Kinetics of Hexane Combustion in a Shock Tube ALEXANDER BURCAT, * ERNA OLCHANSKI, AND CHARA SOKOLINSKI Faculty of Aerospace Engineering, Technion - Israel Institute of Technology, Haifa 32000, Israel (Received 2 May 1996 and in revised form 25 July 1996) Abstract. A detailed shock tube investigation on the ignition delay times in mixtures containing n-hexane and oxygen diluted in argon is reported. The product distribution ofthe pre-ignited mixtures was also investigated and a computer model- ing of the combustion kinetics was performed. The experiments covered the tem- perature range 1020-1725 K at pressures of 1-7 atm. A computer simulation was performed with a large number of elementary reactions (386), which was then reduced to a scheme containing only 76 reactions. INTRODUCTION Gasoline contains 3-6% hexane blended with heptane, octane, and decane isomers. Hexane is therefore an important part of the fuel, and its combustion kinetics is of great importance. To the best of our knowledge, the kinetics of the combustion of hexane has never been studied - neither in a shock tube nor in a flow reactor. We are also unaware of any modeling studies of the mechanism of its combustion. In this paper we discuss and compare modeling results on the ignition of pen- tane':' and heptane.v' EXPERIMENTAL Apparatus Ignition delay times were measured in a single-pulse, stainless steel shock tube, which served as a homogeneous reactor in the millisecond range. The tube is 54 mm i.d. and 4 m long. The driven section is 2.5 m long. Mylar diaphragms of different gauges were used. They were burst by the helium driver pressure. The samplingsection, 0.25 m long, contains three Kistler 603A piezoelectric trans- ducers.Two of them,located0.2 m apart,measuredthe shock velocity. A thirdpiezoelectric gauge,locatedon theend plate, recordedthe pressuretimehistory. Both eventswere sampled by a Nicoletdual tracedigitaloscilloscope. Two l2-bit traces were recordedat 1 IlS intervals. One trace gave the outputof the end plate transducer, while the secondtrace recordedthe outputof the othertwo transducers connectedin parallel. The ignition delay times, which were measured at the end plate of the driven section, were defined as the time interval between the pressure rise caused by. the re- flected shock and the onset of ignition. Materials The hexane used in this study was Riedel de Haen pure reagent. Hexane vapors were expanded into an evacuated container which was then filled with oxygen and argon. The gases were oxygen, 99% pure (Oxygen Storages, Haifa), argon, 99.9% pure (Gas Technologies, Herzliya), and helium, 99.9% pure (Airco). These gases were used without further purification. All the gas mix- tures were prepared manometrically in stainless steel containers and were pressurized to -50 psia. To ensure thorough mixing of the gases, the mixtures were left in the containers for 48 hr before use. Analysis and Calculations Gas chromatographic analyses were performed on post-shock mixtures which were quenched just before ignition. A Packard 427 gas chromatograph with a flame ionization detector was used with a 1/16" PORAPAK N column. Helium served as the carrier gas at 30 mL/min. The reflected shock temperatures were calculated using standard conservation equations and the ideal gas equa- tion of state, assuming frozen chemistry. The enthalpies were taken from a recent compilation.' RESULTS Some 250 experiments were run with various mixtures containing n-hexane and oxygen diluted in argon. Table 1 lists the experimental conditions of three repre- sentative experiments from each of the n-hexane mix- tures used. A Semenov-type equation for n-hexane was applied for the mixtures listed in Table 1. Only 211 experiments *Author to whom correspondence should be addressed. Israel Journal of Chemistry Vol. 36 1996 pp.313-320