Autoignition chemistry is central to predictive modeling of many advanced engine designs that combine high efficiency and low inherent pollutant emissions. This chemistry, and especially its pressure dependence, is poorly known for fuels derived from heavy petroleum and for biofuels, both of which are becoming increasingly prominent in the nation's fuel stream. We have investigated the pressure dependence of key ignition reactions for a series of molecules representative of non-traditional and alternative fuels. These investigations combined experimental characterization of hydroxyl radical production in well-controlled photolytically initiated oxidation and a hybrid modeling strategy that linked detailed quantum chemistry and computational kinetics of critical reactions with rate-equation models of the global chemical system. Comprehensive mechanisms for autoignition generally ignore the pressure dependence of branching fractions in the important alkyl + O{sub 2} reaction systems; however we have demonstrated that pressure-dependent 'formally direct' pathways persist at in-cylinder pressures.
The OH concentration in the Cl-initiated oxidation of cyclohexane has been measured between 6.5-20.3 bar and in the 586-828 K temperature range by a pulsed-laser photolytic initiation--laser-induced fluorescence method. The experimental OH profiles are modeled by using a master-equation-based kinetic model as well as a comprehensive literature mechanism. Below 700 K OH formation takes place on two distinct time-scales, one on the order of microseconds and the other over milliseconds. Detailed modeling demonstrates that formally direct chemical activation pathways are responsible for the OH formation on short timescales. These results establish that formally direct pathways are surprisingly important even for relatively large molecules at the pressures of practical combustors. It is also shown that remaining discrepancies between model and experiment are attributable to low-temperature chain branching from the addition of the second oxygen to hydroperoxycyclohexyl radicals.
The rate coefficient for the self-reaction of vinyl radicals has been measured by two independent methods. The rate constant as a function of temperature at 20 Torr has been determined by a laser-photolysis/laser absorption technique. Vinyl iodide is photolyzed at 266 nm, and both the vinyl radical and the iodine atom photolysis products are monitored by laser absorption. The vinyl radical concentration is derived from the initial iodine atom concentration, which is determined by using the known absorption cross section of the iodine atomic transition to relate the observed absorption to concentration. The measured rate constant for the self-reaction at room temperature is approximately a factor of 2 lower than literature recommendations. The reaction displays a slightly negative temperature dependence, which can be represented by a negative activation energy, (E{sub a}/R) = -400 K. The laser absorption results are supported by independent experiments at 298 K and 4 Torr using time-resolved synchrotron-photoionization mass-spectrometric detection of the products of divinyl ketone and methyl vinyl ketone photolysis. The photoionization mass spectrometry experiments additionally show that methyl + propargyl are formed in the vinyl radical self-reaction, with an estimated branching fraction of 0.5 at 298 K and 4 Torr.
The reactions of HCO and DCO with NO have been measured by the laser photolysis/continuous-wave (CW) laser-induced fluorescence (LIF) method from 296 to 623 K, probing the ({tilde B}{sup 2}A{prime} {l_arrow} {tilde X}{sup 2}A{prime}) HCO (DCO) system. The HCO + NO rate coefficient is (1.81 {+-} 0.10) x 10{sup -11} cm{sup 3} molecule{sup -1} s{sup -1} and the DCO + NO rate coefficient is (1.61 {+-} 0.12) x 10{sup -11} cm{sup 3} molecule{sup -1} s{sup -1} at 296 K. Both rate coefficients decrease with increasing temperature between 296 and 623 K. The kinetic isotope effect is k{sub H}/k{sub D} = 1.12 {+-} 0.09 at 296 K and increases to 1.25 {+-} 0.15 at 623 K. The normal kinetic isotope effect supports abstraction as the principal mechanism for the reaction, in agreement with recent computational results.