(79b) The Reaction of Vinyl Radical with Alkenes: Measured Rates and Predicted Pressure-Dependent Product Distributions

This work presents experimental and theoretical work on the reaction of the vinyl radical (H2C=CH) with the first five alkenes: ethene, propene, 1-butene, 2-butene, and isobutene. A rate rule for the addition and H-abstraction reactions for the vinyl radical with larger alkenes is presented.

The experiments were performed in a 160 cm stainless steel, temperature and pressure-controlled flow-tube reactor. The vinyl radical was generated by laser photolysis of vinyl iodide (C2H3I) at 266 nm. The photolysis pulse was generated by frequency-quadrupling the 1064 nm output of a short-pulse Nd:YAG laser. The vinyl radical was detected by direct absorption at 423.2 nm; for the vinyl + isobutene experiments, the vinyl radical was detected at 475 nm, due to a background absorption at 423.2 nm. The absorption laser path length was increased up to 40 meters by the inclusion of a Herriott-type multiple-pass resonator within the reactor. The absorption wavelength for this laser was generated using a mode-locked Ti:Sapphire laser, pumped by a 532 nm diode-pumped solid-state continuous-wave (CW) laser. The result is a quasi-CW probe laser, with pulse durations of 1.2 ps at 80 MHz. The vinyl + alkene rate constants were measured under pseudo-first order conditions, with the concentration of the alkene gas in sufficient excess of the vinyl radical to guarantee that the pseudo-first order decay constant was at least five times faster than the decay signal under no alkene conditions. The experiments were performed at temperatures between 300 ? 700 K and at a pressure of 100 torr; additionally, the vinyl + ethene and vinyl + propene experiments were performed at 10 torr. At each temperature and pressure, the rate of decay of vinyl concentration was fit to a single exponential, and this calculation was repeated for increased alkene concentrations. The desired first-order rate constant, k2, was calculated from the plot of the pseudo-first order rate constant for vinyl loss, k'' = k2[CnH2n] + k1, versus [CnH2n]. For all five experiments, the slope was linear.

Theoretical calculations were performed to predict the rate of formation for various product channels as a function of temperature and pressure. For each reaction, a potential energy surface (PES) was generated. The electronic energy at 0K was calculated using the G3 method in Gaussian03. The harmonic oscillator frequencies were calculated using b3pw91/6-311++G(3df,3pd) level theory. For internal rotors, the potential for internal rotation was calculated from a relaxed scan at the b3pw91/DIDZ level. The hindered-rotor partition function and density of states were calculated using a semi-classical approximation. The pressure-dependent rate constants were calculated by solving the master equation, which was done using VariFlex. The predicted rates for the total disappearance of vinyl agree with the experimental results within 20% at temperatures greater than 350 K; below 350 K, the agreement is less accurate, presumably due to approximations made for tunneling and hindered internal rotation, which are less accurate at low temperatures.

The rates in descending order are: propene, isobutene, 1-butene, ethene, and 2-butene. This order can be explained by two phenomena: the degree of substitution on the double-bonded carbons, and the rotational constants of the alkene. By replacing an H with an R at the ?tail? (i.e. non-bonding end) of an alkene, the initial adduct becomes increasingly stable, going from a primary to secondary to tertiary radical with each substitution. In contrast, by replacing an H with an R on the bonding CH2 end, the radical stability does not change, but the R-group presents an obstacle for bond formation. Additionally, ethene and propene have one rotational constant that is significantly larger than the largest rotational constant for the three butenes, which thereby increases the rate of these two reactions.

The master equation results predict that at low temperatures, the dominant product will be the initial adduct formed by the addition of the vinyl radical to the least-substituted carbon in the double bond. As the temperature is increased, the dominant products will become a mixture of dienes, with 1,3-dienes the most common. At higher temperatures, the direct H-abstraction reaction will dominate over the addition-isomerization-dissociation reaction, and the dominant product will be ethene plus a resonantly stabilized allyllic radical. The rate of formation of cyclic species is predicted to be several orders of magnitude slower than the rate of formation for the dienes. For all five systems, the temperature at which chemically activated product formation begins to dominate the collisional stabilization of the initial adduct is pressure dependent.

The experimental and computational results are used to develop a new rate rule for the vinyl radical + alkene reaction system. The new rate rule will calculate the rate constant for the two addition reactions as well as H-abstraction reactions.