Introduction

Criegee intermediates, also known as carbonyl oxides, are generated through ozonolysis of alkene in the atmosphere.¹ Criegee intermediates can react with atmospheric gases, such as SO2, NOx, H2O, organic acid, inorganic acid, volatile organic compounds (VOCs), etc. Also, the unimolecular decomposition of Criegee intermediates generates OH radicals, which plays an important role in atmospheric chemistry.²

Before 2012, the direct measurement of the Criegee intermediate in kinetic studies was unachievable. The steady-state concentration of the Criegee intermediate is

estimated by Equation 1 below.

3

3

alkene O 3

s.s.

uni CI+O 3 CI+alkene CI+product

[alkene][O ]

[CI] [O ] [alkene] [product]

k

k k k k

 

   (Equation 1)

where kalkene+O3 represents the reaction rate coefficient of ozone with alkene, which is only ~10^17 cm³ s^1;³ k_uni indicated the unimolecular decay rate of Criegee intermediates, which is ~10¹-10³ s^1 dependent on the structure of Criegee intermediates;^4-6 k_CI+O3 is the reaction rate coefficient of Criegee intermediates with ozone, which has been reported to be 6.7 x 10^14 cm³ s^1 for CH2OO;⁷ kCI+alkene means the reaction rate coefficient of Criegee intermediates with alkene, which has been reported to be 1.5 x

10^15 cm³ s^1 for the reaction of CH₂OO with isoprene;⁸ k_CI+product is the reaction rate

coefficient of Criegee intermediate with products, such as aldehyde, ketone, acid, etc, the reaction rate coefficients of Criegee intermediates with acids has been reported to be

~10^10-10^11 cm³ s^1.^{6, 9}

Assume [alkene] = 10¹⁰ cm^3, [O₃] = 10¹² cm^3, and the reaction rate coefficients of Criegee intermediates + O3 and Criegee intermediates + alkene can be represented by the rate coefficients of CH2OO + O3 and CH2OO + isoprene, the steady-state

concentration of Criegee intermediates is estimated to be:

     

which is too low to be measured directly.

If we increase the concentration of alkene and O3 to obtain higher concentration of Criegee intermediates, the concentration of byproducts, such as aldehyde, ketone, and acid, would also increase. The reaction of Criegee intermediates with byproducts would make the kinetics of Criegee intermediates too complicated to be solved.

The reaction rate coefficients obtained from indirect measurements have a large scatter of several orders of magnitude.^{5, 10} At 2012, Taatjes et al. reported a method to generate the simplest Criegee intermediate (CH2OO) through the reaction of O2 and a iodo-alkyl radical, produced from the photolysis of diiodomethane (CH₂I₂ + h → CH2I

+ I; CH2I + O2 → CH2OO + I), and directly measured the kinetics of Criegee

intermediate for the first time.¹¹ Many researches on small alkyl Criegee intermediates (CH2OO, CH3CHOO, (CH3)2COO, CH3CH2CHOO) emerged since then, and there have been several reviews to summary these works.^{6, 12-18} However, due to the lack of proper precursors, the researches on large and non-alkyl Criegee intermediates are relatively

scarce.

Isoprene is the most abundant alkene in the atmosphere.¹⁹ The ozonolysis of isoprene can generate three kinds of Criegee intermediates, which are methyl vinyl ketone oxide (CH3(C2H3)COO, MVKO, yield = 0.23), formaldehyde oxide (CH2OO, yield = 0.58), and methacrolein oxide (CH3CCH2CHOO, MACRO, yield = 0.19).⁴ Both MVKO and MACRO are resonance-stabilized vinyl substituted Criegee intermediates. However, it is difficult to study their individual kinetics from the ozonolysis of isoprene.^{20, 21}

At 2018, Barber et al. reported a new method to produce methyl vinyl ketone oxide from the reaction of O2 and an iodo-alkenyl radical, which was produced from the

photolysis of 1,3-diiodo-2-butene (ICH₃CHCICH₃ + h → CH3(C₂H₃)CI + I;

CH3(C2H3)CI + O2 → CH3(C2H3)COO (MVKO) + I).⁴ This method made the direct measurement of MVKO in kinetic studies possible. There are four possible conformers of MVKO, which are syn-trans-MVKO, syn-cis-MVKO, anti-trans-MVKO, and

anti-cis-MVKO; syn and anti refer to the orientation of methyl group with respect to the

terminal oxygen, while trans and cis refer to the orientation of the C=C and C=O double bond. The calculated barrier for the trans-cis isomerization is lower than that of the syn-anti isomerization. syn-trans-MVKO is the most stable conformer.⁴ (See Figure 1)

Figure 1. Four conformers of MVKO with their calculated zero-point-corrected energies.

The calculated isomerization barriers are also shown. The calculated values are adapted from Barber et al.⁴ Geometries optimization is done at the B2PLYP-D3/cc-pVTZ level, and energies are calculated at the ANL0-B2F level. The stationary point energy for the transition state of anti-cis-MVKO to syn-cis-MVKO is calculated by

CCSD(T)-F12b/cc-pVTZ-F12 with some corrections.

Barber et al. used an infrared laser to pump MVKO and probed the OH radical product with laser induced fluorescence. IR spectrum of MVKO was measured with this IR pump-UV probe system. All four conformers of MVKO were observed. The

energy-dependent unimolecular decay rate coefficients k(E) of MVKO were also

obtained through the formation rate coefficient of OH radicals from MVKO excited in the region of CH stretch overtone (2CH). The thermal-dependent unimolecular decay

rate coefficient k(T) was predicted through the extension of k(E) with the master

equation modeling. For syn-MVKO, the thermal rate coefficient for unimolecular decay was calculated to be 33 s^1 at 298K, and that for anti-MVKO was reported to be 2140 s^1.⁴

Vansco et al. used the same method as Barber et al. to generate MVKO in pulsed supersonic expansion. MVKO was excited at its first * ←  electronic band with

tunable UV-Vis light at 305-480 nm before being ionized with vacuum ultraviolent photoionization at 118 nm. The UV spectrum of MVKO with peak at 388 nm was measured from the depletion of the mass signal at m/z = 86 channel (MVKO, m = 86).²² At 2020, Caravan et al. (including our group) reported the UV absorption spectrum of MVKO at 298 K with multi-pass broadband probe light. The peak of the spectrum was at 370.6 nm with full width at half maximum (FWHM) of 73.4 nm.⁹ Comparing the UV spectrum of MVKO with that of small alkyl Criegee intermediates,^23-26 we could find

that the spectrum of MVKO was red-shifted. This result is consistent with the previous theoretical research.²⁷

Our group generated MVKO with the method reported by Barber et al

(ICH₃CHCICH₃ + h → CH₃(C₂H₃)CI + I; CH₃(C₂H₃)CI + O₂ → CH₃(C₂H₃)COO (MVKO) + I)⁴ and studied the kinetics of MVKO via its strong UV absorption. The detail of experimental setup is in chapter 2. We measured the reaction rate coefficient of MVKO with SO2 at 4-703 Torr. Also, we observed that the behavior of MVKO

generation depends on pressure and used a model with the formation of MVKO from an adduct to explain this phenomenon. The recorded time traces of MVKO were fitted well with the model considering the adduct kinetics. The results and discussion about the kinetics of the adduct as well as the reaction of MVKO with SO2 are in chapter 3.

Chapter 4 discusses the unimolecular decomposition of MVKO. The unimolecular reaction of MVKO was measured in 100-503 Torr and at 278-319 K.

Chapter 2. Experimental setup