We conducted a comprehensive chemical speciation study of fine particulate matter (PM_2.5) collected from Changping, a suburban district in Beijing over a one-year period to identify and quantify major sources of PM_2.5 and organic aerosol (OA). In particular, more than 115 organic compounds including both primary and secondary OA (SOA) source tracers were measured together with major ions, elements, elemental carbon (EC) and organic carbon (OC). The OC fraction was broken down on the basis of polarity, water solubility and hydrophobicity. An organic matter (OM) to OC ratio of 1.7 was determined using the detailed organic speciation data and facilitated to obtain an annual average OM mass (17 μg m^-3) that accounts for 31% of PM_2.5. Source apportionment of PM_2.5 and OC were perform...[ Read more ]

We conducted a comprehensive chemical speciation study of fine particulate matter (PM_2.5) collected from Changping, a suburban district in Beijing over a one-year period to identify and quantify major sources of PM_2.5 and organic aerosol (OA). In particular, more than 115 organic compounds including both primary and secondary OA (SOA) source tracers were measured together with major ions, elements, elemental carbon (EC) and organic carbon (OC). The OC fraction was broken down on the basis of polarity, water solubility and hydrophobicity. An organic matter (OM) to OC ratio of 1.7 was determined using the detailed organic speciation data and facilitated to obtain an annual average OM mass (17 μg m^-3) that accounts for 31% of PM_2.5. Source apportionment of PM_2.5 and OC were performed using positive matrix factorization (PMF) based on the comprehensive data set. Three SOA sources (SOA I, II and III) are identified by isoprene SOA tracers, β-caryophyllene SOA tracer and toluene SOA tracer, respectively. Vehicle emissions and domestic cooking emissions were found as important as biomass burning, with contributions of 10% to PM_2.5 from each, all higher than that of coal combustion (2%). As for source contributions to OC, the three SOA sources combined, vehicle and cooking emissions were significant sources, contributing ~25% by each of them. Pollution event analysis indicated that higher concentrations of PM_2.5 are normally associated with predominant secondary formation processes. However, primary combustion emissions contributed more to OC even when secondary formation processes were a dominating factor driving up PM_2.5 mass. On low pollution days, PM_2.5 and OC were affected mostly by the primary combustion sources. This work helps enhance the understanding of secondary organic aerosol formation and aerosol evolution in urban atmospheric conditions. It has been demonstrated that comprehensive source identification with organic source tracers considered could provide valuable insights for air pollution control in China.

Qualitative characterization of organics by ultra-high-resolution mass spectrometry was conducted using Fourier-Transform Ion Cyclotron Resonance/mass spectrometry (FT-ICR-MS). We found that by adding ammonium formate (AF), a large number of aromatic compound formulas are detected in electrospray ionization (ESI) in positive mode compared with tradition ESI ionization. Meanwhile, NO₂ substituted CH are also detected with much higher formula numbers than that of CH formula. Pairs of CHN₁O₂ and the respective CH precursor (presumably linked via atmospheric nitration process) are found positively correlated with NO₂ concentration as well as the ambient concentration of polycyclic aromatic hydrocarbons (PAH). It is speculated that combustion sources play important roles in producing the precursors like unsubstituted PAH substances. NO₂ coming from vehicular or other local emission sources would act as oxidizing reagent in most of the CHN₁O₂ formation without requiring excessive O₃. In conclusion, a more comprehensive speciation profile could be created by including both substituted and unsubstituted aromatic compounds to expand our current scope of atmospheric research

At last, comparing the results from both ESI positive and negative modes, more than 2000 common CHO and CHON formulas are detected with molecular mass between 200 and 500 Da. There are 400 unique CHN₁O₂ and CHO₁ formulas detected in the positive mode, which are characterized as nitro-hydrocarbon and hydroxyl-hydrocarbon respectively. It appears that non-polar organics extracted by dichloromethane (DCM) are more favored to be detected in positive mode. In addition, positive correlations were found between the signal to noise ratio (S/N) of substitutional hydrocarbons (i.e., CHN₁O₂ or CHO₁) vs their parent precursors. However, a negative correlation was found between S/N of substitutional hydrocarbons vs O₃, indicating the photochemical degradation may initiate a series of substitutional reactions causing the depletion of hydrocarbons. There are more than 1000 sulfur containing formulas detected in the samples, over half of which are aromatic organosulfates (OSs) determined by their aromaticity equivalent. In conclusion, this study highlights the knowledge gap in current understanding of OA composition and provides guiding insights for future further molecular-level characterization of OA.