PNNL

Abstract

Secondary organic aerosol (SOA) is an important fraction of the fine-mode atmospheric aerosol mass. Frameworks used to develop SOA parameters from laboratory experiments and subsequently used to simulate SOA formation in atmospheric models make many simplifying assumptions about the processes that lead to SOA formation in the interest of computational efficiency. These assumptions can limit the ability of the model to predict aerosol mass, composition, and properties of SOA accurately. In this work, we developed a computationally efficient, process-level model named simpleSOM to represent the chemistry, thermodynamic properties, and microphysics of SOA. simpleSOM simulates multigenerational gas-phase chemistry, phase-state-influenced kinetic gas/particle partitioning, heterogeneous chemistry, oligomerization reactions, and vapor losses to the walls of Teflon chambers. As a case study, we used simpleSOM to simulate the SOA formation from photooxidation of ??-pinene. This was done to demonstrate the ability of the model to develop parameters that can reproduce environmental chamber data, to highlight the chemical and microphysical processes within simpleSOM, and discuss implications for SOA formation in chambers and in the real atmosphere. SOA parameters developed from experiments performed in the chamber at the California Institute of Technology (Caltech) reproduced observations of SOA mass yield, O:C, and volatility distribution gathered from other chambers. Sensitivity simulations suggested that multigenerational gas-phase aging contributed to nearly half of all SOA and that in the absence of vapor wall losses, SOA production in the Caltech chamber could be nearly 50% higher. Simulations performed at atmospherically relevant conditions indicated that the SOA mass yields were sensitive to whether and how oligomerization reactions and the particle phase state were represented in the chamber experiment from which the parameters were developed. simpleSOM provides a comprehensive, process-based framework to consistently model the SOA formation and evolution in box and 3D models.