Cite this: Phys. Chem. Chem. Phys., 2014, 16, 10438
The UV absorption spectrum of the simplest
Criegee intermediate CH
2
OO†
Wei-Lun Ting,aYing-Hsuan Chen,aWen Chao,abMica C. Smithacand Jim Jr-Min Lin*abd
SO2scavenging and self-reaction of CH2OO were utilized for the
decay of CH2OO to extract the absorption spectrum of CH2OO
under bulk conditions. Absolute absorption cross sections of CH2OO
at 308.4 and 351.8 nm were obtained from laser-depletion measure-ments in a jet-cooled molecular beam. The peak cross section is
(1.23 0.18) 1017cm2at 340 nm.
Ozonolysis is a major removal mechanism in the troposphere for unsaturated hydrocarbons which are emitted in large quan-tities from both natural and human sources. Now it is generally accepted that ozonolysis of alkenes proceeds via Criegee inter-mediates, highly reactive species postulated in 1949 by Rudolf Criegee.1,2 In the troposphere, Criegee intermediates are involved in several important atmospheric reactions,3 includ-ing reactions with SO2and NO2,4–7or can be photolyzed by near
UV light,7–10as shown for CH
2OO in (R1)–(R4).
C2H4+ O3- CH2OO + H2CO (R1)
CH2OO + SO2- SO3+ H2CO (R2)
CH2OO + NO2- NO3+ H2CO (R3)
CH2OO + hn- O(1D) + H2CO (R4)
The formation of SO3and NO3, as in (R2) and (R3), plays an
important role in atmospheric chemistry,11,12including aerosol and cloud formation. The formation of O(1D), as in (R4), will result in OH formation through (R5).
O(1D) + H2O- 2OH (R5)
Because CH2OO absorbs strongly at wavelengths longer than
300 nm,7–9tropospheric photolysis of CH2OO would be quite
efficient with an effective photolysis lifetime on the order of 1 second.8 As a result, the OH formation of (R4) + (R5) may contribute significantly to the atmospheric OH concentrations. Despite their importance, the direct detection of Criegee intermediates was not realized until recently.4,13 Welz et al.4 reported an efficient way to prepare Criegee intermediates. For example, CH2OO can be prepared via (R6) + (R7a).
CH2I2+ hn- CH2I + I (R6)
CH2I + O2- CH2OO + I (R7a)
Welz et al.4also demonstrated the direct detection of CH2OO
by using vacuum UV photoionization mass spectrometry. The parent ion CH2O2+ was observed when the photon energy
exceeded the ionization energy of CH2OO (10.0 eV); other
isomers like dioxirane and formic acid were excluded due to their different ionization energies.4At low pressure, the yield of (R7a) is close to unity,14,15while the adduct formation (R7b) may dominate at near atmospheric pressures.14
CH2I + O2+ M- ICH2OO + M (R7b)
The kinetics of CH2OO reactions with SO2 and NO2 were
investigated by Welz et al.4and by Stone et al.5by observing the disappearance of CH2OO and by detecting the H2CO products,
respectively. The rate coefficients of these reactions were found to be unexpectedly rapid and imply a substantially greater role of Criegee intermediates in models of tropospheric sulfate and nitrate chemistry.
Beames et al.8recorded the UV spectrum of CH
2OO through
observing its depletion in a molecular beam upon laser irradia-tion (an acirradia-tion spectrum). Based on their laser pulse energy and spot size, Beames et al.8roughly estimated the peak absorption cross section to be 5 1017 cm2 (at 335 nm with FWHM B40 nm). Lehman et al.10 measured the angular and velocity
distributions of the O(1D) photoproduct arising from UV excita-tion of CH2OO in the 300–365 nm range. From the observed
a
Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 10617, Taiwan
bDepartment of Chemistry, National Taiwan University, Taipei 10617, Taiwan cDepartment of Chemistry, University of California at Berkeley, Berkeley,
CA 94720, USA
dDepartment of Applied Chemistry, National Chiao Tung University, Hsinchu 30010, Taiwan. E-mail: jimlin@gate.sinica.edu.tw
†Electronic supplementary information (ESI) available. See DOI: 10.1039/ c4cp00877d Received 28th February 2014, Accepted 14th April 2014 DOI: 10.1039/c4cp00877d www.rsc.org/pccp
COMMUNICATION
Published on 14 April 2014. Downloaded by National Chiao Tung University on 25/12/2014 03:15:55.
View Article Online
anisotropic angular distribution (bD 0.97), the authors con-cluded that the orientation of the transition dipole moment reflects the p* ’ p character of the electronic transition associated with the COO group. The significant anisotropy of the photofragments also indicates that the dissociation is faster than rotation.
Su et al.16reported an infrared (IR) absorption spectrum of CH2OO. By comparing their experimental results with
high-level ab initio calculations, the authors concluded that the observed vibrational frequencies are more consistent with a zwitterion structure rather than a diradical structure. With IR detection, the same group17 found that the self-reaction of CH2OO is extremely fast, with a rate coefficient of (4 2)
1010cm3s1, which reflects a unique property of the zwitter-ionic character.
Sheps7used a cavity-enhanced technique to measure the UV absorption spectrum of CH2OO and observed significant
vibra-tional structures at the long wavelength side of the absorption band. Moreover, the absorption spectrum7differs significantly
from the action spectrum reported by Beames et al.8 Sheps’
argument7is the following: ‘‘The difference between the
absorp-tion and acabsorp-tion spectra likely arises from excitaabsorp-tion to long-lived B˜ (1A0) vibrational states that relax to lower electronic states by
fluorescence or nonradiative processes, rather than by photo-dissociation.’’ However, the measurement of the photoproduct anisotropy10indicates that the photodissociation is faster than rotation which is in the picosecond time scale. Thus, the slower fluorescence process cannot compete with the fast dissociation. Furthermore, there is no theoretical evidence for the nonradiative processes. To investigate the source of this difference, we re-investigate the UV spectrum of CH2OO using two new methods.
CH2OO was prepared in a pulse-photolysis cell following the
well-established method of CH2I2/O2 photolysis.4,16 CH2I2
mixed with O2and N2was photolyzed at 248 nm (KrF excimer
laser); transient absorption spectra were recorded using a gated intensified CCD camera (1 ms gate width) after the probe light was dispersed using a grating monochromator19,20(see ESI† for the experimental details). Fig. 1a shows examples of the tran-sient absorption spectra. In Fig. 1a the most significant feature is a strong and broad absorption band peaked atB340 nm which showed up quickly upon photolysis and decayed with time. In addition, depletion of the CH2I2precursor near 290 nm
and formation of IO with distinct peaks near 430 nm were clearly observed, especially at long delay times.
Under our experimental conditions, CH2OO reacted quickly
with itself17and with I atoms to form H2CO, O2, and IO. Because
H2CO and O2absorb rather weakly, the transient spectra at long
delay times mainly consist of the absorption changes of CH2I2
(depletion) and IO (formation). Since the spectra of CH2I2and IO
are very different, their contributions to the transient absorption spectra can be extracted and removed (see ESI† for details). The remaining spectra are shown in Fig. 2a. Consistent with ref. 14, we did not observe significant difference in the CH2OO yield for
different O2 mixing ratios. The identical shape of these spectra
under various experimental conditions (delay times, laser fluences, and O2pressures) strongly suggests that the spectral carrier is a
single species. Based on the high yield of CH2OO from the
CH2I2/O2photolysis at low pressure,14,15it is most reasonable
to assign the spectral carrier to CH2OO (see below for
discus-sion on the pressure dependence). If another absorbing species contributes significantly to these spectra, this species must exhibit kinetic behavior similar to that of CH2OO.
It is known that CH2OO reacts quickly with SO2(k2B (3–4)
1011cm3s1).4,5We utilized this kinetic signature of CH2OO
to examine the spectral carrier of band A. As shown in Fig. 1c, it is clear that the intensity of band A decreases at higher SO2
concentrations. Because the absorption cross sections of SO2,
H2CO and SO3 in the 316–450 nm range are much smaller
Fig. 1 (a) Examples of the transient absorption spectra. The [CH2I2]0, [O2]0
and the total number density ntotal(N2balance) are 1.6 1015, 3.4 1017
and 2.0 1018cm3
, respectively. The depletion of CH2I2(o10% of
[CH2I2]0) results in negative absorbance peaked atB290 nm. The broad
and strong absorption band peaked atB340 nm (band A) is most likely due
to CH2OO, which is short-lived. At longer delay times, the formation of IO
gives sharp peaks in the 410–460 nm range. (b) Published spectra of CH2I2
and IO.18(c) Examples of transient absorption spectra at different SO2
concentrations. It is clear that the intensity of band A decreases at higher
SO2concentrations. [CH2I2]0, [O2]0, and ntotalare 1.3 1015, 1.6 1018,
3.3 1018
cm3, respectively; delay time = 10.6 ms.
(o1 1019 cm2) than those of CH
2OO (peak cross section
41 1017cm2),8the differences between the spectra of Fig. 1c should be mostly due to the absorption of CH2OO. The
height-normalized curves of such difference spectra are shown in Fig. 2b. The good agreement between Fig. 2a and b confirms the above assignment.
When O2was absent in the photolysis cell, the absorption of
CH2I was present and band A could not be observed. With the
published absorption cross sections of CH2I,21the initial number
density of CH2I can be estimated. Under the high O2pressures used,
CH2I reacted with O2within a few microseconds.5,7,14,15Therefore
the number density of CH2OO can be estimated based on the
published quantum yield of (R7a) (FCH2OO= 86% at 11 Torr).
14From
the estimated number density and the observed absorbance of CH2OO, its absolute peak cross section can be deduced to be
(1.26 0.25) 1017cm2at 340 nm. The overall error bar is estimated to be 20% mostly due to the uncertainty in FCH2OO
14,15(see ESI† for details).
In Fig. 2a we can see that the shape of band A does not depend on the total pressure in the range of 8–100 Torr.
This observation excludes the contribution of ICH2OO because
its formation is a termolecular process which has strong pressure dependence.14,15At higher pressures (100 Torro P o 760 Torr),
the yield of CH2OO was found to decrease with pressure (see
Table S1 (ESI†) for a typical example) and an additional (weaker) absorption band was observed at l o 290 nm, indicating the formation of a new species (likely ICH2OO).21,22The absorption
of ICH2OO seems to be much weaker than that of CH2OO, such
that the change in spectral shape (not including the yield) with pressure is not very obvious.
The IO peaks are absent in short delay times while the absorption of CH2OO is very significant, indicating that IO is
not a primary product. Based on our signal-to-noise ratio, we further constrain the primary IO yield to be less than 1%, resolving some debate among published results.23–26The forma-tion of IO is likely due to (R8).5,14
CH2OO + I- H2CO + IO (R8)
Detailed kinetic analysis is beyond the scope of this paper and will be published elsewhere.
To further quantify the absolute value of the absorption cross section of CH2OO, we measured the depletion of CH2OO in a
molecular beam upon laser irradiation at 308.4 and 351.8 nm. CH2OO was detected using a quadrupole mass spectrometer
equipped with an electron impact ionizer. This method27–29has been demonstrated to be efficient in determining the photo-dissociation cross section of a species in a mixture without the knowledge of its concentration. Under our experimental condi-tions, the number of molecules N after laser irradiation can be described by eqn (1). N N0 ¼ eIsf DN N0 ¼ 1 eIsf (1)
Where N0is the number of molecules before the laser
irradia-tion, I is the laser fluence in photons per cm2, s is the
absorption cross section in cm2, f is the dissociation quantum
yield and DN = N0 N. For CH2I2, the excitations at 351.8 and
308.4 nm correspond to repulsive (unbound) excited states which dissociate in picosecond time scales,30–32 resulting in 100% dissociation (f = 1).
Fig. 3a shows the arrival-time profiles of CH2OO at various
laser fluences at 308.4 nm. Fig. 3b shows the corresponding saturation curve. A nice fit of eqn (1) to the experimental data indicates that the measurement corresponds to a single species (or multiple species having the same cross section, which is unlikely). The results at 351.8 nm are similar (see ESI†). The complete depletion of CH2OO indicates that its dissociation
yield is unity. The absolute cross section of CH2OO can be
obtained by comparing its saturation curve with that of CH2I2,
for which the cross section is known. A summary of the cross section measurement is shown in Table 1.
With the absolute cross sections of CH2OO (Table 1), we may
set the spectra of Fig. 2 on the absolute scale. However, we need to consider the temperature effect of the absorption cross sections because the temperature of the molecular beam is lower than room temperature. Since the cross section of CH2I2
Fig. 2 (a) Background-corrected height-normalized spectra of CH2OO.
The variation of the experimental parameters includes combinations of
delay time (2–50 ms) and gas composition (O2percentage: 17–99%, N2
balance), total pressure (7.5–100 Torr), laser fluence (8–16 mJ cm2), etc.
A total of 99 spectra (gray lines) and their average (black line) are plotted.
(b) Height-normalized spectra of CH2OO obtained by subtracting the
experimental spectra at different SO2concentrations. A total of 24 spectra
(gray lines) and their average (black line) are plotted. Delay time: 6.5–50 ms;
[SO2]: 2.9 1015 1.2 1016cm3.
at 308.4 nm does not change with temperature,33,34we can use the near-room-temperature value for the cross section of CH2I2
in a molecular beam. The UV absorption band of CH2OO can be
assigned to the intense B˜’ X˜ transition8,22which is analogous to the Hartley band of O3. The cross section of the O3Hartley
band has a quite weak temperature dependence.18,35The peak
cross section (at 254 nm) of O3increases byB1.5% when the
temperature decreases from 293 K to 203 K. For the main region of the Hartley band (215–288 nm, 1 1018cm2o s o 1.1 1017cm2), the temperature effect is within 5% (203–293 K).18,35 Similarly, it is expected that the temperature dependence of the CH2OO cross sections is weak near the peak. Therefore, we
choose the cross section at 308.4 nm to scale the average spectra
of Fig. 2 and to plot the scaled spectra in Fig. 4. We believe that the SO2scavenging method would give a more reliable spectrum
of CH2OO (see Table S3 (ESI†) for numerical values) while the
result of the self-reaction method is very similar. The peak value of the scaled spectrum is (1.23 0.18) 1017cm2at 340 nm.
We assume an error bar of15% to include possible variations due to the temperature effect. This value is consistent with the peak cross section of (1.26 0.25) 1017 cm2 obtained in the transient absorption experiment of this work based on the estimated CH2OO number density.
Previous UV absorption7 and action spectra8 of CH2OO
exhibit significant differences. Beames et al.8 measured the laser depletion of CH2OO in a similar molecular beam and
obtained an action spectrum of CH2OO. However Beames et al.8
only estimated the laser fluence from their laser (a dye laser) pulse energy and spot size. The beam spot of a dye laser is usually highly non-uniform. Without using a laser beam pro-filer, it is difficult to quantify the actual laser fluence. Beames et al.8might underestimate their laser fluence and thus
over-estimate the CH2OO cross section. In this work, we utilized a
reference molecule to effectively calibrate the laser fluence and to cancel the effect of non-uniform laser spot.36Therefore, our results should be more accurate.
Fig. 4 compares our results with those of Sheps7and Beames et al.8The scaled spectrum of Beames et al.8is weaker at l Z 360 nm. The temperature effect may be one possible reason for this difference. While the intense Hartley band of O3 has a
rather weak temperature dependence, the weak Huggins band or the long-wavelength tail of the Hartley band (310–380 nm) has very strong temperature dependence (smaller cross sections at lower temperatures).18,35If the temperature dependence of the CH2OO cross sections at l Z 360 nm is as strong as that of
the Huggins band of O3, this may explain why the spectrum of
Beames et al.8 is weaker in this wavelength range. Another
possibility mentioned by Sheps7is a decrease in the
dissocia-tion yield at long wavelengths. Although this might explain the discrepancy between the absorption and action spectra, it is inconsistent with the product anisotropy measurement by Lehman et al.10which shows that the UV photodissociation of CH2OO is faster than its rotation (Bpicosecond). Thus, a
non-unity dissociation yield would require a fast process that can compete with photodissociation. Fluorescence is too slow to fulfill this condition. Other fast non-radiative processes are unlikely but cannot be fully ruled out at this moment. More evidence and investigations are needed.
Sheps7used a newly-built cavity-enhanced absorption spectro-meter to measure the transient absorption spectra of the CH2I2/
O2photolysis system. Sheps7determined the absolute CH2OO
spectrum based on the measured CH2I spectrum (when O2was
absent) and an estimated (90 10)% yield of transforming CH2I to CH2OO at 5 Torr. Qualitatively, the shape of Sheps’
spectrum7 is similar to ours, particularly the structures at the long-wavelength side (the peak positions are matched). However, the short-wavelength side of Sheps’ spectrum7decays much faster than that in this work. Furthermore, the reported peak cross section and position of the CH2OO spectrum by
Fig. 3 (a) Arrival-time profiles of CH2OO at various laser fluences at
308.4 nm. The parent ion of CH2OO was detected at m/z = 46 amu.
(b) Saturation curve for the laser depletion of CH2OO (m/z = 46) and CH2I2
(m/z = 141, CH2I+, a daughter ion of CH2I2) at 308.4 nm. The x-axis is the
laser pulse energy which is proportional to the laser fluence. The lines are the fit of eqn (1). The nice fit indicates that the laser depletion experiments are single-photon processes.
Table 1 Summary of the cross section measurements of CH2OO in a
jet-cooled molecular beam
Wavelength (nm)
sfðCH2OOÞ
sfðCH2I2Þ s(CH2I2)b(cm2) s(CH2OO) (cm2)
308.4 2.52 0.28a 3.21 1018 (8.09 0.90) 1018
351.8 47.6 5.2 r2.54 1019 r(1.21 0.13) 1017
aThe error bar is 2 standard deviation.bAverage values of ref. 33 and
34 at T = 273 K. The temperature dependence of the UV absorption
cross section of CH2I2 is very weak at 308.4 nm, but moderate at
351.8 nm.34 The actual cross section at 351.8 nm would be smaller
for CH2I2in a jet-cooled molecular beam.
Sheps7[(3.6 0.9) 1017cm2at 355 nm] are different from
our values. The source of the discrepancies is not clear. It might arise from the complexity of the cavity-enhanced measurement. There are at least 7 vibrational peaks observable on the long-wavelength side of the UV absorption band of CH2OO (Fig. 2
and 4). The widths of these vibrational peaks are significantly wider than the instrument resolution of B2 nm. Similar structures have been reported by Sheps7 at a slightly lower
resolution and signal-to-noise ratio. The positions of the most well-defined peaks are 363.7, 372.0, 380.7, 389.2, 399.0, 409.3, 420.5 nm (27495, 26882, 26267, 25694, 25063, 24432, 23781 cm1). The average peak separation is about 620 cm1. Analogous to the Huggins band of O3, these vibrational structures may arise from
some periodic motions on the excited potential energy surface, most likely the B˜(1A0) surface. The widths of the vibrational peaks may
originate from congested vibrational structures (vibrational modes involving O–O stretching and C–O–O bending)22 or rotational contours at room temperature. For the O3Huggins band, the widths
of the vibrational peaks become narrower at low temperatures.18,35 It will be interesting to see how the peak structures change at lower temperatures for CH2OO.
In summary, more accurate UV absorption cross sections of the simplest Criegee intermediate CH2OO are reported. The peak cross
section is determined to be (1.23 0.18) 1017cm2at 340 nm.
This value is significantly smaller than previous reports,7,8 imply-ing slower photolysis rates in the atmosphere than previously expected. Nonetheless, this intense absorption band of CH2OO
overlaps well with the incoming solar spectrum, resulting in efficient photolysis of this Criegee intermediate. The clear vibra-tional structures on the long-wavelength side of the CH2OO
spectrum provide a fingerprint feature for spectroscopic identification of this elusive intermediate.
Acknowledgements
This work was supported by Academia Sinica and Ministry of Science and Technology, Taiwan (NSC100-2113-M-001-008-MY3). The authors thank Miss Shu-Yi Meng for assistance in data acqui-sition and Profs Yuan T. Lee and Yuan-Pern Lee for discussions.
References
1 R. Criegee, Angew. Chem., Int. Ed. Engl., 1975, 14, 745–752. 2 R. Criegee and G. Wenner, Liebigs Ann. Chem., 1949, 564,
9–15.
3 C. A. Taatjes, D. E. Shallcross and C. Percival, Phys. Chem. Chem. Phys., 2014, 16, 1704–1718, DOI: 10.1039/C3CP52842A. 4 O. Welz, J. D. Savee, D. L. Osborn, S. S. Vasu, C. J. Percival, D. E. Shallcross and C. A. Taatjes, Science, 2012, 335, 204–207.
5 D. Stone, M. Blitz, L. Daubney, N. U. M. Howesa and P. Seakins, Phys. Chem. Chem. Phys., 2014, 16, 1139–1149, DOI: 10.1039/c3cp54391a.
6 B. Ouyang, M. W. McLeod, R. L. Jones and W. J. Bloss, Phys. Chem. Chem. Phys., 2013, 15, 17070–17075.
7 L. Sheps, J. Phys. Chem. Lett., 2013, 4, 4201–4205.
8 J. M. Beames, F. Liu, L. Lu and M. I. Lester, J. Am. Chem. Soc., 2012, 134, 20045–20048.
9 J. M. Beames, F. Liu, L. Lu and M. I. Lester, J. Chem. Phys., 2013, 138, 244307.
Fig. 4 Absorption spectrum of CH2OO. The thick orange and thin black lines are the average curves of Fig. 2a and b, respectively. For the thin black line
(numerical values can be found in ESI†), the absorbance due to the reacted SO2has been removed, based on the mass balance of (R2) (see ESI†). Square
symbols are the absolute cross sections from the molecular beam-laser depletion measurements (Table 1). The black and orange lines are scaled to match
the absolute cross section at 308.4 nm. The results of Sheps7and Beames et al.8are scaled by a factor of 0.3 for easier comparison. The temperature of the
photolysis cell was 295 K. The molecular beams of this work and Beames et al.8were jet-cooled (estimated rotational temperatureB10 K).
10 J. H. Lehman, H. Li, J. M. Beames and M. I. Lester, J. Chem. Phys., 2013, 139, 141103.
11 C. J. Percival, et al., Faraday Discuss., 2013, 165, 45–73. 12 M. Boy, D. Mogensen, S. Smolander, L. Zhou, T. Nieminen,
P. Paasonen, C. Plass-Du¨lmer, M. Sipila¨, T. Peta¨ja¨, L. Mauldin, H. Berresheim and M. Kulmala, Atmos. Chem. Phys., 2013, 13, 3865–3879.
13 C. A. Taatjes, G. Meloni, T. M. Selby, A. J. Trevitt, D. L. Osborn, C. J. Percival and D. E. Shallcross, J. Am. Chem. Soc., 2008, 130, 11883–11885.
14 D. Stone, M. Blitz, L. Daubney, T. Ingham and P. Seakins, Phys. Chem. Chem. Phys., 2013, 15, 19119–19124.
15 H. Huang, A. J. Eskola and C. A. Taatjes, J. Phys. Chem. Lett., 2012, 3, 3399–3403; H. Huang, B. Rotavera, A. J. Eskola and C. A. Taatjes, J. Phys. Chem. Lett., 2013, 4, 3824.
16 Y.-T. Su, Y.-H. Huang, H. A. Witek and Y.-P. Lee, Science, 2013, 340, 174–176.
17 Y.-T. Su, H.-Y. Lin, R. Putikam, H. Matsui, M. C. Lin and Y.-P. Lee, Nat. Chem., 2014, DOI: 10.1038/nchem.1890. 18 S. P. Sander, J. Abbatt, J. R. Barker, J. B. Burkholder, R. R.
Friedl, D. M. Golden, R. E. Huie, C. E. Kolb, M. J. Kurylo, G. K. Moortgat, V. L. Orkin and P. H. Wine, Chemical Kinetics and Photochemical Data for Use in Atmospheric Studies, Evaluation Number 17, JPL Publication 10-6, Jet Propulsion Laboratory, Pasadena, 2011, http://jpldataeval. jpl.nasa.gov.
19 M.-N. Su and J. J. Lin, Rev. Sci. Instrum., 2013, 84, 086106.
20 M.-N. Su and J. J. Lin, GSTF Journal of Chemical Sciences (JChem), 2013, 1, 52–57, DOI: 10.5176/2339-5060_1.1.6, ISSN: 2339-5060.
21 J. Sehested, T. Ellermann and O. J. Nielsen, Int. J. Chem. Kinet., 1994, 26, 259–272.
22 E. P. F. Lee, D. K. W. Mok, D. E. Shallcross, C. J. Percival, D. L. Osborn, C. A. Taatjes and J. M. Dyke, Chem. – Eur. J., 2012, 18, 12411–12423.
23 T. J. Gravestock, M. A. Blitz, W. J. Bloss and D. E. Heard, ChemPhysChem, 2010, 11, 3928.
24 T. J. Dillon, M. E. Tucceri, R. Sander and J. N. Crowley, Phys. Chem. Chem. Phys., 2008, 10, 1540.
25 A. J. Eskola, D. Wojcik-Pastuszka, E. Ratajczak and R. S. Timonen, Phys. Chem. Chem. Phys., 2006, 8, 1416–1424. 26 S. Enami, T. Yamanaka, S. Hashimoto, M. Kawasaki,
K. Tonokura and H. Tachikawa, Chem. Phys. Lett., 2007, 445, 152–156.
27 H.-Y. Chen, C.-Y. Lien, W.-Y. Lin, Y. T. Lee and J. J. Lin, Science, 2009, 324, 781.
28 J. J. Lin, A. F. Chen and Y. T. Lee, Chem. – Asian J., 2011, 6, 1664. 29 B. Jin, M.-N. Su and J. J. Lin, J. Phys. Chem. A, 2012, 116,
12082–12088.
30 J. Zhang, E. J. Heller, D. Huber, D. G. Imre and D. Tannor, J. Chem. Phys., 1988, 89, 3602.
31 H. F. Xu, Y. Guo, S. L. Liu, X. X. Ma, D. X. Dai and G. H. Sha, J. Chem. Phys., 2002, 117, 5722.
32 J. H. Lehman, H. Li and M. I. Lester, Chem. Phys. Lett., 2013, 590, 16–21.
33 J. C. Mo¨ssinger, D. E. Shallcross and R. A. Cox, J. Chem. Soc., Faraday Trans., 1998, 94, 1391–1396.
34 C. M. Roehl, J. B. Burkholder, G. K. Moortgat, A. R. Ravishankara and P. J. Crutzen, J. Geophys. Res., 1997, 102, 12819–12829.
35 W. Chehade, B. Gu¨r, P. Spietz, V. Gorshelev, A. Serdyuchenko, J. P. Burrows and M. Weber, Atmos. Meas. Tech., 2013, 6, 1623–1632.
36 B. Jin, I.-C. Chen, W.-T. Huang, C.-Y. Lien, N. Guchhait and J. J. Lin, J. Phys. Chem. A, 2010, 114, 4791.
