Letter Cite This: J. Phys. Chem. Lett. 2019, 10, 3598−3603
pubs.acs.org/JPCL
Pressure-Dependent Kinetics of the Reaction between CH3OO and OH Focusing on the Product Yield of Methyltrioxide (CH3OOOH) Feng Zhang*,† and Can Huang†,‡ †
National Synchrotron Radiation Laboratory, University of Science and Technology of China, Hefei, Anhui 230029, P. R. China Center for Combustion Energy and Key Laboratory for Thermal Science and Power Engineering of MOE, Tsinghua University, Beijing 100084, P. R. China
‡
Downloaded via NOTTINGHAM TRENT UNIV on July 19, 2019 at 19:10:53 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
ABSTRACT: The reaction kinetics of methyl peroxy radical (CH3OO) and hydroxyl radical (OH) has attracted an increasing level of interest in the past decade, while the branching yields of various product channels are still under debate. In this work, a comprehensive theoretical effort was made to investigate the branching yield of the stabilized methyltrioxide (CH3OOOH, TRIOX) adduct, which has recently been a research focus. Our computed branching ratio of TRIOX at 298 K and 760 Torr is ∼0.04, in agreement with the result of multiplexed photoionization mass spectrometry. We show that the large branching yield obtained in an early theoretical study mainly originated from the collision-induced strong stabilization presented in their simulation. Our findings clarify the controversial product yield results for this important species in recent studies. The computed rate constants over wide temperature and pressure ranges allow better integration of this reaction into global atmospheric models and low-temperature combustion kinetic models.
M
311++G(3df,3pd) level and predicted the branching of dominant reaction channels by statistical rate theories. By using the upper limit of the overall rate coefficient (2.8 × 10−10 cm3 molecule−1 s−1)5 and their computed branching ratio of R1:R2:R3 = 0.82:0.107:0.069 at 298 K and 1 atm giving a large uncertainty of 3.5), Müller et al.7 indicated that this reaction could largely contribute to the atmospheric methanol with a direct methanol yield via R3 and an indirect yield from R2 (it was assumed that ∼65% of TRIOX will transform into methanol).7 However, Caravan et al.9 argued that the reaction of hydroxyl and methylperoxy radicals is not a major source of atmospheric methanol by combined multiplexed photoionization mass spectrometry (MPIMS) experiments, continuous photolysis chamber experiments, and global atmospheric modeling. The branching fraction of the direct methanol channel (R3) was measured as 0.03−0.12, and that of the TRIOX channel (R2) is 0.04−0.08. Ferracci et al.8 employed the UM-UKCA chemistry−climate modeling adopting a recently refined overall rate constant (1.6 × 10−10 cm3 molecule−1 s−1),16 indicating the reaction is a major sink for CH3OO and a source of HO2 over remote tropic oceans but makes an insignificant contribution to methanol yield. Very recently, Fittschen et al.15 proposed that the decomposition of ROOOH (R = alkyl radicals), especially when those ROOOH molecules originate from peroxy radicals of large VOCs reacting with OH, could interpret the interference for OH
ethyl peroxy radical (CH3OO) is the most abundant organic peroxy radical in the atmosphere, with a global production of 2500 Tg per year.1,2 The reaction between CH3OO and hydroxyl radical OH, the most important oxidant in the atmosphere, has attracted a great deal of attention.3−9 Scheme 1 illustrates major reaction pathways of CH3OO + Scheme 1. Reaction Pathways of CH3OO + OH
OH, indicating a complex multiwell process involving competition among collisional stabilization of the methyltrioxide (TRIOX) (R2), direct H abstraction (R5), and those well-skipping reactions10,11 (R1, R3, and R4). The negligible contribution of R5 forming Criegee intermediate (CI) has been confirmed by recent studies both experimentally and theoretically.6,7,12,13 Previous studies have shown that this reaction will contribute to the increase in HO2 abundance due to the dominance of R1 under atmospheric conditions.3,6,8 The strong dominance of the HO2 channel (R1) under atmospheric conditions has been widely acknowledged; however, the kinetics of the other nondominant channels and its implication for the atmosphere have been under debate for the past five years.6−9,14,15 Müller et al.7 investigated this reaction at the CCSD(T)-F12/cc-pVTZ-F12//M06-2X-D3/6© 2019 American Chemical Society
Received: March 19, 2019 Accepted: June 13, 2019 Published: June 13, 2019 3598
DOI: 10.1021/acs.jpclett.9b00781 J. Phys. Chem. Lett. 2019, 10, 3598−3603
Letter
The Journal of Physical Chemistry Letters
Figure 1. Simplified potential energy surface of CH3OO + OH.
Table 1. Energy Corrections in High-Level Evaluations and Stationary Point Energies for the CH3OO + OH Potential Energy Surface CH3OO + OH OH CH3OO 1 RC TRIOX TS1 TS4
CCSD(T)_TZa
core−valencea
CBSa
T(Q)/DZa
ZPEa
this workb
−265.5801 −75.6377 −189.9433 −265.6016 −265.6349 −265.5895 −265.5872
−0.2470 −0.0631 −0.1838 −0.2452 −0.2469 −0.2457 −0.2461
−0.1347 −0.0387 −0.0960 −0.1319 −0.1373 −0.1337 −0.1336
−0.0030 −0.0005 −0.0025 0.01158 −0.0033 −0.0030 −0.0079
0.0524 0.0086 0.0438 0.0550 0.0600 0.0539 0.0530
0
−2.7 −30.8 −3.1 −5.4
literatureb 0c
−2.7c −29.0,c −29.5d −2.6c −4.9c
a Absolute energy, in hartrees. bRelative energy, in kilocalories per mole. cTheoretical calculation from ref 7 at the CCSD(T)-F12/cc-pVTZ-F12// M06-2X-D3/6-311++G(3df,3pd) level. dTheoretical calculation from ref 14 at the CCSD(T)/aug-cc-pVTZ//M06-2X/aug-cc-pVTZ level.
measurements by a commonly used technique, fluorescence assay by gas expansion (FAGE). In other words, ROOOH could be a missing piece of the puzzle for OH measurements in low-NOx environments, which has been debated over the past two decades.8,15 The TRIOX branching yield of the title reaction is a focal point for the discrepancies in previous studies in terms of not only the chemical kinetics but also its atmospheric implication.6−9,15,17 In the work presented here, with the simplified PES shown in Figure 1, key energy parameters influencing the TRIOX stabilization rate were traced by bruteforce sensitivity analysis (varying each energy parameter by ±1 kcal/mol) based on the RRKM/master equation (ME) simulation.18,19 Labels in ref 7 for those stationary points were adopted in this work for the sake of convenient comparison hereafter. Detailed descriptions of the RRKM/ ME calculations, including the theory and input parameters, are provided in the Supporting Information. After the sensitivity analysis (see Figure S1), the recognized key energies were then computed by a high-level composite method via the consideration of a series of additive corrections.20 The correction for core−valence interactions was obtained from CCSD(T, full)/CBS calculations based on extrapolation of results for the cc-pVTZ and cc-pVQZ basis sets.21 The complete-basis-set (CBS) limit was estimated from the two-point extrapolation [l/(l + 1)4 formula]22 of CCSD(T) calculations employing aug-cc-pVQZ and aug-ccpV5Z basis sets and CCSD(T)-F1223 calculations employing the cc-pVTZ-F12 and cc-pVQZ-F12 basis sets.24 A correction for higher-order excitations was obtained from CCSDT(Q)/
cc-pVDZ calculations.25 The CCSD(T) calculations presented here generally employed RHF wave functions within the UCCSD(T) formalism implemented in MOLPRO,26 while the CCSDT(Q) calculations employed UHF wave functions as required by the MOLPRO implementation of Kállay’s MRCC code.27 Table 1 shows our computational results for the key stationary points (CH3OO + OH, 1RC, TS1, TRIOX, and TS4) via the computational procedures described above. The uncertainty of the computed energies listed in Table 1 is estimated as ±0.5 kcal/mol. Previous results in the literature7,14 are also shown for comparison. The discrepancies between the calculation of Müller et al.7 and our high-level energies range from 0.5 to 1.8 kcal/mol for the selected saddle points that have large sensitivity coefficients for the TRIOX yield. Such a discrepancy could have notable influences on the computed rate constants. For example, at room temperature the rate constant will increase by a factor of ∼5 if the barrier height is decreased by 1 kcal/mol. The largest improvement was found for the computed well depth of the TRIOX (−1.8 kcal/mol), which will consequently increase the computed branching fraction of TRIOX. On the basis of the PES computed by Müller et al.7 and our refinement at higher theoretical levels, we finally computed the temperature- and pressure-dependent rate constants of the reaction channels (R1−R4) by the RRKM/ME method. The recently measured rate coefficient of the overall reaction, 1.0 × 10−10 cm3 molecule−1 s−1 at 298 K and 760 Torr,28 was used to mimic the entrance flux, with which the branching yields of the competing reaction channels were deduced. 3599
DOI: 10.1021/acs.jpclett.9b00781 J. Phys. Chem. Lett. 2019, 10, 3598−3603
Letter
The Journal of Physical Chemistry Letters
Figure 2. Computed branching ratios of (a) R1, (b) R2, and (c) R3 at various temperatures and pressures.
Figure 2 illustrates our computed branching ratios of R1−R3 over temperature range of 200−500 K and pressure range of 30−760 Torr (the computed branching ratios of R4 and R5 are both
