A shock tube and theoretical study on the pyrolysis of 1,4-dioxane.

The dissociation of 1, 2 and 4% 1,4-dioxane dilute in krypton was studied in a shock tube using laser schlieren densitometry, LS, for 1550-2100 K with 56 ± 4 and 123 ± 3 Torr. Products were identified by time-of-flight mass spectrometry, TOF-MS. 1,4-dioxane was found to initially dissociate via C-O bond fission followed by nearly equal contributions from pathways involving 2,6 H-atom transfers to either the O or C atom at the scission site. The 'linear' species thus formed (ethylene glycol vinyl ether and 2-ethoxyacetaldehyde) then dissociate by central fission at rates too fast to resolve. The radicals produced in this fission break down further to generate H, CH(3) and OH, driving a chain decomposition and subsequent exothermic recombination. High-level ab initio calculations were used to develop a potential energy surface for the dissociation. These results were incorporated into an 83 reaction mechanism used to simulate the LS profiles with excellent agreement. Simulations of the TOF-MS experiments were also performed with good agreement for consumption of 1,4-dioxane. Rate coefficients for the overall initial dissociation yielded k(123Torr) = (1.58 ± 0.50) × 10(59) × T(-13.63) × exp(-43970/T) s(-1) and k(58Torr) = (3.16 ± 1.10) × 10(79) × T(-19.13) × exp(-51326/T) s(-1) for 1600 < T < 2100 K.

Shock tube and theory investigation of cyclohexane and 1-hexene decomposition.

The decomposition of cyclohexane (c-C(6)H(12)) was studied in a shock tube using the laser-schlieren technique over the temperature range 1300-2000 K and for 25-200 Torr in mixtures of 2%, 4%, 10%, and 20% cyclohexane in Kr. Vibrational relaxation of the cyclohexane was also examined in 10 experiments covering 1100-1600 K for pressures below 20 Torr, and relaxation was found to be too fast to allow resolution of incubation times. The dissociation of 1-hexene (1- C(6)H(12)), apparently the sole initial product of cyclohexane decomposition, was also studied over 1220-1700 K for 50 and 200 Torr using 2% and 3% 1-hexene in Kr. On heating, cyclohexane simply isomerizes to 1-hexene, and this then dissociates almost entirely by a more rapid C-C scission to allyl and n-propyl radicals. This two-step reaction results in an initial small density gradient from the slight endothermicity of the isomerization. The gradient then rises strongly as the product 1-hexene dissociates. For the lower temperatures, this behavior is fully resolved here. For the higher pressures, 1-hexene decomposition generates negative gradients (exothermic reaction) as the radicals formed begin to recombine. Cyclohexane also generates such gradients, but these are now much smaller because the radical pool is depleted by abstraction from the reactant. A complete mechanism for the 1-hexene decomposition and for that of cyclohexane involving 79 reactions and 30 species is used in the final modeling of the gradients. Rate constants and RRKM fit parameters for the initial reactions are provided for the entire range of conditions. The possibility of direct reaction to allyl and n-propyl radicals, without stabilization of the intermediate 1-hexene, is examined down to pressures as low as 25 Torr, without a clear resolution of the issue. High-pressure limit rate constants from RRKM extrapolation are k(infinity)(c-C(6)H(12) --> 1-C(6)H(12)) = (8.76 x 10(17)) exp((-91.94 kcal/mol)/RT) s(-1) (T = 1300-2000 K) and k(infinity)(1-C(6)H(12) --> (*)C(3)H(7) + (*)C(3)H(5)) = (1.46 x 10(16)) exp((-69.12 kcal/mol)/RT) s(-1) (T = 1200-1700 K). This high-pressure rate for cyclohexane is entirely consistent with the notion that the isomerization involves initial C-C fission to a diradical. These extrapolated high-pressure rates are in good agreement with much of the literature.

Shock-tube study of relaxation in HCN.

Thermal vibrational relaxation in HCN mixtures with Kr has been observed with the laser-schlieren technique. The experiments cover the temperatures 750-2900 K and a large pressure range of 13-420 Torr in 5% and 20% HCNKr mixtures. Relaxation is extremely fast but appears to occur in two well-separated stages that are assigned to the vibrational transitions (000)-->(010) and (000)-->(100) with perhaps some lesser contribution from (000)-->(001). This interpretation is strongly supported by a comparison of net density changes to thermodynamic calculations. The first and faster process shows near constant relaxation times whereas the latter slower stage has a slight decrease of these with T. Relaxation times in pure HCN obtained by neglecting the small contribution of krypton are as follows: (a) Ptau(HCN-HCN)=27 exp(1.473T(13)) ns atm (000)-->(010); (b) Ptau(HCN-HCN)=11 exp(32.6T(13)) ns atm (000)-->(100). Probabilities suggested by these results are around 0.05 for the fast step and 0.0035 to 0.005 for the slow process. These results are close to those found by laser fluorescence measurements for deactivation of levels involving excitation of the C-H stretch (001) at 3312 cm(-1). These results are also consistent with the notion of a dominance of the fast stage by T,R-V transfer (thermal relaxation) occurring in a weakly bound complex. However, the slow step most likely occurs through a V-V process (03 (1)0)-->(100), DeltaE=27.7 cm(-1), after multiple excitation of the (010) mode. These are the first thermal measurements of relaxation in HCN and the first to see energy transfer involving the low-frequency modes.

Vibrational relaxation in methyl hydrocarbons at high temperatures: propane, isobutene, isobutane, neopentane, and toluene.

Vibrational relaxation has been seen in shock waves in propane, isobutene, isobutane, neopentane, and toluene dilute in krypton with the laser-schlieren technique. These experiments cover about 600-2200 K and post-shock pressures from 5 to 29 Torr. The process cannot be resolved in any for T<600 K, or in any for large molecule fraction. All the ultrasonic measurements of relaxation in these at room temperature show characteristic times in the 1-5 ns atm range, corresponding to fewer than five collisions, whereas the relaxation times in the shock waves range from 20 to 200 ns atm, with a clearly defined negative or "inverted" temperature dependence. It would seem the observed slowdown of relaxation with increasing T is simply a consequence of the much increased energy transfer required at high temperature in such large polyatomics when this is combined with a collision efficiency, here interpreted as down, already so large it cannot much increase. The simple method for the extraction of a down from relaxation data offered here by consideration of the energy relaxation equation for Evib=0 appears to be original and should prove quite useful in connecting thermal relaxation data to values obtained from spectroscopy and master-equation analyses. Here it is found that the derived down extrapolate well to room temperature ultrasonic measurements, showing a slight increase with temperature.

Tunable laser flash absorption: a new technique for measuring rates and yields of chemical reactions at high temperatures.

The flash absorption technique, whereby light from an excimer laser is used to measure the kinetic behavior of absorbing species in the high temperature region behind a shock front with a linear array detector, has been extended by using tunable light from a high resolution, pulsed dye laser. The use of narrowband, tunable light allows us to access isolated rovibronic transitions and, thereby, obtain state-specific kinetic information. If the oscillator strength of the transition and the absorption line profile are known, the absolute concentration may be determined. We demonstrate the technique by measuring the temporal development of the hydroxyl radical as it is formed after propane has been thermally dissociated in the presence of oxygen. We conclude that accurate kinetic measurements can be made with hydroxyl concentrations of 10(15) cm(-3) This technique may also be applied to study any species which absorbs below 50,000 cm(-1).

Physical optics of the laser-schlieren shock tube technique.

The physical optics of the laser-schlieren technique for the measurement of rate processes in shock waves is examined in detail. The method utilizes the Kirchhoff-Huygens integral with the usual thin lens, paraxial, and Fresnel approximations, all of which are appropriate for the typical laser schlieren experiment. The resolution and sensitivity of the technique are defined for all detector separations, and a reliable method for locating the time origin in the schlieren signal is provided. Diffraction is found to have a significant effect on the shock front generated signal, and geometrical optics treatments of this signal are shown to be inadequate.