Abstract
Eighteen polycyclic aromatic hydrocarbons (PAHs), 24 n-alkanes, 7 hopanes, 2 cholestanes, inorganic ions, elements and carbon fractions were analyzed in real-world source samples of PM2.5 (fine particulate matter) from traffic emissions (gasoline vehicles-TGV, diesel vehicles-TDV, diesel ship-TDS, and heavy oil ships-THOS), coal combustion (coal-fired industrial boilers-CIB, power plants-CPP, and residential stoves-CRS), industrial process emissions (cement industry-IPCI, and steel industry-IPSI), and dust (soil dust-DSD, road dust-DRD, and construction dust-DCD). High molecular weight (sum of five to seven rings) PAHs accounted for higher fractions for TGV (80%) and THS (61%) than for TDV, TDS and coal combustion sources (31%–47%). Hopane ratios (C29αβ/C30αβ) in coal related sources were mostly higher than 1, whereas that of traffic emissions was lower than 1. The homohopane index [S/(S + R)], which is a useful index for identifying the maturity of fuels, ranked as TGV > THS > TDV and TDS > coal combustion. For n-alkane profiles, coal related sources showed peaks at C16–C19, TDV, TDS and THS showed similar peaks at C17–C25, but peaks for DSD (C30–C32), DRD (C17–C20, C24–25 and C30–C31), CRS (C16–C18 and C28–C29) and TGV (C24–C26) are different. Organic markers were selected which can best differentiate the subtypes within source categories by considering the component levels and variations. Through a comprehensive review, we showed that it is inadvisable to directly use diagnostic ratios for source attribution, although their trends can assist in identifying influential sources.
•Chemical profiles of residential coal stoves largely different from other coal combustion.•It is recommended not to directly use diagnostic ratios for source attribution.•N-alkanes from coal and traffic sources peaking at C16–C19 and C17–C25.•Hopane ratio (C29αβ/C30αβ) > 1 for coal sources but < 1 for traffic emission.