δ13C signatures of organic aerosols: Measurement method evaluation and application in a source study

Analysis of the stable carbon isotope ¹³C in organic carbon (OC) can give insight into sources and atmospheric processing of carbonaceous aerosols, provided the ¹³C source signatures are known. However, only few data on ¹³C signatures of OC emitted by common sources of carbonaceous aerosol are available in Europe. We present and evaluate an improved version of a measurement method to obtain δ¹³C signatures on organic aerosols desorbed from filter samples at three different desorption temperatures (200 °C, 350 °C and 650 °C) and apply it in a source study. With our calibration approach, the reproducibility of a L-Valine reference material desorbed at a single temperature step of 650 °C shows a standard deviation of 0.19‰ over a period of more than one year. The average δ¹³C value for this reference material over 248 measurements is −24.10‰, which shows only a slight bias to the nominal value of −24.03‰. Repeated analysis of ambient filter samples desorbed at three temperature steps show typical standard deviations of about 0.3‰ for all temperature steps (200 °C, 350 °C and 650 °C). Isotopic fractionation due to partial thermal desorption during the individual temperature steps was tested on single compound reference materials. It showed significant isotopic fractionation only at temperature steps, in which a very minor fraction of the compound was desorbed. Possible isotope effects caused by charring of organic material were investigated and found to be not significant. The thermal desorption method was applied to various source filter samples from the region of Naples, Italy. We analyzed two different biomass burning sources, exhaust from a city bus and traffic emissions collected in a tunnel and compared these to ambient filter samples from the same region. δ¹³C signatures of the total OC show values in a narrow range of about −28‰ to −26‰ for all sources, which does not allow a source apportionment only based on ¹³C. Nevertheless, the results add information to a source inventory of δ¹³C, where information of ¹³C in organic aerosol from specific emission sources are rare. City bus emissions show little variation of δ¹³C over the temperature steps, whereas biomass burning aerosol is enriched in ¹³C for OC desorbed at 650 °C. For PM10 samples in the urban tunnel an enrichment in δ¹³C at the 650 °C temperature steps was observed, which is likely caused by the contribution of carbonate carbon to the carbonaceous material desorbed at this temperature step.