Articles | Volume 9, issue 1
Atmos. Meas. Tech., 9, 103–114, 2016
Atmos. Meas. Tech., 9, 103–114, 2016

Research article 18 Jan 2016

Research article | 18 Jan 2016

Measurement of nonvolatile particle number size distribution

G. I. Gkatzelis1,2, D. K. Papanastasiou1,2, K. Florou1,2, C. Kaltsonoudis1,2, E. Louvaris1,2, and S. N. Pandis1,2,3 G. I. Gkatzelis et al.
  • 1Institute of Chemical Engineering Sciences, ICE-HT, Patras, Greece
  • 2Department of Chemical Engineering, University of Patras, Patras, Greece
  • 3Department of Chemical Engineering, Carnegie Mellon University, Pittsburgh, USA

Abstract. An experimental methodology was developed to measure the nonvolatile particle number concentration using a thermodenuder (TD). The TD was coupled with a high-resolution time-of-flight aerosol mass spectrometer, measuring the chemical composition and mass size distribution of the submicrometer aerosol and a scanning mobility particle sizer (SMPS) that provided the number size distribution of the aerosol in the range from 10 to 500 nm. The method was evaluated with a set of smog chamber experiments and achieved almost complete evaporation (> 98 %) of secondary organic as well as freshly nucleated particles, using a TD temperature of 400 °C and a centerline residence time of 15 s.

This experimental approach was applied in a winter field campaign in Athens and provided a direct measurement of number concentration and size distribution for particles emitted from major pollution sources. During periods in which the contribution of biomass burning sources was dominant, more than 80 % of particle number concentration remained after passing through the thermodenuder, suggesting that nearly all biomass burning particles had a nonvolatile core. These remaining particles consisted mostly of black carbon (60 % mass contribution) and organic aerosol (OA; 40 %). Organics that had not evaporated through the TD were mostly biomass burning OA (BBOA) and oxygenated OA (OOA) as determined from AMS source apportionment analysis. For periods during which traffic contribution was dominant 50–60 % of the particles had a nonvolatile core while the rest evaporated at 400 °C. The remaining particle mass consisted mostly of black carbon with an 80 % contribution, while OA was responsible for another 15–20 %. Organics were mostly hydrocarbon-like OA (HOA) and OOA. These results suggest that even at 400 °C some fraction of the OA does not evaporate from particles emitted from common combustion processes, such as biomass burning and car engines, indicating that a fraction of this type of OA is of extremely low volatility.

Short summary
A method for the measurement of the nonvolatile atmospheric particle size distribution is developed and tested. The tests include laboratory experiments with biogenic and anthropogenic secondary organic aerosol as well as nucleation experiments with ambient air. The method is then further tested during an ambient campaign.