Photodissociation of CH2BrI using cavity ring-down spectroscopy: in search of a BrI elimination channel

Balaganesh Muthiah ab, Denís Paredes-Roibás c, Toshio Kasai bd and King-Chuen Lin *ab
aDepartment of Chemistry, National Taiwan Univeristy, Taipei 106, Taiwan. E-mail:
bInstitute of Atomic and Molecular Sciences, Academia Sinica, Taipei 106, Taiwan
cDepartment of CCTT Fisicoquímicas, Fac. Ciencias, UNED, Paseo de la Senda del Rey, 9, 28040-Madrid, Spain
dInstitute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka 567-0047, Japan

Received 29th June 2018 , Accepted 6th August 2018

First published on 7th August 2018

Photodissociation of CH2BrI was investigated in search of unimolecular elimination of BrI via a primary channel using cavity ring-down absorption spectroscopy (CRDS) at 248 nm. The BrI spectra were acquired involving the first three ground vibrational levels corresponding to A3Π1 ← X1Σ+ transition. With the aid of spectral simulation, the BrI rotational lines were assigned. The nascent vibrational populations for v′′ = 0, 1, and 2 levels are obtained with a population ratio of 1[thin space (1/6-em)]:[thin space (1/6-em)](0.58 ± 0.10)[thin space (1/6-em)]:[thin space (1/6-em)](0.34 ± 0.05), corresponding to a Boltzmann-like vibrational temperature of 713 ± 49 K. The quantum yield of the ground state BrI elimination reaction is determined to be 0.044 ± 0.014. The CCSD(T)//B3LYP/MIDI! method was employed to explore the potential energy surface for the unimolecular elimination of BrI from CH2BrI.

1 Introduction

Oceans are the primary natural sources of CH2BrI and other iodocarbons in the Earth's atmosphere. Among these organic iodine species, those containing two chromophores (CH2ClI, CH2BrI and CH2I2) are the most photo-labile. With a tropospheric lifetime ranging from a few days (CH3I) to a few minutes (CH2I2), that of CH2BrI is estimated to be on the order of 2.5 h. Iodomethanes are of importance mainly for tropospheric chemistry. Although iodocarbons are found at lower levels than their brominated counterparts, the combined effects of reactive iodine and bromine chemistry enhance the oxidative capacity, and subsequently cause efficient tropospheric ozone depletion. In addition, iodine is involved in new particle formation, which may grow to become cloud condensation nuclei, thus impacting the radiative balance of the atmosphere and hence the climate.1–4 Understanding the UV photochemistry of CH2BrI is therefore extremely important. The primary dissociation channels for CH2BrI upon photo-irradiation are given in eqn (1a)–(1f).
CH2BrI → CH2Br + I/I*(1a)
CH2BrI → CH2I + Br/Br*(1b)
CH2BrI → CHBr + HI(1c)
CH2BrI → CHI + HBr(1d)
CH2BrI → CH2 + BrI(1e)
CH2BrI → CH2 + Br + I(1f)

In eqn (1a) and (1b), the I or Br and I* or Br* represent the spin orbit ground (2P3/2) and spin orbit excited (2P1/2) states of iodine and bromine atoms, respectively. As reported, three UV absorption bands were identified. A broad band (nI, σC–I*) with the maximum near 270 nm (named the A band) corresponds to the excitation of non-bonding electrons of the I atom to the anti-bonding orbital of the C–I bond, while similarly a broad band (nBr, σC–Br*) has the maximum near 215 nm and a sharp feature around 190 nm. For the A band absorption, the I atom may be released rapidly from the excited state. In contrast, the molecular elimination (eqn (1c)–(1e)) may probably stem from the energetic ground state via internal conversion or intersystem crossing.

Thus far, most prior studies were focused on either I/I* or Br/Br* product channels resulting from the transitions of n(I) → σ*(C–I) and n(Br) → σ*(C–Br) in the photo-dissociation of CH2BrI.5–10 Butler et al.6,7 studied the primary photo-dissociation channels of CH2BrI following excitation at 193.3, 210, and 248.5 nm using the crossed laser-molecular beam technique. They reported the following observations: (i) excitation at 248.5 nm results in both C–I and C–Br bond fissions, (ii) excitation at 193.3 nm leads to C–Br bond fission, some amount of C–I bond fission, and some amount of concerted elimination of BrI, and (iii) excitation at 210 nm results in selective breaking of the C–Br bond and some amount of BrI product, but no fission of the C–I bond.

Attar et al.8 investigated the photo-dissociation of CH2BrI at 266 nm using XUV transient absorption spectroscopy. They observed both the C–I or C–Br reaction coordinates in real time and concluded that C–I dissociation is the dominant channel and C–Br dissociation is a minor pathway. They have measured C–I dissociation lifetimes leading to I and I* atomic products to be 48 ± 12 fs and 44 ± 4 fs, respectively, while the C–Br dissociation lifetime releasing atomic Br is 114 ± 17 fs. The induction of both C–I and C–Br bond breaking is consistent with those reported at 248.5 nm by the Butler group. Thus far, the BrI product was observed only at shorter wavelengths such as 210 and 193.3 nm. From the theoretical aspects, Abrashkevich et al.9 and Liu et al.10 looked into the bond and state selective photo-dissociation of CH2BrI.

Although, there are a great number of studies on the photo-dissociation of CH2BrI, none of them focus on the characterization and quantification of BrI formation. Even during shorter wavelength photolysis, Butler and coworkers observed BrI traces (<6% at 210 nm) but they were not able to quantify BrI because of the moderate sensitivity of their experimental technique. This work is aimed to perform single-photon photo-dissociation of CH2BrI at 248 nm using the cavity ring-down absorption spectroscopy (CRDS) technique which is well known for its high sensitivity, especially for the detection of halogen molecules.11–17 In this work, the ro-vibrational spectra of the fragmented molecular BrI are acquired and the BrI product is verified to stem from the primary photo-dissociation of the precursor.

Then, the quantum yields for BrI production along with the vibrational branching ratios of BrI(v′′ = 0, 1, and 2) are determined. With the aid of ab initio calculations, this dissociation channel is anticipated to proceed via internal conversion to the electronic ground state followed by dissociation through a transition state. Finally, the BrI production process is compared with Br2 and I2 elimination from CH2Br2 and CH2I2, respectively.

2 Experimental setup

The detailed description of the CRDS instrumental setup was explained elsewhere.11–17 Our CRDS is divided into three major units which are (i) a ring-down flow cell unit, (ii) a probe and photolysis laser system, and (iii) light detecting and signal analysis.

A KrF excimer laser was used to photolyze CH2BrI, while a Nd:YAG laser-pumped dye laser was adopted to probe the BrI fragment in the A3Π1 ← X1Σ+ transition. The probe laser beam was regulated through a spatial filter to retain the TEM00 mode to the most extent. Then, the probe beam and the photolysis beam, which were focused with a 30 cm focal-length cylindrical lens, were guided to overlap at right angle in the center of the ring down cell. A 20 ns delay time of the BrI probe with respect to the photolysis laser was retained. Two dyes pyridine I (667–736 nm) and coumarin 503 (485–546 nm) were used in probing the BrI fragment. The energies of photolysis and probe lasers were controlled in the ranges of 10–30 mJ and 2–4 mJ, respectively.

A crossed-shape stainless steel ring-down cell was designed with two short and two long arms; the later was sealed by a set of 99.95% reflective mirrors at either 700 nm or 550 nm with a diameter of 25.4 mm, and a radius of curvature of 1 m. To prevent these mirrors from chemical adsorption and contamination, argon gas was continuously purged. The samples were injected into the center of the cavity where photolysis and probe beams overlapped each other. Prior to injection of CH2BrI (Sigma aldrich) and BrI (Acros), the samples were purified by freeze–pump–thaw cycles at 77 K up to 3–4 times.

After the probe beam was injected through the front mirror, it will make several round trips inside the cavity. At each round trip, a tiny amount of probe beam leaked out of the ring-down cell and was monitored using a photo-multiplier tube which was positioned behind the rear mirror. The temporal profile of the ring-down signal was recorded on a transit digitizer. The ring-down time (τ) for each laser pulse may be determined by a best fit of the acquired exponential decay. The spectra of the samples can be obtained using the following equation,

image file: c8cp04130j-t1.tif(2)
where α is the wavelength (λ) dependent absorption coefficient, d is the length of the cavity, l is length of light interacting with the samples, c is the velocity of light, and τ0 and τ denote the ring down time without and with the sample present in the ring-down cell, respectively. Accordingly, the BrI absorption spectra with a 0.1 cm−1 spectral resolution were acquired and analyzed with the aid of a laboratory-developed program based on a MATLAB environment.

3 Theoretical methods

The BrI dissociation channel on the adiabatic singlet ground-state potential energy surface (PES) of CH2BrI is characterized. The geometries and the harmonic frequencies of reactants, transition states, and products were obtained at the level of hybrid density functional theory, B3LYP18,19/MIDI!.20,21 To obtain vibrational frequencies and zero-point energy, the same method was used. The reactant and products were identified with all positive frequencies, while transition states with one imaginary frequency. Intrinsic reaction coordinate22,23 (IRC) calculation was carried out using a second-order algorithm of Gonzalez and Schlegel, to confirm that the transition state structure connects the designated reactant and product. To further improve the accuracy of energies of B3LYP/MIDI! optimized minima and transition states, the CCSD(T)/MIDI! method was employed. The Gaussian09 program24 was utilized in the electronic structure calculations.

4 Results and discussion

4.1 Rovibrational spectra of the BrI fragment

The rovibrational spectra of BrI in v′′ = 0, 1 and 2 levels in the A3Π1 ← X1Σ+ transition are obtained by the photodissociation of CH2BrI at 248 nm, using CRDS. The spectra of BrI were acquired in different ranges from 705 to 720 nm, accounting for the vibrational levels (v′′ = 0, 1, and 2). The measured spectrum was confirmed by comparing with that of the pure BrI molecule and further verified by spectral simulation, as shown in Fig. 1. The comparison shows good agreement with each other, thereby confirming the formation of BrI.
image file: c8cp04130j-f1.tif
Fig. 1 A portion of BrI spectra acquired in the photolysis of CH2BrI at 248 nm. (a) BrI spectrum from photolysis of CH2BrI at 248 nm, (b) pure BrI molecule, (c) simulation and (d) background spectrum.

To inspect whether any secondary reaction probably contributed to the formation of BrI/Br2/I2, we carried out photo-dissociation of CH2BrI under identical conditions in the region (522.2 to 522.6 nm) where Br2 has the maximum absorption. The spectrum of the BrI fragment is compared with the Br2 spectrum acquired for the pure Br2 molecule, as shown in Fig. 2.

image file: c8cp04130j-f2.tif
Fig. 2 Comparison of the BrI spectrum with the Br2 spectrum. (a) Pure Br2 molecule, (b) pure BrI molecule, (c) BrI spectrum from photolysis of CH2BrI at 248 nm and (d) background spectrum.

It is found that no significant Br2 product was formed during photolysis of CH2BrI even after increasing the sample concentration to 500 mTorr and the photolysis-probe delay time to 100 ns. As a result, one can conclude that the BrI fragment is eliminated by the primary photodissociation of CH2BrI and its contribution from the secondary reaction should be negligible. If the secondary reaction does occur, one might observe the formation of either BrI, Br2 or I2 product as per the following secondary reaction.

I + CH2BrI → I2/BrI + other fragments(3a)
I + CH2Br → BrI + other fragments(3b)
I + CH2I → I2 + other fragments(3c)
Br + CH2BrI → Br2/BrI + other fragments(3d)
Br + CH2Br → Br2 + other fragments(3e)
Br + CH2I → BrI + other fragments(3f)

In addition, the dependence of photolysis laser energy and pressure of CH2BrI on the formation of BrI was studied at 519.72 nm. The plot of the concentration of BrI against the photolysis laser energy yields a straight line which supports the single-photon involvement in the molecular elimination of BrI and the plot is shown in Fig. 3.

image file: c8cp04130j-f3.tif
Fig. 3 Absorption coefficient of the BrI fragment as a function of photolysis laser energy varied from ∼8 to 18 mJ. The rotational line at 519.72 nm is selected for this measurement.

Given a constant photolysis laser energy, the plot of the pressure dependence of CH2BrI (17–110 mTorr) against the produced BrI concentration yields a straight line extrapolated through the origin point, as shown in Fig. 4. This supports that there is no or minimal probability to form BrI by the secondary recombination process.

image file: c8cp04130j-f4.tif
Fig. 4 The amount of the BrI fragment produced as a function of CH2BrI concentration varied from 3 × 1015 to 5 × 1015 molecules cm−3. The rotational line at 519.72 nm is selected for this measurement.

In case a pseudo-first order reaction occurs in eqn (3a)–(3f), then a single BrI molecule can be produced by consuming only one CH2BrI molecule, because the number density of produced halogen atoms is much smaller than that of the precursor. The plot of the pressure dependence also yields a unity. However, the photodissociation of CH2BrI and its chemical analogs is dominated by the elimination of halogen atoms, and thus the pseudo-first order reaction cannot be applied to the secondary reactions.

We have carried out carefully the experiment in different regions containing v′′ = 0, 1 and 2 bands. A portion of the BrI spectrum in the range of 705.5–705.9 nm was acquired, corresponding to the vibrational bands (v′,v′′) = (20,0) and (28,1) in the A3Π1 ← X1Σ+ transition. The measured spectrum with partially assigned peaks was compared with the pure BrI and simulated counterpart, as shown in Fig. S1 of the ESI. Their comparison is in good agreement. Similarly, Fig. 5 shows the BrI spectra in the range of 719.25–719.6 nm corresponding to the bands of (16,0), (20,1) and (28,2). In Fig. 5, we have included the pure BrI spectrum and simulated counterparts for individual vibrational bands and their sum with appropriate adjustment of the population ratio of BrI(v′′ = 1)/BrI(v′′ = 0) and BrI(v′′ = 2)/BrI(v′′ = 1). When BrI(v′′ = 1)/BrI(v′′ = 0) is fixed at 0.58 and BrI(v′′ = 2)/BrI(v′′ = 1) is adjusted to 0.34, the spectral simulation becomes appreciably consistent with the experimental finding.

image file: c8cp04130j-f5.tif
Fig. 5 A portion of BrI spectra acquired in the photolysis of CH2BrI at 248 nm. (a) Trace acquired experimentally for the bands at v′′ = 0, 1, and 2, (b) the spectrum of the pure BrI molecule, (c) the simulated counterpart with the population ratio of BrI(v′′ = 1)/BrI(v′′ = 0) fixed at 0.58 and BrI(v′′ = 2)/BrI(v′′ = 0) adjusted to 0.34, (d) simulated counterpart with the transition involving only v′′ = 0, (e) simulated counterpart with the transition involving only v′′ = 1, and (f) simulated counterpart with the transition involving only v′′ = 2.

4.2 Simulation and vibrational populations

Since the average ring down time was about 800 ns, the nascent vibrational population may be preserved approximately because the vibrational energy transfer is slow, but the rotational population was rapidly thermally equilibrated. There is no difference in the rovibrational spectra, even though the photolysis-probe delay time is prolonged to hundred ns. Due to the lack of enough data on Franck–Condon factors (FCFs), it is required to construct potential energy curves and subsequently evaluate FCFs for the simulation of the ro-vibrational spectrum of BrI. Given the spectroscopic parameters of the desired electronic states, the potential energy curve of a diatomic molecule can be calculated using the Rydberg25,26–Klein27–Rees28 (RKR) method. The information on the spectroscopic parameters of X1Σ+ and A3Π1 electronic states of BrI were taken from the literature29–35 and used to construct the outer and inner turning points of the potential curves for both X1Σ+ and A3Π1 states with the RKR-16 program.36 The FCFs are the extent of overlap between the ro-vibrational wavefunctions of two different electronic states and directly associated with the intensity of transitions as can be seen from eqn (4). Given the potential information of X1Σ+ and A3Π1 states, the FCFs of various rovibrational states were calculated37 and are listed in Table S1 (ESI). Then, the BrI spectrum was simulated with the aid of PGOPHER software.38 The BrI spectral intensities I can be further evaluated according to the following equation
image file: c8cp04130j-t2.tif(4)
where k is a factor associated with instrument and experimental conditions, HLF the Hönl–London factor, J′′ the rotational quantum number of the ground state BrI, NJ′′ the Boltzmann distribution of the rotational population for which 300 K is assumed, riso the ratio of isotopic variants, 79BrI[thin space (1/6-em)]:[thin space (1/6-em)]81BrI equal to 0.4999[thin space (1/6-em)]:[thin space (1/6-em)]0.5001, and f the intensity ratio of strong and weak rotational lines. Accordingly, the BrI spectra may be simulated for both spectral positions and intensities, as displayed in Fig. 1 and 5.

The intensities of simulated spectra were best fit with experimental results by optimizing the population ratio of BrI(v′′ = 1)/BrI(v′′ = 0) to be 0.58 ± 0.10 in the range of 705.5–705.9 nm. A similar approach was applied to obtain the population ratio of BrI(v′′ = 2)/BrI(v′′ = 0) to be 0.34 ± 0.05 in the range of 719.37–719.44 nm while the BrI(v′′ = 1)/BrI(v′′ = 0) ratio was fixed at 0.58 ± 0.10. An attempt to evaluate the population ratio of v′′ = 3 was made, but its intensities are buried in the congested spectra composed of v′′ = 0, 1, and 2 population and cannot be estimated accurately. Fig. 6 shows a detailed comparison of the experimental spectrum with the simulated counterpart to evaluate the simulation uncertainty by varying the population ratios. Consequently, the resulting population ratio of the first three ground vibrational levels is 1[thin space (1/6-em)]:[thin space (1/6-em)](0.58 ± 0.10)[thin space (1/6-em)]:[thin space (1/6-em)](0.34 ± 0.05), corresponding to the Boltzmann vibrational temperature of 713 ± 49 K.

image file: c8cp04130j-f6.tif
Fig. 6 (a) Comparison between the observed (solid line) and simulated spectra (dashed line) of BrI in which the v′′ = 1/v′′ = 0 vibrational population ratio is optimized to 0.58 ± 0.10. (b) Comparison between the observed and simulated spectra in which the v′′ = 2/v′′ = 0 population ratio is optimized to 0.34 ± 0.05, as the v′′ = 1/v′′ = 0 ratio is fixed at 0.58.

4.3 Quantum yield

The quantum yield for molecular elimination of BrI from the photolysis of CH2BrI was determined as follows:
image file: c8cp04130j-t3.tif(5)
where [BrI] indicates the BrI concentration produced in the beam-crossed region of photolysis/probe lasers and Np is the number density of photons absorbed in the same region. Following our previous work,11,12,15,17 the above equation may be modified to:
image file: c8cp04130j-t4.tif(6)
where E0 is the photolysis laser energy prior to the cell, h is the Planck constant, ν is the radiation frequency, ΔV is the volume of the beam-crossed region, n is the number density of samples in the cell, σ is the absorption cross section of CH2BrI at 248 nm, and l1 (or l2) indicates the distance between the inner side of the entrance cell window and the front (or rear) edge of the beam-crossed region.

The quantum yield for the formation of BrI elimination channels can also be determined by a relative method, when the quantum yields of similar reactions were well established. Br2 formation from CH2Br2 was chosen as a reference reaction with a quantum yield of 0.21 ± 0.06.17 The following equation is used for calculating the current quantum yield:

image file: c8cp04130j-t5.tif(7)
where [BrI] and [Br2] are the concentrations obtained from the experiment, nCH2BrI and nCH2Br2 are the number density and σCH2BrI and σCH2Br2, the absorption cross sections at 248 nm, are measured to be 1.6 × 10−18 cm2 and 3.7 × 10−19 cm2,13 respectively. Given the ring-down time τ (or τ0) with (or without) photolysis of CH2BrI in the ring-down cell, the absorption cross section σ of BrI and Br2, and other parameters reported previously, the BrI and Br2 concentrations are determined by:
image file: c8cp04130j-t6.tif(8)

Substituting [BrI] and [Br2] along with the above-mentioned data into eqn (7) gives rise to a quantum yield of 0.044 ± 0.014 for the BrI fragment eliminated from CH2BrI. In comparison with this result, the formation quantum yield of Br2 and I2 from the photodissociation of CH2Br213,17 and CH2I216 at the same wavelength yielded values of 0.21 ± 0.06 and 0.004 ± 0.0025, respectively; as listed in Table 1. The quantum yield for the BrI channel from CH2BrI is smaller by 5 times with respect to Br2 from CH2Br2 and larger by 10 times compared with I2 from CH2I2. In addition, the BrI quantum yield of 4.3% in this work is consistent with the trend of the Butler et al. observation (<6% at 210 nm).

Table 1 The values of quantum yields for the three different reactions
Reaction Quantum yield
*This study
CH2Br2 → CH2 + Br2 0.21 ± 0.0613,17
CH2I2 → CH2 + I2 0.004 ± 0.002516
CH2BrI → CH2 + BrI 0.044 ± 0.014*

4.4 Photodissociation mechanism

The PES analysis for the formation of BrI from CH2BrI was carried out using density functional theory. The geometries of the reactant (CH2BrI), the transition state (tsBrI) and products (CH2 and BrI) were optimized at the B3LYP/MIDI! level of theory. The reactant and products were identified with no negative frequencies but the transition state was identified with one negative frequency. The intrinsic reaction coordinate (IRC) calculations were carried out on the tsBrI geometry to confirm the transition states connected with the designated reactant and products. The optimized geometries of these components involved in the reaction pathway are displayed in Fig. 7 with internal coordinates.
image file: c8cp04130j-f7.tif
Fig. 7 The optimized geometries of reactant, transition state and products for the unimolecular dissociation of CH2BrI at the B3LYP/MIDI! level of theory.

The C–Br and C–I bond distances in tsBrI are elongated up to 2.391 Å and 2.995 Å as compared with the reactant where the C–Br and C–I bond distances are 1.995 Å and 2.185 Å, respectively. In addition, the Br–I bond distance in tsBrI is 2.582 Å, while it is 3.487 Å in the reactant. The Br–C–I bond angle is shortened to 55.9° in tsBrI from 114.6° as in the reactant. This reveals that the three-center concerted dissociation mechanism from the transition state may yield a vibrationally hot BrI product.

The high level CCSD(T) calculations were carried out on the B3LYP optimized geometry to get more accurate energy. The B3LYP/MIDI!, zero-point and CCSD(T)/MIDI! energies of all geometries are tabulated in Table 2. The ground state CH2BrI decomposes into CH2 and BrI via tsBrI with a barrier height of 409.4 kJ mol−1. The reaction scheme is endothermic with an energy of 398.7 kJ mol−1. The dissociation pathway for CH2BrI to form BrI was evaluated at the CCSD(T)//B3LYP/MIDI! level of theory and is shown in Fig. 8.

Table 2 The calculated energies of reactant, transition state and products at various levels of theory
  B3LYP/midix (Hartree) E zpc (Hartree) CCSD(T)/midix (Hartree) E (kJ mol−1)
CH2BrI −9491.188022 0.027717 −9486.8150796 0.0
ts-BrI −9491.036649 0.021495 −9486.6529203 409.4
CH2 −38.88066 0.016225 −38.727781309  
BrI −9452.148011 0.000612 −9447.9245541  
CH2 + BrI −9491.028671 0.016837 −9486.6523354 398.72

image file: c8cp04130j-f8.tif
Fig. 8 The potential energy surface diagram for the formation of BrI from the CH2BrI dissociation channel, computed at the CCSD(T)/MIDI! level of theory with B3LYP/MIDI! zero-point energy corrections.

In addition to PES exploration, the density and sum of states were calculated for the reactant and transition states using the Baeyer–Swinehart algorithm implemented in the multiwell program.39,40 The density of states of CH2BrI evaluated to be 7.75 × 105/cm−1 is in between those of CH2I2 (1.15 × 106/cm−1) and CH2Br2 (5.61 × 105/cm−1) at the same excitation energy at 248 nm. The energy dependent density of states for CH2BrI, CH2I2 and CH2Br2 are plotted and shown in Fig. S2 (ESI). The unimolecular rate constant for the dissociation channel (CH2BrI → CH2 + BrI) was also calculated to be 2.12 × 109 s−1 using Rice–Ramsperger–Kassel–Marcus (RRKM) theory.

5 Conclusion

Photodissociation of CH2BrI to eliminate the BrI fragment at 248 nm was investigated using the CRDS technique. The BrI(v′′ = 0, 1 and 2) spectra were acquired in the A3Π1 ← X1Σ+ transition. Generation of the primary BrI product was confirmed in the photolysis, that is, CH2BrI + → CH2 + BrI, while the contribution from the secondary reactions such as I + CH2BrI → BrI + CH2I was negligible. Due to the lack of Franck–Condon factors in the A3Π1 ← X1Σ+ transition in the literature, the RKR potential curves were constructed and the FCFs were subsequently evaluated. Given the FCFs along with some other molecular parameters, the BrI spectra were simulated and the results were used for the peak assignment of the experimental findings. Furthermore, with the aid of spectral simulation, we have calculated the vibrational population ratio of v′′ (0, 1 and 2) to be 1[thin space (1/6-em)]:[thin space (1/6-em)](0.58 ± 0.10)[thin space (1/6-em)]:[thin space (1/6-em)](0.34 ± 0.05), corresponding to a Boltzmann vibrational temperature of 713 ± 49 K. The quantum yield for the formation channel of BrI was also estimated as 0.044 ± 0.014, which lies between those of the Br2 and I2 fragments eliminated from CH2Br2 and CH2I2, respectively. The related dissociation pathway was theoretically explored. This is the first case to observe the heterohalogen fragment in the photolysis of halo-hydrocarbons which can be well characterized.

Conflicts of interest

There are no conflicts to declare.


This work is supported by the National Science Council of Taiwan, Republic of China, under Contract No. NSC 102-2113-M-002-009-MY3 and the ERASMUS MUNDUS EURASIACAT project (Advanced Education European-Asiatic Exchange Programme in Materials Science and Catalysis, with ref. no. 552067). Computer resources of the National Center for High-performance Computer of Taiwan were utilized in the calculations.


  1. 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 CrossRef CAS.
  2. A. J. Eskola, D. Wojcik-Pastuszka, E. Ratajczak and R. S. Timonen, Phys. Chem. Chem. Phys., 2006, 8, 1416–1424 RSC.
  3. A. Masaki, S. Tsunashima and N. Washida, J. Phys. Chem., 1995, 99, 13126–13131 CrossRef CAS.
  4. R. Vogt, R. Sander, R. Von Glasow and P. J. Crutzen, J. Atmos. Chem., 1999, 32, 375–395 CrossRef CAS.
  5. S. Solomon, R. R. Garcia and A. Ravishankara, J. Geophys. Res.: Atmos., 1994, 99, 20491–20499 CrossRef.
  6. L. Butler, E. Hintsa and Y.-T. Lee, J. Chem. Phys., 1986, 84, 4104–4106 CrossRef CAS.
  7. L. Butler, E. Hintsa, S. F. Shane and Y.-T. Lee, J. Chem. Phys., 1987, 86, 2051–2074 CrossRef CAS.
  8. A. R. Attar, L. Piticco and S. R. Leone, J. Chem. Phys., 2014, 141, 164308 CrossRef.
  9. D. G. Abrashkevich, M. Shapiro and P. Brumer, J. Chem. Phys., 2002, 116, 5584–5592 CrossRef CAS.
  10. K. Liu, H. Zhao, C. Wang, A. Zhang, S. Ma and Z. Li, J. Chem. Phys., 2005, 122, 044310 CrossRef.
  11. C.-C. Wu, H.-C. Lin, Y.-B. Chang, P.-Y. Tsai, Y.-Y. Yeh, H. Fan, K.-C. Lin and J. Francisco, J. Chem. Phys., 2011, 135, 234308 CrossRef.
  12. H.-Y. Huang, W.-T. Chuang, R. C. Sharma, C.-Y. Hsu, K.-C. Lin and C.-H. Hu, J. Chem. Phys., 2004, 121, 5253–5260 CrossRef CAS.
  13. P.-Y. Wei, Y.-P. Chang, W.-B. Lee, Z. Hu, H.-Y. Huang, K.-C. Lin, K. Chen and A. Chang, J. Chem. Phys., 2006, 125, 133319 CrossRef.
  14. P.-Y. Wei, Y.-P. Chang, Y.-S. Lee, W.-B. Lee, K.-C. Lin, K. Chen and A. Chang, J. Chem. Phys., 2007, 126, 034311 CrossRef.
  15. C.-Y. Hsu, H.-Y. Huang and K.-C. Lin, J. Chem. Phys., 2005, 123, 134312 CrossRef.
  16. S.-Y. Chen, P.-Y. Tsai, H.-C. Lin, C.-C. Wu, K.-C. Lin, B. Sun and A. Chang, J. Chem. Phys., 2011, 134, 034315 CrossRef.
  17. B.-J. Chen, P.-Y. Tsai, T.-K. Huang, Z.-H. Xia, K.-C. Lin, C.-J. Chiou, B.-J. Sun and A. Chang, Phys. Chem. Chem. Phys., 2015, 17, 7838–7847 RSC.
  18. A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652 CrossRef CAS.
  19. C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785 CrossRef CAS.
  20. D. Feller, J. Comput. Chem., 1996, 17, 1571–1586 CrossRef CAS.
  21. K. L. Schuchardt, B. T. Didier, T. Elsethagen, L. Sun, V. Gurumoorthi, J. Chase, J. Li and T. L. Windus, J. Chem. Inf. Model., 2007, 47, 1045–1052 CrossRef CAS.
  22. C. Gonzalez and H. B. Schlegel, J. Chem. Phys., 1989, 90, 2154–2161 CrossRef CAS.
  23. C. Gonzalez and H. B. Schlegel, J. Phys. Chem., 1990, 94, 5523–5527 CrossRef CAS.
  24. M. Frisch, G. Trucks, H. B. Schlegel, G. Scuseria, M. Robb, J. Cheeseman, G. Scalmani, V. Barone, B. Mennucci and G. Petersson, et al., Gaussian 09, Revision A.02, Gaussian Inc., Wallingford, CT, 2009 Search PubMed.
  25. R. Rydberg, Z. Phys., 1932, 73, 376–385 CrossRef CAS.
  26. R. Rydberg, Z. Phys., 1933, 80, 514–524 CrossRef CAS.
  27. O. Klein, Z. Phys., 1932, 76, 226–235 CrossRef CAS.
  28. A. Rees, Proc. Phys. Soc., 1947, 59, 998 CrossRef CAS.
  29. R. E. Willis Jr and W. W. Clark III, J. Chem. Phys., 1980, 72, 4946–4950 CrossRef.
  30. J. Campbell and P. Bernath, J. Mol. Spectrosc., 1993, 158, 339–346 CrossRef CAS.
  31. D. R. Appadoo, P. F. Bernath and R. J. L. Roy, Can. J. Phys., 1994, 72, 1265–1272 CrossRef CAS.
  32. B. N. V. S. M. Dulick and V. B. P. Bernath, Mol. Phys., 1998, 93, 137–144 CrossRef.
  33. T. Yukiya, N. Nishimiya and M. Suzuki, J. Mol. Spectrosc., 2002, 214, 132–143 CrossRef CAS.
  34. J. Pittner and P. Jungwirth, Chem. Phys. Lett., 2000, 321, 281–286 CrossRef CAS.
  35. S. Patchkovskii, Phys. Chem. Chem. Phys., 2006, 8, 926–940 RSC.
  36. R. J. Le Roy, J. Quant. Spectrosc. Radiat. Transfer, 2017, 186, 158–166 CrossRef.
  37. R. J. Le Roy, J. Quant. Spectrosc. Radiat. Transfer, 2017, 186, 167–178 CrossRef.
  38. C. M. Western, J. Quant. Spectrosc. Radiat. Transfer, 2017, 186, 221–242 CrossRef CAS.
  39. J. Barker, T. Nguyen, J. Stanton, C. Aieta, M. Ceotto, F. Gabas, T. Kumar, C. Li, L. Lohr and A. Maranzana, et al., Millimeter wave measurements of the rotational spectra of ClF, BrF, BrCl, ICl, and IBr, Ann Arbor, Michigan, USA, 2016 Search PubMed.
  40. J. R. Barker, Int. J. Chem. Kinet., 2001, 33, 232–245 CrossRef CAS.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c8cp04130j

This journal is © the Owner Societies 2019