Interplay of Gas-phase Reactions and Multi-phase Processes on Phase State and Growth Dynamics of Secondary Organic Aerosols

Abstract

The formation, growth and evolution of secondary organic aerosols (SOA) are complex multiphase chemical processes, which represent one of the most challenging and demanding problems in research on atmospheric aerosol processes. In almost all current climate and air quality models, simplified gas-phase organic chemistry is described kinetically, while gas-particle partitioning is approximated by instantaneous equilibrium partitioning, which assumes that SOA particles are homogeneously well-mixed liquids. When particles are in glassy or viscous semisolid states, however, considering particle phase state is essential to accurately describe gas-particle interactions and particle size distribution dynamics in SOA growth. To date, however, few aerosol process models resolve both detailed gas-phase reactions and multiphase processes explicitly; there is a strong and urgent need to develop such models for further elucidation and quantitative assessment of SOA processes in the atmosphere.

The research goal of this project is to advance the fundamental understanding of the interplay between gas-phase chemistry and multiphase processes on phase state and growth dynamics of SOA for better predictions of aerosol effects on climate and air quality. The first objective is to develop a novel aerosol process model by combining the explicit gas-phase chemistry model GECKO-A (Generator of Explicit Chemistry and Kinetics of Organics in the Atmosphere) with the kinetic multi-layer model of gas-particle interactions in aerosols and clouds (KM-GAP). Our recently developed methods of viscosity prediction and effective mass accommodation will be implemented into GECKO-A. We propose to build a mechanism synthesizer that will use machine learning to reduce the full explicit chemical mechanism using unsupervised learning algorithms. We will implement the reduced mechanism into KM-GAP, which can simulate the evolution of particle viscosity and particle size distribution dynamics. The model is also uniquely capable of treating the formation of viscous surface layers with gradients and discontinuities of bulk diffusivity, which may control SOA partitioning kinetics.

The second objective is to apply the new model to chamber experiments and ARM field data to evaluate the interplay between gas-phase reactions and multiphase processes on SOA formation, growth, and chemical transformation. We will run the model for SOA formation by oxidation of anthropogenic and biogenic precursors in chamber experiments to compare with the evolution of particle size distributions and chemical compositions in the gas and particle phases. The modeling results will be compared with an existing DOE-supported aerosol model. We will also model nanoparticle growth events observed in the GoAmazon2014/15 aerial campaign to investigate how gas-phase chemistry and aerosol phase state affect aerosol concentrations and size distributions as well as to assess kinetic limitations of SOA formation and partitioning in the atmosphere.