A Shock-Tube Laser-Schlieren Study of the Decomposition of Diacetyl

The dissociation of diacetyl dilute in krypton has been studied in a shock tube using laser schlieren densitometry, LS, at 1200-1800 K and at two reaction pressures, 55±2 Torr and 120±3 Torr. The experimentally determined rate coefficients show falloff. An ab-initio/Master Equation/VRC-TST analysis was used to determine pressure dependent rate coefficient expressions that are in good agreement with the experimental data. k(T)120Torr = 1.32 x 10 (T/300 K) exp(–50482/T) s and k(T)55Torr = 1.59 x 10 (T/300 K) exp(–48972/T) s. The LS profiles were simulated using a model for methyl recombination with appropriate additions for diacetyl. Excellent agreement is found between the simulations and experimental profiles. * Corresponding author: [email protected] Proceedings of the 6th U.S. National Combustion Meeting Introduction The dissociation of diacetyl, 2,3-butadione, is initiated by C-C fission, R1, to form two acetyl radicals. CH3COCOCH3 → CH3CO + CH3CO (1) The earliest reports on the thermal decomposition of diacetyl are by Rice and Walters [1] (420-470 K, 38458 Torr) and Walters [2] (383-436 K, 147-287 Torr) who studied the reaction in bulb experiments. Product analyses were performed and a reaction mechanism proposed along with rate coefficients for (1). Subsequent thermal experiments were carried out in a stirred flow reactor (677-776 K, 0.6-45 Torr) by Hole and Mulcahy [3] and in a flow tube by Scherzer and Plarre [4] (822-905 K, 0.6-430 Torr). Knoll et al [5] investigated R1 in static cells (648-690 K, 43-183 Torr). The rate coefficients for reaction (1) obtained by Knoll et al., Hole and Mulcahy and, Scherzer and Plarre are in good agreement. Based on product analyses from the above investigations and experimental studies by Guenther et al. [6], Blacet [7] and Bell and Blacet [8] a reaction mechanism for the low temperature pyrolysis of diacetyl has been elucidated that satisfactorily explains the main products ketene, methane, acetone, ethane and CO. The previous studies indicate that (1) is the sole dissociation path for diacetyl but that the rapid decomposition of the acetyl radical via (2) promotes a chain reaction mechanism propagated by methyl radicals. CH3CO → CH3 + CO (2) CH3+CH3COCOCH3→CH4 +CH2COCOCH3 (3) CH3 +CH3COCOCH3 → CH3COCH3 + CH3CO (4) CH3+CH3 → C2H6 (5) Methyl radicals attack the parent molecule via (3) and (4) and the CH2COCOCH3 radical formed in (3) readily dissociates to ketene and CH3CO. At the low temperatures of these studies methyl recombination (5) is the main termination step. There are no high temperature studies of diacetyl pyrolysis in the literature and at elevated temperatures the mechanism may be complicated by reactions of H atoms generated from secondary reactions of CH3 and dissociation of ketene, Although, from Frank et al. [9] ketene may be relatively stable in the temperature range of the current work . The dissociation of acetyl radicals, (2), is the primary source of CH3 radicals in diacetyl pyolysis. At shock tube temperatures the only experimental value for dissociation of the acetyl radical is a recent estimate by Yasunaga et al. [10] which was derived from a shock tube study of acetaldehyde pyrolysis. Recommended rate coefficients for (2) also appear in compilations of kinetic data such as those of Baulch et al. [11]. Reaction (2) has also been the subject of three recent theoretical investigations by Huynh et al. [12], Senosiain et al. [13] and Lee and Bozzelli [14], with calculated rate coefficients covering the range 200-2500 K. Senosiain et al. estimated pressure dependent rate coefficients that are in good agreement with the low temperature experimental results and demonstrate that dissociation of acetyl via (2) is the only viable route at high temperatures with all other pathways having significantly higher barriers. Given the above considerations dicetyl pyrolyis appears to be an attractive, clean pyrolytic source of methyl radicals at shock tube temperatures and it may be superior to some other sources we have used: ethane, acetaldehyde and acetone. The easiest of these other precursors to dissociate, acetone, has about a 10 kcal/mol higher bond strength, so diacetyl should be usable for methyl formation at somewhat lower temperatures. Also, with the diacetyl, its shorter lifetime should reduce the effect of interfering abstraction reactions. Specific Objectives The study of the diacetyl dissociation and the subsequent recombination of the resulting methyl radicals are well suited to investigation by the laserschlieren shock tube technique. The measured beam deflections are proportional to the net endothermic rate and will generate large initial positive gradients from (1) followed by strong negative gradients arising mainly from methyl recombination, (5). The two processes are well-separated in time as well as sign and can be clearly seen and differentiated. Both initial diacetyl decomposition rates and an expanded and verified mechanism for the methyl recombination and its associated chain mechanism are presented here. Some theory of the decomposition and RRKM modeling are also presented Experimental The LS experiments were performed in a diaphragmless shock tube, DFST, which has been fully described elsewhere [15]. The driver section of the DFST contains a fast acting valve which replaces the more traditional diaphragm, Fig. 1. When the valve is closed by pressurizing the inside of the bellows, the driver and driven sections are separated and can be filled to the desired loading pressures and the DFST is fired by rapidly opening the valve. By varying both the driver section pressure, P4, and the driven section pressure, P1, the pressure behind the incident shock wave, P2, can be constrained to very narrow ranges, typically < ±3%, over a wide range of temperatures [15]. BellowsExhaust Bellows Gas Inlet Bellows Exhaust Bellows Shaft Seal plate Linear Bearing To driven