A Journey from n-Heptane to Liquid Transportation Fuels. 1. The Role of the Allylic Radical and Its Related Species in Aromatic Precursor Chemistry

ion from 1-hexene, changes in reactions involving 1-hexene also affect the prediction of 1,3-butadiene. For example, the predicted peak concentration of 1,3-butadiene increases by 13% of the MPC due to a higher concentration profile of 1-hexene after a slower 1-hexene thermal decomposition reaction (using the generic rate of reaction 1 in Table 1) is selected. 1-Butylene has an important effect on the concentration of 1,3-butadiene, since it is the major source of 3-butenyl (allylic) or 4-butenyl radicals, the decomposition reactions of which are the most important formation routes of 1,3-butadiene. A higher 1-butylene concentration after the modification of its thermal decomposition rate (reactions 2 and 3 in Table 1) increases the predicted peak concentration of 1,3-butadiene by 42% of the MPC. Besides 1-hexene and 1-butylene, changes involving the allyl radical also affect the profile of 1,3-butadiene. The lower level of allyl radical due to various modifications, for example, reactions 1–3 in Table 1, reduces the predicted peak concentration of 1,3-butadiene by a total of 11% of the MPC in the n-heptane flame. With those modifications, the peak concentration of 1,3-butadiene in the n-heptane flame is overpredicted by 15%, as seen in Figure 4. Allene. The predicted peak concentrations of allene in the n-decane and n-heptane flames using the Utah heptane mechanism are higher than the measured values by 12 and 231%, respectively. No modifications specifically targeted at allene were intended in the extended mechanism, but the concentration profiles of allene are affected significantly by changes in reactions involving other species, especially those involving the allyl radical. When the formation of various olefins is favored after the additions of olefin formation from scission, or from larger olefin species via hydrogen addition followed by decomposition, the resulting higher levels of the allyl radical lead to the higher predicted peak concentrations of allene. These higher concentrations are results of hydrogen abstraction or thermal decomposition of the allyl radical and exceed the values measured in the n-decane and n-heptane flames by 42 and 33% of the MPC, respectively.28 Although the gains in allene level lead to larger numerical deviations from the measured concentration profiles, they are offset by changes in reactions also involving the allyl radical after the formation rates of the allyl radical from thermal decomposition or hydrogen abstraction of olefins (reactions 1–3 in Table 1) are reduced. The decreased levels of the allyl radical lead to declines of allene concentrations in the n-decane and n-heptane flames of 66 and 143% of the MPC, respectively. Those changes in reactions involving the allyl radical yield better-predicted allene concentration profiles, especially in the n-heptane flame where the peak concentration of allene is now overpredicted by 100% compared to 230% before modifications. Besides the allyl radical, changes in reactions involving other C3H5 isomers also affect the concentrations of allene. The predicted peak concentrations of allene in the n-decane and n-heptane flames are 16 and 39% of the MPC lower, respectively, after the enhanced formation of propyne from 2-propenyl radical (CH3sC•dCH2) was selected (leaving out the third body species, as mentioned above). The modified reaction competes with that of allene from the same C3H5 isomer. The allene concentrations are slightly reduced when a new combination reaction of 2-propenyl and methyl radicals (reaction 5 in Table 1) is added to enhance the formation of isobutylene. Although the predicted maximum concentration of allene in the n-heptane flame has been greatly improved by modifying or adding reactions involving C3H5 isomers, it is still higher than the measured value by 97%. The prediction of allene in the n-decane flame is much better, with the peak concentration 18% lower than that of the experimental data. The effects of major modifications are shown in Figure 4.