Theoretical reinvestigation of the O(3P) + C6H6 reaction: quantum chemical and statistical rate calculations

The lowest-lying triplet and singlet potential energy surfaces for the O(3P) + C6H6 reaction were theoretically characterized using the "complete basis set" CBS-QB3 model chemistry. The primary product distributions for the multistate multiwell reactions on the individual surfaces were then determined by RRKM statistical rate theory and weak-collision master equation analysis using the exact stochastic simulation method. It is newly found that electrophilic O-addition onto a carbon atom in benzene can occur in parallel on two triplet surfaces, 3A' and 3A' '; the results predict O-addition to be dominant up to combustion temperatures. Major expected end-products of the addition routes include phenoxy radical + H*, phenol and/or benzene oxide/oxepin, in agreement with the experimental evidence. While c-C6H5O* + H* are nearly exclusively formed via a spin-conservation mechanism on the lowest-lying triplet surface, phenol and/or benzene oxide/oxepin are mainly generated from the lowest-lying singlet surface after inter-system crossing from the initial triplet surface. CO + c-C5H6 are predicted to be minor products in flame conditions, with a yield < or = 5%. The O + C6H6 --> c-C5H5* + *CHO channel is found to be unimportant under all relevant combustion conditions, in contrast with previous theoretical conclusions (J. Phys. Chem. A 2001, 105, 4316). Efficient H-abstraction pathways are newly identified, occurring on two different electronic state surfaces, 3B1 and 3B2, resulting in hydroxyl plus phenyl radicals; they are predicted to play an important role at higher temperatures in hydrocarbon combustion, with estimated contributions of ca. 50% at 2000 K. The overall thermal rate coefficient k(O + C6H6) at 300-800 K was computed using multistate transition state theory: k(T) = 3.7 x 10-16 x T 1.66 x exp(-1830 K/T) cm(3) molecule(-1) s(-1), in good agreement with the experimental data available.