Characterization of carbonaceous aerosols in Singapore: insight from black carbon fragments and trace metal ions detected by a soot-particle aerosol mass spectrometer

Abstract

<p><strong>Abstract.</strong> Understanding sources and atmospheric processes that can influence the physio-chemical properties of carbonaceous aerosols are essential to evaluate their impacts on air quality and climate. However, resolving the sources, emission characteristics and aging processes of carbonaceous aerosols in complex urban environments remain challenging. In this work, a soot-particle aerosol mass spectrometer (SP-AMS) was deployed to characterize organic aerosols (OA), refractory BC (rBC) and trace metals in Singapore, a highly urbanized city with multiple local and regional air pollution sources in the tropical region. rBC (C<sub>1</sub><sup>+</sup>–C<sub>9</sub><sup>+</sup>) fragments and trace metals ions (K<sup>+</sup>, Na<sup>+</sup>, Ni<sup>+</sup>, V<sup>+</sup> and Rb<sup>+</sup>) were integrated into our positive matrix factorization of OA. Two types of fossil fuel combustion-related OA with different degree of oxygenation were identified. This work provides evidence that over 90 % of rBC was originated from local combustion sources with ~ 30 % of them associated with the fresh secondary organic aerosol (SOA) produced under the influences of industrial emissions during daytime. The results also show that ~ 43 % of the total rBC was emitted from local traffic, and the rest of rBC fraction due to multiple sources, including vehicular, shipping and industrial emissions, being not fully resolved. There was only a weak association between the cooking-related OA component and rBC. Although there was no observable biomass burning episode during the sampling period, tracers for biomass burning, K<sup>+</sup> and Rb<sup>+</sup>, were mainly associated with the more-oxidized oxygenated OA component (~ 32 % of the total OA), indicating significant contributions of regional biomass burning emissions to this aged OA component. Furthermore, the aerosol pollutants transported from the industrial area and shipping ports gave higher C<sub>1</sub><sup>+</sup>/C<sub>3</sub><sup>+</sup> and V<sup>+</sup>/Ni<sup>+</sup> ratios than those associated with traffic and biomass burning. The observed association between Na+ and rBC suggests that the contribution of anthropogenic emissions to total particulate sodium should not be ignored in coastal urban environments. Overall, this work demonstrates that rBC fragments and trace metal ions can improve our understanding on the sources, emissions characteristics and aging history of carbonaceous aerosol (OA and rBC) in this type of complex urban environments.</p>

Light Absorbing Properties of Primary and Secondary Brown Carbon in a Tropical Urban Environment

“…It should be noted that the mass spectral features of HOA were uniquely different to those of typical BBOA mass spectra. 44 The MAC of the HOA component is similar to the MAC of mixed urban emissions reported in Beijing, where both local and distant sources of fossil fuel combustion from vehicular and industrial emissions and occasional biomass fires were reported. 58 Cappa et al 57 reported limited absorptivity for HOA in Fresno, California during the wintertime, whereas the SOA factor associated with nitrate showed MAC values comparable to that of our LO-OOA factor.…”

“…MO-OOA represented a relatively aged fraction of SOA (O/C = 0.94) that was suspected to be influenced by aged regional biomass burning and/or coal combustion emissions, together with other types of SOAs that formed due to various types of gas-and particle-phase chemistry, based on its moderate associations with Rb + (R-0.58) and K + (R-0.71) and the air mass back trajectory analysis. 44 Note that influence of transboundary biomass burning smoke on OOA factors has been previously observed in this region. 52 3.2.…”

“…Mass concentrations of nonrefractory aerosol components, including organics, sulfate, nitrate, ammonium, and chloride, were measured by a co-located soot particle-aerosol mass spectrometer (SP-AMS, Aerodyne Research). The specific operating conditions, calibration procedures, and data analysis approaches of this study can be found in the work of Rivellini et al 44 Note that the composition dependent collection efficiency (CDCE) approach was applied in our data analysis. 45 The OA concentrations measured by the SP-AMS were compared to the organic matter (OM) concentrations obtained by a co-located semicontinuous organic carbon− elemental carbon (OC−EC) analyzer (Sunset Laboratories).…”

“…The concentrations of OA were ∼20% higher than those of OM (OM/OC = 1.6 for urban OA) 46 (see detail in S2 in the SI). Rivellini et al 44 identified five OA factors: referred to as hydrocarbon-like OA (HOA), oxidized-HOA (O-HOA), cooking-related OA (COA), less oxidized-oxygenated OA (LO-OOA), and more oxidized-OOA (MO-OOA), based on positive matrix factorization (PMF) of organic fragments. The detailed interpretation of each OA factor can be found in Rivellini et al 44 2. toward shorter wavelengths, 3,47 whereas BC absorption has a relatively weak wavelength dependence with AAE BC approximately equal to 1.…”

“…Rivellini et al 44 identified five OA factors: referred to as hydrocarbon-like OA (HOA), oxidized-HOA (O-HOA), cooking-related OA (COA), less oxidized-oxygenated OA (LO-OOA), and more oxidized-OOA (MO-OOA), based on positive matrix factorization (PMF) of organic fragments. The detailed interpretation of each OA factor can be found in Rivellini et al 44 2. toward shorter wavelengths, 3,47 whereas BC absorption has a relatively weak wavelength dependence with AAE BC approximately equal to 1. Although this usual assumption may be true for externally mixed BC particles with a diameter <50 nm, 1,48 several studies have highlighted that this value may vary for ambient BC (from 0.8 to 1.4) due to particle sizes, morphology, and mixing conditions.…”