Reaction kinetics of OH + HNO₃ under conditions relevant to the upper troposphere/lower stratosphere

Abstract

The OH initiated oxidation of HNO3 in the UT/LS plays an important role in controlling the O3 budget, removing HOx radicals whilst driving NOx/y partitioning chemistry by yielding NO3 radicals: OH + HNO3 → H2O + NO3. In this paper, k1(T, P) was measured using OH (A ← X) Laser Induced Fluorescence (LIF) and the data was modelled over the 223-298 K temperature and 25-750 Torr pressure ranges, using the modified Lindemann-Hinshelwood expression , where k0 = 5.2 × 10-14 exp(200/T) cm3 s-1, k2 = 8.4 × 10-17 exp(1900/T) cm3 s-1 and k3 = 1.6 × 10-34 exp(1745/T) cm3 s-1. A significant source of experimental uncertainty derives from accurate determination of HNO3 concentration, which is impacted by heterogeneous uptake of the low volatility HNO3 onto cold surfaces of the reactors. Our results represent the determination of k1(T, P) using two different in situ [HNO3] measurements: VUV absorption and a new two photon Photolysis Induced Fluoresence (PIF). Experimental results are discussed along with a computational master equation calculation (MESMER), which highlight the need for further theoretical study into the OH + HNO3 mechanism and potential energy surface. The atmospheric impact of these new rate constants were modelled using the STOCHEM-CRI chemistry transport global model, which have shown a small reduction in global budgets of key atmospheric species, with more significant changes in the NOx/HNO3 ratio, peaking in the tropical upper troposphere regions.

Pressure-Dependent Rate Constant Caused by Tunneling Effects: OH + HNO3 as an Example

“…⟨ ΔE ⟩ d is varied from 50 to 175 cm –1 with a step size of 25 cm –1 . As seen in Figure , all calculated k ( T , P ) are in good agreement (within 20%) with the experimental results. − Theory with ⟨ ΔE ⟩ d = 50 cm –1 agrees with the experiment of Dulitz et al, while a higher value of ⟨ ΔE ⟩ d = 175 cm –1 is desirable to match the experimental results of Brown et al The overall best fit between theory and experiment in this scenario is found with ⟨ ΔE ⟩ d ∼ 100 cm –1 . Following the same strategy, we did the same at other temperatures (from 210 to 350 K), where experimental data are available, − and a good agreement between theory and experiment has been seen (see Figure S1 in the Supporting Information).…”

“…The purpose of this work is to examine the largely unexplored pressure dependence of κ­( T ) and to check whether there is a connection between quantum mechanical tunneling and the falloff curve. The reaction of OH and HNO 3 , which significantly affects stratospheric ozone, the HO x budget, as well as NO x partitioning in the upper troposphere and lower stratosphere (UTLS), − is chosen as an example because this bimolecular reaction (which also displays a pressure dependence) has been well-characterized experimentally. − …”

“…The reaction of OH and HNO 3 has been heavily studied but with relatively low levels of theory. ,− In this work, to obtain high-accuracy energies for kinetics calculations, we used mHEAT-345­(Q), which is a composite method mainly based on high-level coupled-cluster calculations including full triple excitations and perturbative noniterative treatment of quadruple excitations. It is reported elsewhere that mHEAT-345­(Q) can provide high accuracy for relative energies, typically within 0.5 kcal/mol .…”

“…Figure displays thermal rate constants calculated at 298 K as a function of pressure (i.e., falloff curves) for the OH + HNO 3 reaction. Three sets of recent experimental data − are also included for comparison. Because both the average vibrational (⟨ ΔE vib ⟩ d ) and rotational (⟨ ΔE rot ⟩ d ) energies transferred in a downward direction by collisions between the vibrationally excited PRC and the bath gas are unknown, a trial and error process is used to determine them.…”

“…Here, we choose ⟨ ΔE ⟩ d = ⟨ ΔE vib ⟩ d = ⟨ ΔE rot ⟩ d (results for different values of ⟨ ΔE vib ⟩ d and ⟨ ΔE rot ⟩ d are provided in the Supporting Information). Where possible, experimental data − are also included for the purpose of comparison. Note that pressure dependence in the calculated rate constants is conspicuous; that for the experimental data is simply suggestive, owing to the limited range of pressure studied.…”