Rate constants and branching ratios for the reaction of CH radicals with NH3: a combined experimental and theoretical study.

The reaction between CH radicals and NH(3) molecules is known to be rapid down to at least 23 K {at which temperature k = (2.21 ± 0.17) × 10(-10) cm(3) molecule(-1) s(-1): Bocherel ; et al. J. Phys. Chem. 1996, 100, 3063}. However, there have been only limited theoretical investigations of this reaction and its products are not known. This paper reports (i) ab initio quantum chemical calculations on the energy paths that lead to various reaction products, (ii) calculations of the overall rate constant and branching ratios to different products using transition state and master equation methods, and (iii) an experimental determination of the H atom yield from the reaction. The ab initio calculations show that reaction occurs predominantly via the initial formation of a datively bound HC-NH(3) complex and reveal low energy pathways to three sets of reaction products: H(2)CNH + H, HCNH(2) + H, and CH(3) + NH. The transition state calculations indicate the roles of "outer" and "inner" transition states and yield rate constants between 20 and 320 K that are in moderate agreement with the experimental values. These calculations and those using the master equation approach show that the branching ratio for the most exothermic reaction, to H(2)CNH + H, is ca. 96% throughout the temperature range covered by the calculations, with those to HCNH(2) + H and CH(3) + NH being (4 ± 3)% and <0.3%, respectively. In the experiments, multiple photon dissociation of CHBr(3) was used to generate CH radicals and laser-induced fluorescence at 121.56 nm (VUV-LIF) was employed to observe H atoms. By comparing signals from CH + NH(3) with those from CH + CH(4), where the yield of H atoms is known to be unity, it is possible to estimate that the yield of H atoms from CH + NH(3) is equal to 0.89 ± 0.07 (2σ), in satisfactory agreement with the theoretical estimate.

Low temperature kinetics: the association of OH radicals with O2.

We report the first measurements of rate constants for the reaction in which OH radicals associate with O(2) to form HO(3). Our recent measurements (Science, 2010, 328, 1258) have shown that the HO-O(2) bond dissociation energy is only (12.3 ± 0.3) kJ mol(-1). Consequently, above ca. 90 K under attainable experimental conditions, the rate of the reverse dissociation of HO(3) becomes comparable to, and then greater than, the rate of the forward association reaction. We have used the CRESU (Cinétique de Réaction en Ecoulement Supersonique Uniforme) method to access low temperatures and have explored the kinetics of OH + O(2) + M → HO(3) + M in two series of experiments. At temperatures between 55.9 and 79.2 K, the OH radicals, created by pulsed laser photolysis of H(2)O(2) and observed by laser-induced fluorescence, decayed by pseudo-first-order kinetics to effectively zero concentration at longer times. The third-order rate constants derived from these experiments fit the expression: k(3rd)(o) (T) = (4.2 ± 1.9) × 10(-34) (T/298 K)(-(3.5 ± 0.3)) cm(6) molecule(-2) s(-1). At temperatures between 87.4 and 99.8 K, rate constants for the association reaction were determined allowing for the significant occurrence of the reverse dissociation reaction. The values of the derived rate constants are consistent with those obtained in the lower temperature range, though the errors are larger. The experimental values of k(3rd)(o) (T) are compared with (a) those for other association reactions involving species of similar complexity, and (b) values of k(3rd)(o) (T) estimated according to both the energy transfer (ET) and the radical-complex (RC) mechanisms. We conclude that the RC mechanism probably makes the major contribution to the association of OH + O(2) at the low temperatures of our experiments.

The thermodynamics of the elusive HO3 radical.

The role of HO3 as a temporary reservoir of atmospheric OH radicals remains an open question largely because of the considerable uncertainty in the value of the dissociation energy of the HO-O2 bond (D0) or, equivalently, the standard enthalpy of formation of HO3 (Delta(f)H;{\overline{);\circ }}$$). Using a supersonic flow apparatus, we have observed by means of laser-induced fluorescence the decay of OH radicals in the presence of O2 at temperatures between 55.7 and 110.8 kelvin (K). Between 87.4 and 99.8 K, the OH concentration approached a nonzero value at long times, allowing equilibrium constants for the reaction with O2 to be calculated. Using expressions for the equilibrium constant from classical and statistical thermodynamics, and values of partition functions and standard entropies calculated from spectroscopic data, we derived values of D0 = (12.3 +/- 0.3) kilojoules per mole and Delta(f)H;{\overline{);\circ }}$$ (298 K) = (19.3 +/- 0.5) kilojoules per mole. The atmospheric implications of HO3 formation are therefore very slight.

An experimental confirmation of the products of the reaction between CN radicals and NH3.

By comparing H-atom yields from the reactions of CN with C(2)H(2) and NH(3), it has been confirmed that the latter reaction produces insignificant amounts of H-atoms, implying that it proceeds exclusively to HCN + NH(2).

A theoretical analysis of the reaction between CN radicals and NH3.

The reaction between CN radicals and NH3 molecules has been studied experimentally over an unusually wide range of temperature (25-716 K). Below 295 K, the rate constant exhibits a strong negative dependence on temperature; that is, it increases sharply as the temperature is lowered. The present work analyses the kinetics of this reaction theoretically, both to explain this unusual temperature-dependence and to identify the major products of the reaction--which have not been well established by experiment. Quantum chemical calculations at the CCSD(T) theoretical level show that the minimum energy path for reaction proceeds: (a) first, via a potential well, which is 39.3 kJ mol(-1) below the energy of the separated reactants, when allowance is made for zero-point energies, corresponding to a quite strongly bound NC-NH3 complex, and (ii) then over a 'submerged' barrier with a crest 10.9 kJ mol(-1) below the energy of the reactants to the products HCN + NH2. These ab initio calculations also demonstrate that there is no low energy path to the products NCNH2 + H. The dynamics of the main reaction have been further investigated using the two transition state model of Klippenstein and co-workers, in which transition state theory is applied at the selected E, J microcanonical level. The rate constants calculated for temperatures between 25 and 200 K are in excellent agreement with the experimental values.

The temperature-dependence of elementary reaction rates: beyond Arrhenius.

The rates of chemical reactions and the dependence of their rate constants on temperature are of central importance in chemistry. Advances in the temperature-range and accuracy of kinetic measurements, principally inspired by the need to provide data for models of combustion, atmospheric, and astrophysical chemistry, show up the inadequacy of the venerable Arrhenius equation--at least, over wide ranges of temperature. This critical review will address the question of how to reach an understanding of the factors that control the rates of 'non-Arrhenius' reactions. It makes use of a number of recent kinetic measurements and shows how developments in advanced forms of transition state theory provide satisfactory explanations of complex kinetic behaviour (72 references).

Understanding reactivity at very low temperatures: the reactions of oxygen atoms with alkenes.

A remarkable number of reactions between neutral free radicals and neutral molecules have been shown to remain rapid down to temperatures as low as 20 kelvin. The rate coefficients generally increase as the temperature is lowered. We examined the reasons for this temperature dependence through a combined experimental and theoretical study of the reactions of O(3P) atoms with a range of alkenes. The factors that control the rate coefficients were shown to be rather subtle, but excellent agreement was obtained between the experimental results and microcanonical transition state theory calculations based on ab initio representations of the potential energy surfaces describing the interaction between the reactants.

Rate coefficients for reaction and for rotational energy transfer in collisions between CN in selected rotational levels (X 2Sigma+, v=2, N=0, 1, 6, 10, 15, and 20) and C2H2.

Rate coefficients (ktot,Ni) are reported (a) for total removal (reactive+inelastic) of CN(X2Sigma+,v=2,Ni) radicals from selected rotational levels (Ni=0, 1, 6, 10, 15, and 20) and (b) for state-to-state rotational energy transfer (ki-->f) between levels Ni and other rotational levels Nf in collisions with C2H2. CN radicals were generated by pulsed laser photolysis of NCNO at 573 nm. A fraction of the radicals was then promoted to a selected rotational level in v=2 using a tunable infrared "pump" laser operating at approximately 2.45 microm, and the subsequent fate of this subset of radicals was monitored using pulsed laser-induced fluorescence (PLIF). Values of ktot,Ni were determined by observing the decay of the PLIF signals as the delay between pump and probe laser pulses was systematically varied. In a second series of experiments, double resonance spectra were recorded at a short delay between the pump and probe laser pulses. Analysis of these spectra yielded state-to-state rate coefficients for rotational energy transfer, ki-->f. The difference between the sum of these rate coefficients, Sigmafki-->f, and the value of ktot,Ni for the same level Ni is attributed to the occurrence of chemical reaction, yielding values of the rotationally selected rate coefficients (kreac,Ni) for reaction of CN from specified rotational levels. These rate coefficients decrease from (7.9+/-2.2)x10(-10) cm3molecule-1 s-1 for Ni=0 to (0.8+/-1.3)x10(-10) cm3 molecule-1 s-1 for Ni=20. The results are briefly discussed in the context of microcanonical transition state theory and the statistical adiabatic channel model.

Energy transfer in thermal and hyperthermal collisions between CN(X2Sigma+, v = 2) in selected rotational levels (Ni = 0, 1, 6, 10, 15 and 20) and N2.

We report rate coefficients (k(tot,N(i))) for total removal of CN(X(2)Sigma(+), v = 2, N(i)) radicals from selected rotational levels (N(i) = 0, 1, 6, 10, 15 and 20) and for state-to-state rotational energy transfer (k(i-->f)) between levels N(i) and other rotational levels N(f) in single collisions with N(2). CN radicals have been generated using two sources: (a) the pulsed laser photolysis of ICN at 266 nm, which generates translationally 'hot' CN radicals; and (b) the pulsed laser photolysis of NCNO at 570 nm, which generates CN radicals with translational energies close to the average value at 298 K. Comparison of the values of k(tot,N(i)) obtained using these two sources of CN demonstrates: firstly, that the same results are obtained as long as time is allowed for the translationally hot CN radicals generated from ICN to be thermalised before radicals are promoted to a specific rotational level in v = 2 using a tuneable infrared 'pump' laser operating at ca. 2.45 micro m; and secondly, that the rate coefficients decrease, but the averaged cross-sections remain approximately constant, as the excess translational energy in CN radicals is moderated by collisions. With NCNO as the source of CN radicals, the observed values of k(tot,N(i)) do not depend on the delay between the pulses from the photolysis and pump lasers. Finally, we demonstrate that, for the non-reactive collision partner N(2) and with allowances made for the rate coefficients that are too small to measure directly, the sum of the state-to-state rate coefficients, Sigma(f)k(i-->f), for rotational energy transfer from a selected initial level N(i) agrees quite well with the value of k(tot,N(i)) for total transfer from the same initial level. The values of k(tot,N(i)) and of the state-to-state rate coefficients are compared with similar, earlier, results in which helium and argon were the collision partners. The relevance of these results to the study of collisions of CN with reactive collision partners is briefly discussed.

The temperature-dependence of rapid low temperature reactions: experiment, understanding and prediction.

Despite the success of the CRESU (Cinétique de Réaction en Ecoulement Supersonique Uniforme) method in measuring rate coefficients for neutral-neutral reactions of radicals down close to the very low temperatures prevalent in dense interstellar clouds (ISCs), there are still many reactions of potential importance in the chemistry of these objects for which there have been no measurements of low temperature rate coefficients. One important class of reactions is that between atomic and molecular free radicals and unsaturated hydrocarbons; that is, alkynes and alkenes. Based on semi-empirical arguments and correlations of 'room temperature' rate coefficients, k(298 K), for reactions of this type with the difference between the ionisation energy of the alkyne/alkene and the electron affinity of the radical, we suggest which reactions between the radicals, C(3P), O(3P), N(4S), CH, C2H and CN, and carbon chain molecules (Cn) and cyanopolyynes (HC2nCN and NCC2nCN) are likely to be fast at the temperature of dense ISCs. These reactions and rate coefficients have been incorporated into a purely gas-phase model (osu2005) of ISC chemistry. The results of these calculations are presented and discussed.

Reactions at very low temperatures: gas kinetics at a new frontier.

Advances in experimental techniques, especially the development of the CRESU (Cinétique de Réaction en Ecoulement Supersonique Uniforme) method, allow many gas-phase molecular processes to be studied at very low temperatures. This Review focuses on the reactions of molecular and atomic radicals with neutral molecules. Rate constants for almost 50 such reactions have been measured at temperatures as low as 13 K by using the CRESU method. The surprising demonstration that so many reactions between electrically neutral species can be extremely rapid at these very low temperatures has excited interest both from theoreticians and from those seeking to understand the chemistry that gives rise to the 135 or so molecules that are present in low-temperature molecular clouds in the interstellar medium. Theoretical treatments of these reactions are based on the idea that a reaction occurs when the long-range potential between the reagent species brings them into close contact. The astrochemical context, theoretical studies, and the determination of the rate constants of these low-temperature reactions are critically discussed.

Kinetics of the radical-radical reaction, O(3P(J)) + OH(X2Pi omega) --> O2 + H, at temperatures down to 39 K.

The kinetics of the reaction between O atoms and OH radicals, both in their electronic ground state, have been investigated at temperatures down to ca. 39 K. The experiments employed a CRESU (Cinétique deRéaction en Ecoulement Supersonique Uniforme) apparatus to attain low temperatures. Both reagents were created using pulsed laser photolysis at 157.6 nm of mixtures containing H2O and O2 diluted in N2 carrier gas. OH radicals were formed by both direct photolysis of H2O and the reaction between O(1D) atoms and H2O. O(3P) atoms were formed both as a direct product of O2 photolysis and by the rapid quenching of O(1D) atoms formed in that photolysis by N2 and O2. The rates of removal of OH radicals were observed by laser-induced fluorescence, and concentrations of O atoms were estimated from a knowledge of the absorption cross-section for O2 at 157.6 nm and of the measured fluence from the F2 laser at this wavelength. To obtain a best estimate of the rate constants for the O + OH reaction, we had to correct the raw experimental data for the following: (a) the decrease in the laser fluence along the jet due to the absorption by O2 in the gas mixture, (b) the increase in temperature, and consequent decrease in gas density, as a result of energy released in the photochemical and chemical processes that occurred, and (c) the formation of OH(v = 0) as a result of relaxation, particularly by O2, of OH radicals formed in levels v > 0. Once these corrections were made, the rate constant for reaction between OH and O(3P) atoms showed little variation in the temperature range of 142 to 39 K and had a value of (3.5 +/- 1.0) x 10(-11) cm3 molecule(-1) s(-1). It is recommended that this value is used in future chemical models of dense interstellar clouds.

Rotational energy transfer in collisions between CO(X 1Sigma+, v= 2 , J = 0, 1, 4, and 6) and He at temperatures from 294 to 15 K.

Infrared-vacuum ultraviolet double resonance experiments have been implemented in the ultracold environment provided by a Cinetique de Reaction en Ecoulement Supersonique Uniforme apparatus. With this technique rate coefficients of two kinds have been measured for rotational energy transfer in collisions between CO and He: (a) those for total removal from the selected rotational states J = 0, 1, 4, and 6 in the vibronic state X 1Sigma+, v = 2, and (b) those for transfer between selected initial and specific final states. Using different Laval nozzles, results have been obtained at several different temperatures: 294, 149, 63, 27, and 15 K. The thermally averaged cross sections for total removal by collisions with He show only slight variations both with initial rotational state and with temperature. The variation of state-to-state rate coefficients with DeltaJ show several general features: (i) a decrease with increasing DeltaJ; (ii) a propensity to favor odd DeltaJ over even DeltaJ; and (iii) at lower temperatures, the distribution of rate coefficients against DeltaJ becomes narrower, and decreases in J are increasingly favored over increases in J, a preference which is most strongly seen for higher initial values of J. The results are shown to be in remarkably good agreement with those obtained in ab initio scattering calculations by Dalgarno and co-workers [Astrophys. J. 571, 1015 (2002)].

Relaxation of H2O from its /04>- vibrational state in collisions with H2O, Ar, H2, N2, and O2.

We report rate coefficients at 293 K for the collisional relaxation of H2O molecules from the highly excited /04>(+/-) vibrational states in collisions with H2O, Ar, H2, N2, and O2. In our experiments, the mid R:04(-) state is populated by direct absorption of radiation from a pulsed dye laser tuned to approximately 719 nm. Evolution of the population in the (/04>(+/-)) levels is observed using the combination of a frequency-quadrupled Nd:YAG laser, which selectively photolyses H2O(/04>(+/-)), and a frequency-doubled dye laser, which observes the OH(v=0) produced by photodissociation via laser-induced fluorescence. The delay between the pulse from the pump laser and those from the photolysis and probe lasers was systematically varied to generate kinetic decays. The rate coefficients for relaxation of H2O(/04>(+/-)) obtained from these experiments, in units of cm3 molecule(-1) s(-1), are: k(H2O)=(4.1+/-1.2) x 10(-10), k(Ar)=(4.9+/-1.1) x 10(-12), k(H2)=(6.8+/-1.1) x 10(-12), k(N2)=(7.7+/-1.5) x 10(-12), k(O2)=(6.7+/-1.4) x 10(-12). The implications of these results for our previous reports of rate constants for the removal of H2O molecules in selected vibrational states by collisions with H atoms (P. W. Barnes et al., Faraday Discuss. Chem. Soc. 113, 167 (1999) and P. W. Barnes et al., J. Chem. Phys. 115, 4586 (2001).) are fully discussed.

The Liversidge Lecture 2001-02. Chemistry amongst the stars: reaction kinetics at a new frontier.

Over the past decade, experiments in the Universities of Rennes and Birmingham have provided rate constants for over 40 reactions of molecular and atomic radicals with neutral molecules at temperatures down to 13 K using the CRESU (cinétique de réaction en écoulement supersonique uniforme) technique. The demonstration that reactions between electrically neutral species can be extremely rapid at these very low temperatures has excited interest both from theoreticians and from those seeking to understand the chemistry that gives rise to the 120 or so molecules that have been identified as being present in dense interstellar clouds. This laboratory work, and its astrochemical and theoretical contexts, are reviewed here. In addition, I deal briefly with the present limitations of the experiments and how they might be overcome in future work.