Dynamic oxidative potential of organic matter and their source-specific key toxicants during haze episodes in North...

Haze episodes pose substantial health risks to human-beings, yet the connections between the sources and chemical composition of fine particulate matter (PM_2.5) and health effects remain inadequately understood. This study employed the high time-resolution (1-h) offline sampling and investigated the chemical characteristics and oxidative potential (OP) of the methanol-soluble organic matter (MSOM) in PM_2.5 during three haze events in the North China Plain in 2020. The average volume-normalized and mass-normalized OP (OP_v and OP_m) during the three episodes (EP1-EP3) were 4.1 ± 1.6, 3.4 ± 1.2, 5.5 ± 1.9 pmol/min/m³, and 20.3 ± 4.9, 27.9 ± 6.7, 22.4 ± 5.3 pmol/min/μg, respectively. Positive matrix factorization model analysis revealed that the contributions of primary combustion (vehicle emission, biomass burning and coal combustion) accounted for ∼60% of total OP_m, and the first- and secondary-generation aqueous-phase secondary organic aerosol formation processes (aqSOA I and aqSOA Ⅱ) contributed ∼36% of total OP_m. The average OP_m in nighttime showed higher values than that in daytime, with the nighttime OP_m being predominantly influenced by primary emissions, whereas daytime SOA processes making a major contribution to OP_m. Non-targeted screening based on Fourier-transform ion cyclotron resonance mass spectrometry (FT-ICR MS) analysis suggested that nitrogen-containing organic compounds were the dominant contributors to OPm (61%), especially those nitro-/oxygenated-condensed aromatics (Nitro-/O-ConA) and high-oxygen highly unsaturated compounds (HO-HUPC). By integrating the source information with molecular fingerprinting, we find that nitro- ConA were key toxic components in primary emissions, whereas secondary sources were enriched with HO-HUPC. Overall, our results improved the molecular-level understanding of the sources and evolution of key toxic components during haze events.