Investigation of OC and EC fractions of aerosol samples in Sydney area by radiocarbon analysis
Dr Bin Yang1, Dr Melita Keywood2, Dr Fabienne Reisen2, Dr A. M. Smith1, Dr Vladimir Levchenko1
1Australian Nuclear Science & Technology Organisation, Lucas Heights, Australia, 2CSIRO Oceans and Atmosphere, Aspendale, Australia
Secondary Organic Aerosols (SOA) can be a major component of atmospheric PM.2.5 pollution, emitted from natural and anthropogenic sources. SOA is formed by the oxidation of volatile organic compounds (VOC) which have biogenic and anthropogenic sources. Measurement of the radiocarbon activity of SOA allows to discriminate between these sources, as biogenic sources have a near-modern activity and anthropogenic sources are generally depleted in ¹⁴C. As part of the Sydney Particle Study [1,2], aerosol samples were collected on quartz filters using a high volume sampler fitted with a PM2.5 size selective inlet during the summer months of 2011 and autumn months of 2012. In order to estimate the apportionment of the SOA sources, we measured the radiocarbon content of organic carbon (OC) and elemental carbon (EC) fractions, using the novel method described below.
We combusted strips (90 × 35 mm) of the quartz filters strip inside a quartz tube filled with high purity oxygen at ~300 mbar at 375°C to collect the OC fraction and then at 780°C to collect the EC fraction. CO₂ gas produced during each combustion was collected in a cold trap at -170°C, volumetrically measured and transferred into a Micro Conventional Furnaces (MCF)  for graphitisation. This method was shown to be reproducible for EC and OC filter densities from the same filter. We processed 25 air filters in this way to produce 50 samples for Accelerator Mass Spectroscopy (AMS) measurement with an average of 58 µg carbon (range 10 µg to 220 µg). Our densities compared well with OC and EC densities obtained using a standard thermal desorption method at CSIRO .
We combined the measured radiocarbon activity with sophisticated chemical transport modelling, using the EC tracer method  to determine SOA. Levoglucosan was used as a tracer to allow for biomass burning events. Our results suggested that i) biogenic SOA comprised around 50% of the SOA in summer and autumn, ii) higher radiocarbon activities for OC are associated with higher SOA concentrations, supporting the model theory  that that biogenic VOCs are an important contributor to SOA in the Sydney airshed, iii) the formation of SOA involves both anthropogenic and biogenic VOC, iv) the lowest EC and OC radiocarbon activities were for summer mornings, indicating high fossil fuel carbon (i.e. vehicle emissions). Afternoons in summer and autumn displayed the highest ratios, indicating low fossil fuel carbon.
 Keywood M. et al., https://www.environment.nsw.gov.au/-/media/OEH/Corporate-Site/Documents/Air/sydney-particle-study-2010-13.pdf (2014)
 Keywood, M. et al., Earth System Science Data, 11(4), (2019). pp. 1883-1903. doi:10.5194/essd-11-1883-2019
 B. Yang, A.M. Smith, Radiocarbon V59 Issue 3 (2017) pp. 859-873, DOI: https://doi.org/10.1017/RDC.2016.89.
 Keywood M. et al., Environmental Chemistry 8(2): (2011), pp. 115-126. doi: 10.1071/en10100
Dr. Bin Yang; PhD in Physics; University of Bologna, Italy
Dr Yang has been working in the Centre of Accelerator Sciences of the Australian Nuclear Science and Technology Organisation (ANSTO) since 2008. He has developed Mark II Laser Heated Furnace and Micro Conventional Furnace for processing thousands of samples, including important and valuable samples from Antarctica. The above-mentioned apparatus have enhanced ANSTO’s capacity in the graphitisation of micro-gram samples and led to the publication of many high quality articles. He is interested in radiocarbon analysis for different types of samples such as water, land gas and aerosol for environmental research.