Model development for surrogate fuel components by auto-generation and rate rule optimization

This work presents the development of an automated computational framework for chemical kinetic modelling of large surrogate fuel components, with a focus on large alkenes. The MAMOX++ framework is upgraded and extended to enable the automated generation of detailed kinetic models for large alkenes. By incorporating 52 alkene reaction classes, the framework allows systematic construction of sub-models for a wide range of linear and branched alkenes (> C4) with minimal manual intervention. In combination, the Optima++ framework is employed to perform global, data-driven optimization of rate rules within their prior uncertainty bounds. A total of 323 rate rules are optimized against a comprehensive experimental database, including IDTs and species profiles, achieving improved agreement across a wide range of operating conditions. This framework significantly enhances the accuracy and efficiency of kinetic models for complex hydrocarbon fuels. The developed models are applied to investigate structure–reactivity relationships in alkene oxidation. Shifting the double bond from terminal to internal positions is found to reduce fuel reactivity at low and intermediate temperatures due to the suppression of hydroperoxyl alkenyl radical formation, which weakens low-temperature chain-branching pathways. For linear 1-alkenes, reactivity increases with increasing carbon chain length (C5–C12) due to the increasing number of hydrogen abstraction sites, particularly at secondary carbon sites and diminishing effect of the primary carbon and the C=C double bond, thereby approaching a limit at higher carbon numbers of approximately C12. Notably, highly branched alkenes, represented by tetramethyl ethylene (XC6D2), exhibit a distinct behaviour compared to 1-alkenes, showing a monotonic increase in reactivity with temperature and exceeding the reactivity of their alkane counterparts (XC6H14) at intermediate temperatures. This is attributed to the absence of inhibiting β-scission pathways and the lack of negative temperature coefficient (NTC) behaviour. In contrast, 1-alkenes consistently exhibit lower reactivity than their corresponding alkanes. This work further examines interactive oxidation chemistry in surrogate fuel systems. In methane (CH4)/n-decane (nC10H22) mixtures, non-linear reactivity enhancement for CH4 is observed with the addition of nC10H22 due to radical interactions, where early ȮH formation from nC10H22 oxidation promotes methane consumption, while CH3Ȯ2 radicals formed during CH4 oxidation, in turn, accelerate nC10H22 oxidation. Furthermore, through a systematic kinetic modelling approach for the oxygenated fuel (ethyl tert-butyl ether, ETBE), this study elucidates the competitive decomposition pathways of ETBE under high-temperature conditions between alcohol elimination and C–O bond scission pathways, identifying this competition as a key factor governing the prediction of key intermediate species, particularly carbon monoxide (CO), which is essential for understanding the combustion chemistry of sustainable oxygenated fuels. Overall, this work establishes a systematic and extensible framework for automated kinetic model development and optimization of large alkene fuels. The results provide new insights into alkene oxidation chemistry and fuel interactions, contributing to improved surrogate fuel design and more accurate predictive combustion modelling.

Publisher

University of Galway

Rights

CC BY-NC-ND

cbn

Collections

University of Galway Theses (PhD Theses)