Ozonolysis of Oleic Acid Aerosol Revisited: Multiphase Chemical Kinetics and Reaction Mechanisms

Abstract

The chemical processing of organic aerosol particles is important for atmospheric chemistry, climate, and public health. The heterogeneous oxidation of oleic acid particles by ozone is one of the most frequently investigated model systems. The available kinetic data span a wide range of particle size and ozone concentration and are obtained with different experimental techniques including electrodynamic balance (EDB), optical tweezers, environmental chamber, and aerosol flow tube reactors using mass spectrometry and Raman spectroscopy as detection methods. Existing kinetic and mechanistic analyses, however, reveal systematic differences and inconsistencies that are a matter of ongoing debate. We developed and applied an inverse modeling approach using a kinetic multilayer model (KM-SUB) and Monte Carlo-based global optimization algorithms to 11 literature data sets and an additional new set of EDB data. We were able to reconcile most experimental data with consistent sets of multiphase chemical kinetic parameters. For a unique determination of these parameters, however, further experiments with simultaneous measurement of multiple observables at specific, insightful reaction conditions are required. We tested three different reaction mechanisms and conclude that secondary chemistry involving Criegee intermediates appears crucial to resolve the discrepancies found in earlier studies. Primary ozone chemistry occurs close to the particle surface and secondary reactions seem to dominate in the particle bulk, involving OH formation and radical chain reactions.

Coupled interfacial and bulk kinetics govern the timescales of multiphase ozonolysis reactions

“…[9][10][11] To understand these chemical transformations, we need to consider multiphase reaction dynamics in concentrated to dilute aqueous phases relevant to aqueous aerosol and cloud water, which may not reflect the purely organic aerosol systems that are often the focus of mechanistic laboratory studies. [12][13][14][15] In tropospheric multiphase reactions, O3 is an important oxidant that enters the condensed phase through partitioning from the gas phase. This Henry's Law partitioning into atmospheric aqueous phases is governed by two coupled adsorption-desorption and solvation-desolvation equilibria; connecting the gas, interface and bulk phases.…”

“…16,17 The position of these two coupled equilibria will respond dynamically to the consumption of O3 by both interfacial and bulk reactions, such that O3 transport and chemical reaction are inherently coupled. 15 A large number of reaction rate coefficients from beaker-scale laboratory experiments are available for atmospherically relevant aqueous phase ozonolysis reactions. 18 However, to use these bulk rate coefficients to predict reaction timescales in atmospheric aqueous phases we must be able to accurately represent how adsorption, desorption, solvation, desolvation and diffusion processes control multiphase reaction rates.…”

“…Even for well-studied pure organic systems, such as oleic acid, 15 challenges arise in reconciling inversemodelled kinetic parameters with experimental values. A third approach is to build kinetic models with computational and experimental constraints on elementary reaction steps.…”

A Kinetic Model for Predicting Trace Gas Uptake and Reaction

“…However, despite the relative ease with which modern aerosol techniques are used to determine reaction probabilities, there remains a substantial challenge in connecting γ with a bimolecular reaction rate coefficient of a single elementary step. 5 This challenge arises because γ is an aggregate of many kinetic steps, leading to cases where the observed uptake coefficient and decay kinetics depend in complex ways on the gas, interface and condensed phase environment of the aerosol or droplet. 6 This complexity requires the development and application of models to interpret γ, with the goal of linking the physical properties of the aerosol or droplet with its multiphase reactivity.…”

“…Kinetic multilayer models [31][32][33][34][35][36][37][38] employ a flux-based representation, numerically solving the coupled differential equations for mass transport and chemical reactions. Many multilayer models require a comprehensive set of variables for each molecule and so are often employed in inverse modeling studies 5,39 of large data sets, where they can resolve the fine details of surface and bulk processes as well as the formation of chemical gradients. Kinetic descriptions of multiphase chemistry, implemented in stochastic reaction-diffusion simulations by Houle, 40 Wilson 41 and coworkers, 6,28,[42][43][44][45][46][47] have been used to describe multiphase transformations using a set of elementary kinetic and diffusion steps, with the goal of obtaining physically realistic, albeit simple descriptions, of reactive uptake.…”