Thermochemistry Prediction and Automatic Reaction Mechanism Generation for Oxygenated Sulfur Systems: A Case Study of Dimethyl Sulfide Oxidation

Abstract

Automatic mechanism generation is a powerful approach to understanding complex chemical mechanisms. Here we describe the expansion of an open source mechanism creating software, the Reaction Mechanism Generator or RMG, to oxygenated sulfur systems. This principally involved the expansion of thermochemistry estimation techniques for higher valence oxygenated sulfur molecules. As a demonstration of this new tool, we present an automatically generated mechanism describing the oxidation of dimethyl sulfide. This mechanism is then compared to several experimental studies of dimethyl sulfide oxidation. This case study provides insight into the transformation of sulfur species in the atmosphere, especially the formation pathways of sulfoxides, sulfones, and sulfur dioxide. However, the applications of the newly enhanced mechanism generation tool extend far beyond this single system with obvious use in understanding a variety of atmospheric, petrochemical, and other industrial chemistries.

Reaction Mechanism Generator v3.0: Advances in Automatic Mechanism Generation

“…40 Furthermore, the group additivity estimator in RMG has been significantly expanded for sulfur compounds, with the addition of 200 new group values for various C/H/O/S groups. 31 These values were fitted from a collection of thermochemical data derived from quantum chemistry calculations.…”

“…The inclusion of these new atom types enables RMG to model a wider range of chemical systems. The new nitrogen and sulfur atom types were applied in recent models for ethylamine oxidation and dimethyl sulfide oxidation . More broadly, the new sulfur atom types allow RMG to handle zwitterionic sulfur motifs and hypervalent sulfur motifs such as sulfone, sulfoxide, and sulfonanilide groups, enabling future work on oxidation of organosulfur compounds.…”

Gas‐phase kinetic of boron carbide chemical vapor deposition using BCl3+CH4+H2 mixture

“…17 It has been extensively used to generate gas-phase combustion of hydrocarbons species, including carbon, nitrogen, hydrogen, oxygen, chlorine, and sulfur elements. 23,24 The RMG saves the species in the form of graphs, with atoms as nodes and chemical bonds as edges. 17 It uses the notion of reaction families to create all the plausible elementary reactions through reaction templates and recipe actions.…”

“…This work employs an RMG, open‐source software, to build a detailed reaction mechanism for the BCl 3 +CH 4 +H 2 mixture 17 . It has been extensively used to generate gas‐phase combustion of hydrocarbons species, including carbon, nitrogen, hydrogen, oxygen, chlorine, and sulfur elements 23,24 . The RMG saves the species in the form of graphs, with atoms as nodes and chemical bonds as edges 17 .…”

Moving from postdictive to predictive kinetics in reaction engineering

“…For example, the predictions for the thermochemistry of molecules containing fused rings or multiple substitutions on an aromatic ring are still unreliable, despite significant efforts to improve them 63‐65 . Also, the group values are unknown for many common functional groups involving heteroatoms, though in recent years many of these values have been filled in using quantum chemistry calculations 66‐69 . Several researchers have proposed replacing Benson‐type estimates with machine‐learning models, this has shown some promise, but so far these models also have limited scope and significant uncertainties 42,70‐72 …”