Integration of optical and chemical parameters to improve the particulate matter characterization

Abstract Integrating nephelometer measurements have been combined with co-located in space and time PM10 and PM1 mass concentration measurements to highlight the benefits of integrating aerosol optical properties with the chemical speciation of PM1 and PM10 samples. Inorganic ions (SO 4 2− , NO 3 − , NH 4 + , Cl − , Na + , K + , Mg 2+ , and Ca 2+ ), metals (Fe, Al, Zn, Ti, Cu, V, Mn, and Cr), and the elemental and organic carbon (EC and OC, respectively) have been monitored to characterize the chemical composition of PM1 and PM10 samples, respectively. The scattering coefficient ( σ p ) at 450 nm, the scattering Angstrom coefficient ( A ) calculated at the 450–635 nm wavelength pair, and the scattering Angstrom coefficient difference (∆ A ) retrieved from nephelometer measurements have been used to characterize the optical properties of the particles at the surface. The frequency distribution of the A daily means during the one-year monitoring campaign, performed at a southeastern Italian site, has allowed identifying three main A variability ranges: A  ≤ 0.8, 0.8  A  ≤ 1.2, and A  > 1.2. We found that σ p and ∆ A mean values and the mean chemical composition of the PM1 and PM10 samples varied with the A variability range. σ p and ∆ A reached the highest (149 Mm −1 ) and the smallest (0.16) mean value, respectively, on the days characterized by A  > 1.2. EC, SO 4 2− , and NH 4 + mean mass percentages also reached the highest mean value on the A  > 1.2 days, representing on average 8.4, 9.8, and 4.2%, respectively, of the sampled PM10 mass and 12.4, 10.6, and 7.7%, respectively, of the PM1 mass. Conversely, σ p and ∆ A mean values were equal to 85 Mm −1 and 0.55, respectively, on the days characterized by A  ≤ 0.8 and the EC, SO 4 2− , and NH 4 + mean mass percentages reached smaller values on the A  ≤ 0.8 days, representing 4.5, 6.0, and 1.9% of the PM10 mass and 9.4, 7.3, and 5.8% of the PM1 mass, respectively. Primary and secondary OC (POC and SOC, respectively) contributions also varied with the A variability range. POC and SOC mean mass percentages reached the highest and the smallest value, respectively, on the days characterized by A  > 1.2. Conversely, POC and SOC mean mass percentages reached the smallest and the highest value, respectively, on the days characterized by A  ≤ 0.8. It has also been shown that the PM, OC, OC + EC, POC, and SOC mass scattering cross sections varied significantly with the A variability range, because of the A dependence on aerosol sources and/or emission, transport, and transformation mechanisms. Therefore, it has been shown that A daily mean values can represent a good tool to better differentiate the chemical speciation of size-fractioned PM samples.