Abstract. An intercomparison of different aerosol chemical characterization techniques has been performed as part of a chamber study of biogenic secondary organic aerosol (BSOA) formation and aging at the atmosphere simulation chamber SAPHIR (Simulation of Atmospheric PHotochemistry In a large Reaction chamber). Three different aerosol sampling techniques – the aerosol collection module (ACM), the chemical analysis of aerosol online (CHARON) and the collection thermal-desorption unit (TD) were connected to proton transfer reaction time-of-flight mass spectrometers (PTR-ToF-MSs) to provide chemical characterization of the SOA. The techniques were compared among each other and to results from an aerosol mass spectrometer (AMS) and a scanning mobility particle sizer (SMPS). The experiments investigated SOA formation from the ozonolysis of β-pinene, limonene, a β-pinene–limonene mix and real plant emissions from Pinus sylvestris L. (Scots pine). The SOA was subsequently aged by photo-oxidation, except for limonene SOA, which was aged by NO₃ oxidation.

Despite significant differences in the aerosol collection and desorption methods of the PTR-based techniques, the determined chemical composition, i.e. the same major contributing signals, was found by all instruments for the different chemical systems studied. These signals could be attributed to known products expected from the oxidation of the examined monoterpenes. The sampling and desorption method of ACM and TD provided additional information on the volatility of individual compounds and showed relatively good agreement.

Averaged over all experiments, the total aerosol mass recovery compared to an SMPS varied within 80 ± 10, 51 ± 5 and 27 ± 3 % for CHARON, ACM and TD, respectively. Comparison to the oxygen-to-carbon ratios (O : C) obtained by AMS showed that all PTR-based techniques observed lower O : C ratios, indicating a loss of molecular oxygen either during aerosol sampling or detection. The differences in total mass recovery and O : C between the three instruments resulted predominantly from differences in the field strength (E∕N) in the drift tube reaction ionization chambers of the PTR-ToF-MS instruments and from dissimilarities in the collection/desorption of aerosols. Laboratory case studies showed that PTR-ToF-MS E∕N conditions influenced fragmentation which resulted in water and further neutral fragment losses of the detected molecules. Since ACM and TD were operated in higher E∕N than CHARON, this resulted in higher fragmentation, thus affecting primarily the detected oxygen and carbon content and therefore also the mass recovery. Overall, these techniques have been shown to provide valuable insight on the chemical characteristics of BSOA and can address unknown thermodynamic properties such as partitioning coefficient values and volatility patterns down to a compound-specific level.