Direct gas-liquid interfacial dynamics: the reaction between O(3P) and a liquid hydrocarbon

We report the first measurements of internal energy distributions of the OH produced via a direct mechanism, isolated from other components on the basis of time-of-flight, in the interfacial reaction between gas-phase O((3)P) atoms and the liquid hydrocarbon squalane, C(30)H(62). O((3)P) atoms were generated by laser photolysis of NO(2) above the liquid. Resulting hydroxyl radicals that escape from the surface were detected by laser-induced fluorescence. Time-of-flight profiles demonstrate that the kinetic energy of the fastest OH (nu' = 1) is lower than that of (nu' = 0). Rotational distributions were measured at the rising edge of their appearance for both OH (nu' = 0) and (nu' = 1). They were found to differ substantially more than at the peak of their profiles. They were also less dependent on the bulk liquid temperature. We conclude that the new data confirm strongly that at least two mechanisms contribute to the production of OH. The higher-velocity component has translational and rotational energy distributions, observed cleanly for the first time, consistent with a direct mechanism. The close correspondence of these rotational distributions to those from the corresponding homogeneous gas-phase reaction of O((3)P) with smaller hydrocarbons suggests a very similar, near collinear direct abstraction. This is accompanied by a slower component with kinetic energy and rotational (but not vibrational) distributions reflecting the temperature of the liquid, consistent with a distinct trapping-desorption mechanism.     

The reaction of O(1D) with CH4

The room-temperature photolysis of N₂O (10–100 torr) at 2139 Å to produce O(¹D) has been studied in the presence of CH₄ (10–891 torr). The reactions of O(¹D) with CH₄ were found to be The method of chemical difference was used to measure the rate constant ratio k₄/(k₂ + k₃), where reactions (2) and (3) are The CH₃ radicals produced in reaction (4) react with the O₂ and NO produced in reactions (2) and (3). Thus, near the endpoint of the internal titration, ?{C₂H₆} gives an accurate measure of k₄/(k₂ + k₃). For the translationally energetic O(¹D) atoms produced in the photolysis, k₄/(k₂ + k₃) = 2.28 ± 0.20. However, if He is added to remove the excess translational energy, then k₄/(k₂ + k₃) drops to 1.35 ± 0.3.     

Multiple channel dynamics in the O(1D) reaction with alkanes

In this article, we briefly review the recent experimental studies of the multiple channel dynamics of the O((1)D) reaction with alkane molecules using the significantly improved universal crossed molecular beam technique. In these reactions, the dominant reaction mechanism is found to be an O atom insertion into the C-H bond, while a direct abstraction mechanism is also present in the OH formation channel. While the reaction mechanism is similar for all of these reactions, the product channels are quite different because of the significantly different energetics of these reaction channels. In the O((1)D) reaction with methane, OH formation is the dominant process while H atom formation is also a significant process. In the O((1)D) reaction with ethane, however, the CH(3) + CH(2)OH is the most important process, OH formation is still significant and H atom formation is of minor importance. A new type of O atom insertion mechanism (insertion into a C-C bond) is also inferred from the O((1)D) reaction with cyclopropane. Through these comprehensive studies, complete dynamical pictures of many multiple channel chemical reactions could be obtained. Such detailed studies could provide a unique bridge between dynamics and kinetics research.     

The effects of surface temperature on the gas-liquid interfacial reaction dynamics of O(3P)+squalane

OH/OD product state distributions arising from the reaction of gas-phase O(3P) atoms at the surface of the liquid hydrocarbon squalane C30H62/C30D62 have been measured. The O(3P) atoms were generated by 355 nm laser photolysis of NO2 at a low pressure above the continually refreshed liquid. It has been shown unambiguously that the hydroxyl radicals detected by laser-induced fluorescence originate from the squalane surface. The gas-phase OH/OD rotational populations are found to be partially sensitive to the liquid temperature, but do not adapt to it completely. In addition, rotational temperatures for OH/OD(v'=1) are consistently colder (by 34+/-5 K) than those for OH/OD(v'=0). This is reminiscent of, but less pronounced than, a similar effect in the well-studied homogeneous gas-phase reaction of O(3P) with smaller hydrocarbons. We conclude that the rotational distributions are composed of two different components. One originates from a direct abstraction mechanism with product characteristics similar to those in the gas phase. The other is a trapping-desorption process yielding a thermal, Boltzmann-like distribution close to the surface temperature. This conclusion is consistent with that reached previously from independent measurements of OH product velocity distributions in complementary molecular-beam scattering experiments. It is further supported by the temporal profiles of OH/OD laser-induced fluorescence signals as a function of distance from the surface observed in the current experiments. The vibrational branching ratios for (v'=1)/(v'=0) for OH and OD have been found to be (0.07+/-0.02) and (0.30+/-0.10), respectively. The detection of vibrationally excited hydroxyl radicals suggests that secondary and/or tertiary hydrogen atoms may be accessible to the attacking oxygen atoms.     

The reaction of O(1D) with H2O and the reaction of OH with C3H6

N₂O was photolyzed at 2139 Å to produce O(¹D) atoms in the presence of H₂O and CO. The O(¹D) atoms react with H₂O to produce HO radicals, as measured by CO₂ production from the reaction of OH with CO. The relative importance of the various possible O(¹D )–H₂O reactions is The relative rate constant for O(¹D) removal by H₂O compared to that by N₂O is 2.1, in good agreement with that found earlier in our laboratory. In the presence Of C₃H₆, the OH can be removed by reaction with either CO or C₃H₆: From the CO₂ yield, k₃/k₂ = 75,0 at 100°C and 55.0 at 200°C to within ± 10%. When these values are combined with the value of k₂ = 7.0 × 10^?13exp (–1100/RT) cm³/sec, k₃ = 1.36 × 10^?11 exp (–100/RT) cm³/sec. At 25°C, k₃ extrapolates to 1.1 × 10^?11 cm³/sec.     

H2 formation in the reaction of O(1D) with CH4

The reaction of O(¹D) with CH₄ was studied to determine the efficiency of H₂ production in a direct process, and it was found to be 0.11 ± 0.02. Thus the two channels which account for all of the reaction between O(¹D) and CH₄ in the gas phase are      

Energy partitioning in the reaction O(1D) + H2O → OH + OH.: The influence of O(1D) translational energy on the reaction rate constant

For the exothermic reaction O (¹D) + H₂O → OH + OH the rate constant and its energy dependence were determined by monitoring the concentration of the OH product. This was done by integrating the distribution of product molecules over all accessible states. The rate constant, determined at different velocity distributions of the reacting metastable over a wide range, is energy independent.     

Kinetic and mechanistic study of the reaction of O(1D) with CF2HBr

A laser flash photolysis–resonance fluorescence technique has been employed to investigate the kinetics and mechanism of the reaction of electronically excited oxygen atoms, O(¹D), with CF₂HBr. Absolute rate coefficients (k₁) for the deactivation of O(¹D) by CF₂HBr have been measured as a function of temperature over the range 211–425 K. The results are well described by the Arrhenius expression k₁(T) = 1.72 × 10^?10 exp(+72/T) cm³molecule^?1 s^?1; the accuracy of each reported rate coefficient is estimated to be ±15% (2σ). The branching ratio for nonreactive quenching of O(¹D) to the ground state, O(³P), is found to be 0.39 ± 0.06 independent of temperature, while the branching ratio for production of hydrogen atoms at 298 K is found to be 0.02_?0.02^+0.01. The above results are considered in conjunction with other published information to examine reactivity trends in O(¹D) + CF₂XY reactions (X,Y = H, F, Cl, Br). © 2001 John Wiley & Sons, Inc. Int J Chem Kinet 33: 262–270, 2001     

Reaction of O(1D) with N2O

O(¹D), produced from the photolysis of N₂O at 2139 Å, reacts with N₂O in accord with: We have used the method of chemical difference to obtain an accurate measure of k₂/k₃ = 0.59 ± 0.01. Furthermore, the quantum yield of production of O(³P), either on direct photolysis or on deactivation of O(¹D) by N₂O, is less than 0.02 and probably zero.     

IR kinetic spectroscopy investigation of the CH4 + O(1D) reaction

The branching of the title reaction into several product channels has been investigated quantitatively by laser infrared kinetic spectroscopy for CH(4) and CD(4). It is found that OH (OD) is produced in 67 +/- 5% (60 +/- 5%) yield compared to the initial O((1)D) concentration. H (D) product is produced in 30 +/- 10%(35 +/- 10%). H(2)CO is produced in 5% yield in the CH(4) system (it was not possible to measure the CD(2)O yield in the CD(4) case). D(2)O is produced in 8% yield in the CD(4) system (it was not feasible to measure the H(2)O yield). The ratio of the overall rate constant of the CD(4) reaction to the overall rate constant of the O((1)D) + N(2)O reaction was determined to be 1.2(5) +/- 0.1. A measurement of the reaction of O((1)D) with NO(2) gave 1.3 x 10(-10) cm(3) molecule(-1) s(-1) relative to the literature values for the rate constants of O((1)D) with H(2) and CH(4). Hot atom effects in O((1)D) reactions were observed.     

The reaction of OH with NO2 and the deactivation of O(1D) by CO

NO₂ was photolyzed with 2288 Å radiation at 300° and 423°K in the presence of H₂O, CO, and in some cases excess He. The photolysis produces O(¹D) atoms which react with H₂O to give HO radicals or are deactivated by CO to O(³P) atoms The ratio k₅/k₃ is temperature dependent, being 0.33 at 300°K and 0.60 at 423°K. From these two points, the Arrhenius expression is estimated to be k₅/k₃ = 2.6 exp(?1200/RT) where R is in cal/mole – °K. The OH radical is either removed by NO₂ or reacts with CO The ratio k₂/k^α is 0.019 at 300°K and 0.027 at 423°K, and the ratio k₂/k⁰ is 1.65 × 10^?5M at 300°K and 2.84 × 10^?5M at 423°K, with H₂O as the chaperone gas, where k^α = k₁ in the high-pressure limit and k⁰[M] = k₁ in the low-pressure limit. When combined with the value of k₂ = 4.2 × 10⁸ exp(?1100/RT) M^?1sec^?1, k^α = 6.3 × 10⁹ exp (?340/RT)M^?1sec^?1 and k⁰ = 4.0 × 10¹²M^?2sec^?1, independent of temperature for H₂O as the chaperone gas. He is about 1/8 as efficient as H₂O.     

Vibrational excitation of OH(X2Π) produced in the reaction of O(1D) with H2

A lower limit to the OH(X²Π) vibrational excitation produced by the reaction O(¹D) + H₂ has been observed using a low-pressure infrared chemiluminescence apparatus. The O(¹D) was generated by laser photolysis of O₃. The measured OH(v') vibrational distribution is inverted; it peaks at v' = 2.     

Quantum scattering calculation of the O(1D)+HBr reaction

Three-dimensional time-dependent quantum wave packet calculation for the O((1)D)+HBr reaction has been carried out using an accurate ab initio global potential energy surface [K. A. Peterson, J. Chem. Phys. 113, 4598 (2000)]. The calculations show that the initial state-selected reaction probabilities are dominated by resonance structures, and the lifetime of the resonance is generally in the subpicosecond time scale. The energy dependence of the reaction cross section is computed, which manifests still resonance structures, and is a decreasing function of the translational energy. The thermal rate constants are also computed, which are nearly independent on the temperature. The calculation results are discussed and compared to similar reaction with deep well.     

A highly sensitive method for time-resolved detection of O(1D) applied to precise determination of absolute O(1D) reaction rate constants and O(3P) yields

We demonstrate detection, in the gas-phase, of O(1D2) at concentrations down to 10(7) cm(-3) and develop this new method for time-resolved kinetic studies allowing both the total removal rate of O(1D2), of up to 1.5 x 10(6) s(-1), and the fraction quenched to O(3P(J)) by species X, k(q)/k(X), to be determined precisely from a single time profile: at 295 K we find, k(O(1D2) + N2O) = (1.43 +/- 0.08) x 10(-10) cm3 s(-1) with k(q)/k(N2O) = 0.056 +/- 0.009; k(O(1D2) + C2H2) = (3.1 +/- 0.2) x 10(-10) cm3 s(-1) with k(q)/k(C2H2) = 0.020 +/- 0.010; k(q)/k(H2O) < 0.003 for O(1D2) + H2O.     

Kinetics of O(1D) reactions with bromocarbons

The rate coefficients for the removal of O(¹D) by reaction with the following seven bromocarbons were measured using time-resolved vacuum UV atomic resonance fluorescence detection of O(³P): CH₃Br, CH₂Br₂, CHBr₃, CF₃Br (Halon-1301), CF₂Br₂ (Halon-1202), CF₂ClBr (Halon-1211), and CF₂BrCF₂Br (Halon-2402). The branching ratios for the quenching of O(¹D) to O(³P) by the above molecules were also determined. From these measurements, the rate coefficients for reactive and nonreactive loss of O(¹D) were calculated. These results are discussed in terms of their atmospheric relevance and trends in reactivity of O(¹D) with hydrocarbons upon halogen substitution. © 1993 John Wiley & Sons, Inc.     

A crossed molecular beam study of the O(1D) + C(3)H(8) reaction: multiple reaction pathways.

The O((1)D) + C(3)H(8) reaction has been reinvestigated using the universal crossed molecular beam method. Three reaction channels, CH(3) + C(2)H(4)OH, C(2)H(5) + CH(2)OH, and OH + C(3)H(7), have been observed. All three channels are significant in the title reaction with the C(2)H(5) formation process to be the most important, while the CH(3) formation and the OH formation channels are about equal. Product kinetic energy distributions and angular distributions have been determined for the three reaction channels observed. The oxygen-containing radicals in the CH(3) and C(2)H(5) formation pathways show forward-backward symmetric angular distribution relative to the O atom beam, while the OH product shows a clearly forward angular distribution. These results indicate that the OH formation channel seems to exhibit different dynamics from the CH(3) and C(2)H(5) channels.     

Computational study of carbon atom (3P and 1D) reaction with CH2O. Theoretical evaluation of 1B1 methylene production by C (1D)

Singlet and triplet free energy surfaces for the reactions of C atom ((3)P and (1)D) with CH(2)O are studied computationally to evaluate the excited singlet ((1)B(1)) methylene formation from deoxygenation of CH(2)O by C ((1)D) atom as suggested by Shevlin et al. Carbon atoms can react by addition to the oxygen lone pair or to the C=O double bond on both the triplet and singlet surfaces. Triplet C ((3)P) atoms will deoxygenate to give CO plus CH(2) ((3)B(1)) as the major products, while singlet C ((1)D) reactions will form ketene and CO plus CH(2) ((1)A(1)). No definitive evidence of the formation of excited singlet ((1)B(1)) methylene was found on the singlet free energy surface. A conical intersection between the (1)A' and (1)A' ' surfaces located near an exit channel may play a role in product formation. The suggested (1)B(1) state of methylene may form via the (1)A' ' surface only if dynamic effects are important. In an effort to interpret experimental observation of products trapped by (Z)-2-butene, formation of cis- and trans-1,2-dimethylcyclopropane is studied computationally. The results suggests that "hot" ketene may react with (Z)-2-butene nonstereospecifically.     

The kinetics of the reaction of O(3P) and D atoms with cyclohexane

The kinetics of the reactions of O(³P) and D atoms with cyclohexane have been investigated using fast-flow techniques. The rates of reaction were computed by monitoring changes in both atom and cyclohexane concentrations using electron spin resonance and mass spectrometric methods, respectively. The O(³P) + C₆H₁₂ reaction was studied over a temperature range of 344 to 513°K and we obtain a specific rate constant of (3.2±0.6) × 10¹⁴ exp (?4400±400/RT) cm³/mole˙sec for this reaction. The only reaction product detectable mass spectrometrically under flow conditions of excess oxygen atoms is formaldehyde. The D + C₆H₁₂ reaction was studied over a temperature range of 297 to 596°K. A specific rate constant of (4.1±1.0) × 10¹³ exp (?4000±300/RT) cm³/mole˙sec was obtained for this reaction. On the basis of the results obtained in these studies, the important primary process in both the O(³P) and D atom reactions is concluded to be abstraction of a hydrogen atom from the cyclohexane molecule.     

Crossed molecular beam studies on the reaction dynamics of O(1D)+N2O

The reaction of oxygen atom in its first singlet excited state with nitrous oxide was investigated under the crossed molecular beam condition. This reaction has two major product channels, NO+NO and N2+O2. The product translational energy distributions and angular distributions of both channels were determined. Using oxygen-18 isotope labeled O(1D) reactant, the newly formed NO can be distinguished from the remaining NO that was contained in the reactant N2O. Both channels have asymmetric and forward-biased angular distributions, suggesting that there is no long-lived collision complex with lifetime longer than its rotational period. The translational energy release of the N2+O2 channel (fT = 0.57) is much higher than that of the NO+NO channel (fT = 0.31). The product energy partitioning into translational, rotational, and vibrational degrees of freedom is discussed to learn more about the reaction mechanism. The branching ratio between the two product channels was estimated. The 46N2O product of the isotope exchange channel, 18O+44N2O-->16O+46N2O, was below the detection limit and therefore, the upper limit of its yield was estimated to be 0.8%.     

Dynamics of the C(1D)+D2 reaction: a comparison of crossed molecular-beam experiments with quasiclassical trajectory and accurate statistical calculations

In this paper we report a combined experimental and theoretical study on the dynamics of the insertion reaction C((1)D)+D(2) at 15.5 kJ mol(-1) collision energy. Product angular and velocity distributions have been obtained in crossed beam experiments and quasiclassical trajectory (QCT) and rigorous statistical calculations have been performed on the recent and accurate ab initio potential energy surface of Bussery-Honvault, Honvault, and Launay at the energy of the experiment. The molecular-beam results have been simulated using the theoretical calculations. Good agreement between experiment and both QCT and statistical predictions is found.