Variations of source-specific risks for inhalable particles-bound PAHs during long-term air pollution controls in a Chinese megacity: impact of gas/particle partitioning

Gas/particle (G/P) partitioning is a determining factor that drives the behavior of polycyclic aromatic hydrocarbons (PAHs) in the atmosphere. To comprehend the variations and impacts of PAHs G/P partitioning on inhalation risks from different sources, atmospheric PAHs in Chengdu were studied from January 2009 to January 2021. Chengdu's air quality demonstrated improvement, with a noticeable decline in PM10 (from 238 to 103 μg·m-3), PM10-bound PAHs mass concentrations (from 0.08 to 0.01 μg·m-3), as well as PAHs inhalation non-cancer (from 63.5 to 7.1) and cancer risks (from 6.2 × 10−5 to 7.7 × 10−6). Decreasing trends were also observed in the concentrations of organic carbon (OC) and the values of absorptive G/P partitioning coefficient (Kp, OM) from 2009 to 2020, indicating a slight decrease in the PM absorptive capacity. A positive matrix factorization (PMF) model utilized three distinct datasets: PMFpp (particle-phase PAHs only), PMFtot (total PAHs concentrations), and PMFpp/gas (particle-phase in parallel with gas-phase PAHs). PMFpp and PMFpp/gas provide reasonable source apportionment and source-specific risk results. However, daily variations in source contributions of PMFtot were found to be unreasonable. PMFpp/gas and PMFtot could identify a source of low molecular weight PAHs (LPAHs) related to temperature, and a significantly positive correlation between LPAHs and temperature was discovered. Gas/particle partitioning of PAHs showed a notable impact on source-specific risks associated with coal and biomass combustion (CC&BC). The findings contribute to our understanding of the variation in PAHs G/P partitioning and the PM absorptive capacity resulting from effective air pollution control in China. Additionally, they offer prerequisite information for incorporating semi-volatile organic compounds in source apportionment and source-specific risk assessment.