Abstract
The construction of high-dimensional global potential energy surfaces (PESs) from ab initio data has been a major challenge for decades. Advances in computer hardware, electronic structure theory, and PES fitting methods have greatly alleviated many challenges in PES construction, but building fitting sets has remained a bottleneck so far. We present the robosurfer program system that completely automates the generation of new geometries, performs ab initio computations, and iteratively improves the PES under development. Unlike previous efforts to automate PES development, robosurfer does not require any uncertainty estimate from the PES fitting method and thus it is compatible with the permutationally invariant polynomial (PIP) method. As a demonstration we have developed five related but different global reactive PIP PESs for the CH3Br + F- system and used them to perform quasiclassical trajectory (QCT) reaction dynamics simulations over a wide range of collision energies. The automatically developed PESs show good to excellent accuracy at known stationary points without any manual sampling, and QCT results indicate the lack of unphysical minima on the fitted surfaces. We also present evidence suggesting that the breakdown of single reference electronic structure theory may contribute significantly to the fitting errors of global reactive PESs.
Highlights
The reactive collision of atoms and molecules has been a cornerstone of chemical kinetics ever since the proposal of collision theory[1] in the early 20th century
To address the shortcomings of the traditional techniques of the fitting set generation we have developed the ROBOSURFER program system that integrates the permutationally invariant polynomial (PIP) potential energy surfaces (PESs) fitting and quasiclassical trajectory (QCT) programs previously used by our group and several new pieces of software
We have developed five full-dimensional reactive PESs for the CH3Br + F− system and used them to perform QCT reaction dynamics simulations over a wide range of collision energies
Summary
The reactive collision of atoms and molecules has been a cornerstone of chemical kinetics ever since the proposal of collision theory[1] in the early 20th century. Employed sparse sampling methods include generating systematic or random displacements along the normal modes of the reactants, known stationary points and known products,[5,6,19] sampling along known minimum energy paths (MEPs),[6,14,19] Sobol sequences,[16] running direct classical dynamics,[5,19] and running classical[6,14] or quantum dynamics[29] using a preliminary PES While these methods (or a combination of them) can provide reasonably accurate PESs that are suitable for QCT and quantum dynamics simulations,[5,18,30−35] they suffer from a number of downsides. It remains to be seen if this approach is suitable for creating high-dimensional reactive PESs for reaction dynamics purposes After this brief survey of literature, it appears that almost all examples of automated PES development tools rely on the uncertainty supplied by the fitting method to choose new geometries. The paper is arranged as follows: Section II details the structure and operation of the ROBOSURFER program system and its components, Section III presents a rebuilding scheme, Section IV contains computational details pertaining to the developed PESs and the QCT results, Section V discusses the properties and accuracy of the PESs and presents the QCT results, and Section VI gives a condensed summary of the results and conclusions and an outlook on further research
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
