Shock Tube Study on the Thermal Decomposition of CH 3 OH Ku-We Lu, Hiroyuki Matsui,* Ching-Liang Huang, P. Raghunath, Niann-Shiah Wang,* and M. C. Lin Department of Applied Chemistry, National Chiao Tung UniVersity, 1001, Ta Hsuch Road, Hsinchu 30010 Taiwan ReceiVed: January 19, 2010; ReVised Manuscript ReceiVed: March 15, 2010 H atom produced in the thermal decomposition of CH 3 OH highly diluted in Ar (0.48-10 ppm) was monitored behind reflected shock waves by atomic resonance absorption spectrometry (ARAS) at fixed temperatures (and pressures), that is, 1660 (1.73 atm), 1760 (2.34 atm), 1860 (2.04 atm), 1950 (2.18 atm), and 2050 K (1.76 atm) ( (10 K, respectively). High sensitivity for the H atom has been attained by signal averaging of the ARAS signals down to the concentrations of ∼1 × 10 11 atoms/cm 3 and enables us to determine the branching fraction for the direct H atom production channel, CH 3 OH f CH 2 OH + H (channel 1c) in a mixture of 1 ppm CH 3 OH. Channel 1c is confirmed to be minor, that is, branching fraction for channel 1c is expressed by Log( k 1c / k 1 ) ) ( - 2.88 ( 1.88) × 10 3 / T - (0.23 ( 1.02), which corresponds to k 1c / k 1 < 0.03 for the present temperature range. By using 0.48 and 1.0 ppm CH 3 OH with (100-1000) ppm H 2 , the total decomposition rate k 1 for CH 3 OH f products is measured from the time dependence of H atom, where the radical products of main channels 1a and 1b, that is, OH, CH 3 , and CH 2 , were converted rapidly into H atoms. The experimental result is summarized as Log( k 1 /cm 3 molecule -1 s -1 ) ) ( -12.82 ( 0.71) × 10 3 / T - (8.5 ( 0.38). A theoretical study based on ab initio/TST calculations with high accuracy has been conducted for the reaction: 3 CH 2 + H 2 f CH 3 + H (reaction 3). The rate is given by k 3 /cm 3 molecule -1 s -1 ) (7.32 × 10 -19 ) T 2.3 exp ( -3699/ T). This result is used for numerical simulations to evaluate k 1 . Present experimental results on the thermal decomposition rate of CH 3 OH are found to be consistent with previous works. It is also found that time dependence of [H] observed in the 10 ppm CH 3 OH in Ar can be reproduced very well by kinetic simulations by using a reaction mechanism composed of 36 elementary reactions. 1. Introduction Alcohol fuels are recognized as the most promising renewable energy resources. Therefore, it is important to investigate the reaction mechanisms and kinetics for their thermal decomposi- tion and combustion. Thermal decomposition of CH 3 OH has been studied extensively 1-14,57 but understanding the details of the reaction mechanism of pyrolysis at an elevated concentration still remains to be below an acceptable level. It is suggested in some of the theoretical studies 13,14 that CH 3 OH decomposition has the following possible product channels. The main reaction channel is concluded in most of the experimental and theoretical studies to be reaction 1a, and some contribution from 1b is also indicated under the ambient conditions. Direct production of the H atom from scission of a C-H bond (reaction 1c) has been indicated to be minor by observing time dependence of H atom production, 4,5 but some of the bulk experimental studies indicated its non-negligible contribution in the pyrolysis of methanol. 7,8 Trials in the previous experimental studies to examine the branching fractions have been obscured more or less by the secondary reactions because of the high concentration of CH 3 OH employed. Time depen- dence of the main product OH from (reaction 1a) was measured in recent shock tube studies at relatively low concentrations of CH 3 OH (<50 ppm CH 3 OH). 12,57 Even with such a low concen- tration of CH 3 OH, the profile of OH is strongly affected by many secondary reactions. Therefore, non-negligible uncertain- ties are still left in the reported rate parameters and the mechanism for the title reaction. The main issue of this study is to conduct an experimental study to derive the kinetic information on the thermal decom- position by using sufficiently low concentration of CH 3 OH so as to minimize the effect of the secondary reactions. By utilizing the advantage of an excellent reproducibility of the diaphragmless shock tube, signal averaging of the experi- mental data at fixed temperatures and pressures has brought improvement in the detection limit for H atoms down to 10 11 atoms cm -3 : examination of the direct H atom production channel 1c is possible by lowering the initial concentration of CH 3 OH to 1 ppm level (or even lower). To get information of the radical species X (X ) OH, CH 3 , and/or CH 2 ) produced in reactions 1a and 1b, they are converted to H atoms by the reactions with H 2 in the presence of a much excess of H 2 (100-1000 ppm). Reactions of 3 CH 2 become important under the shock tube condition because 1 CH 2 produced in reaction 1b is very quickly quenched to 3 CH 2 by collisions with Ar. 4 * To whom correspondence should be addressed. CH 3 OH + M f products + M (1) CH 3 OH + M f CH 3 + OH + M (1a) f 1 CH 2 + H 2 O + M (1b) f CH 2 OH + H + M (1c) f CH 2 O + H 2 + M (1d) f cis-HOCH + H 2 + M (1e) f trans-HOCH + H 2 + M (1f) J. Phys. Chem. A 2010, 114, 5493–5502 5493 10.1021/jp100535r  2010 American Chemical Society Published on Web 04/12/2010