Study on the Reaction of CH2 with H2 at High Temperature

Direct measurement of site-specific rates of reactions of H with C3H8, i-C4H10, and n-C4H10

We measured the rates of abstraction of a hydrogen atom from specific sites in propane C₃H₈, 2-methyl propane (i-C₄H₁₀), and butane (n-C₄H₁₀); the sites are a primary hydrogen of C₃H₈ and i-C₄H₁₀ and a secondary hydrogen of n-C₄H₁₀. The excellent reproducibility of conditions of a diaphragm-less shock tube enabled us to conduct comparative measurements of the evolution of H atoms in three mixtures-(i) 0.5 ppm C₂H₅I + Ar, (ii) 0.5 ppm C₂H₅I + 50-100 ppm alkane as C₃H₈ or i-C₄H₁₀ or n-C₄H₁₀ + Ar, and (iii) the same concentrations of alkane + Ar without C₂H₅I-in the temperature range 1000-1200 K and at a pressure of 2.0 bars. The net profile of rise and decay of H atoms in the C₂H₅I + alkane mixture was derived on subtracting the absorbance of (iii) from that of (ii). Measurements of the mixture (iii) are important because the absorption of alkanes at 121.6 nm is not negligible. In the temperature range 1000-1100 K, the rate of decomposition of C₂H₅I was evaluated directly on analyzing the exponential growth of H atoms in the mixture (i). The rate of decomposition of C₂H₅I is summarized as ln(k/s^-1) = (33.12 ± 1.4) - (25.23 ± 1.5) 10³/T (T = 1000-1100 K, P = 2.0 bars); the broadening factor F(T) in the Lindemann-Hinshelwood formula was evaluated in the fall-off region. The site-specific rates of H + (C₃-C₄) alkanes are summarized as follows: H + C₃H₈ → H₂ + 1-C₃H₇, ln(k_1a) = -(21.34 ± 0.86) - (5.39 ± 0.93)10³/T, H + i-C₄H₁₀ → H₂ + i-C₄H₉, ln(k_2a) = -(20.50 ± 1.36) - (6.14 ± 0.13)10³/T, H + n-C₄H₁₀ → H₂ + 2-C₄H₉, ln(k_3b) = -(21.37 ± 1.15) - (4.83 ± 1.26)10³/T. The present experimental results are compared with published results from quantum-chemical calculations of potential-energy surfaces and transition-state theory. The present experiments are consistent with those calculations for the reaction rates for the attack at the primary site for H + C₃H₈ and H + i-C₄H₁₀, but for the attack at the secondary site of n-C₄H₁₀, our results are substantially smaller than the computational prediction, which might indicate a hindrance by the C-H bonds of the primary sites that serves to decrease the rate of abstraction from the secondary site of n-C₄H₁₀. The influence on the total rates of reactions H + alkane and the group additivity rule are discussed.