Volatility Parametrization of Atmospheric Organic Compounds Based on Molecular Formula

Saturation vapor pressure of atmospheric organic compounds is crucial for understanding gas-particle partitioning. Yet measurements of such a property are quite incomplete, and an accurate and efficient volatility parametrization remains challenging. Here, we utilized a Long Time-of-Flight Chemical Ionization Mass Spectrometer (LToF-CIMS) coupled with a Filter Inlet for Gases and Aerosols (FIGAERO) to measure molecular composition and saturation vapor pressure of organic compounds on particulate samples collected from Dianshan Lake and Tai Lake, China across different seasons with a wide temperature coverage. Two desorption procedures, i.e., a high-temperature range of 20–160 °C and a low-temperature range of 5–35 °C, were implemented. The obtained saturation vapor pressures of organic components were integrated with previous results (Yang et al., 2023) to compose our data set. By incorporating the concept of mole fraction (X) and addressing data imbalance across different volatility bins via Synthetic Minority Oversampling (SMOTE) and Random Under-sampling (RUS), a 4-parameter volatility parametrization of a commonly used volatility formula was developed, which outperforms previous ones for volatility prediction of both CHO and CHON compounds when validated against the NCI database, and potentially improves atmospheric models that use volatility as a key input.