Product branching fractions for the reaction of O(
3
P) with ethenew
Akira Miyoshi,*
aJun-ichi Yoshida,
aNaoya Shiki,
aMitsuo Koshi
aand Hiroyuki Matsui
bcReceived 23rd March 2009, Accepted 22nd May 2009
First published as an Advance Article on the web 19th June 2009 DOI: 10.1039/b905787k
The product branching fractions for the reaction of atomic oxygen with ethene,
O(3P) + C2H4- CH3+ HCO (1a),- H + CH2CHO (1b),- H2+ CH2CO (1c), have been investigated at room temperature (295 4 K) and pressures from 1 to 4 Torr (with N2or He buffer) by a laser photolysis–photoionization mass spectrometry method. From the yield of CH3 radical, f(CH3), the branching fraction for (1a) was determined to be 0.53 0.04 and no apparent pressure dependence was found from 1.5 to 4.0 Torr (N2buffer). The ratio of the HCO yield to that of CH3, f(HCO)/f(CH3), was measured to be less than unity and increased as pressure increased (B0.7 at 1 Torr and B0.9 at 4 Torr [He]) suggesting prompt dissociation of the hot HCO radical (to H + CO) formed by channel (1a) at low pressures. An interpretation which reduces the large discrepancy among branching fractions reported for low pressure region is presented. The existence of the molecular H2-elimination channel (1c) was confirmed. The branching fraction for channel (1c) was determined to be 0.019 0.001 by the yield of CH2CO and was independent of pressure from 1.0 to 4.0 Torr (He buffer). As a side result, the yield of CH3radical from O(1D) + C2H4reaction was also determined.
Introduction
The reactions of atomic oxygen with alkenes play an important role in the combustion of hydrocarbons as one of the dominant degradation processes of alkenes. Extensive kinetic measurements have revealed small, or even negative, activation energies,1suggesting that the reactions involve the initial electrophilic addition of atomic oxygen to the double bonds.2 The product branching has been subject of various investigations for decades,3–20and has been discussed in terms of the site preference of the initial oxygen-atom attack and the subsequent isomerization and/or dissociation processes of the biradical adducts.
As a reaction with the simplest and prototypical alkene, the reaction of ethene (C2H4) has been most extensively investigated. The rate constant has been measured experimentally from room temperature to 2300 K.5After a long argument on the dominant channels,6–20 the reaction is now known to proceed via two major channels,
O(3P) + C2H4- CH3+ HCO, (1a)
-H + CH2CHO, (1b)
maybe with some contribution of minor channels,
-H2+ CH2CO, (1c)
-CH2+ H2CO. (1d)
Theoretical investigation also supported the initial formation of the triplet biradical.21,22A schematic energy diagram for the reaction is shown in Fig. 1. Earlier theoretical investigation23 was focused on the triplet dissociation route of the biradical and showed that the simple H-elimination to produce H + CH2CHO (1b) proceeds without a barrier higher than the reactants, while a significantly high barrier was found for the hydrogen shift to produce CH3CHO, which may further
Fig. 1 Energy diagram for the O(3P) + C2H4reaction based on the
quantum chemical calculations.24,25 Dotted and solid lines denote
triplet and singlet surfaces, respectively.
a
Department of Chemical System Engineering, University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-8656, Japan.
E-mail: miyoshi@chemsys.t.u-tokyo.ac.jp
bDepartment of Ecological Engineering, Toyohashi University of
Technology, Tempaku-cho, Toyohashi 441-8580, Japan
c
Department of Applied Chemistry, National Chiao Tung University, Hsinchu, Taiwan
w Electronic supplementary information (ESI) available: Detailed experimental results for O(1D) + C
2H4and for the determination of
CH2CO yields, as well as additional information on quantum chemical
calculations. See DOI: 10.1039/b905787k
PAPER www.rsc.org/pccp | Physical Chemistry Chemical Physics
dissociate to CH3+ HCO (1a). Later theoretical investigations24,25 including singlet surfaces supported the earlier calculation, and also indicated that, on the singlet surface, a hydrogen shift to CH3CHO proceeds easily while an H-elimination process involves a significantly high barrier.
The theoretical investigations suggest two major product channels, namely, direct dissociation of the triplet biradical to H + CH2CHO (1b) and a process involving ISC (intersystem crossing) to singlet biradical followed by the hydrogen-shift isomerization to acetaldehyde and further by the C–C bond fission to CH3+ HCO (1a). One of the main questions of the experimental investigations in the last three decades was how the reaction involves the ISC, that is, whether the ISC is induced by collisions or not. The experimental studies in the early 1980s at moderate pressures8 (80–760 Torr) and collision-free molecular beam condition9seemed to be consis-tent with the collision induced ISC mechanism, that is, the triplet channel (1b) dominates at collision free condition while both singlet (1a) and triplet (1b) channels are significant at higher pressures, although this interpretation contradicted an earlier qualitative measurement6 which had reported a significant formation of CH3and HCO even at the collision free condition.
However, a later low-pressure (30 mTorr) experiment12and revised molecular beam experiment15 showed significant branching to the singlet channel (1a) which involves ISC, suggesting the fast ISC without collisions. The argument seemed to be settled after the report16 which suggested the collision-induced ISC mainly from the pressure dependence of the yield of hydrogen atoms observed in the authors’ and earlier measurements.10,11,14 More recently, two molecular beam experiments have been reported.18,20 The reported branching fractions for (1b) significantly differ from each other, f1b= 0.62 (ref. 18) and 0.27 (ref. 20).
The discrepancy of f1bat lower pressures seems to be left unresolved since no crucial flaw could be found in either experimental investigations. However, an inspiring experi-mental information was reported by Quandt et al.,3 who observed a direct formation of CO molecule in the reaction of O(3P) + C2H4under single collision conditions. Since the C–H bond dissociation energy of HCO (D1298= 65.7 kJ mol1) is smaller than the exothermicity of channel (1a), 113.2 kJ mol1, the most probable source of the CO will be the dissociation of hot HCO formed via channel (1a), and they estimated that about half of H atoms are produced by the dissociation of HCO. Considering that the strong pressure dependence of f1b has been seen only in the measurement of H-atoms, the mechanism including the hot HCO (HCO*) dissociation and the competing stabilization,
HCO*- H+CO, (2)
HCO* + M- HCO + M, (3)
seems to explain the contradiction between H-atom measure-ments and others.
In the present study, the product yield from O(3P) + C 2H4 has been reinvestigated since few measurements for product other than H-atom have been reported in the intermediate pressure range (0.1–100 Torr). In the pressure range from 1 to
4 Torr, the yields were determined for CH3 and HCO for channel (1a), and CH2CO for channel (1c).
Experimental
Experiments were carried out on a laser photolysis– photoionization mass spectrometry apparatus. The apparatus is similar to those developed by Washida and co-workers.26In a tubular Pyrex glass reactor (15.6 mm id), flowing sample gas containing the precursor molecule for atom or radical was irradiated by pulsed 193-nm ArF excimer laser (Lambda Physik, COMPex 102) light. Gases were introduced to an ionization chamber through a pinhole (200 mm id) located at the side wall of the reactor. Reactant or product radicals were ionized by vacuum-ultraviolet light from a resonance lamp powered by microwave discharge. An H-lamp with MgF2 window (10.20 eV, 22PJ 1
2
S1/2) was used for the ionization of CH3radical (ionization potential [IP] = 9.84 eV), CH2CO (IP = 9.62 eV), and NO (IP = 9.26 eV), and Kr-lamp with CaF2window (10.03 eV, 5s2[3/2]01 4p6 1S) was used for CH3 and HCO (IP = 8.14 eV). Ions were mass-selected by a quadrupole mass filter (Anelva, AQA-200) and detected by stacked multichannel plates. Pulsed ion signals were dis-criminated and counted by a multichannel scaling circuit interfaced to a microcomputer. The gas flow velocity was kept high enough to assure the gas replacement between laser shots (repetition rate: 4–9 Hz).
O(3P) atoms were generated either by 193-nm photolysis of SO2,
SO2+ hn (l = 193 nm)- O(3P) + SO, (4) or by 193-nm photolysis of N2O followed by rapid quenching with N2,
N2O + hn (l = 193 nm)- O(1D) + N2, (5) O(1D) + N2- O(3P) + N2. (6) The initial O-atom concentration was kept low (3–6 1011atoms cm3) to minimize the effect of subsequent radical– radical or radical–atom reactions. For the determination of CH2CO yield, the initial O-atom concentration was determined by measuring NO formed by the reaction,
O(3P) + NO2- O2+ NO. (7)
All experiments were performed at room temperature (295 4 K). The error limits reported with the experimental results are two standard deviations throughout the paper. The gases used were obtained from Nippon Sanso (He, 499.9999%; N2, 499.9999%; SO2/He standard gas, 4.97%; N2O/He standard gas, 9.98%, NO2/He standard gas, 5.0%), Takachiho (C2H4, 499.9%; CH4, 499%), ISOTEC (13C2H4, 499% isotopic purity), and Katayama (CCl4, 499%; CH3COCH3, 499%). Formaldehyde (H2CO) was prepared by the thermal decomposition of paraformaldehyde27 (Wako, 495%) and purified by trap-to-trap distillation. Ketene (CH2CO) was prepared by the thermal decomposition of diketene (Tokyo Kasei, 499%).
Results
Branching fraction for (1a) [-CH3+ HCO]
The determination of the branching fractions for the major channels (1a) and (1b) might be influenced by the dissociation of HCO* (2) when the yield of HCO or H atom was measured, as described above. Thus, in the present study, the branching fraction for channel (1a), f1a, was determined by measuring the CH3 radical yield, f(CH3), against the initial O-atom concentration under the condition of excess C2H4.
The CH3yield was determined by comparing CH3 signal intensity with that from the reference reaction,
O(1D) + CH
4- OH + CH3, (8a)
-H + CH2OH (or CH3O). (8b) The O(1D) atoms were generated by 193-nm photolysis of N2O in He buffer. For the observation of O(3P) + C2H4 products, O(1D) was quenched by using N2buffer. The ratio of the CH3 concentration observed in N2O/C2H4/N2 mixture to that in N2O/CH4/He mixture, at the same N2O con-centration and laser fluence, gives f1a/f8a. The experiments with two gas mixtures were alternately repeated to cancel the error due to gradual change of the sensitivity, laser fluence, etc. The quenching of O(1D) by He (rate constant o3 1016cm3molecule1s1)28was negligible under the present experimental conditions.
Examples of the observed CH3signals are shown in Fig. 2. Since the detection sensitivity was significantly different under He and N2buffer, the signal intensities were corrected against this effect. The sensitivity difference was calibrated by separate experiments using the photolysis of acetone, CH3COCH3 + hn (l = 193 nm) - 2 CH3 + CO, in He and N2 buffer. Though minor, some further corrections were also made against: (1) O(1D) loss by O(1D) + N2O, (2) O(1D) reaction with C2H4, and (3) O(3P) heterogeneous loss at the reactor wall. The rate constants used for the correction were taken from the literature.5,29,30The CH3yield from O(
1
D) + C2H4 reaction was derived to be 0.54 0.03, from a separate experiment. [See ESIw for details.] The heterogeneous loss rate of O(3P) was also derived from a separate experiment.
The branching fraction for 8a, f8a= 0.71 0.05, was taken from a recent direct measurement of OH radical yield31using O(1D) + H2- OH + H as a reference reaction. The CH3 radical yield from O(3P) + C
2H4was measured in the pressure range from 1.5 to 4.0 Torr (N2), and the results are shown in Fig. 3. No apparent pressure dependence was found. As an average, the CH3yield, that is, the branching fraction for (1a) was determined to be f1a= 0.53 0.04.
HCO Yield from O(3P) + C2H4
In order to examine the possibility of the dissociation/ stabilization mechanism, reactions (2) and (3), respectively, of hot HCO* formed via channel (1a), the ratio of the yields of HCO to CH3, f(HCO)/f(CH3) was measured as a function of the pressure (He buffer) in the present study.
Since the HCO+ signal at m/z = 29 overlapped with the C2H5+from ethyl radical formed by
H + C2H4+ M- C2H5+ M, (9)
carbon-13 ethene (13C2H4) was used for this experiment. O( 3
P) atoms were generated by 193-nm photolysis of SO2. The signals at m/z = 30 (H13CO+) and m/z = 16 (13CH3+) were alternately recorded in the repeated experiments. In order to minimize the possibility of interference from fragment ions, a Kr-lamp with CaF2 window, which has the lowest photon energy sufficient to ionize both CH3 and HCO, was used. Sensitivity difference between HCO and CH3 was calibrated by a separate experiment, using
Cl + H2CO- HCO + HCl, (10)
Cl + CH4- CH3+ HCl, (11)
where Cl atoms were generated by the 193-nm photolysis of CCl4,
CCl4+ hn (l = 193 nm)- CCl3+ Cl, (12a) -CCl2+ 2 Cl. (12b) From the alternately repeated experiments observing HCO in the CCl4/H2CO/He mixture and observing CH3in CCl4/CH4/He, the sensitivity ratio was determined to be S(HCO)/S(CH3) = 0.63 0.07 for the Kr lamp. To minimize the influence of the heterogeneous loss of Cl atoms, the concentration of H2CO or
Fig. 2 Measurement of the yield of CH3radical from O(3P) + C2H4.
Left trace: CH3 signal observed in the photolysis of N2O/CH4/He
mixture ([N2O] = 13.6 mTorr, [CH4] = 129 mTorr, total pressure =
4.0 Torr, ArF laser fluence = 8.6 mJ cm2). Right trace: CH3signal
observed in the photolysis of N2O/C2H4/N2 mixture ([C2H4] =
32.9 mTorr, Other conditions are the same as the left trace).
Fig. 3 Yield of CH3 from O( 3
P) + C2H4 reaction, f(CH3).
Experimental conditions were similar to Fig. 2 except for the N2pressure.
CH4was chosen so that the time constant for reaction (10) or (11) is less than 0.2 ms. Since reaction (11) is slow and high concentration of CH4 was required, a correction was made against the sensitivity change due to the large amount of CH4 (probably caused by the VUV absorption of CH4 in the ionization chamber) by a separate calibration experiment.
The yield ratio, f(HCO)/f(CH3), was determined in the pressure range 0.6–4.5 Torr (He buffer), and is shown in Fig. 4. The determined ratio was less than unity and pressure-dependent. The observed behavior is consistent with the dissociation/stabilization mechanism of hot HCO* radical. By assuming a simple Lindemann–Hinshelwood type mechanism for reactions (2) and (3), the yield ratio can be expressed as
fðHCOÞ fðCH3Þ ¼ f0þ ð1 f0Þ ½M ðk2=k3Þ þ ½M ð13Þ where f0is the yield ratio at zero-pressure, that is, the fraction of HCO which survives even at zero-pressure. Here, k2and k3 denote the rate constants for reactions (2) and (3), respectively, and [M] is the concentration of buffer gas. The solid line in Fig. 4 shows a least-squares fit to eqn (13). The parameters in eqn (13) were derived to be f0= 0.60 and k2/k3= 1.0 1017molecule cm3(=3.0 Torr).
Branching fraction for (1c) [- H2+CH2CO]
Since few branching-fraction measurements7,19 have been reported for this channel so far, experimental determination was done by measuring the yield of CH2CO. The initial concentration of O(3P) was measured by observing NO from the reaction of O(3P) with NO2 (7). Since the subsequent reaction of CH2CHO [formed via the channel (1b)] with other radicals,
CH2CHO + X- CH2CO + XH, (14) where X = O(3P), CH3, etc., may produce CH2CO as well, the experimental conditions were carefully chosen to avoid the subsequent reactions. The signal intensity of CH2CO linearly depended on the initial concentration of O(3P) at lower concentration while it significantly curved at higher con-centration aboveB1012atoms cm3. The initial concentration was kept low enough,B3 1011 atoms cm3. Further, the
analysis of the rise rate of CH2CO by changing the concentra-tion of C2H4 gave a rate constant for O(3P) + C2H4 (1) as 7.3 1013cm3molecule1s1, which is in good agreement with the reported5rate constant for reaction (1) (7.2 1013at 295 K), which also supports the fact that the observed CH2CO is the primary product of reaction (1). Details of these confirmation experimental results are shown in the ESI.w
Experiments of CH2CO measurements in the photolysis of SO2/C2H4/He mixtures and NO measurements in the photo-lysis of SO2/NO2/He mixtures were repeated alternately to obtain the signal intensity ratio of CH2CO to NO. The ratio of the sensitivity of CH2CO to that of NO, S(CH2CO)/S(NO) was determined by a calibration experiment as 22.6 1.5 for H-lamp. The signal intensity ratios were thus converted to the concentration ratios and are plotted in Fig. 5. No apparent pressure dependence was found in the pressure range 1.0 to 4.0 Torr. The branching fraction for the channel (1c) was determined to be f1c= 0.019 0.001, which is in reasonable agreement with the recent theoretical prediction, 0.024.25
Discussion
In the present study, the branching fractions for channels (1a) and (1c) were determined as; f1a (-CH3 + HCO) = 0.53 0.04 and f1c(-H2+ CH2CO) = 0.019 0.001, and no apparent pressure dependence was found in the pressure range 1–4 Torr (N2or He) in either channel.
HCO from reaction (1)
The yield of HCO from O(3P) + C2H4 reaction, f(HCO), derived from the present results is plotted in Fig. 6 with the previous reports. Although not mentioned in the report, the HCO yield determined by LMR14 also showed pressure dependence and was in good agreement with the present result. Also the f(HCO) measured at higher pressure by visible absorption8is in reasonable agreement with the extrapolation of the present results. These also support the possibility of the prompt dissociation of HCO* (2). By comparing the LIF intensity of CO and H against those observed for the reaction O(3P) + C2H2, Quandt et al.
3
estimated the yield ratio
Fig. 4 Ratio of the yield of H13CO to that of 13CH3 from
O(3P)+13C
2H4reaction measured by 193-nm photolysis of SO2/C2H4/He
mixtures. Experimental conditions: [C2H4] = 28 mTorr, [SO2] =
6.2 mTorr, laser fluence = 0.5 mJ cm2.
Fig. 5 The yield of ketene from O(3P) + C2H4reaction, f(CH2CO),
measured by 193-nm photolysis of SO2/C2H4/He mixtures in
compar-ison with SO2/NO2/He mixtures. Experimental conditions: [C2H4] =
20 mTorr, [NO2] = 1.4 mTorr, [SO2] = 2 mTorr, laser fluence =
1.0 mJ cm2.
f(CO)/f(H) to beB0.5 at single collision conditions, which is in reasonable agreement with f(CO)/f(H) = 0.35 expected from the present measurement by extrapolation to the zero pressure by assuming f1b= 0.40.25
The vibrational structure of the HCO molecule has been well understood from recent spectroscopic32and theoretical33 studies. The vibrational state density is small at the low energy region of HCO. Only 15 vibrational states lie below the tunneling thresholds (=the H + CO asymptote), while 9 resonances were found between the classical threshold (the barrier height) and tunneling thresholds. It seems to be reason-able to expect that a significant portion of HCO is formed with the vibrational energies around the two thresholds, though a further quantitative argument requires detailed information on the energy distribution and collisional vibrational energy transfer processes, which is, however, unavailable. From the result of a simple analysis by eqn (13) (k2/k3 = 1.0 1017molecule cm3) with the estimated collision rate, k3 B 4 1010 cm3 molecule1 s1, the rate of unimolecular dissociation of HCO* was estimated to be k2B 4 107s1, which is in the allowable range for the resonances calculated33 in the tunneling region [o2 107for (0,1,3) and (0,0,5), 1.6 108for (1,1,1), 3 109for (1,0,3), and 2 107for (0,3,0)].
Ketene channel:- H2+CH2CO (1c)
In the present study, precise kinetic measurements of the ketene (CH2CO) indicated that it is a primary product of the reaction of O(3P) + C
2H4. Since the product pair can only correlate the singlet potential energy surface, the molecular H2 elimination most likely occurs on the singlet surface. Although
the major routes to H2 + CH2CO have been characterized by Nguyen et al,25 direct 1,1-H2 elimination from the singlet biradical suggested in a previous work15has not been calculated.
Therefore, the direct 1,1-H2elimination transition state was searched by quantum chemical calculations by using Gaussian 98.34The results are shown in Fig. 7, and additional details are shown in the ESI.w The energies of the stationary points were estimated by G3(MP2) method.35It should be noted that the UHF calculations of the singlet biradical and the correlated transition states indicated a large spin contamination,hS2i = 0.5B1.0, and the energy calculated by the UHF-based method might not be accurate enough.
The zero-point energy corrected G3(MP2) energy of the singlet biradical lies 99 kJ mol1 below the reactants, O(3P) + C2H4, and 2 kJ mol1below the triplet biradical. A saddle point, SP1, correlating the singlet biradical and H2+ CH2CO was found at the energy only 7 kJ mol1above the singlet biradical. Other features were essentially similar to the previous theoretical calculation.25Though Nguyen et al,25 predicted the branching fraction for H2+ CH2CO (1c) to be 2.4% at 298 K as a sum of other three indirect channels, the direct elimination via SP1 may be of some importance. The lack of apparent pressure dependence in the CH2CO yield, f(CH2CO), suggests that the ISC is fast enough even in the present experimental pressure range (1–4 Torr).
Pressure dependence of the branching fraction and other channels
The present experimental results indicate no apparent pressure dependence of the branching fraction for (1a) or (1c). The
Fig. 6 Yields of HCO radical from O(3P) + C2H4reaction, f(HCO).
Closed circles (K) denote present results. Open triangles (n) and squares (&) denote previous measurements by LMR14 and visible absorption,8respectively.
Fig. 7 Part of the energy diagram for O(3P) + C2H4system relevant
to CH2CO formation calculated by G3(MP2) method. The plotted
energies are the potential energy corrected for the zero point energy. Dotted and solid lines denote triplet and singlet surfaces, respectively.
Table 1 Summary of the branching fractions for O(3P) + C 2H4
Channel Present study (295 4 K) Theoreticala(298 K) Evaluationb(298 K)
(1a) CH3+CHO 0.53 0.04 0.477 0.6 0.1 (1b) H+CH2CHO 0.401 0.35 0.05 (1c) H2+CH2CO 0.019 0.001 0.024 0.05 0.1 (1d) CH2+H2CO 0.052 (1e) CH3CO+H 0.022 (1f) CH4+CO 0.023 a Ref. 25.bRef. 5.
pressure dependence of HCO yield suggests prompt dissociation of HCO* formed via (1a). This implies that, at least some portion of the H-atom yield, which is observed to increase toward lower pressure, can be attributed to the result of HCO* dissociation. This interpretation seems to reduce the large discrepancy among the reported branching fractions at low pressures. However, the larger branching fraction for (1b) in a molecular beam experiment18 (0.62) as well as the smaller value (0.36) at moderate pressures8(80 and 760 Torr) may still imply some pressure dependence in a wide pressure range. It seems to be obvious that ISC occurs without collision, though this does not contradict the possibility of further ISC enhancement by collisions.
In summary, the branching fractions determined in the present study are compared with the results of the theoretical investigation25 and the evaluation5 in Table 1. Though not measured in the present study, reaction (1d) producing CH2 + H2CO has been recognized as one of the major channels.20,25
Conclusions
(1) The branching fractions for O(3P) + C
2H4 - CH3 + HCO (1a) and - H2+ CH2CO (1c) have been determined to be 0.53 0.04 [1.5–4.0 Torr (N2)] and 0.019 0.001 [1.0–4.0 Torr (He)], respectively, at room temperature. No apparent pressure dependence was observed in either branch-ing fraction in the experimental pressure range.
(2) The yield ratio, f(HCO)/f(CH3), was less than unity and pressure dependent, in agreement with a previous HCO measurement. This result, together with the reported direct CO formation, indicates the dissociation of hot HCO* (-H + CO) formed in (1a). The pressure dependence of f(HCO)/f(CH3) could be explained by the competition between dissociation and collisional stabilization. This inter-pretation seems to reduce the discrepancy among previous branching fractions at low pressures.
(3) Precise kinetic measurements indicated the existence of the molecular elimination channel, (1c) (-H2+ CH2CO).
Acknowledgements
The authors thank Prof. Nobuaki Washida for useful dis-cussions in regard to this study. H.M. acknowledges support from the National Science Council of Taiwan and National Chiao Tung University, Taiwan.
References
1 R. J. Cvetanovic´, J. Phys. Chem. Ref. Data, 1987, 16, 261. 2 R. J. Cvetanovic´ and D. L. Singleton, Rev. Chem. Intermediates,
1984, 5, 183.
3 R. Quandt, Z. Min, X. Wang and R. Bersohn, J. Phys. Chem. A, 1998, 102, 6; Z. Min, T.-H. Wong, R. Quandt and R. Bersohn, J. Phys. Chem. A, 1999, 103, 10451.
4 N. Washida, S. Inomata and M. Furubayashi, J. Phys. Chem. A, 1998, 102, 7924.
5 D. L. Baulch, C. T. Bowman, C. J. Cobos, R. A. Cox, Th. Just, J. A. Kerr, M. J. Pilling, D. Stocker, J. Troe, W. Tsang, R. W. Walker and J. Warnatz, J. Phys. Chem. Ref. Data, 2005, 34, 757, and references therein.
6 J. R. Kanofsky and D. Gutman, Chem. Phys. Lett., 1972, 15, 236; J. R. Kanofsky, D. Lucas and D. Gutman, Proc. Combust. Inst., 1972, 14, 285.
7 F. J. Pruss, Jr, I. R. Slagle and D. Gutman, J. Phys. Chem., 1974, 78, 663.
8 H. E. Hunziker, H. Kneppe and H. R. Wendt, J. Photochem., 1981, 17, 377.
9 R. J. Buss, R. J. Baseman, G. He and Y. T. Lee, J. Photochem., 1981, 17, 389.
10 U. C. Sridharan and F. Kaufman, Chem. Phys. Lett., 1983, 102, 45. 11 J. F. Smalley, F. L. Nesbitt and R. B. Klemm, J. Phys. Chem.,
1986, 90, 491.
12 Y. Endo, S. Tsuchiya, C. Yamada, E. Hirota and S. Koda, J. Chem. Phys., 1986, 85, 4446; S. Koda, Y. Endo, S. Tsuchiya and E. Hirota, J. Phys. Chem., 1991, 95, 1241.
13 S. Koda, Y. Endo, E. Hirota and S. Tsuchiya, J. Phys. Chem., 1987, 91, 5840.
14 U. Bley, P. Dransfeld, B. Himme, M. Koch, F. Temps and H. Gg. Wagner, Proc. Combust. Inst., 1988, 22, 997.
15 A. M. Schmoltner, P. M. Chu, R. J. Brudzynski and Y. T. Lee, J. Chem. Phys., 1989, 91, 6926.
16 V. D. Knyazev, V. S. Arutyunov and V. I. Vedeneev, Int. J. Chem. Kinet., 1992, 24, 545.
17 C. Anastasi, M. G. Sanderson, P. Pagsberg and A. Sillesen, J. Chem. Soc., Faraday Trans., 1994, 90, 3625.
18 M. L. Morton, D. E. Szpunar and L. J. Butler, J. Chem. Phys., 2001, 115, 204.
19 J. Peeters and D. Maes, Tenth International Symposium on Gas Kinetis, University College of Swansea, July 1988, p. A31. 20 P. Casavecchia, G. Capozza, E. Segoloni, F. Leonori,
N. Baulucani and G. G. Volpi, J. Phys. Chem. A, 2005, 109, 3527. 21 K. Yamaguchi, Y. Yabushita, T. Fueno, S. Kato and
K. Morokuma, Chem. Phys. Lett., 1980, 70, 27.
22 M. Dupuis, J. J. Wendoloski, T. Takada and W. A. Lester, Jr, J.Chem. Phys., 1982, 76, 481.
23 C. F. Melius, J. S. Binkley, M. L. Koszykowski, as cited in K. Mahmud, P. Marshall and A. Fontijn, J. Phys. Chem., 1987, 91, 1568.
24 T. Fueno, Y. Takahara and K. Yamaguchi, Chem. Phys. Lett., 1990, 167, 291.
25 T. L. Nguyen, L. Vereecken, X.-J. Hou, M. T. Nguyen and J. Peeters, J. Phys. Chem. A, 2005, 109, 7489.
26 M. Masaki, S. Tsunashima and N. Washida, Chem. Phys. Lett., 1994, 218, 523; A. Miyoshi, H. Matsui and N. Washida, J. Phys. Chem., 1990, 94, 3016.
27 R. Spence and W. Wild, J. Chem. Soc., 1935, 338. 28 K. Schofield, J. Photochem., 1978, 9, 55.
29 R. Atkinson, D. L. Baulch, R. A. Cox, R. F. Hampson, Jr, J. A. Kerr, M. J. Rossi and J. Troe, J. Phys. Chem. Ref. Data, 1997, 26, 521.
30 O. Kajimoto and T. Fueno, Chem. Phys. Lett., 1979, 64, 445. 31 F. Ausfelder, H. Hippler and F. Striebel, Z. Phys. Chem., 2000,
214, 403.
32 D. W. Neyer, X. Luo, I. Burak and P. L. Houston, J. Chem. Phys., 1995, 102, 1645; C. Sto¨ck, X. Li, H.-M. Keller, R. Schinke and F. Temps, J. Chem. Phys., 1997, 106, 5333, and references therin. 33 H.-M. Keller, H. Floethmann, A. J. Dobbyn, R. Schinke, H.-J. Werner,
C. Bauer and P. Rosmus, J. Chem. Phys., 1996, 105, 4983.
34 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle and J. A. Pople, GAUSSIAN 98 (Revision A.7), Gaussian Inc., Pittsburgh PA, 1998. 35 L. A. Curtiss, P. C. Redfern, K. Raghavachari, V. Rassolov and
J. A. Pople, J. Chem. Phys., 1999, 110, 4703.
