Simulating Diffusion-driven Phase Separation in Atmospheric Aerosols
ANDREAS ZUEND, Zixuan Shen, Ying Li, Meredith Schervish, Manabu Shiraiwa, McGill University
Abstract Number: 439
Working Group: Aerosol Physics
Abstract
Airborne particles consisting of organic and inorganic species have been shown to frequently exhibit phase separation. Liquid–liquid phase separation (LLPS) has impacts on size-dependent particle morphology, hygroscopicity, heterogeneous reactions, and phase viscosities. In this presentation, we will introduce a new approach for simulating the time-resolved, diffusion-driven liquid–liquid phase separation and/or phase merging in particles of various sizes exposed to changing environmental conditions. Traditionally, the prediction of liquid–liquid phase separation in bulk aerosol systems can be conducted using a thermodynamic equilibrium framework, such as a liquid–liquid equilibrium model based on Aerosol Inorganic–Organic Mixtures Functional groups Activity Coefficients (AIOMFAC). However, in the context of the nonequilibrium evolution of particles that either start in a phase-separated core–shell configuration or adopt one over time, e.g., due to evaporation of water, phases of distinct chemical compositions and viscosities may emerge or disappear, affecting nonideal mixing, growth kinetics, and equilibration times. This is of particular interest in viscous, semi-solid particles that evolve slowly under typical indoor or outdoor conditions. Furthermore, understanding a potentially size-dependent evolution of LLPS and bulk–surface partitioning (e.g. comparing ultrafine vs. fine mode particles) is of interest in practice due to associated effects on cloud droplet activation and heterogeneous chemistry – aspects that are not represented in current large-scale atmospheric models. Our simulations make use of the recently developed ONION box model. ONION is a multicomponent, multilayer kinetic model based on a combination of the kinetic multilayer gas–aerosol partitioning (KM-GAP) framework and the AIOMFAC model for chemical thermodynamics. By introducing a composition-dependent model for interfacial energy contributions between distinct model phases (and layers), we can model dynamic phase separation of the core–shell type with ONION. We will discuss simulations of spinodal decomposition LLPS under humidification or drying conditions, providing insights into dynamic aerosol growth.