Kinetics of CH₂OO reactions with SO₂, NO₂, NO, H₂O and CH₃CHO as a function of pressure

Abstract

Kinetics of CH₂OO Criegee intermediate reactions with SO₂, NO₂, NO, H₂O and CH₃CHO and CH₂I radical reactions with NO₂ are reported as a function of pressure at 295 K. Measurements were made under pseudo-first-order conditions using flash photolysis of CH₂I₂–O₂–N₂ gas mixtures in the presence of excess co-reagent combined with monitoring of HCHO reaction products by laser-induced fluorescence (LIF) spectroscopy and, for the reaction with SO₂, direct detection of CH₂OO by photoionisation mass spectrometry (PIMS). Rate coefficients for CH₂OO + SO₂ and CH₂OO + NO₂ are independent of pressure in the ranges studied and are (3.42 ± 0.42) × 10⁻¹¹ cm³ s⁻¹ (measured between 1.5 and 450 Torr) and (1.5 ± 0.5) × 10⁻¹² cm³ s⁻¹ (measured between 25 and 300 Torr), respectively. The rate coefficient for CH₂OO + CH₃CHO is pressure dependent, with the yield of HCHO decreasing with increasing pressure. Upper limits of 2 × 10⁻¹³ cm³ s⁻¹ and 9 × 10⁻¹⁷ cm³ s⁻¹ are placed on the rate coefficients for CH₂OO + NO and CH₂OO + H₂O, respectively. The upper limit for the rate coefficient for CH₂OO + H₂O is significantly lower than has been reported previously, with consequences for modelling of atmospheric impacts of CH₂OO chemistry.

---

1. Introduction

Criegee intermediates, carbonyl oxide biradicals with the general formula CR₂OO, are principally produced in the atmosphere following ozonolysis of unsaturated volatile organic compounds (VOCs) and are key species in the tropospheric oxidation of both biogenic and anthropogenic compounds.^1,2 The exothermicity of ozonolysis reactions leads to production of vibrationally excited Criegee intermediates with sufficient energy to undergo unimolecular decomposition to products including OH and HO₂,^3–6 representing a significant source of these key oxidising species in certain important environments.^7–9 However, collisional quenching of the nascent excited Criegee intermediate by N₂ or O₂, to produce stabilised Criegee intermediates, is competitive with the unimolecular decomposition processes at ambient pressures,^1,5 and reactions of stabilised Criegee intermediates have the potential to impact atmospheric budgets of NO_x (NO_x = NO + NO₂), NO₃, O₃, HO_x (HO_x = OH + HO₂), SO₂, H₂SO₄, sulfate aerosol and secondary organic aerosol (SOA).^5,10–17

Despite their potential importance in atmospheric chemistry, and thus in the assessment and prediction of issues such as air quality and climate change, direct observations of Criegee intermediates have only recently been achieved.^{10–12,18–20} Kinetics and product yields of Criegee intermediate reactions currently used in atmospheric models are subject to large uncertainties, owing to the reliance of previous investigations on indirect techniques involving measurements of stable species in complex ozonolysis experiments, in which there are several potential sources and sinks of the measured species.^1,2 Welz et al.¹⁰ reported the first direct measurements of Criegee intermediate kinetics, where the photolysis of CH₂I₂ in the presence of O₂ was used to generate the CH₂OO Criegee intermediate at low pressure (4 Torr) and, using synchrotron photoionisation mass spectrometry (PIMS) at the Advanced Light Source (ALS), demonstrated unequivocally that the Criegee intermediate, CH₂OO, was being monitored:

CH₂I₂ + hν → CH₂I + I (R1)

CH₂I + O₂ → CH₂OO + I (R2a)

While reactions of CH₂OO with NO and water vapour were reported to be slow, the reactions of CH₂OO with SO₂ and NO₂ were shown to be significantly faster than indicated by the indirect methods. Rate coefficients for both CH₂OO + SO₂ and CH₂OO + NO₂, measured at a pressure of 4 Torr and temperature of 298 K, were both approximately 1000 times greater than previously assigned, implying a more significant role of Criegee intermediate chemistry in the atmosphere than expected.

The ability to produce CH₂OO following photolysis of CH₂I₂ in the presence of O₂¹⁰ has also facilitated spectroscopic investigations of CH₂OO in the infrared¹⁹ and ultraviolet,²⁰ and has been used to demonstrate the production of NO₃ in the reaction of CH₂OO with NO₂.²¹ Subsequent work at the ALS has investigated the reactions of CH₂OO with acetone, acetaldehyde and hexafluoroacetone at low pressures,¹¹ with theoretical investigation²² of the reaction between CH₂OO and acetaldehyde (CH₃CHO) indicating pressure dependence of the reaction and collisional stabilisation of nascent reaction adducts to produce secondary ozonides (SOZs) at higher pressures which subsequently decompose to generate organic acids.

Taatjes et al.¹² have also recently demonstrated production of the CH₃CHOO Criegee intermediate following photolysis of CH₃CHI₂ in the presence of O₂. The structure of the CH₃CHOO Criegee intermediate gives rise to the possibility of syn- and anti-conformers, with the conformers sufficiently different in energy, and with a barrier to conversion, leading to the potential for their behaviour as distinct species. Using the synchrotron PIMS technique, Taatjes et al.¹² were not only able to identify both the syn- and anti-CH₃CHOO conformers, but were also able to assign separate rate coefficients for reactions of the two conformers with SO₂ and water vapour. The anti-conformer was shown to display greater reactivity towards both SO₂ and H₂O compared to the syn-conformer, with rate coefficients for reactions of both syn- and anti-conformers with SO₂ greater than previously expected.¹²

Field observations in a boreal forest in Finland have provided further evidence for rapid reactions between Criegee intermediates and SO₂, with measurements identifying the presence of oxidising species other than OH which are able to oxidise SO₂ to SO₃ and ultimately to produce H₂SO₄.²³ The presence of the unknown oxidising species was shown to be related to emissions of biogenic alkenes, and it was postulated that Criegee intermediates may be responsible, with laboratory measurements of H₂SO₄ production during alkene ozonolysis reactions in the presence of SO₂ and OH scavengers providing further support for the action of Criegee intermediates as atmospheric oxidants of SO₂.²³

Implementation of increased Criegee intermediate + SO₂ reaction rates in atmospheric models has been shown to improve model simulations of H₂SO₄ in forested regions in Finland and Germany,¹⁴ and global modelling has shown that while global production of H₂SO₄ increases by only 4%, there are increases of up to 100% in the boundary layer in tropical forests.¹⁵ Further modelling work has shown that reactions of Criegee intermediates with SO₂ can compete with OH + SO₂ in a number of regions, and that Criegee + SO₂ reactions may be the dominant removal mechanism for SO₂ in certain areas and are major contributors to sulfate aerosol formation on a regional scale.¹⁷ Air quality modelling over the U.S. displayed limited impacts of increased Criegee + SO₂ reaction rates on sulfate aerosol production in this region, but the impacts were shown to be highly dependent on the competition between Criegee + SO₂ and Criegee + H₂O, with a combination of increased Criegee + SO₂ and decreased Criegee + H₂O reaction rates leading to enhanced sulfate aerosol concentrations.¹⁶ However, such studies have largely been based on the low pressure data for CH₂OO + SO₂ reported by Welz et al.¹⁰ and there is considerable uncertainty regarding the upper limit for CH₂OO + H₂O.^2,17

Theoretical work has provided support for rapid reactions between Criegee intermediates and SO₂,^13,24 with reactions proceeding via the initial barrierless formation of a cyclic secondary ozonide, and has enabled prediction of potential effects of pressure.¹³ For CH₂OO + SO₂, it has been predicted that the reaction products at atmospheric pressure will be a mixture of HCHO + SO₃ (∼68%), formyl sulfinic ester (HC(O)OS(O)OH) (∼15%) and a singlet bisoxy diradical (CH₂(O)O) + SO₂ (∼17%).¹³ In contrast, reactions of larger Criegee intermediates, including CH₃CHOO, at ambient pressures are expected to result in production of stabilised secondary ozonide species, with little formation of SO₃, and therefore little impact on H₂SO₄ and sulfate aerosol.¹³ Investigation of the reaction products and pressure dependence of Criegee intermediate reactions is thus essential to the accurate determination of their atmospheric impacts.

The yield of CH₂OO Criegee intermediates following CH₂I₂ photolysis in O₂ was studied by Huang et al.,²⁵ and in our previous work,²⁶ as a function of pressure. Both investigations indicate that the initial reaction between CH₂I radicals and O₂(R2) produces a chemically activated species, CH₂IO₂^#, which decomposes at low pressures to produce CH₂OO + I (R2a), but is collisionally stabilised at higher pressures to produce the CH₂IO₂ peroxy radical (R2b).

CH₂I₂ + hν → CH₂I + I (R1)

CH₂I + O₂ → CH₂IO₂^# (R2)

CH₂IO₂^# → CH₂OO + I (R2a)

CH₂IO₂^# + M → CH₂IO₂ + M (R2b)

Our previous work²⁶ indicates a yield of ∼18% CH₂OO following photolysis of CH₂I₂ in air at 760 Torr, with recent results from Huang et al.²⁷ in reasonable agreement. This result has potential significance for modelling of atmospheric chemistry in iodine-rich regions,^28–31 and also indicates potential for pressure dependent studies of CH₂OO kinetics using photolysis of CH₂I₂ in O₂.

In this work, we report kinetics of CH₂OO reactions with SO₂, NO₂, NO, H₂O and CH₃CHO at pressures between 25 and 450 Torr at a temperature of 295 K, using photolysis of CH₂I₂–O₂–N₂ mixtures under pseudo-first-order conditions combined with monitoring of the HCHO reaction products by laser-induced fluorescence (LIF) spectroscopy, and, for the CH₂OO + SO₂ reaction at ∼1.5 Torr, direct monitoring of CH₂OO by photoionisation mass spectrometry (PIMS). We also report kinetics of the CH₂I + NO₂ reaction at pressures between 25 and 300 Torr at 295 K.

2. Experimental

2.1 Laser-induced fluorescence experiments

Apparatus and experimental procedures for the laser-induced fluorescence (LIF) experiments have been described elsewhere in detail,^26,32 therefore only a brief description is given here. Kinetics of CH₂OO reactions were studied by monitoring of HCHO reaction products by LIF spectroscopy. Radicals were generated by the laser flash photolysis of CH₂I₂–O₂–N₂ gas mixtures (R1 and R2) with the addition of excess co-reagent (NO₂, NO, SO₂, H₂O or CH₃CHO) to ensure pseudo-first-order conditions. Experiments to investigate CH₂I + NO₂ kinetics were performed in the absence of O₂, while those to investigate CH₂OO + NO₂ were performed using a limited range of NO₂ concentrations in order to avoid production of HCHO through the reaction of CH₂I with NO₂ (see Section 3.1), whilst maintaining pseudo-first-order conditions.

CH₂I₂ (Sigma-Aldrich, 99%) was used as a dilute gas in N₂ either by filling a glass bulb containing liquid CH₂I₂ with N₂ or by bubbling a slow flow of N₂ through liquid CH₂I₂. Reagents (NO, NO₂, SO₂, CH₃CHO) were prepared at known concentrations in N₂ and stored in glass bulbs. NO (BOC Special Gases, 99.5%) was purified prior to use by a series of freeze–pump–thaw cycles. CH₂I₂, CH₃CHO (Sigma-Aldrich, 99.5%), NO₂ (Sigma-Aldrich, 99.5%), SO₂ (Sigma-Aldrich, 99.9%), N₂ (BOC, 99.99%) and O₂ (BOC, 99.999%) were used as supplied. Water vapour was added to the gas mixture by bubbling a known flow of N₂ gas through a bubbler containing deionised water at a known temperature. Gases were mixed in a gas manifold and passed into a six-way cross reaction cell at known flow rates (determined by calibrated mass flow controllers). The pressure in the reaction cell was monitored by a capacitance manometer (MKS Instruments, 626A) and controlled by throttling the exit valve to the reaction cell. The total gas flow rate through the reaction cell was adjusted with total pressure to maintain an approximately constant gas residence time in the cell (∼0.1 s). All experiments were performed at T = (295 ± 2) K unless stated otherwise.

For experiments using NO₂, NO, CH₃CHO or H₂O as co-reagents, initiation of chemistry within the cell was achieved using an excimer laser (KrF, Tui ExciStar M) operating at λ = 248 nm with typical laser fluence in the range 30–80 mJ cm⁻². Experiments in which SO₂ was present as the co-reagent were performed at a photolysis wavelength of 355 nm (typical fluence ∼ 150 mJ cm⁻²), generated by frequency tripling the output of a Nd:YAG laser (Spectron Laser Systems) to avoid potential multi-photon photolysis of SO₂ at shorter wavelengths.^33–35

Production of HCHO was monitored by laser-induced fluorescence (LIF) of HCHO at λ ∼ 353.1 nm.³⁶ Approximately 2 to 4 mJ pulse⁻¹ of laser light at ∼353.1 nm was generated by a dye laser (Lambda Physik, FL3002) operating on DMQ/dioxirane dye and pumped by a 308 nm excimer laser generating ∼50 mJ pulse⁻¹ (XeCl, Lambda Physik LPX100). The output of the dye laser was passed through the reaction cell in an orthogonal axis to the 248 nm/355 nm photolysis laser output, with HCHO fluorescence detected in the visible region of the spectrum by a channel photomultiplier (CPM, Perkin-Elmer C1943P) orthogonal to both the photolysis laser and the LIF excitation laser beams. A Perspex filter was used to prevent scattered laser light from the photolysis laser and the LIF excitation laser reaching the CPM. The HCHO fluorescence signal was monitored as a function of time following photolysis of CH₂I₂ by varying the time delay between firing the photolysis laser and the LIF excitation laser through use of a delay generator (SRS DG535). Results from between 5 and 20 photolysis shots were typically averaged prior to analysis.

2.2 Photoionisation mass spectrometry experiments

Photoionisation mass spectrometry (PIMS) experiments were performed in this work to determine the kinetics of CH₂OO + SO₂ at low pressure (∼1.5 Torr) and 295 K by direct monitoring of CH₂OO in reactions performed under pseudo-first-order conditions. The PIMS apparatus has been described previously in detail^32,37,38 and only a brief description is given here. Gas mixtures of CH₂I₂–O₂–N₂ and CH₂I₂–O₂–N₂–SO₂ were prepared in a gas handling line, with reagents and reagent preparation as described above for the LIF experiments, and introduced to the steel reaction flow tube (10.5 mm internal diameter, 70 cm in length) via calibrated mass flow controllers. The pressure in the reaction flow tube was monitored by a capacitance manometer (MKS Instruments, 626A) and controlled by throttling the exit valve to the flow tube.

Chemistry was initiated by a pulsed excimer laser (Lambda Physik, Compex 205) at a wavelength of 248 nm, with typical fluence of ∼50 mJ cm⁻², through reactions (R1) and (R2). A representative sample from the reaction mixture effused into a high vacuum chamber (<10⁻⁵ Torr, maintained by diffusion and turbo pumps) via a 1 mm pinhole situated in the sidewall of the reaction flow tube. Components of the gas mixture were photoionised using 118 nm vacuum ultraviolet (VUV) laser light (typically 10¹¹ photons pulse⁻¹), generated by frequency tripling of the third harmonic of a Nd:YAG laser (Continuum Powerlite, 8010) in a Xe gas cell, and passed across the effusing gas flow within 2–3 mm of the sampling pinhole. VUV light of 118 nm (equivalent to 10.5 eV) is sufficiently energetic to ionise CH₂OO (threshold = 10.02 eV), but is below the threshold required to ionise other isomers at m/z = 46 (dioxirane, threshold = 10.82 eV; formic acid, threshold = 11.33 eV).¹⁰ Ions were sampled by the time of flight mass spectrometer (TOF-MS, Kore Technology Ltd), and detected by an electron multiplier. The ion signals were amplified and boxcar averaged on an oscilloscope and then stored on the control computer. The ion signals were monitored as a function of time following photolysis of CH₂I₂ by varying the time delay between the excimer laser and the Nd:YAG laser, used to generate the VUV radiation, through use of a delay generator (SRS DG35). These kinetic traces consisted of typically 200 time points, with typically between 10 and 25 shot averaging per time point.

3. Results and discussion

3.1 Photolysis of CH₂I₂–O₂–N₂ mixtures

Fig. 1 shows the HCHO fluorescence signal following photolysis of CH₂I₂–O₂–N₂ mixtures (i.e. in the absence of any additional co-reagent), resulting in production of HCHO through reactions (R1)–(R6):^26,32

CH₂I₂ + hν → CH₂I + I (R1)

CH₂I + O₂ → CH₂OO + I (R2a)

CH₂I + O₂ + M → CH₂IO₂ + M (R2b)

CH₂OO + I → HCHO + IO (R3)

CH₂IO₂ + I → CH₂IO + IO (R4)

CH₂IO₂ + CH₂IO₂ → 2CH₂IO + O₂ (R5)

CH₂IO → HCHO + I (R6)

Previous work in this laboratory²⁶ has shown that the yields of CH₂OO and CH₂IO₂ from (R2) are dependent on pressure, owing to initial formation of the excited species CH₂IO₂^#, which can either decompose to produce the CH₂OO Criegee intermediate and iodine atoms (R2a) or can be collisionally stabilised to produce the peroxy radical CH₂IO₂(R2b). Since subsequent reactions of both CH₂OO and CH₂IO₂ in the absence of any additional co-reagent result in production of HCHO, there is no change in the total HCHO yield as a function of pressure following photolysis of CH₂I₂–O₂–N₂ mixtures.

Fig. 1 HCHO fluorescence signals at 200 Torr following photolysis of CH₂I₂ in the presence of O₂ in the absence of any co-reagent (black open squares) and in the presence of NO₂ (red open circles). The fits to eqn (1) are shown by the solid lines, and give k_g′ = (460 ± 30) s⁻¹ in the absence of any additional co-reagent and k_g′ = (1490 ± 50) s⁻¹ in the presence of NO₂. The ratio of S₁ (eqn (1)) in the presence of NO₂ to that in the absence of NO₂ is 0.37.

Production of HCHO in reactions (R1)–(R6) can be approximated by eqn (1):^26,32

where S_HCHO,t is the HCHO signal at time t, S₀ is the height of the HCHO signal at time zero, S₁ is the maximum HCHO signal, k_g′ is the pseudo-first-order rate coefficient for HCHO growth, and k_loss is the rate coefficient representing the slow loss of HCHO from the detection region via diffusion. Although the HCHO growth through reactions (R1)–(R6) is not strictly first-order, our previous work²⁶ demonstrates that eqn (1) can faithfully reproduce the HCHO growth kinetics. In the presence of excess co-reagent (e.g. SO₂, NO₂) the kinetics of HCHO production from CH₂OO are under pseudo-first-order conditions. Fig. 1 shows the fits to HCHO production in the absence and presence of additional co-reagent, indicating the fidelity of the fit to the analytical equation.

In the absence of any additional co-reagent, the first-order rate coefficient approximating the production of HCHO, k_g′, was found to vary from ∼300 s⁻¹ to ∼3500 s⁻¹, depending on the concentration of CH₂I₂, and thus of I atoms, in the system, in keeping with the work of Welz et al.¹⁰ and Taatjes et al.¹¹ Some initial HCHO production was observed owing to multi-photon photolysis of CH₂I₂ and the subsequent rapid reaction of ³CH₂ with O₂, with S₀ typically no greater than 5–10% of S₁.^39–43

3.2 CH₂OO + SO₂

The reaction of CH₂OO with SO₂(R7) was investigated in separate experiments using the PIMS method to monitor CH₂OO and the LIF method to monitor HCHO production.

CH₂OO + SO₂ → HCHO + SO₃ (R7)

Experiments using the PIMS method were performed at a total pressure of 1.5 Torr. Fig. 2 shows a typical decay for CH₂OO observed in the presence of excess SO₂, with the pseudo-first-order rate coefficient for CH₂OO decay found by least-squares fitting to eqn (2):

where S_CH2OO,t is the CH₂OO ion signal at time t, S_max is the maximum CH₂OO ion signal, k′ is the pseudo-first-order rate coefficient for CH₂OO decay, and k_sampling is the rate coefficient representing the transport of molecules in the reactor to the ionisation region (∼30 000 s⁻¹, described in detail by Baeza-Romero et al.³⁸).

Fig. 2 CH₂OO ion signals at 1.5 Torr following photolysis of CH₂I₂–O₂–N₂ in the presence of SO₂, with the fit to eqn (2) (solid red line). For these data, k′ = (3310 ± 450) s⁻¹.

The bimolecular rate coefficient for CH₂OO + SO₂ (k₇) determined using the PIMS method at 1.5 Torr was (3.6 ± 0.5) × 10⁻¹¹ cm³ s⁻¹ (Fig. 3), similar to the value of (3.9 ± 0.7) × 10⁻¹¹ cm³ s⁻¹ at 4 Torr reported by Welz et al.¹⁰ and several orders of magnitude greater than the values typically used in atmospheric models.

Fig. 3 (a) Pseudo-first-order rate coefficients (k′) at 1.5 Torr, derived from fits to eqn (2), for the decay of the CH₂OO ion signal (m/z = 46, ionised using VUV radiation at 118 nm) following photolysis of CH₂I₂–O₂–N₂ in the presence of SO₂. Error bars are 1σ. The fit to the data (shown in red) gives the bimolecular rate coefficient for CHOO + SO₂ (k₇); (b) pseudo-first-order rate coefficients (k_g1′) for the rapid HCHO production at 250 Torr following photolysis of CH₂I₂–O₂–N₂ in the presence of SO₂ derived from fits to eqn (3). Error bars are 1σ. The fit to the data (shown in red) gives the bimolecular rate coefficient for CHOO + SO₂ (k₇).

The LIF experiments monitoring HCHO production from CH₂OO + SO₂ were performed over the pressure range 50–450 Torr, with SO₂ concentrations in the range 2.4 × 10¹⁴ to 1.6 × 10¹⁵ cm⁻³. The HCHO growth (Fig. 4) was observed to display biexponential behaviour, with no decrease in the total HCHO yield compared to experiments performed in the absence of any co-reagent, indicating complete titration of both CH₂OO and CH₂IO₂ to HCHO. Kinetic parameters were determined by fitting to eqn (3):

where S_HCHO,t is the HCHO signal at time t, S₀ is the height of the HCHO signal at time zero, S₁ is the maximum HCHO signal, k_g1′ is the pseudo-first-order rate coefficient for the fast HCHO growth, k_g2′ is the pseudo-first-order rate coefficient for the slower HCHO growth, f is the fractional contribution of the fast growth process to the total HCHO yield (hence (1 − f) is the fractional contribution of the slower growth process to the total HCHO yield), and k_loss is the rate coefficient representing the slow loss of HCHO from the detection region via diffusion. For the SO₂ experiments (conducted using a photolysis wavelength of 355 nm) there was no contribution from S₀ (i.e. S₀ = 0).

Fig. 4 HCHO fluorescence signals at 250 Torr following photolysis of CH₂I₂–O₂–N₂ in the presence of SO₂, with the fit to eqn (3) (solid red lines). The inset panel shows the evolution of the signal to longer times. For these data, k_g1′ = (45 500 ± 2240) s⁻¹; k_g2′ = (3580 ± 280) s⁻¹; k_loss = (40 ± 9) s⁻¹; f = (0.49 ± 0.01); S₁ = (0.43 ± 0.01).

The initial fast growth of HCHO displayed a linear dependence on [SO₂], while the slower growth was independent of [SO₂] and at a similar rate to the observed HCHO production in the absence of any additional co-reagent. The yields of HCHO from the faster growth process were consistent with production from CH₂OO + SO₂, while those from the slower process were consistent with production from reactions of CH₂IO₂ (i.e.reactions (R4)–(R6)). We thus determine k₇ from linear fits of k_g1′ (eqn (3)) against [SO₂]. The validity of describing the system using eqn (3) is discussed in our previous work.²⁶

Fig. 5 and Table 1 show the values of k₇ as a function of pressure. No significant dependence of k₇ on pressure was observed, with an average value of (3.42 ± 0.42) × 10⁻¹¹ cm³ s⁻¹ for all experiments (PIMS and LIF) described in this work (all errors are 1σ unless stated otherwise). Moreover, there is no significant change in the HCHO yield from the reaction of CH₂OO with SO₂ as a function of pressure, indicating there is little stabilisation of reaction products. These results are consistent with the low pressure results obtained by Welz et al.¹⁰ and theoretical work by Vereecken et al.,¹³ and support arguments for an increased role of CH₂OO + SO₂ in the atmosphere. Taatjes et al.¹² have also shown that the reaction of the C₂ Criegee intermediate, CH₃CHOO, with SO₂ at a pressure of 4 Torr is also significantly faster than previously expected, potentially indicating an increased role for CH₃CHOO + SO₂ in the atmosphere. However, theoretical calculations predict that reactions of larger Criegee intermediates will exhibit pressure dependence,¹³ and that production of SO₃ in reactions of larger Criegee intermediates at atmospheric pressures is unlikely owing to stabilisation of SO₂–Criegee intermediate complexes to produce secondary ozonide species, thus reducing the impacts of SO₂ + Criegee intermediate reactions on H₂SO₄ and sulfate aerosol production.¹³ Field observations and laboratory studies by Mauldin et al.²³ indicate that larger Criegee intermediates, such as those produced in the ozonolysis of monoterpenes, do impact on atmospheric concentrations of H₂SO₄ through oxidation of SO₂, but that the impacts may not be as great as those reported for CH₂OO, potentially owing to stabilisation of reaction products. Further work is thus required to investigate the effects of pressure on the reactions of larger Criegee intermediates. Moreover, modelled impacts of increases in the rates of Criegee intermediate reactions with SO₂ are highly dependent on the competition with rates of Criegee intermediate reactions with water vapour. We thus investigate CH₂OO + H₂O in Section 3.6.

Fig. 5 Bimolecular rate coefficients for CH₂OO + SO₂ (k₇) as a function of pressure. Error bars are 1σ. The plot includes results from the PIMS experiments (at 1.5 Torr) and the LIF experiments (pressures ≥ 50 Torr). The data point shown by the red open circle is that determined by Welz et al.¹⁰

Table 1 Bimolecular rate coefficients for CH₂OO + SO₂ (k₇) as a function of pressure. Errors are 1σ

Pressure (Torr)	k 7 (10^−11 cm^3 s^−1)	Ref.
a Data at 1.5 Torr are from the PIMS experiments.
1.5^a	3.6 ± 0.5	This work
4	3.9 ± 0.7	Welz et al.^10
50	3.04 ± 0.66	This work
100	3.11 ± 0.57	This work
150	3.17 ± 0.34	This work
250	3.68 ± 0.21	This work
350	3.19 ± 0.53	This work
450	4.18 ± 0.30	This work

---

3.3 CH₂I + NO₂

Production of HCHO following photolysis of CH₂I₂–NO₂–N₂ mixtures was examined as a function of pressure to facilitate assessment of the competition between CH₂I + O₂(R2) and CH₂I + NO₂(R8) in CH₂OO + NO₂ experiments (Section 3.4).

CH₂I + NO₂ → HCHO + products (R8)

The production of HCHO could be described by eqn (1) (above), where k_g′ = k₈[NO₂], with concentrations of NO₂ between 1 × 10¹⁴ and 9 × 10¹⁴ cm⁻³. Pseudo-first-order rate coefficients (k_g′) were in the range ∼5000 to 45 000 s⁻¹, and typically large compared to the rate coefficients describing HCHO production in the absence of any additional co-reagent (Section 3.1). The bimolecular rate coefficient k₈ was determined from plots of k_g′ against [NO₂] at each pressure (Fig. S1, ESI†), and was found to increase with increasing pressure (Fig. S2 and Table S1, ESI†), with a corresponding decrease in the HCHO yield as the pressure was increased (Fig. S3, ESI†).

A previous investigation of CH₂I + NO₂ at pressures of 2 to 5 Torr gave a value of k₈ = (2.2 ± 0.1) × 10⁻¹¹ cm³ s⁻¹.⁴⁴ Results of this work show k₈ to be (2.56 ± 0.17) × 10⁻¹¹ cm³ s⁻¹ at 50 Torr, increasing to (5.07 ± 0.28) × 10⁻¹¹ cm³ s⁻¹ at 300 Torr.

The rate coefficient for reaction of CH₂I radicals with O₂(R2), has been shown previously to be ∼1.6 × 10⁻¹² cm³ s⁻¹.^45,46 Experiments to investigate HCHO production in the reaction of CH₂OO (produced by CH₂I + O₂) with NO₂ must therefore be conducted at sufficiently high [O₂] to avoid complications owing to HCHO production from CH₂I + NO₂.

3.4 CH₂OO + NO₂

Experiments to investigate CH₂OO + NO₂(R9) kinetics were performed with sufficient NO₂ concentrations (1.0 × 10¹⁴ to 1.4 × 10¹⁵ cm⁻³) to ensure pseudo-first-order conditions for CH₂OO loss whilst also ensuring that k₂[O₂] > k₈[NO₂] at all times to avoid potential complications owing to HCHO production through CH₂I + NO₂.

CH₂I₂ + hν → CH₂I + I (R1)

CH₂I + O₂ → CH₂OO + I (R2a)

CH₂I + NO₂ → HCHO + products (R8)

CH₂OO + NO₂ → HCHO + NO₃ (R9)

Fig. 1 shows the evolution of the HCHO signal following photolysis of CH₂I₂–O₂–N₂–NO₂ mixtures. Experiments in which NO₂ was used as a co-reagent resulted in a decrease in the total HCHO yield when compared to experiments performed in the absence of any co-reagent. We attribute this to the formation of the peroxy nitrate species CH₂IO₂NO₂ which inhibits formation of HCHO through reactions (R4)–(R6).

Experiments performed at 273 K to increase the lifetime of CH₂IO₂NO₂ with respect to dissociation to CH₂IO₂NO₂ did not result in any significant decrease in the HCHO yield compared to equivalent experiments at 295 K, indicating that the CH₂IO₂NO₂ lifetime at 295 K is sufficiently long to minimise production of HCHO from CH₂IO₂. Thus, while there is a small contribution to the HCHO signal owing to rapid chemistry following multi-photon photolysis of CH₂I₂, the growth of HCHO observed following photolysis of CH₂I₂–O₂–N₂–NO₂ mixtures can be attributed to CH₂OO + NO₂(R9) exclusively.

The pseudo-first-order rate coefficient for the reaction of CH₂OO with NO₂ was determined by least-squares fitting to eqn (1), with k_g′ = k₉[NO₂]. The bimolecular rate coefficient for CH₂OO + NO₂ (k₉) was subsequently determined from plots of k_g′ against [NO₂], as shown in Fig. 6. Fits to experimental data using the numerical integration package Kintecus⁴⁷ to determine k₉, detailed in the ESI,† gave results within 10% of those obtained using the analytical expression (eqn (1)).

Fig. 6 Pseudo-first-order rate coefficients (k_g′) for HCHO production at 50 Torr, derived from fits to eqn (1), following photolysis of CH₂I₂–O₂–N₂ in the presence of NO₂. Error bars are 1σ. The fit to the data (shown in red) gives the bimolecular rate coefficient for CH₂OO + NO₂ (k₉).

Values for k₉ as a function of pressure are shown in Fig. 7 and Table 2. No significant dependence of k₉ on total pressure was observed over the pressure range investigated (25 to 300 Torr), with an average value of k₉ = (1.5 ± 0.5) × 10⁻¹² cm³ s⁻¹. Errors in k₉ include the 1σ errors in the fits to the bimolecular plots at each pressure and an error of ±10% to account for any differences between fits using the analytical expression and those obtained by numerical integration (see ESI†).

Fig. 7 Bimolecular rate coefficients for CH₂OO + NO₂ (k₉) as a function of pressure. Error bars are 1σ. The data point shown by the red open circle is that determined by Welz et al.¹⁰

Table 2 Bimolecular rate coefficients for CH₂OO + NO₂ (k₉) as a function of pressure. Errors include the 1σ in the fits to the bimolecular plots and an error of ±10% to account for any differences between the fits using the analytical expression and those obtained by numerical integration

Pressure (Torr)	k 9 (10^−12 cm^3 s^−1)	Ref.
a Measured using N2 as the bath gas. b Measured using O2 as the bath gas.
4	7^+3−2	Welz et al.^10
25^a	1.70 ± 0.38	This work
50^a	1.04 ± 0.27	This work
50^b	0.94 ± 0.16	This work
75^a	1.69 ± 0.28	This work
100^a	1.38 ± 0.33	This work
150^a	1.19 ± 0.30	This work
200^a	2.00 ± 0.56	This work
250^a	0.96 ± 0.29	This work
300^a	2.53 ± 0.47	This work

---

Yields of HCHO in the presence of NO₂, determined relative to experiments performed in the absence of NO₂ (i.e. production through reactions (R3)–(R6)), were consistent with the yields of CH₂OO determined in our previous work²⁶ (Fig. 8). This result demonstrates that ∼100% of CH₂OO is titrated to HCHO by CH₂OO + NO₂, indicating a lack of pressure dependence in k₉, and that there is insignificant HCHO production from CH₂IO₂ in the presence of NO₂. Recent measurements by Ouyang et al.²¹ have demonstrated the production of NO₃ at atmospheric pressure from the reaction of CH₂OO with NO₂, thus also suggesting little stabilisation of reaction products to a secondary ozonide species in this system.

Fig. 8 Stern–Volmer plot showing (inverse) yields of CH₂OO as a function of pressure from the reaction of CH₂I with O₂. Results from our previous work are shown for experiments monitoring iodine atom production in the system (black squares), and monitoring of HCHO production in experiments with SO₂ (blue triangles) and NO (red circles), with the best fit line (red). Yields of HCHO from the reaction of CH₂OO with NO₂ (this work, green diamonds), determined relative to the HCHO yields in the absence of NO₂ (i.e. through reactions (R3)–(R6)), suggest that there is 100% titration of CH₂OO to HCHO in the presence of NO₂ at all pressures (i.e. there is no stabilisation of reaction products), and that there is little production of HCHO from CH₂IO₂ in the system. The fit to our previous work (comprising data from the I atom, NO and SO₂ experiments) gives an intercept of 1.10 ± 0.23 and a slope of (1.90 ± 0.22) × 10⁻¹⁹ cm³. The NO₂ experiments give an intercept of 1.05 ± 0.12 and a slope of (1.70 ± 0.18) × 10⁻¹⁹ cm³.

No significant difference in k₉ or in yields of HCHO were observed between experiments performed in O₂ bath gas and N₂ bath gas (results shown in Table 2), providing further evidence for similar quenching of the nascent excited CH₂IO₂^# species (produced in (R2)) by O₂ and N₂, as discussed in our previous work.²⁶

Results for k₉ obtained in this work, while lower than those reported by Welz et al.,¹⁰ are on the same order of magnitude, and demonstrate a significantly faster reaction between CH₂OO and NO₂ than suggested by previous indirect measurements.¹

3.5 CH₂OO + NO

Production of HCHO following photolysis of CH₂I₂–O₂–N₂ mixtures in the presence of excess NO (3.6 × 10¹⁴ to 1.7 × 10¹⁵ cm⁻³) exhibits biexponential growth, as shown in Fig. 9, similar to experiments with SO₂. Again, no decrease in the total HCHO yield compared to experiments performed in the absence of any co-reagent, indicating complete titration of both CH₂OO and CH₂IO₂ to HCHO. Kinetic parameters for the processes contributing to HCHO production were obtained by fitting to eqn (3) (above).

Fig. 9 HCHO fluorescence signals at 250 Torr following photolysis of CH₂I₂–O₂–N₂ in the presence of NO, with the fit to eqn (3) (solid red lines). The inset panel shows the evolution of the signal to longer times. For these data, k_g1′ = (24 800 ± 1400) s⁻¹; k_g2′ = (2660 ± 320) s⁻¹; k_loss = (10 ± 2) s⁻¹; f = (0.70 ± 0.02); S₁ = (1.33 ± 0.01).

The rate coefficient describing the fast HCHO growth process, k_g1′, was observed to increase linearly with increasing [NO], with the slope of a plot of k_g1′ against [NO] giving a bimolecular rate coefficient of (1.07 ± 0.06) × 10⁻¹¹ cm³ s⁻¹ at 250 Torr (Fig. S4, ESI†). The rate coefficient describing the slower HCHO growth, k_g2′, was found to be independent of [NO], and similar to the rate coefficient for HCHO production obtained in the absence of NO. Reactions of peroxy radicals (RO₂) with NO are well established, and are typically on the order of 10⁻¹² to 10⁻¹¹ cm³ s⁻¹,^48,49 with a rate coefficient for CH₃O₂ + NO of 7.2 × 10⁻¹² cm³ s⁻¹,⁴⁹ while Welz et al.¹⁰ reported an upper limit of 6 × 10⁻¹⁴ cm³ s⁻¹ for the rate coefficient for CH₂OO with NO. Thus, in contrast to the experiments with SO₂, we attribute the fast HCHO growth to the rapid decomposition of CH₂IO (R6), produced in the reaction of CH₂IO₂ with NO (R10) and assign k₁₀ = (1.07 ± 0.06) × 10⁻¹¹ cm³ s⁻¹ at 250 Torr.

CH₂IO₂ + NO → CH₂IO + NO₂ (R10)

CH₂IO → HCHO + I (R6)

The slower HCHO growth thus contains contributions from CH₂OO + I (R3) and potentially CH₂OO + NO (R11). In the absence of NO, production of HCHO was observed with a pseudo-first-order rate coefficient of 1860 ± 100 s⁻¹ (eqn (1)). On addition of up to 1.7 × 10¹⁵ cm⁻³ NO, the average value for the rate coefficient describing the slow HCHO growth (k_g2′ in eqn (3)) was 1800 ± 340 s⁻¹. Any potential influence of NO on the observed rates of HCHO production is assumed to be within the error of the experiment, and we thus place an upper limit of 2 × 10⁻¹³ cm³ s⁻¹ on the rate coefficient for reaction of CH₂OO + NO (k₁₁).

CH₂OO + NO → HCHO + NO₂ (R11)

The upper limit for k₁₁ determined here is higher than that reported by Welz et al. (k₁₁ < 6 × 10⁻¹⁴ cm³ s⁻¹), owing to increased uncertainties associated with the biexponential fit, relatively low concentrations of NO, and higher concentrations of CH₂I₂ used in these experiments compared to those performed by Welz et al., which lead to increased iodine atom concentrations in this work and thus increased rates of HCHO production through CH₂OO + I (R3). In subsequent experiments (notably those used to investigate the kinetics of CH₂OO + H₂O) lower CH₂I₂ concentrations were used by changing the delivery method for CH₂I₂. There are also additional uncertainties in the rate coefficients for reactions with NO owing to the potential for production of NO₂ in the gas lines leading to the reaction cell through oxidation of NO by O₂ (the gas mixture has a residence time of ∼1 s in the gas lines leading from the mixing line to the reaction cell), leading to the potential for contributions to the observed HCHO growth from reactions involving NO₂.

3.6 CH₂OO + H₂O

Welz et al. did not observe any change in the rate of CH₂OO decay on addition of water vapour to the system, and reported an upper limit of 4 × 10⁻¹⁵ cm³ s⁻¹ for the rate coefficient for reaction of CH₂OO with H₂O (R12):

CH₂OO + H₂O → HCHO + H₂O₂ (R12)

Similarly to the results of Welz et al., the addition of water vapour to the LIF experiments in this work did not result in any significant change to the rate of HCHO production. The total HCHO yield was also unaffected by the presence of water vapour, indicating complete titration of CH₂OO and CH₂IO₂ to HCHO through reactions (R3)–(R6). Fig. 10 shows the HCHO fluorescence signals following photolysis of CH₂I₂–O₂–N₂ in the absence and presence of water vapour. While the HCHO signal is reduced in the presence of water vapour, there is no change in the kinetics and the reduction in signal is attributed to increased fluorescence quenching by water vapour.

Fig. 10 HCHO fluorescence signals at 200 Torr following photolysis of CH₂I₂–O₂–N₂ in the absence (black open circles) and presence of water vapour (red open triangles), with the fits to eqn (1) (solid lines). The differences in the amplitude of the signal result from the quenching of the fluorescence signal by H₂O. For these data, k_g′ = (41 ± 15) s⁻¹ in the absence of water vapour and k_g′ = (52 ± 13) s⁻¹ in the presence of water vapour.

At 200 Torr the pseudo-first-order rate coefficient for HCHO production was determined to be 41 ± 15 s⁻¹ by fitting to eqn (1), and was lower than the typical values reported in Section 3.1 as a result of lower concentrations of CH₂I₂ to reduce the rate of HCHO production through radical–radical reactions in the absence of water vapour. On addition of up to 1.7 × 10¹⁷ cm⁻³ water vapour to the system, a value of 52 ± 13 s⁻¹ was obtained, with no obvious dependence on the concentration of water vapour added. Owing to the higher total pressures used in this work, enabling the addition of a higher number density of water vapour to the system compared to the low pressure experiments by Welz et al., we are able to place an upper limit of 9 × 10⁻¹⁷ cm³ s⁻¹ on k₁₂ at 295 K by assuming any influence of water vapour is within the error of the experiment. Ouyang et al.²¹ have reported a value for k₁₂ of (2.5 ± 1) × 10⁻¹⁷ cm³ s⁻¹ at 760 Torr, determined in a relative rate experiment monitoring NO₃ production and using the absolute value for k₉ (CH₂OO + NO₂) reported by Welz et al.¹⁰ (7 × 10⁻¹² cm³ s⁻¹). Using the relative rate coefficient ratio reported by Ouyang et al., with the value for k₉ determined in this work (1.5 × 10⁻¹² cm³ s⁻¹), a value of k₁₂ = 5.4 × 10⁻¹⁸ cm³ s⁻¹ can be obtained.

Modelling studies investigating the impacts of CH₂OO chemistry on the atmospheric oxidation of SO₂ may therefore be underestimating the effects of increasing the rate coefficient for CH₂OO + SO₂ owing to overestimation of the competition with CH₂OO + H₂O, resulting in more significant impacts on atmospheric production of H₂SO₄ and sulfate aerosol than indicated thus far. However, Taatjes et al.¹² have shown that the anti-CH₃CHOO Criegee intermediate does react with water vapour (k = (1.0 ± 0.4) × 10⁻¹⁴ cm³ s⁻¹), and the lack of reaction between CH₂OO and water vapour may not be representative of all Criegee intermediates. Modelling of Criegee chemistry in forested regions in Finland and Germany has indicated that concentration of the CH₂OO Criegee intermediate is only ∼20–33% of the concentrations of larger Criegee intermediates derived from monoterpenes,¹⁴ with global modelling indicating that the production rate of CH₂OO comprises ∼40% of the total global production rate of all Criegee intermediates.¹⁵ The chemistry of larger Criegee intermediates warrants further attention.

3.7 CH₂OO + CH₃CHO

The reactions of Criegee intermediates with carbonyl compounds are of interest not only for their potential atmospheric relevance, but also to facilitate the use of carbonyl compounds as scavengers of Criegee intermediates in alkene ozonolysis experiments, enabling the determination of product yields of ozonolysis reactions.

Horie et al.⁵⁰ studied the relative rates of CH₂OO reactions with CH₃CHO (R13) and CF₃COCF₃(R14) at 730 Torr in synthetic air using FT-IR spectroscopy to monitor the decay of CF₃COCF₃ and the production of the secondary ozonide propene ozonide (methyl-1,2,4-trioxolane) from the reaction with CH₃CHO, and found the reaction with CF₃COCF₃ to be 13 times faster than that with CH₃CHO.

CH₂OO + CH₃CHO → products (R13)

CH₂OO + CF₃COCF₃ → products (R14)

Secondary ozonide products were observed by Horie et al. for both (R13) and (R14) at 730 Torr, while photoionisation mass spectrometry experiments by Taatjes et al.¹¹ at 4 Torr observed a secondary ozonide product for (R14) but not for (R13). Absolute rate coefficients for CH₂OO + CH₃CHO and CH₂OO + CF₃COCF₃ were measured by Taatjes et al.¹¹ at 4 Torr in He by direct monitoring of CH₂OO, with results indicating the reaction with CF₃COCF₃ to be ∼32 times faster than that with CH₃CHO and k₁₃ = (9.4 ± 0.7) × 10⁻¹³ cm³ s⁻¹ at 4 Torr. As discussed by Taatjes et al.,¹¹ the differences between the results of Horie et al. and Taatjes et al. may arise from differences in the fall-off behaviour of the two reactions, indicating pressure dependence of one or both of the reactions over the range of pressures investigated. Differences in product observations between the two studies also suggest pressure dependence in k₁₃. In the low pressure experiments, Taatjes et al. do not observe formation of secondary ozonide products. At 730 Torr, propene ozonide was observed as the major product of (R13), indicating collisional stabilisation of the nascent secondary ozonide at high pressures. Recent theoretical work²² has investigated the potential energy surface for the reaction of CH₂OO with CH₃CHO, and supports the observed pressure dependence of the reaction. Reaction products are predicted to be collisionally stabilised to a secondary ozonide (SOZ) species, with significant production of the SOZ at atmospheric pressure (760 Torr) and the SOZ dominating the reaction products at pressures above 1000 Torr.

Pressure dependent kinetics are expected to be typical for reactions of larger Criegee intermediates with atmospherically relevant species, including SO₂, and investigation of the CH₂OO + CH₃CHO system may therefore provide insight to the behaviour of other Criegee intermediates.

In this work, we investigate HCHO production from CH₂OO + CH₃CHO (R13) at total pressures between 25 and 300 Torr and concentrations of CH₃CHO in the range 2 × 10¹⁴ to 1 × 10¹⁵ cm⁻³. Production of HCHO displayed single exponential growth, and the HCHO fluorescence signal was fitted to eqn (1) (Fig. 11). Fig. 12 shows the bimolecular plot used to determine k₁₃ at 25 Torr, giving k₁₃ = (1.48 ± 0.04) × 10⁻¹² cm³ s⁻¹ at 25 Torr. The HCHO yield from (R13) (corrected for any HCHO production from CH₂IO₂ in reactions (R4)–(R6) using the results of our previous work) was observed to decrease with increasing pressure, indicating stabilisation of the CH₂OO + CH₃CHO reaction product at higher pressures (R13b) and pressure dependence in k₁₃.

CH₂OO + CH₃CHO → CH₂OO–CH₃CHO^#

CH₂OO–CH₃CHO^# → HCHO + CH₃C(O)OH (R13a)

CH₂OO–CH₃CHO^# + M → propene ozonide + M (R13b)

Fig. 11 HCHO fluorescence signals at 25 Torr following photolysis of CH₂I₂–O₂–N₂ in the presence of CH₃CHO, with the fit to eqn (1) (solid red line). For these data, k_g′ = (2040 ± 120) s⁻¹.

Fig. 13 shows the Stern–Volmer plot for HCHO yields from (R13), giving an intercept of 1.19 ± 0.39 and slope (k_13b/k_13a) of (1.09 ± 0.08) × 10⁻¹⁸ cm³. Using an intercept of 1, at 4 Torr we estimate a yield of HCHO of 88%, with a yield of 4% at 730 Torr, reconciling the results of Taatjes et al.¹¹ and Horie et al.⁵⁰ and in agreement with theoretical work of Jalan et al.²²

Fig. 12 Pseudo-first-order rate coefficients (k_g′) for HCHO production at 25 Torr, derived from fits to eqn (1), following photolysis of CH₂I₂–O₂–N₂ in the presence of CH₃CHO. Error bars are 1σ. The fit to the data (shown in red) gives the bimolecular rate coefficient for CH₂OO + CH₃CHO (k₁₃).

Fig. 13 Stern–Volmer analysis for HCHO yields from CH₂OO + CH₃CHO (R13) (corrected for HCHO production from CH₂IO₂ chemistry) as a function of total pressure, with the fit to the data (red). Error bars are 1σ.

Owing to the decrease in HCHO yield with increasing pressure, assignment of the kinetics of (R13) at pressures above 25 Torr is challenging. Using the results of Taatjes et al.¹¹ at 4 Torr (k₁₃ = (9.5 ± 0.7) × 10⁻¹³ cm³ s⁻¹), together with those determined here at 25 Torr (k₁₃ = (1.48 ± 0.04) × 10⁻¹² cm³ s⁻¹), 50 Torr (∼2.2 × 10⁻¹² cm³ s⁻¹) and the determination of k_13b/k_13a from the Stern–Volmer plot ((1.09 ± 0.08) × 10⁻¹⁸ cm³), we estimate a low pressure limit (k_13,0) of ∼1.6 × 10⁻²⁹ cm⁶ s⁻¹ and a high pressure limit (k_13,∞) of ∼1.7 × 10⁻¹² cm³ s⁻¹ (see ESI†).

4. Conclusions

Reactions of the CH₂OO Criegee intermediate with NO₂, NO, SO₂, H₂O and CH₃CHO have been investigated over a range of pressures. The reactions of CH₂OO with NO₂, SO₂ and CH₃CHO are rapid, in agreement with recent measurements by Welz et al.¹⁰ and Taatjes et al.¹¹ but in contrast to recommendations for atmospheric modelling based on indirect measurements. Rate coefficients for reactions of CH₂OO with NO₂ and SO₂ are essentially independent of pressure over the pressure ranges studied in this work. The rate coefficient for CH₂OO + CH₃CHO is pressure dependent, with stabilisation to form the secondary ozonide reaction products at high pressures.

We observe no evidence for reactions of CH₂OO with NO or H₂O under the conditions employed in this work, and place upper limits on rate coefficients for these reactions of 2 × 10⁻¹³ cm³ s⁻¹ and 9 × 10⁻¹⁷ cm³ s⁻¹, respectively. The upper limit for the rate coefficient for CH₂OO + H₂O is significantly lower than has been reported previously. Earlier assessments^2,14,15,17 of the impacts of increased reaction rates for CH₂OO + SO₂ and CH₂OO + NO₂ will therefore be lower limits owing to overestimation of the impacts of CH₂OO + H₂O.

Acknowledgements

The authors are grateful to the National Centre for Atmospheric Science (NCAS) and the Engineering and Physical Sciences Research Council (EPSRC, grant reference EP/J010871/1) for funding.

References

1. D. Johnson and G. Marston, Chem. Soc. Rev., 2008, 37, 699–716.
2. C. A. Taatjes, D. E. Shallcross and C. Percival, Phys. Chem. Chem. Phys., 2013 10.1039/C3CP52842A.
3. N. M. Donahue, G. T. Drozd, S. A. Epstein, A. A. Presto and J. H. Kroll, Phys. Chem. Chem. Phys., 2011, 13, 10848–10857.
4. M. S. Alam, M. Camredon, A. R. Rickard, T. Carr, K. P. Wyche, K. E. Hornsby, P. S. Monks and W. J. Bloss, Phys. Chem. Chem. Phys., 2011, 12, 11002–11015.
5. L. Vereecken, Science, 2013, 340, 154–155.
6. T. L. Malkin, A. Goddard, D. E. Heard and P. W. Seakins, Atmos. Chem. Phys., 2010, 10, 1441–1459.
7. D. E. Heard, L. J. Carpenter, D. J. Creasey, J. R. Hopkins, J. D. Lee, A. C. Lewis, M. J. Pilling, P. W. Seakins, N. Carslaw and K. M. Emmerson, Geophys. Res. Lett., 2004, 31, L18112.
8. R. M. Harrison, J. Yin, R. M. Tilling, X. Cai, P. W. Seakins, J. R. Hopkins, D. L. Lansley, A. C. Lewis, M. C. Hunter, D. E. Heard, L. J. Carpenter, D. J. Creasey, J. D. Lee, M. J. Pilling, N. Carslaw, K. M. Emmerson, A. Redington, R. G. Derwent, D. Ryall, G. Mills and S. A. Penkett, Sci. Total Environ., 2006, 360, 5–25.
9. D. Stone, L. K. Whalley and D. E. Heard, Chem. Soc. Rev., 2012, 41, 6348–6404.
10. 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.
11. C. A. Taatjes, O. Welz, A. J. Eskola, J. D. Savee, D. L. Osborn, E. P. F. Lee, J. M. Dyke, D. W. K. Mok, D. E. Shallcross and C. J. Percival, Phys. Chem. Chem. Phys., 2012, 14, 10391–10400.
12. C. A. Taatjes, O. Welz, A. J. Eskola, J. D. Savee, A. M. Scheer, D. E. Shallcross, B. Rotavera, E. P. F. Lee, J. M. Dyke, D. K. W. Mok, D. L. Osborn and C. J. Percival, Science, 2013, 340, 177–180.
13. L. Vereecken, H. Harder and A. Novelli, Phys. Chem. Chem. Phys., 2012, 14, 14682–14695.
14. M. Boy, D. Mogensen, S. Smolander, L. Zhou, T. Nieminen, P. Paasonen, C. Plass-Duelmer, M. Sipila, T. Petaja, L. Mauldin, H. Berresheim and M. Kulmala, Atmos. Chem. Phys., 2013, 13, 3865–3879.
15. J. R. Pierce, M. J. Evans, C. E. Scott, S. D. D'Andrea, D. K. Farmer, E. Swietlicki and D. V. Spracklen, Atmos. Chem. Phys., 2013, 13, 3163–3176.
16. G. Sarwar, K. Fahey, R. Kwok, R. C. Gilliam, S. J. Roselle, R. Mathur, J. Xue, J. Yu and W. P. L. Carter, Atmos. Environ., 2013, 68, 186–197.
17. C. J. Percival, O. Welz, A. J. Eskola, J. D. Savee, D. L. Osborn, D. O. Topping, D. Lowe, S. R. Utembe, A. Bacak, G. McFiggans, M. C. Cooke, P. Xiao, A. T. Archibald, M. E. Jenkin, R. G. Derwent, I. Riipinen, D. W. K. Mok, E. P. F. Lee, J. M. Dyke, C. A. Taatjes and D. E. Shallcross, Faraday Discuss., 2013 10.1039/C3FD00048F.
18. 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.
19. Y.-T. Su, Y.-H. Huang, H. A. Witek and Y.-P. Lee, Science, 2013, 340, 174–176.
20. J. M. Beames, F. Liu, L. Lu and M. I. Lester, J. Am. Chem. Soc., 2012, 134, 20045–20048.
21. B. Ouyang, M. W. McLeod, R. L. Jones and W. J. Bloss, Phys. Chem. Chem. Phys., 2013, 15, 17070–17075.
22. A. Jalan, J. W. Allen and W. H. Green, Phys. Chem. Chem. Phys., 2013, 15, 16841–16852.
23. R. L. Mauldin, III, T. Berndt, M. Sipilae, P. Paasonen, T. Petaja, S. Kim, T. Kurten, F. Stratmann, V. M. Kerminen and M. Kulmala, Nature, 2012, 488, 193–197.
24. T. Kurten, J. R. Lane, S. Jorgensen and H. G. Kjaergaard, J. Phys. Chem. A, 2011, 115, 8669–8681.
25. H. Huang, A. J. Eskola and C. A. Taatjes, J. Phys. Chem. Lett., 2012, 3, 3399–3403.
26. D. Stone, M. Blitz, L. Daubney, T. Ingham and P. Seakins, Phys. Chem. Chem. Phys., 2013, 15, 19119–19124.
27. H. Huang, B. Rotavera, A. J. Eskola and C. A. Taatjes, J. Phys. Chem. Lett., 2013, 4, 3824.
28. C. M. Roehl, J. B. Burkholder, G. K. Moortgat, A. R. Ravishankara and P. Crutzen, J. Geophys. Res.: Atmos., 1997, 102, 12819–12829.
29. L. J. Carpenter, W. T. Sturges, S. A. Penkett, P. S. Liss, B. Alicke, K. Hebestreit and U. Platt, J. Geophys. Res., [Atmos.], 1999, 104, 1679–1689.
30. L. J. Carpenter, Chem. Rev., 2003, 103, 4953–4962.
31. L. J. Carpenter, S. D. Archer and R. Beale, Chem. Soc. Rev., 2012, 41, 6473–6506.
32. T. J. Gravestock, M. A. Blitz, W. J. Bloss and D. E. Heard, ChemPhysChem, 2010, 11, 3928–3941.
33. R. A. Cox, J. Phys. Chem., 1972, 76, 814–820.
34. J. L. Jourdain, G. Lebras and J. Combourieu, Int. J. Chem. Kinet., 1979, 11, 569–577.
35. K. J. Hughes, M. A. Blitz, M. J. Pilling and S. H. Robertson, Proc. Combust. Inst., 2002, 29, 2431–2437.
36. D. T. Co, T. F. Hanisco, J. G. Anderson and F. N. Keutsch, J. Phys. Chem. A, 2005, 109, 10675–10682.
37. M. A. Blitz, A. Goddard, T. Ingham and M. J. Pilling, Rev. Sci. Instrum., 2007, 78, 034103.
38. M. Teresa Baeza-Romero, M. A. Blitz, A. Goddard and P. W. Seakins, Int. J. Chem. Kinet., 2012, 44, 532–545.
39. G. Hancock and V. Haverd, Chem. Phys. Lett., 2003, 372, 288–294.
40. H. M. Su, W. T. Mao and F. N. Kong, Chem. Phys. Lett., 2000, 322, 21–26.
41. U. Bley, F. Temps, H. G. Wagner and M. Wolf, Ber. Bunsen-Ges. Phys. Chem., 1992, 96, 1043–1048.
42. R. A. Alvarez and C. B. Moore, J. Phys. Chem., 1994, 98, 174–183.
43. M. A. Blitz, C. Kappler, M. J. Pilling and P. W. Seakins, Z. Phys. Chem., 2011, 225, 957–967.
44. A. J. Eskola, D. Wojcik-Pastuszka, E. Ratajczak and R. S. Timonen, J. Phys. Chem. A, 2006, 110, 12177–12183.
45. A. J. Eskola, D. Wojcik-Pastuszka, E. Ratajczak and R. S. Timonen, Phys. Chem. Chem. Phys., 2006, 8, 1416–1424.
46. A. Masaki, S. Tsunashima and N. Washida, J. Phys. Chem., 1995, 99, 13126–13131.
47. J. C. Ianni, Kintecus, Windows Version 2.80, www.kintecus.com, 2002.
48. P. D. Lightfoot, R. A. Cox, J. N. Crowley, M. Destriau, G. D. Hayman, M. E. Jenkin, G. K. Moortgat and F. Zabel, Atmos. Environ., Part A, 1992, 26, 1805–1961.
49. R. Atkinson, D. L. Baulch, R. A. Cox, J. N. Crowley, R. F. Hampson, R. G. Hynes, M. E. Jenkin, M. J. Rossi and J. Troe, Atmos. Chem. Phys., 2006, 6, 3625–4055.
50. O. Horie, C. Schafer and G. K. Moortgat, Int. J. Chem. Kinet., 1999, 31, 261–269.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c3cp54391a

---