Secondary organic aerosol (SOA) is an important yet not fully characterized constituent of atmospheric particulate matter. A number of different techniques and chromatographic methods are currently used for the analysis of SOA, so the comparison of results from different laboratories poses a challenge. So far, tentative structures have been suggested for many organosulfur compounds that have been identified as markers for the formation of SOA, including isoprene-derived organosulfates. Despite the effectiveness and robustness of LC-MS/MS analyses, the structural profiling of positional isomers of recently discovered organosulfates with molecular weights (MWs) of 214 and 212 from isoprene was entirely unsuccessful. Here, we developed a UHPLC combined with high-resolution tandem mass spectrometric method that significantly improves the separation efficiency and detection sensitivity of these compounds in aerosol matrices. We discovered that selection of the proper solvent for SOA extracts was a key factor in improving the separation parameters. Later, we took advantage of the enhanced sensitivity, combined with a short scan time window, to perform detailed structural mass-spectrometric studies. For the first time, we elucidate a number of isomers of the MW 214 and the MW 212 organosulfates and provide strong evidence for their molecular structures. The structure of trihydroxyketone sulfate MW 214 that we propose has not been previously reported. The methods we designed can be easily applied in other laboratories to foster an easy comparison of related qualitative and quantitative data obtained throughout the world.

In-cloud processing of volatile organic compounds is one of the significant routes leading to secondary organic aerosol (SOA) in the lower troposphere. In this study, we demonstrate that two atmospherically relevant α,β-unsaturated carbonyls, i.e., but-3-en-2-on (methyl vinyl ketone, MVK) and 2-methylopropenal (methacrolein, MACR), undergo sulfate radical-induced transformations in dilute aqueous systems under photochemical conditions to form organosulfates previously identified in ambient aerosols and SOA generated in smog chambers. The photooxidation was performed under sun irradiation in unbuffered aqueous solutions containing carbonyl precursors at a concentration of 0.2 mmol and peroxydisulfate as a source of sulfate radicals (SO) at a concentration of 0.95 mmol. UV-vis analysis of solutions showed the fast decay of unsaturated carbonyl precursors in the presence of sulfate radicals. The observation confirms the capacity of sulfate radicals to transform the organic compounds into SOA components in atmospheric waters. Detailed interpretation of high-resolution negative ion electrospray ionization tandem mass spectra allowed to assign molecular structures to multiple aqueous organosulfate products, including an abundant isoprene-derived organosulfate CHSO detected at m/z 199. The results highlight the solar aqueous-phase reactions as a potentially significant route for biogenic SOA production in clouds at locations where isoprene oxidation occurs. A recent modelling study suggests that such processes could likely contribute to 20-30 Tg year production of SOA, referred to as aqSOA, which is a non-negligible addition to the still underestimated budget of atmospheric aerosol.

Abstract. Concentration of radon (222Rn) in the near-ground atmosphere has been measured quasi-continuously from January 2005 to December 2009 at two continental sites in Europe: Heidelberg (south-west Germany) and Krakow (southern Poland). The atmosphere was sampled at ca. 30 and 20 m above the local ground. Both stations were equipped with identical instruments. Regular observations of 222Rn were supplemented by measurements of surface fluxes of this gas in the Krakow urban area, using two different approaches. The measured concentrations of 222Rn varied at both sites in a wide range, from less than 2.0 Bq m−3 to approximately 40 Bq m−3 in Krakow and 35 Bq m−3 in Heidelberg. The mean 222Rn content in Krakow, when averaged over the entire observation period, was 30% higher than in Heidelberg (5.86 ± 0.09 and 4.50 ± 0.07 Bq m−3, respectively). Distinct seasonality of 222Rn signal is visible in the obtained time series of 222Rn concentration, with higher values recorded generally during late summer and autumn. The surface 222Rn fluxes measured in Krakow also revealed a distinct seasonality, with broad maximum observed during summer and early autumn and minimum during the winter. When averaged over a 5-year observation period, the night-time surface 222Rn flux was equal to 46.8 ± 2.4 Bq m−2 h−1. Although the atmospheric 222Rn levels at Heidelberg and Krakow appeared to be controlled primarily by local factors, it was possible to evaluate the "continental effect" in atmospheric 222Rn content between both sites, related to gradual build-up of 222Rn concentration in the air masses travelling between Heidelberg and Krakow. The mean value of this build-up was equal to 0.78 ± 0.12 Bq m−3. The measured minimum 222Rn concentrations at both sites and the difference between them was interpreted in the framework of a simple box model coupled with HYSPLIT (Hybrid Single Particle Lagrangian Integrated Trajectory) analysis of air mass trajectories. The best fit of experimental data was obtained for the mean 222Rn flux over the European continent equal to 52 Bq m−2 h−1, the mean transport velocity of the air masses within the convective mixed layer of the planetary boundary layer (PBL) on their route from the Atlantic coast to Heidelberg and Krakow equal to 3.5 m s−1, the mean rate constant of 222Rn removal across the top of the PBL equal to the 222Rn decay constant and the mean height of the convective mixed layer equal to 1600 m.