometries and which bonds are breaking and forming that allow initial placements to be optimized using the GFN1-xTB semiempirical method to create reliable saddle point guesses. This method is reaction class agnostic and fast, allowing Pynta to consider all possible adsorbate site placements efficiently. We demonstrate Pynta on 11 diverse reactions involving monodenate, bidentate, and gas-phase species, many distinct reaction classes, and both a low and a high index facet of Cu. Our results suggest that it is very important to consider reactions between adsorbates adsorbed in all unique configurations for interadsorbate group transfers and reactions on high index surfaces.

Computational Design of 2D Phosphorus Nanostructures for Renewable Energy Applications: A Review

“…For example, the solid electrode interfacial layer formation and effect of electrodes are not adequately considered in energy storage devices, while the intricacy of photoexcitation process with distinct light intensity and wavelength indicates that the generation and migration of charge must be simulated indirectly for photocatalytic and solar cell devices. In addition, the development automated workflows such as Pynta, [ 356 ] GRMM, [ 357 ] and autodE [ 358 ] are crucial for high‐throughput studies of reaction energy profiles in the catalytic community. Overall, a synergistic approach from theoretical and experimental frameworks is required to rationally design and optimize phosphorus‐based nanomaterials in a target application.…”

Diffusion-Limited Kinetics in Reactive Systems

“…Diffusion occurs as a series of individual steps that move a species from one site to another. The rate coefficients for these diffusion steps can be calculated using transition-state theory; however, the flux depends not just on the amount of A but also on the number of empty sites available for A to move to, which depends on the concentrations of all species on the surface.…”

A machine learning based approach to reaction rate estimation

“…Much progress is being made on automatic high accuracy rate calculations based on transition state finding algorithms such as Kinbot, nudged elastic band, HFSP, AFIR, and string methods [6][7][8][9][10][11][12] and workflow codes such as ARC, Pynta, KinBot, and AutoMech. 6,9,13,14 However, for the moment these methods are computationally expensive and not entirely robust, with only limited benchmarking.…”

“…Much progress is being made on automatic high accuracy rate calculations based on transition state finding algorithms such as Kinbot, nudged elastic band, HFSP, AFIR, and string methods [6][7][8][9][10][11][12] and workflow codes such as ARC, Pynta, KinBot, and AutoMech. 6,9,13,14 However, for the moment these methods are computationally expensive and not entirely robust, with only limited benchmarking. 15,16 But even if these methods were robust, it would probably be impractical to use them to compute all the rate coefficients of interest (though perhaps this assessment could change after exascale computers become widely available).…”