Intercomparison between a single particle soot photometer and evolved gas analysis in an industrial area in Japan: Implications for the consistency of soot aerosol mass concentration measurements

摘要

Mass concentrations of soot (typically comprising black and elemental carbon; BC and EC, respectively) aerosols, were measured at Yokosuka city, an industrial region in Japan in the early summer of 2014. The results of laser-induced incandescence (LII) and evolved gas analysis (EGA) techniques were compared using a single particle soot photometer (SP2) and semi-continuous elemental/organic carbon analyzer (EC/OC analyzer), respectively. We revisited the procedure of SP2 calibration with a focus on investigating the relationship between LII intensity (S-LII) and refractory BC (rBC) mass per particle (m(PP)) for some BC-proxies in the laboratory, as well as for ambient rBC particles in order to discuss the uncertainty of the SP2. It was found that the m(PP)-S-LII for the fullerene soot and carbon black particles agreed well within 3% and 10%, respectively, with that for ambient rBC particles. This is the first time to suggest the use of carbon black as a reference material. We also found that the m(PP)-S-LII for the aqueous deflocculated Acheson graphite particles with the correction factor given by Baumgardner et al. (2012) was still biased by around +20% to that for ambient rBC particles. EC quantified by the semi-continuous EC/OC analyzer using a thermal-protocol similar to that of Interagency Monitoring of Protected Visual Environments (IMPROVE-like), systematically showed higher concentrations than rBC measured by the SP2. The uncertainties related to SP2 cannot fully account for this difference. This result was likely caused by the contribution of charred organic materials to EC, which can be affected significantly by thermal-protocols for the EGA, The consistency and differences between rBC and EC are discussed with regard to comparing their respective mass concentrations. (C) 2015 Elsevier Ltd. All rights reserved.