Photolysis (193 nm) of SO2: Nascent Product Energy Distribution

ARTICLE pubs.acs.org/JPCA

Photolysis (193 nm) of SO2: Nascent Product Energy Distribution Examined through IR Emission Jianqiang Ma, Michael J. Wilhelm, Jonathan M. Smith, and Hai-Lung Dai* Department of Chemistry, Temple University, Philadelphia, Pennsylvania, United States ABSTRACT: Infrared emission following the photolysis of SO2 by a 193 nm laser pulse (20 ns duration) was recorded with 500 ns time and 10 cm
1 spectral resolution. Spectral analyses of the time-resolved spectra revealed the vibrationally excited nascent SO population distribution as (v = 1)/ (v = 2)/(v = 3)/(v = 4)/(v = 5) = 0.54 ( 0.04, 1.00 ( 0.03, 0.00 ( 0.03, 0.01 ( 0.03, and 0.10 ( 0.03. The nascent SO was found to be rotationally excited with an average rotational temperature around 1000 K for v = 1 and v = 2 levels and 300 K for the v = 5 level. The vibrationally excited SO likely originates from two distinct dissociation mechanisms; the v = 1 and 2 populations are generated through ~ state and a repulsive state (23A0 ), and the v = 5 population is generated through internal intersystem crossing between the C ~ ~ conversion from the C to the X state. Efficient V
V energy transfer from nascent vibrationally excited SO to SO2(ν1) is also observed. The appearance of the SO2(ν1) ν1 = 2 emission, before that from the ν1 = 1 population is consistent with the previous report that the Δν =
2 channel is more efficient than the Δν =
1 channel.

emission spectroscopy study2 between 197 and 212 nm as well as the theoretical calculation of the vibrational states up to the dissociation limit.8 Kanamori et al. used tunable infrared diode laser spectroscopy to characterize the nascent SO product from 193 nm photolysis of SO2.4 They observed correlations between the SO levels with different electronic spin and rotational angular momentum quantum numbers and suggested that this result was indicative ~ state and a 3A0 repulsive of spin
orbit mixing between the C state of SO2 that facilitates its dissociation. Several other studies have used a variety of methods to examine the population distribution of the SO radical from photolysis at 193 nm.1,3,7,9,11 A summary on the population distribution of the nascent SO was presented by Yamasaki et al.3 The nascent SO was determined to be in an inverted vibrational population with the maximum at v = 2. Populations of v > 2, on the other hand, remain unclear as one report indicated no substantial population for v > 2,11 while others have reported that there is small but non-negligible (99.9%) and Ar (Airgas, >99.999%) were used directly without further purification. Experiments were performed with 10
180 mTorr of SO2 and 0
3 Torr of Ar buffer gas in a continuous flow reaction cell.

III. RESULTS A. Time-Resolved FTIR Emission Spectrum. Time-resolved FTIR emission spectra at selected times following the 193 nm pulse, for samples of 100 mTorr pure SO2 and 50 mTorr SO2 in 2 Torr Ar, are shown in Figure 1. The initial rise of the IR intensity (before 1 μs) is due to the response time of the system. Three main features appear in both spectra: (1) the feature in the 1000
1200 cm
1 region, which is assigned to emission from vibrationally excited SO (X3Σ
) radical; (2) the feature in the 1200
1400 cm
1 region, assigned to SO2(ν3) emission; and (3) a very weak feature in the 2200
2400 cm
1 region, assigned to the overtone (Δν =
2) emission of the SO radical. In both the pure SO2 and the SO2/Ar cases, the center of the SO emission is slightly red-shifted from the 1 f 0 fundamental transition at 1137.94 cm
1. The red shift indicates the presence of emission from v > 1 populations of SO radicals. As observed in the early time spectra, the bandwidth of the rotational branches

II. EXPERIMENTAL SECTION The experimental setup for the time-resolved FTIR emission spectroscopy has been described previously.16,17 A brief summary is outlined here. The 193 nm output from an Excimer laser (Lambda Physik LP200, 20 Hz repetition rate, beam spot size 1  2 cm2) was collimated into a flow cell as the photolysis source. The typical laser pulse was 0) is generated through near-resonant V
V energy transfer between SO and SO2, so only ν1 = 1 or 2 will have significant population from this process. In Figure 2, simulations of emissions from SO and SO2(ν1) bands are shown. The relative intensity of the emission bands of SO and SO2(ν1) are scaled with the Einstein A coefficients of SO and SO2(ν1), which are calculated using the Gaussian 09 program (B3LYP/6-31G(d)).25,26 The spectral feature observed at 1000
1200 cm
1 is fitted as a sum of emission bands from each of the modeled SO and SO2(ν1) vibrational levels, weighted by their population. Nonlinear leastsquares fitting of the observed spectral shape results in a measure of the time-dependent population distributions. For the spectral fitting, two kinds of population distributions were tried: (1) the population in each level was allowed to vary individually in the fitting (the socalled “free-for-all”), and (2) Gaussian functions were used for describing the vibrational population distribution of SO but individual SO2 levels were allowed to vary independently. Briefly, the total intensity is formulated as a sum of the emissions from SO levels up to v = 5 and SO2(ν1) up to ν1 = 2: I ¼

5

2

∑ ai I½SOðν Þ þ i∑¼ 1 bi I½SO ðν Þðν Þ i¼1 i

2

1

i

ð1Þ

where the ai coefficients depict the relative population of SO for the v = i vibrational levels and bi are the relative populations of SO2(ν1) 168

dx.doi.org/10.1021/jp2061943 |J. Phys. Chem. A 2012, 116, 166–173

The Journal of Physical Chemistry A

ARTICLE

Figure 3. Spectral fittings of the emission feature between 1000 and 1200 cm
1 (TR-FTIR spectrum from 100 mTorr SO2 and 2 Torr Ar at 1 μs after the laser pulse). The best result from the free-for-all procedure is shown in the left panel: experimental data (solid line) and fitted spectrum (dotted line); the χ2 is 2.05  106. The two-Gaussian functions procedure result is in the right panel: experimental data (solid line) and fitted spectrum (dotted line); the χ2 is 8.34  106.

for ν1 = 1 or 2. In the first fitting procedure, both ai and bi were allow to vary independently with the only restrictions that ai, bi > 0. In the second fitting procedure, one or more Gaussian functions were assumed for the population distribution for SO vibration levels. For example, ai for the two Gaussian functions form is set as ! 1 ðνi
x1 Þ2 exp
ai ¼ pffiffiffiffiffiffi 2σ 1 2 2π 3 σ1 3 ! 1 ðνi
x2 Þ2 þ pffiffiffiffiffiffi exp
ð2Þ 2σ2 2 2π 3 σ2 3

1.35  107. The χ2 values provide comparisons on the goodness of the fittings only when the degrees of freedom of the fittings are the same. To compare different fittings, the reduced χ2 values must be normalized by the degrees of freedom, which are determined by the number of data points and the number of fitting parameters. The reduced χ2 are calculated to be 5.0  103, 2.0  104, and 3.2  104 for the free-for-all, two-Gaussian functions peaking at ν = 0, and twoGaussians peaking at ν = 2 and ν = 5, respectively. From the analysis above, the free-for-all fitting generated the smallest reduced χ2 values therefore the best quality of the spectral fittings. The free-forall fitting results are then adopted here. Nonlinear least-squares fittings were performed for several experiments with different (ranging from 20 to 200 mTorr) partial pressures of SO2 and the nascent SO vibrational population distributions extracted from these experiments differ by less than 10%. Here we report the result as an average of all fittings. The nascent SO vibrational distribution from ν = 1
5 as 0.54 ( 0.04, 1.00 ( 0.03, 0.00 ( 0.03, 0.01 ( 0.03, and 0.10 ( 0.03 and 0.00 ( 0.01 and 0.00 ( 0.01 for the SO2(ν1) (ν1 = 1, 2) levels, with the SO (ν = 2) population set as 1.00. Figure 4 shows the time-resolved spectra following the photolysis of both pure SO2 and the SO2/Ar mixture together with the free-for-all fit spectra and the determined SO vibrational populations. In both cases, the vibrational populations of the nascent SO were found to have a maximum at ν = 2. This result is consistent with previous studies.1
10 An experimentally significant population is also found at ν = 5 (with a population equal to one tenth of the ν = 2 level). This population was observed in all spectral fittings regardless of the SO2 sample pressures or the presence of the Ar buffer gas. This result is consistent with the works of Hirota and co-workers4 and Tsuchiya and co-workers.3 Further, in the fitting results the populations measured for the ν = 3 and 4 levels of SO are determined to be negligible. In Figure 4, the best fit of the 1 μs spectrum from the pure SO2 experiment gives a rotational temperature of ∼1000 K for the ν = 1 and 2 populations of the nascent SO radicals; the rotational temperature of the ν = 5 population of the nascent SO radials on the other hand remains at 300 K. This rotational temperature difference is in agreement with the previous report of the observation of rotationally hot nascent SO radicals at ν = 1 and 2, but not in ν = 5.4 Analyses of spectra from the pure SO2 experiment show that at the early times after the photolysis, nascent SO radicals

where σ1 and σ2 are the widths of the respective Gaussian functions, x1 and x2 are the centers of the Gaussian functions, and νi is the SO vibrational quantum number. Previous studies have shown that there are two possible channels for the generation of vibrationally excited SO, hence the sum of two Gaussian functions (eq 2) is used here to describe the ai coefficients. We have also tried fittings using one or three Gaussian components. The single Gaussian fitting yields very similar results to the two-Gaussian fitting, while the threeGaussian results show that adding an additional Gaussian does not significantly change the quality of the fitting. Note that in both the free-for-all and the Gaussian fitting for SO, SO2 (ν1) populations are allowed to vary independently. Figure 3 shows one example of the fitting results (of the FTIR spectrum at 1 μs after the laser pulse from the mixture of 100 mTorr SO2 and 2 Torr Ar) using the two procedures. Best nonlinear least-squares fitting results for the freefor-all procedure (Figure 3, left panel) yield a χ2 value of 2.05  106 and the fitted parameters (value ( one standard deviation), ai and bi as 0.54 ( 0.04(a1), 1.00 ( 0.02(a2), 0.00 ( 0.02(a3), 0.02 ( 0.01(a4), and 0.13 ( 0.01(a5) for SO (v = 1
5) levels; and 0.00 ( 0.01(b1) and 0.00 ( 0.01(b2) for the SO2(ν1) (ν1 = 1, 2) levels, respectively, with the SO (v = 2) population set as 1.00. Alternatively, the best nonlinear least-squares fitting results for the twoGaussian functions (Figure 3, right panel) gave a χ2 value of 8.34  106. In the two-Gaussian model fitting, the centers of the two Gaussian functions were either allowed to float (and in the fitting ended at ν = 0 for both), or fixed at ν = 2 for one and ν = 5 for the other according to the free-for-all fitting results. Interestingly, the χ2 value of the latter increased nearly an order of magnitude to 169

dx.doi.org/10.1021/jp2061943 |J. Phys. Chem. A 2012, 116, 166–173

The Journal of Physical Chemistry A

ARTICLE

Figure 5. Temporal profile of SO2 (ν1) ν = 1 and 2 populations following 193 nm photolysis of 100 mTorr SO2 in 2 Torr Ar. The shaded columns are the populations of the ν = 2 level, and the clear columns those of ν = 1 at indicated times.

significant population in the SO2(ν1) levels. The emission collected immediately after the photolysis in this spectral region must therefore contain little contribution from SO2(ν1), suggesting that all emission in this feature originates from SO. At later times, however, we observe that the emission from the ν1 = 1 and 2 levels of SO2 (ν1) grows in. This can be interpreted as the result of the near resonant V
V energy transfer between the fundamental SO vibration (1137.9 cm
1) and the SO2(ν1) mode (1151.3 cm
1). The temporal profile of the SO2(ν1) level population, determined from spectral fittings, is shown in Figure 5. The fitting results show that the ν1 = 2 population appears before the ν1 = 1 population. After the initial increase, ν1 = 2 population reaches its maximum and starts to decrease at later time. The ν1 = 1 population of SO2 on the other hand keeps increasing in our experiment. Due to the overall low IR intensity, a quantitative analysis to deduce the relaxation rate of each of the vibrational levels of the SO radical was not attempted.

IV. DISCUSSION A. Is There Vibrationally Excited SO2 Immediately Following the 193 nm Excitation? It is generally accepted1
10 that at

~ 1B2 electronic 193 nm the SO2 molecule is excited into the C state. Following electronic excitation, SO2 can either go through predissociation or radiative decay. LIF and high-resolution absorption studies by Katagiri et al.10 reported that at 193 nm the fluorescence quantum yield is less than 10
4 and the ~ 1B2 state photodissociation lifetime of the vibronic levels of the C is ∼10 ps. Based on this understanding, it is expected that nearly all photoexcited SO2 will dissociate to SO + O, while vibrationally excited SO2, generated from the radiative decay, would be several orders lower in population than the nascent SO. The previous time-resolved FTIR study by Gong et al. has assigned the emission peak in the 1150 cm
1 region, detected right after the photolysis, to the SO2(ν1) mode.1 They then concluded that the Einstein A coefficient for the SO2(ν1) to be 2
3 orders of magnitude larger than that of the SO mode. We have performed theoretical (Gaussian09, B3LYP/6-31G(d)) calculations for both SO and SO2(ν1). The calculations show that the Einstein coefficient of SO2(ν1) is no more than 1 order of magnitude larger than that of SO (Figure 2). Based on the analysis above, it is estimated that the IR emission from vibrationally hot SO2, generated by the radiative decay, will be much weaker than that from the nascent SO. This is consistent

Figure 4. Deduced SO vibrational populations obtained from spectral fittings shown in the top-right. From top to bottom: the 1 μs spectrum obtained for the 30 mTorr pure SO2 sample, and the 1 and 5 μs spectra from 100 mTorr SO2 in 2 Torr Ar. In each graph, the experimental spectra are show as solid lines and fittings as dotted lines.

undergo mainly rotational cooling while the vibration population remains unchanged. For the case of SO2 in Ar, as shown in Figure 4, on the other hand, the rotational temperature of the SO in the early time spectra is found to be at 300 K due to fast rotational cooling by the Ar buffer gas.19,23 The relative vibrational populations of the nascent SO determined are consistent with what is summarized in the previous studies.3 Conversely, spectral analyses did not reveal any 170

dx.doi.org/10.1021/jp2061943 |J. Phys. Chem. A 2012, 116, 166–173

The Journal of Physical Chemistry A

ARTICLE

with the nonlinear least-squares fitting of the emission feature in the 1000
1200 cm
1 region, detected immediately following the photolysis pulse (less than 1 μs), in which all IR emission intensity can be attributed solely to vibrationally excited SO generated from the photolysis. B. Rotational Energy Content of the Nascent SO. In experiments of pure SO2 (with pressure ranging from 30 to 150 mTorr), the emission from the nascent SO right after photolysis is about 50% broader (as in fwhm) than in experiments run with an Ar buffer gas. This observation suggests that the nascent SO is generated with excess rotational excitation. In the presence of a buffer gas, the rotational energy content is collision quenched much faster and the bandwidth of the observed IR emission is rotationally narrowed. The rotational temperature of nascent SO, deduced from fittings of early time spectra in the pure SO2 experiments, is about 1000 K for ν = 1 and ν = 2 population. Detection of rotationally hot nascent SO radicals has been reported in previous studies.4,19 The tunable IR diode-laser spectroscopy study4 reported that the “rotational distribution in each vibrational state differed significantly from the thermal distribution, and was shifted towards higher rotational levels in lower vibrational states”. Significant population in rotational levels up to N = 38 (ν = 2) was found for the nascent SO after 193 nm photolysis, whereas for a room temperature distribution Nmax is 11. This study has also reported that the N = 38 (ν = 2) population is rapidly quenched as the absorption intensity quickly decays following the initial photolysis. The millimeter-wave rotational spectroscopy study19 reported a rotational relaxation time of ∼0.6 μs for 20 mTorr of pure SO2. In our experiments, the IR detector has a response time of 500 ns. Therefore, the rotational temperature deduced may not accurately reflect the nascent rotational population of the SO radical. The rotational temperature of ∼1000 K for the nascent SO ν = 1 and ν = 2 should be considered as a lower limit. Our results also indicate fast rotational relaxation of the nascent SO. Specifically, around 1 μs following the arrival of the photolysis pulse, the SO rotational population is cooled to the room temperature distribution. The ν = 5 population of the nascent SO radical has a rotational temperature of 300 K from the data analysis, which is also consistent with previous observations.4 The different rotational temperatures of these two populations may indicate that different photolysis mechanisms are responsible for each population. This point will be discussed later. Our experiments on the SO2/Ar mixture, on the other hand, show that the SO radicals in early time emission spectra are not rotationally hot. In our experiments 2
3 Torr Ar buffer gas was used. These pressures correspond to an average collision time of 30
50 ns between SO and Ar. Even within the detector response time, there were >10 collisions for SO with Ar. Rotationalto-translational (R-T) energy transfer is commonly considered to be a very efficient process where only several collisions are required to quench rotational excitation.23 Under the current experimental conditions, rotational excitation in the nascent SO should be quenched during the detector response time, suggesting that the IR emission detected would only show ambient temperature rotational excitation. C. Predissociation Channels of SO2 at 193 nm. The spectral fittings have shown that there are two major vibrational populations for the nascent SO: One encompasses ν = 1 and 2, which accounts for over 90% of the total population; the other is ν = 5, which accounts for less than 10% of the total population. Hirota and co-workers’ study of the electronic spin states of individual

Figure 6. Schematic energy diagram of the relevant electronic state potential curves to the predissociation process of SO2 adapted from ref 7 (Figure 1) with minor adjustments.

ro-vibration levels4 found that “the electron spin is polarized in the ν = 1 and 2 states in a direction either parallel or antiparallel to the rotational angular momentum, and this selective population is not observed for v = 5”. The bimodal vibrational distribution, combined with the different spin state distributions, suggest that there are likely two different predissociation mechanisms resulting in the two populations, respectively.4,24 According to previous studies,7,10 from which the resulted potential energy curves have been summarized into Figure 6, following 193 nm excitation, there are several possible pre-dissociation channels available to SO2, specifically, in Figure 6: ~ 1B2 to the X ~ 1A1 states, (2) an (1) internal conversion from the C 1 ~ avoided crossing between the C B2 and the 31A0 states, and (3) ~ to the 23A0 states. The intersystem intersystem crossing from the C crossing channel was suggested in the spin study4 for the generation of ν = 1 and 2 in the nascent SO. Houston and coworkers5 studied the dependence of the excitation wavelength between 202
207 nm in the vibrational population of the nascent SO. Their data suggested that the predissociation mechanism changes at 203 nm. For wavelengths shorter than 203 nm, they observed that the majority of the nascent SO population is in ν = 0. For wavelengths longer than 203 nm, they found the majority of nascent SO radicals to be vibrationally excited in the ν = 1 and 2 levels. Houston and co-workers suggested that at 206.71 nm, >75% of the nascent SO are vibrationally excited and the ν = 2 level has the highest population.5 At 193 nm, the total available energy increases by about 2600 cm
1 compared to 206.71 nm. This increase of excitation energy enables vibrationally excited SO to be generated through the same mechanism at the ν = 5 level. Our results show that there are relatively much smaller populations at ν = 3 and 4. This observation is consistent with the internal conversion mechanism.5 According to ref 5, the highest vibrational level (ν = 2) is preferable in population compared to lower vibrational levels (ν = 0, 1) through the internal conversion mechanism. At 193 nm, it is expected that the highest plausible vibrational level (ν = 5) is more populated than the lower ones (ν, = 3, 4). The total excess energy of the photodissociation of SO2 at 193 nm may be distributed into translational, rotational, and vibrational degrees of freedom. By energy conservation, the available translational and rotational energies for the ν = 5 SO is estimated to be small, consistent with the observation that the rotational temperature of the SO ν = 5 population is 300 K. The avoided crossing mechanism proposed by ref 5, on the other hand, will produce almost all nascent SO radical in the ν = 0 level as the repulsive singlet 31A0 state distributes most of the excess energy into 171

dx.doi.org/10.1021/jp2061943 |J. Phys. Chem. A 2012, 116, 166–173

The Journal of Physical Chemistry A the translational degrees of freedom of the SO and O fragments. Unfortunately, as the vibrational ground state level produces no emission, we cannot probe the ν = 0 population in the experiments reported here. Therefore, we cannot offer useful comment on this predissociation mechanism. The avoided crossing mechanism is thus not included here to account for the observed nascent SO population. An interesting question to consider is that, in the internal conversion mechanism, dissociation occurs through the highly ~ state molecules that energetically can vibrationally excited X create SO up to ν = 5, but why is the ν = 5 level favored? Here we provide some speculations that may answer this question. The ~ state SO2 has the C2v symmetry with the S
O structure of the X ~ state structure on the other hand bond length at 1.432 Å.10 The C has the Cs symmetry with two very unequal S
O bonds with bond lengths of 1.491 and 1.639 Å.10 The dramatic structure difference in these two electronic states of SO2 may be the reason for this unique nascent product distribution. The highly vibra~ state potential energy surface tionally excited SO2 on the X ~ state following the internal conversion should retain the C asymmetric structure with two very unequal S
O bonds. This will place the starting point of the dissociation motion at an asymmetric point on the ground state potential energy surface. A likely scenario is that the trajectory on the mass weighted surface will lead to a highly vibrationally excited SO with little translational energy, thus, favoring the higher vibrationally excited level. Another speculation can be made using the momentum gap law,27 which depicts that in dissociation through vibrational excitation the dissociation rate is faster for channels with a smaller difference between the reactant vibrational momentum and the product translational momentum. This means more products will be produced in the channels with low translational energy and higher vibrational excitation. This momentum gap law may be particularly relevant to this case of dissociation because of the highly asymmetric structure of the starting point of the dissociation; the dissociation occurs along only one of the two bonds with the other as the speculator. D. Collision Relaxation of Vibrationally Excited SO. Yamasaki et al. have reported the vibrational relaxation rate constants of the various vibrational levels of the nascent SO.3 63% of the vibrational deactivation of SO v = 2 by SO2 were suggested to go through two-quantum transitions. The temporal profile of SO2(ν1) that we have deduced is consistent with this finding (Figure 5). We have observed that the ν1 = 2 population grows in before the ν1 = 1 population, which indicates a very efficient Δν =
2 energy transfer channel from SO to SO2. This nonintuitive result is actually consistent with ref 3, which reported an efficient Δν =
2 channel between SO and SO2(ν1). It was proposed that an attractive potential or the formation of an intermediate could be responsible for this type of energy transfer.3 The delayed ν1 = 1 population of SO2(ν1) is attributed to the sum of the V
V energy transfer between SO and SO2 through the Δν =
1 channel and the depletion of the ν1 = 2 population.

ARTICLE

the ν = 1 and 2 vibrational levels, while the ν = 5 population has a rotational temperature ∼300 K. This bimodal energy distribution, with relatively much smaller populations at ν = 3 and 4, can be best explained by two dissociation mechanisms operating at 193 nm: the lower excited majority population generated through an intersystem crossing ~ state and a repulsive triplet state mechanism between the C (23A0 ) and the much smaller high population generated through ~ and X ~ states. an internal conversion mechanism between the C Contrary to a previous IR emission study,1 quantitative fitting of the emission spectra did not reveal any contribution from vibrationally excited SO2 generated immediately following the photolysis. Efficient V
V energy transfer from the nascent vibrationally excited SO to SO2(ν1) has been observed. The time-resolved appearance of the SO2(ν1) ν1 = 1 and 2 populations is consistent with the previous report3 that Δν =
2 channel is faster than the Δν =
1 channel.

’ ACKNOWLEDGMENT This work was supported in part through the U.S. Department of Energy, Basic Energy Sciences, Grant No. DEFG 02-86ER 134584. ’ REFERENCES (1) Gong, Y.; Marakov, V. I.; Weiner, B. R. Chem. Phys. Lett. 2003, 378, 493. (2) Ray, P. C.; Arendt, M. F.; Butler, L. J. Chem. Phys. 1998, 109, 5221. (3) Yamasaki, K.; Taketani, F.; Sugiura, K.; Tokue, I.; Tsuchiya, K. J. Phys. Chem. A 2004, 108, 2382. (4) Kanamori, H.; Butler, J. E.; Kawaguchi, K.; Yamada, C.; Hirota, E. J. Chem. Phys. 1985, 83, 611. (5) Cosofret, B. R.; Dylewski, S. M.; Houston, P. L. J. Phys. Chem. A 2000, 104, 10240. (6) Hydutsky, D. P.; Bianco, N. J.; Castleman, A. W., Jr. Chem. Phys. 2008, 350, 212. (7) Brouard, M.; Cireasa, R.; Clark, A. P.; Preston, T. J.; Vallance, C.; Groenenboom, G. C.; Vasyutinskii, O. S. J. Phys. Chem. A 2004, 108, 7965. (8) Bludsky, O.; Nachitigall, P.; Hrusak, J.; Jensen, P. Chem. Phys. Lett. 2000, 318, 607. (9) Chen, X.; Asmar, F.; Wang, H.; Weiner, B. R. J. Phys. Chem. 1991, 95, 6415. (10) Katagiri, H.; Sako, T.; Hishikawa, A.; Yazaki, T.; Onada, K.; Yamanouchi, K.; Yoshino, K. J. Mol. Struct. 1997, 413
414, 589. (11) Felder, P.; Efenhauser, C. S.; Haas, B. M.; Huber, J. R. Chem. Phys. Lett. 1988, 148, 417. (12) Burkholder, J. B.; Lovejoy, E. R.; Hammer, P. D.; Howard, C. J.; Mizushima, M. J. Mol. Struct. 1987, 124, 379. (13) Okabe, H. J. J. Am. Chem. Soc. 1971, 93, 7095. (14) Brand, C.; Kuepper, J.; Pratt, D. W.; Meerts, W. L.; Kruegler, D.; Tatchen, J.; Schmitt, M. Phys. Chem. Chem. Phys. 2010, 12 (19), 4968. (15) Kommandeur, J.; Majewski, W. A.; Meerts, W. L.; Pratt, D. W. Annu. Rev. Phys. Chem. 1987, 38, 433. (16) Qin, D.; Harland, G. V.; Dai, H. L. J. Phys. Chem. A 2000, 104, 10460. (17) Nikow, M.; Wilhelm, M. J.; Smith, J. M.; Dai, H. L. Phys. Chem. Chem. Phys. 2010, 12, 2915. (18) de Pater, I.; Roe, H.; Graham, J. R.; Strobel, D. F. ICARUS 2002, 156, 295. Zolotov, M. Y.; Fegley, B., Jr. ICARUS 1998, 132, 431. (19) Kolbe, W. F.; Leskovar, B. J. Chem. Phys. 1986, 85, 7117. (20) Burkholder, J. B.; Lovejoy, E. R.; Hammer, P. D.; Howard, C. J.; Mizushima, M. J. Mol. Spectrosc. 1987, 124, 379–392.

V. CONCLUSIONS Analyses of the time-resolved IR emission spectra, recorded following photolysis of SO2 by a 193 nm laser pulse, revealed the nascent SO product population distribution as (ν = 1)/(ν = 2)/ (ν = 3)/(ν = 4)/(ν = 5) = 0.54 ( 0.04, 1.00 ( 0.03, 0.00 ( 0.03, 0.01 ( 0.03, and 0.10 ( 0.03. The nascent SO is rotationally excited with an average rotational temperature of ∼1000 K for 172

dx.doi.org/10.1021/jp2061943 |J. Phys. Chem. A 2012, 116, 166–173

The Journal of Physical Chemistry A

ARTICLE

(21) Guelachvili, G.; Naumenko, O. V.; Ulenikov, O. N. J. Mol. Spectrosc. 1987, 125, 128–139. (22) Judge, R. H.; Clouthier, D. J. Comput. Phys. Commun. 2001, 135, 293–311. (23) Yardley, J. T. Introduction to Molecular Energy Transfer; Academic Press: New York, 1980. (24) Kawasaki, M.; Kasatani, K.; Sato, H.; Shinohara, H.; Nishi, N. Chem. Phys. 1982, 73, 377. (25) The Midwest Undergraduate Computational Chemistry Consortium cluster, 2011; http//:www.chem.hope.edu. (26) Gaussian 09, Revision A.1, Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, € Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. A. D.; Farkas, O.; Gaussian 09; Gaussian, Inc., Wallingford CT, 2009. (27) Beswick, J. A.; Jortner, J. Chem. Phys. Lett. 1977, 49, 13. J. Chem. Phys. 1978, 68, 2277.

173

dx.doi.org/10.1021/jp2061943 |J. Phys. Chem. A 2012, 116, 166–173