6586
J. Phys. Chem. A 2010, 114, 6586–6593
Excited Cl(2P1/2) Atoms: Yield from the Photodissociation of SOCl2 and Collisional Deactivation by NO2, CCl3H, C2H4, C3H6, and SOCl2 Asylkhan Rakhymzhan and Alexey Chichinin* Institute of Chemical Kinetics and Combustion, 630090, NoVosibirsk, Russia ReceiVed: February 1, 2010; ReVised Manuscript ReceiVed: May 5, 2010
Rate constants for the collisional deactivation of spin-orbitally excited Cl* (≡Cl(2P1/2)) atoms by some selected gases at T ) 298 K have been determined using time-resolved laser magnetic resonance (LMR) techniques. Cl* atoms were produced by photodissociation of SOCl2 at 248 nm, and the relative quantum yield of Cl* atoms is determined to be 0.52 ( 0.03. This yield is much larger than the yield at 235 nm (0.35 ( 0.06). The rate constants for the relaxation of Cl* (×10-11 cm3/s, ( 2σ) by NO2(1.5 ( 0.4), C2H4 (18 ( 5), CCl3H (1.8 ( 0.4), CH3-CH)CH2 (16 ( 4), and SOCl2 (0.62 ( 0.2) are reported for the first time. All of them are pressure-independent, and in all cases the dominant channel is physical quenching. It was established that the spin-orbital excitation of chlorine atoms decreases the probability of chemical reactions in collisions with propylene molecules. The rate constants of the reactions of ground state Cl(2P3/2) atoms with C2H4, CH3-CH)CH2, and SOCl2 at T ) 298 K were found to be (5.0 ( 1) × 10-30 cm6/s (PAr ) 8-15 Torr), (5.2 ( 1) × 10-11 cm3/s (PAr ) 9-12 Torr), and (6 ( 4) × 10-14 cm3/s, respectively; the first one is termolecular, the last two are bimolecular, and the buffer gas is Ar. Introduction We continue the study of chemical properties of spin-orbitally excited Cl[3p5(2P1/2)] atoms; hereafter, these excited atoms will be denoted Cl*, and the ground state Cl(2P3/2) will be denoted as Cl (or Cl(X)). Note that the ground state is Cl[3p5(2P3/2)] and the spin-orbit splitting is ∆E1/2-3/2 ) 882 cm-1. In the past two decades our group made considerable work on chemical properties of these atoms by the time-resolved laser magnetic resonance (LMR) technique.1,2 The study of the properties of spin-orbitally excited halogen atoms has a long history; fortunately, there is an extensive review on this subject,3 in which all chemical information on these atoms published before the middle of 2004 is presented. Since then, many experimental results have been obtained on yield of these atoms from the photodissociation of different molecules;4 also, there have been several detailed investigations of the dynamics of benchmark reactions such as Cl* + H2.5,6 However, we know of only few works on collisional deactivation of these atoms: Matsumi et al. studied the quenching of Cl* by N2, He, Ne, and Ar;7 H2O, D2O, and H2O2;8 and CH3OH, C2H5OH, n-C3H7OH, and i-C3H7OH9 by the laser-induced fluorescence (LIF) technique. In the present work we report first measurements of rate constants for the quenching of Cl* by 5 simple gases. We have chosen collisional partners that can easily react with excited Cl* atoms (NO2, C2H4, propylene); the idea was to try to distinguish between chemical reaction and physical quenching of the Cl* atoms; hereafter, propylene will be denoted as C3H6. Note that almost nobody has tried to do that before; the only exceptions are the LIF studies of Matsumi et al., in which the deactivation of Cl* atoms by several alkanes (CH4, CH2D2, CD4, C2H6, C3H8, C3D8, n-C4H10, and i-C4H10) have been studied.10–12 However, we soon found that the aim of the present paper requires changing photolytical source of Cl* atoms. * To whom correspondence should be addressed E-mail: chichinin@ kinetics.nsc.ru.
Note that in all previous studies we used photolysis of the ICl molecule as a source of the Cl* atoms. This source has several disadvantages: first, there is a considerable yield of spin-orbitally excited I(2P1/2) atoms from the photodissociation of ICl at 248 nm ([I*]0/([I*]0 + [I]0) ) 0.41), these atoms were observed by Tonokura et al.13 The I(2P1/2) atoms react easily with ICl molecules and produce mainly Cl* atoms;2,14 this reaction makes the study of the kinetics of Cl* atoms more complicated. Second, ICl has a rather small absorption cross section in the ultraviolet (4.86 × 10-19 cm2 at 248 nm).15 Third, ICl is a liquid with low vapor pressure (26.3 Torr at T ) 25 °C), which makes it difficult to have large concentrations in the reactor; manipulations with this corrosive liquid are also not easy. For example, the Cl2, I2, and ICl3 impurities should be taken into account. Fourth, ground state Cl atoms also react with ICl rather quickly (8 × 10-12 cm3/s),16 and this makes difficult to distinguish between chemical reaction and physical deactivation in collisions of Cl* atoms with different molecules. We attempted, in the initial stage of the present study, to find another more convenient source of Cl* atoms; as a surprising result, we found SOCl2. Note that the yield of Cl* atoms from the photodissociation of SOCl2 is not large at 235 nm: it was found to be 35 ( 3%,17,18 and in earlier work an even smaller value (24%)19 was reported. In the present work we found that in 248 nm photolysis the yield of Cl* atoms is much larger. In other respects SOCl2 seems to be a good choice: it has a large absorption cross section at 248 nm (7.05 × 10-18 cm2),20 it does not react quickly with Cl* or Cl, and it has larger vapor pressure (116 Torr at T ) 25 °C). The disadvantage of this choice is the unclear chemistry of SOCl radical, which should be taken into account. Experimental Section The intracavity LMR apparatus and technique have been previously described in detail1,2 and will be only summarized here. The LMR apparatus consisted of a sealed-off CO2-laser
10.1021/jp100965c 2010 American Chemical Society Published on Web 05/26/2010
Excited Cl(2P1/2) Atoms
J. Phys. Chem. A, Vol. 114, No. 24, 2010 6587
gain tube (2.5 m long) and an intracavity absorption cell (Teflon, 2.9 cm i.d.) located inside the laser cavity. The laser cavity, which is about 3.6 m in length, was formed by grating (100 grooves/mm) and a concave mirror. The absorption cell was placed between the poles caps of a 20 cm electromagnet and self-made modulation coils. The electromagnet was used to turn transitions in Cl atoms into coincidence with the CO2-laser frequency. With the modulation coils, the constant magnetic field was modulated at 150 kHz; the modulation amplitude was up to 65 G. The coils were cooled by air flow. Gas mixtures M/SOCl2/Ar (M ) CH4, C2H4, C3H6, NO2, CCl3H) were slowly pumped through the absorption cell. The cell incorporated two NaCl Brewster windows for CO2-laser radiation and one quartz window for UV radiation. The spaces between the windows and reaction zone of the cell (∼30 cm) were continuously flushed with Ar in order to avoid contamination of the NaCl windows. Although the photolysis-reaction zone of the cell was about 18 cm long, the actual CO2-laser absorption zone was limited by the modulation field length (12 cm). Excited Cl* atoms were produced by pulsed 248 nm laser photolysis of SOCl2. The unfocused KrF-excimer laser beam was directed into the cell at a small angle (about 3°) to the CO2-laser radiation beam. This geometry ensured a large overlap area of the beams. The CO2-laser output was focused on to a Ge-Hg photoresistor and the resulting signals were processed by a lock-in amplifier operating at the magnetic field modulation frequency. The photoresistor was cooled by solid N2 (53 K). Cl atoms were detected by fine structure absorption using the 11 P(36) line of a 13CO2-laser (882.287 cm-1) in E⊥B polarization.21 The intense line of the LMR spectrum at ≈3.1 kG was employed. All gases were commercial grades, SOCl2 was cleaned by fractional distillation. Data Analysis. Kinetic Scheme. The experimental procedure consisted in monitoring the kinetics of the LMR signal of chlorine atoms following 248 nm photolysis of SOCl2/M/Ar mixtures. In the present data analysis, which extends the previous analysis,1 we use a more detailed scheme describing the time evolution of the chlorine atom concentrations after photodissociation: k*
kq+2σ0J
where f and f* are the statistical populations of the sublevels probed relative to the total populations of the 2P3/2 and 2P1/2 states, respectively (f ) 1/16, f* ) 1/8); the Θ factor converts the chlorine atom concentration into the LMR signal amplitude. From the kinetic scheme, two differential equations are obtained governing the temporal behavior of Cl* and Cl. The solution of these equations is
S(t) ) Θ[Cl*]0(Cse-λst + Cqe-λqt)
(2)
λs ≡ r - σ0J
(3)
λq ≡ kq + k* + (3 - D)σ0J
(4)
Cs ) 1/Γ - D
(5)
Cq ) -3 + D
(6)
where
where D ≡ (k* - r)/(kq + k* - r), Γ ≡ [Cl*]0/([Cl*]0 + [Cl(X)]0), [Cl*]0 and [Cl(X)]0 are concentrations of atomic chlorine just after the laser pulse (t ) 0), σ0J , kq is assumed, 1/λq and 1/λs are the quick rise time and slow fall time, respectively, and S(t) ) 0 when t < 0 is assumed. Usually, the analysis of the experimental data starts from decomposition of the LMR signal kinetics into the sum of two exponentials, C˜se-λst + C˜qe-λqt; the first several microseconds are not used in the fitting, the fitting starts from time moment ti. Hence the experimentally measured amplitudes C˜s and C˜q are related to amplitudes from eq 2 as
C˜s ) Cse-λsti
(7)
C˜q ) Cqe-λqti
(8)
r
products 79 Cl* {\} Cl 98 products
Using Eqs 5-8, we obtain the final expression for the ratio of the experimental exponentials amplitudes:
σ0J
here kq and k* are the pseudofirst-order rate constants for nonreactive and reactive deactivation of Cl*, respectively; r is the pseudofirst-order rate constant for decay of the ground state Cl atoms, J is the photon flux density, and σ0 is the absorption cross section of an individual transition. These rate constants are related with bimolecular rate constants as
kq ) kqM[M] + kqAr[Ar] + kqSOCl2[SOCl2] k* ) k*M[M] + k*D r ) rSOCl2[SOCl2] + rM[M] + kD where k*D and kD correspond to diffusion of chlorine atoms from the beam of the CO2-laser. We also introduce a notation for the deactivation rate constant, which is k*qM ≡ kqM + k*M. Generally, the LMR signal amplitude may be expressed as1
S(t) ) Θ[Cl(t)] - (f*/f)[Cl*(t)]
(1)
3-D -C˜q /C˜s ) e-(λq-λs)teff 1/Γ - D
(
)
(9)
where teff ) ti. Note that for the mixtures of slow-reactive quenchers the relations k* , kq and r , kq are obeyed, that means |D| , 1. Hence, in this case we can obtain the value of Γ from the intercept of the plot -C˜q/C˜s versus λq - λs; this plot should be exponential with the intercept of 3Γ. In summary, the experimentally measured values of λq, λs, and - C˜q/C˜s are used for the data analysis: the first two are used to obtain k*qM and rM rate constants, respectively; the last one and the difference λq - λs give the quantum yield Γ and the ratio D. Apparatus Function. When we used eq 9 first to determine the quantum yield Γ, we found that the value teff is not equal to ti. This stimulated us to take into account the apparatus function of our spectrometer. We assume the simplest apparatus function which includes roughly time shift and time resolution. That is, the LMR
6588
J. Phys. Chem. A, Vol. 114, No. 24, 2010
Rakhymzhan and Chichinin
Figure 1. Typical LMR signal kinetics of Cl atoms obtained in the photolysis of SOCl2 at various concentrations of the CCl3H. Left: small time scale, right: large time scale. [CCl3H] ) 0.70, 1.23, 1.68, and 4.49 (in units of 1015 cm-3) for A, B, C, and D curves, respectively. Typical concentrations were: [SOCl2] ) 3 × 1014 cm-3 and [Ar] ) 2-5 × 1017 cm-3.
spectrometer distorts the signal by delaying and smoothing. The time shift and the time resolution are characterized by times tsh and tr, respectively:
S˜(t + tsh) )
∫-∞t S(τ)e-(t-τ)/t dτtr r
(10)
here we denote the distorted signal as S˜(t), and the undistorted signal S(t) is given by eq 2. Upon integrating this expression we obtain the result as
(
S˜(t) ) Θ[C1*]0
)
Cq(e-λqt' - e-t'/tr) Cs(e-λst' - e-t'/tr) + 1 - λstr 1 - λqtr (11)
where t′ ) t - tsh. Using Eqs 2, 5-8, and 11 and assuming λqtr, λstr , 1, we came again to eq 9, where
teff ≡ ti - tr - tsh
(12)
That is, taking into account of the apparatus function changes the expression for time teff in eq 9. Note that the time teff may be predicted with eq 12 very approximately, hence it was treated as an unknown parameter. Results Deactivation of Cl* and Reactions of Cl Atoms. Figure 1 shows examples of the LMR signal kinetics at various pressures of a quencher M (CCl3H in this case) under conditions where [SOCl2] and [Ar] were constant. Such temporal profiles are decomposed into the sum of two exponentials; the inverse lifetimes of these exponentials are plotted versus [M]. The quick exponents correspond to the removal of excited Cl* atoms mainly due to deactivation by CCl3H, and the slow exponent corresponds mainly to reaction of unexcited Cl atoms with CCl3H. According to eqs 3 and 4, these plots should be linear; their slopes are the reaction rate constant rM and the deactivation rate constant k*qM, respectively. Figure 2 shows examples of such plots in experiments where M ) CCl3H and M ) CH4. Note that intercept of the plots of λq versus [CCl3H] (or [CH4]) at Figure 2 is rather large; it is equal to 16 ms-1. At the beginning, we tried to reduce the intercept by decreasing the
Figure 2. Left: typical plots of reciprocal time λq for removal of Cl* atoms vs [CCl3H] and [CH4]. Right: typical plots of reciprocal time λs for removal of ground state Cl atoms vs [CCl3H] and [CH4]. Lines are obtained from least-squares analyses and give the deactivation rate constant k*qM and reaction rate constant rM, respectively. The intercepts of the plots in the left panel correspond mainly to deactivation of Cl* by SOCl2 and transitions induced by CO2-laser, the intercepts at right panel correspond mainly to diffusion of chlorine atoms from the laser beams.
concentration of SOCl2 and by using extra-clean argon. We have found that in our conditions the intercept can not be decreased below 10 ms-1; this stimulated us to develop the new kinetic scheme that is described above. It was found from the analysis of the scheme that the radiation transitions contribute strongly to the intercept. Interesting, in experiments where the fastmagnetic-field-jump version of time-resolved LMR was used,22,23 the radiation contribution to the intercept was measured directly to be ≈5 ms-1. The value depends on photon flux density J and the absorption cross section σ0; in the earlier studies we used a cavity configuration with larger photon flux density, but detected chlorine atoms at an LMR transition with smaller absorption cross section. Hence, a direct quantitative comparison is hindered. All the rate constants measured in the present work are summarized in Table 1. Note that the quoted error bounds reflect the reproducibility of the data obtained from repeated determinations at different conditions, while each determination gave the rate constants with very good accuracy. Unfortunately, in some cases, such as Cl + NO2, our uncertainty is very large, and we still can not explain its origin. Note that the absorption cross section of NO2 at 248 nm is rather small (1.6 × 10-20 cm2),15 the main photodissociation products are NO and O(3P). In our experimental conditions it means that that the fraction of photodissociated NO2 molecules is ≈10-3. The rate constant for deactivation of Cl* by NO is even smaller than the rate constant k*qNO2 (k*qNO ) 6.6 × 10-12 cm3/s);3 the rate constant k*qO is unknown. But even if the rate constant k*qO is close to the gas kinetic limit, the maximum contribution of this process to the first-order rate constant k*q is 20 Torr), and it is very sensitive to the temperature: even at T ) 500 K the probability of the channel Cl + C3H6 + CO2 f C3H6Cl + CO2 decreases more than 10 times in comparison with T ) 300 K.42,43 The literature rate constants for the reaction Cl + C3H6 at low pressures and T ) 300 K are collected in Table 2. In the present work the rate constant for Cl atoms removal by propylene was measured several times between 9 and 12
6592
J. Phys. Chem. A, Vol. 114, No. 24, 2010
Torr of Ar, the range is too small to determine a pressure dependence. To the best of our knowledge this is the first measurement in argon, hence it is difficult to compare the result with the literature. However, it looks unlikely that rise in pressure from 12 to 20 Torr will rise the rate constant by a factor of 3 times, see Table 2. Hence, one could suggest that there is a significant difference between rate constants measured in Ar and He. We prefer, however, another explanation: one should notice that the data of Albaladejo et al.44 clearly contradict the IUPAC data.29 The IUPAC recommends the high pressure limit for rC3H6 to be ≈2.8 × 10-10 cm3/s, and it is approached at total pressures greater than approximately 1 bar of air, whereas Albaladejo et al. report no effect of increasing pressure by an order of magnitude from 20 to 200 Torr of He and the high pressure limit to be ≈1.5 × 10-10 cm3/s. Hence, we feel that the data of from Albaladejo et al. are unreliable and should not be used. Cl* + C3H6. In the present study we have found that Cl* reacts with propylene slower than unexcited Cl atoms, k*C3H6 e rC3H6/2. The most reasonable possible explanation of this fact is that the spin-orbital excitation of chlorine atoms decreases the probability of the collisional stabilization of the (C3H6Cl)* collisional complex. This pathway is the most important reaction channel at high pressures; even at 10 Torr of CO2 it is responsible for ≈73% of the rate constant rC3H6.42,43 It is likely that the lifetime of the (C3H6Cl)* collisional complex should be smaller in the case of Cl* + C3H6 system, since the energy of spin-orbit excitation of chlorine atoms is transferred to the energy of the complex; roughly, it may be treated as a heating of the complex. Hence, the collisional stabilization of the complex becomes less probable. This explanation is in agreement with experiments of Pilgrim and Taatjes who found that the rate constant raddit decreases rapidly with the temperature of the gas mixture.42,43 It is tempting to propose that the spin-orbital excitation of chlorine atoms also decreases the probability of abstraction pathway which leads to HCl + C3H5 products via collisional complex. Note that the rate constant for abstraction pathway does not depend on pressure in the range 3-10 Torr and temperature in the range 300-800 K. Hence, understanding of spin-orbital energy of Cl atoms as a heating of the collisional complex in this case can not explain the decrease in the probability of the abstraction pathway. Matsumi et al. have determined that reactivity the of Cl* atoms in reactions with alkanes RH is lower than reactivity of unexcited Cl atoms.10–12 This observation suggests that the reactions of Cl* and Cl with alkanes proceed via different mechanisms. The reactant state Cl + RH adiabatically correlates to the product ground state HCl(1Σ+) + R, and Cl* + RH correlates to a highly excited product state. Hence, the reaction Cl* + RH should occur via nonadiabatic coupling between the two surfaces.12 Since abstraction pathway starts with the approach of a Cl atom to an H atom, the difference between alkenes and alkanes may not be significant. Hence, it is tempting to propose that the excited Cl* atoms also react with alkenes slower than unexcited Cl atoms. Summary In summary, the rate constants for the collisional deactivation of spin-orbitally excited Cl* atoms by NO2, SOCl2, CCl3H, C2H4, and propylene have been determined for first time using time-resolved laser magnetic resonance techniques. The present experiments indicate that physical quenching plays a dominant
Rakhymzhan and Chichinin role in all these deactivation processes. The rate constants of the reactions of ground state Cl atoms with these molecules have been determined, for the almost thermoneutral reaction of Cl with SOCl2 it is the first reported determination. It was qualitatively established that unexcited Cl atoms react with propylene faster than excited Cl* atoms. We explain this fact in terms of a decreased lifetime of the intermediate collisional complex (C2H6Cl)*, which decreases the probability of collisional stabilization of the complex. The decreased lifetime is due to spin-orbital energy of chlorine atoms, which is randomized in the complex. Also, we have determined the relative quantum yield of Cl* atoms from the photodissociation of SOCl2 at 248 nm to be 0.52 ( 0.03. We propose that this yield is larger than the yield at 235 nm because the photoexcitation occurs via another excited state 21A′′. Acknowledgment. This work was supported by Russian Foundation for Basic Research through Grant 07-03-00873a. A.I.C. gratefully thanks the support of the Alexander von Humboldt Foundation (V-Fokoop-RUS/1065971), and both authors acknowledge the support of the Deutsche Forschugsgemeinschaft (DFG). We thank Dr. C. Maul for carefully reading the manuscript. References and Notes (1) Chichinin, A. I.; Chasovnikov, S. A.; Krasnoperov, L. N. Chem. Phys. Lett. 1987, 138, 371. (2) Chichinin, A. I. J. Chem. Phys. 2000, 112, 3772. (3) Chichinin, A. I. J. Phys. Chem. Ref. Data 2006, 35, 869. (4) Taketani, F.; Takahashi, K.; Matsumi, Y. J. Phys. Chem. A 2005, 109, 2855. (5) Ferguson, M. J.; Meloni, G.; Gomez, H.; Neumark, D. M. J. Chem. Phys. 2002, 117, 8181. (6) Garand, E.; Zhou, J.; Manolopolos, D. E.; Alexander, M. H.; Neumark, D. M. Science 2008, 319, 72. (7) Taketani, F.; Yamasaki, A.; Takahashi, K.; Matsumi, Y. Chem. Phys. Lett. 2005, 406, 259. (8) Kono, M.; Takahashi, K.; Matsumi, Y. Chem. Phys. Lett. 2006, 418, 15. (9) Taketani, F.; Takahashi, K.; Matsumi, Y.; Wallington, T. J. J. Phys. Chem. A 2005, 109, 3935. (10) Matsumi, Y.; Izumi, K.; Skorokhodov, V.; Kawasaki, M.; Tanaka, N. J. Phys. Chem. A 1997, 101, 1216. (11) Hitsuda, K.; Takahashi, K.; Matsumi, Y.; Wallington, T. J. J. Phys. Chem. 2001, 105, 5131. (12) Hitsuda, K.; Takahashi, K.; Matsumi, Y.; Wallington, T. J. Chem. Phys. Lett. 2001, 346, 16. (13) Tonokura, K.; Matsumi, Y.; Kawasaki, M.; Kim, H. L.; Yabushita, S.; Fujimura, S.; Saito, K. J. Chem. Phys. 1993, 99, 3461. (14) Nadkhin, A. I.; Gordon, E. B. Khim. Fiz. 1994, 13, 3. (15) Sander, S. P.; Friedl, R. R.; Ravishankara, A. R.; Golden, D. M.; Kolb, C. E.; Kurylo, M. J.; Molina, M. J.; Moortgat, G. K.; Keller-Rudek, H.; Finlayson-Pitts, B. J.; Wine, P. H.; Huie, R. E.; Orkin, V. L. Chemical Kinetics and Photochemical Data for use in Stratospheric Modeling, Technical Report JPL Publication 06-2; NASA, Jet Propulsion Laboratory, California Institute of Technology: Pasadena, CA, 2006. (16) Clyne, M. A. A.; Cruse, H. W. J. Chem. Soc. Faraday Trans. 2 1972, 68, 1377. (17) Roth, M.; Maul, C.; Gericke, K.-H. Phys. Chem. Chem. Phys. 2002, 4, 2932. (18) Chichinin, A. I.; Einfeld, T.; Gericke, K.-H.; Grunenberg, J.; C.Maul; L.Scha¨fer, Phys. Chem. Chem. Phys. 2005, 7, 301. (19) Kawasaki, M.; Suto, K.; Sato, Y.; Matsumi, Y.; Bersohn, R. J. Phys. Chem. 1996, 100, 19853. (20) Uthmann, A. P.; Demlein, P. J.; Aliston, T. D.; Withiam, M. C.; McClements, M. J.; Takacs, G. A. J. Phys. Chem. 1978, 82, 2252. (21) Dagenais, M.; Johns, J. W. C.; McKellar, A. R. W. Can. J. Phys. 1976, 54, 1438. (22) Chichinin, A. I.; Krasnoperov, L. N. Chem. Phys. Lett. 1986, 124, 8. (23) Chichinin, A. I.; Krasnoperov, L. N. Chem. Phys. Lett. 1989, 160, 448. (24) Avallone, L. J. Photochem. Photobiol. A 2003, 157, 231. (25) Kim, Z. H.; Alexander, A. J.; Bechtel, H. A.; Zare, R. N. J. Chem. Phys. 2001, 115, 179.
Excited Cl(2P1/2) Atoms (26) Baum, G.; Effenhauser, C. S.; Felder, P.; Huber, J. R. J. Phys. Chem. 1992, 96, 756. (27) Wang, H.; Chen, X.; Weiner, B. R. J. Phys. Chem. 1993, 97, 12260. (28) Gross, E. U. K.; Dobson, J. F.; Petersilka, M. In Density Functional Theory II. Nalewajski, R. F., Ed.; Springer Series in Topics in Current Chemistry: Springer: Heidelberg, 1996. (29) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Crowley, J. N.; Hampson, R. F.; Hynes, R. G.; Jenkin, M. E.; Rossi, M. J.; Troe, J. Atmos. Chem. Phys. 2006, 6, 3625. (30) Stutz, J.; Ezell, M. J.; Finlayson-Pitts, B. J. J. Phys. Chem. A 1997, 101, 9187. (31) Stutz, J.; Ezell, M. J.; Ezell, A. A.; Finlayson-Pitts, B. J. J. Phys. Chem. A 1998, 102, 8510. (32) Kaiser, E. W.; Wallington, T. J. J. Phys. Chem. A 1998, 102, 6054. (33) Ravishankara, A. R.; Smith, G. J.; Davis, D. D. Int. J. Chem. Kinet. 1988, 20, 811. (34) Leu, M.-T. Int. J. Chem. Kinet. 1984, 16, 1311.
J. Phys. Chem. A, Vol. 114, No. 24, 2010 6593 (35) Mellouki, A.; Poulet, G.; Le Bras, G. J. Geophys. Res. 1987, 92, 4217. (36) Seeley, J. V.; Jayne, J. T.; Molina, M. J. J. Phys. Chem. 1996, 100, 4019. (37) Bryukov, M. G.; Slagle, I. R.; Knyazev, V. D. J. Phys. Chem. A 2002, 106, 10532. (38) Szori, M.; Csizmadia, I. G.; Fittschen, C.; Viskolcz, B. J. Phys. Chem. A 2009, 113, 9981. (39) Okabe, H. Photochemistry of Small Molecules; Wiley: New Yorkc 1978. (40) Kaiser, E. W.; Wallington, T. J. J. Phys. Chem. 1996, 100, 9788. (41) Brana, P.; Sordo, J. A. J. Comput. Chem. 2003, 24, 2044. (42) Pilgrim, J. S.; Taatjes, C. A. J. Phys. Chem. A 1997, 101, 5776. (43) Pilgrim, J. S.; Taatjes, C. A. J. Phys. Chem. A 2003, 107, 5368. (44) Albaladejo, J.; Notario, A.; Cuevas, C. A.; Ballesteros, B.; Martı´nez, E. J. Atmos. Chem. 2003, 45, 35.
JP100965C
