Predicting isotopologue abundances in the products of organic catagenesis with a kinetic Monte-Carlo model

Thermal cracking of complex organic matter can occur under many geological settings. However, the role of thermal cracking vs. other chemistries (e.g., metallic catalysis or Fischer-Tropsch-type reactions) in petroleum formation remains controversial. Realistic modeling of isotope effects in chemical reaction networks involving thermal cracking might shed light on this problem, especially given the recent progress on measurements of intramolecular stable isotope distributions in organic compounds. Previously published models of thermal cracking in petroleum formation are incapable of predicting the intramolecular isotope patterns of products because they do not incorporate realistic precursors, elementary reactions and patterns of inheritance. In this study, we develop a kinetic Monte-Carlo (kMC) model to address this problem. We simulate thermal breakdown of different types of organic matter that is represented by molecular models. At the onset of each simulation, we initialize the model parent organic molecules with isotopic substitutions, and then subject them to free radical chain reactions in a many-step process. Every simulation captures a possible route of thermal degradation and tallies the numbers of each unique isotopologue of all product molecules at the end. Although this model produces data that contain information for all molecules and isotopic forms, in this study we focus on the proportions of many of the isotopologues of all of the C1-C7 n-alkanes. We use two chemistry schemes that differ in complexity. The basic scheme (scheme A) includes only homolytic cleavage and capping of metastable radicals by hydrogen atoms. The more sophisticated scheme (scheme B) includes most reactions of importance in the free radical chain mechanism of thermal cracking. We find significant differences between the schemes A and B in predicted molecular and isotopic compositions of products. Scheme B is more consistent with natural data than scheme A, suggesting that full radical mechanisms should be considered in models of thermal cracking in natural hydrocarbon formation. Our model is further validated by reproducing results from hydrous pyrolysis experiments. For simulating natural petroleum formation, we found that the kMC model is acceptable at low maturity, but cannot match compositions and time estimates of mature gas formation, suggesting the influence of alternative mechanisms in late-stage, high-thermal-maturity catagenesis. Using our model, we provide mechanistic explanations for some of the existing observations, such as the evolution of intramolecular and intermolecular carbon isotope compositions with thermal maturity. Our study also makes predictions of intramolecular isotope compositions for higher-order alkanes (C4+) that have been little studied to date but present attractive targets for future measurements.