PM2.5-bound organosulfates in two Eastern Mediterranean cities: The dominance of isoprene organosulfates

Abstract

PM_2.5 samples were collected during 2017-2018 at two Eastern Mediterranean urban sites in Greece, Athens and Patra, in order to study the abundances, the seasonal trends, the sources and the possible impact of gas phase pollutants on organosulfate formation. Each of the studied groups, except that of aromatic organosulfates, presented higher concentrations in Patra compared to those measured in Athens, from 1.1 (nitro-oxy organosulfates) to 3.6 times (isoprene organosulfates). At both sites, isoprene organosulfates was the dominant group which accounted on average for more than 50% of the total measured organosulfates, with the contribution being more than 80% during summer. Strong seasonality was observed at both sites, regarding the isoprene organosulfates, with an almost 21-fold increase from winter to summer. The same pattern, but to a lesser extent, was also observed for monoterpenes organosulfates at both sites. Alkyl organosulfates followed an identical seasonal trend with the highest mean concentrations observed during spring followed by autumn. The seasonality of anthropogenic organosulfates, multisource organosulfates and nitro-oxy organosulfates differed among the two sites or presented a more compound-specific variation. The isoprene-epoxydiol pathway appeared to be the dominant pathway of isoprene transformation, with the compounds iOS211, iOS213 and iOS215 being the major isoprene organosulfate compounds at both sites. Organosulfate contribution to the concentration of particulate matter presented common variation at both sites, ranging from 0.20 ± 0.14% (winter) to 2.5 ± 1.2% (summer) and from 0.21 ± 0.13% (winter) to 5.0 ± 2.5% (summer) for Athens and Patra, respectively. The increased NO_x levels in Athens, appeared to affect isoprene organosulfate formation as well as the formation of monoterpene and decalin nitro-oxy organosulfates. Principal component analysis followed by multiple linear regression analysis highlighted the dominance of isoprene organosulfates. In Athens, the possible impact of transportation emissions on the formation of monoterpene nitro-oxy organosulfates is indicated while the correlation of naphthalene organosulfates with low molecular weight polycyclic aromatic hydrocarbons suggests that vehicle emissions may be a significant source. In Patra, the possible contribution of sea on methyl sulfate levels is denoted.

Keywords: Organosulfates; PCA; SOA; Seasonal distribution.