Molecular characterization of dissolved organic matters in winter atmospheric fine particulate matters (PM2.5) from a coastal city of northeast China

Dissolved organic matters (DOM5) in fine particulate matters (PM2.5) play a crucial role in global climate change and carbon cycle. However, the chemical components of DOMs are poorly understood due to its ultra-complexity. In this study, DOM5 in atmospheric PM2.5 collected during the heating period in coastal city Dalian were analyzed with ultrahigh resolution Fourier transform ion cyclotron resonance mass spectrometer, and the molecular composition was characterized. A large number of monoisotopic molecular formulas were assigned to DOM5, which could be classified into CHO, CHNO, CHOS, and CHNOS subgroups. A total 014228 molecular formulas were identified in DOM5 collected in hazy days, while only 2313 components were found in DOM5 collected in normal days. CHO group was the dominated components in normal days, whereas CHNO group gave significantly higher contributions in hazy days. The S-containing (CHOS and CHNOS) groups posed the highest relative percentages in both normal and hazy days. In addition, potential emission sources were discussed according to the chemical component analysis. The van Kreve lent diagram illustrated that lignin-like and protein amino sugar family species were the most abundant subclasses in DOMs; and 78% and 94% of DOMs in atmospheric PM2.5 collected from Dalian could come from biogenic origins in hazy and normal days, respectively. More compounds in hazy days were derived from anthropogenic emissions. (C) 2019 Elsevier B.V. All rights reserved.