Black carbon and elemental carbon measurements in Milan (Po Valley, Italy) by different optical and thermal-optical methods

Carbonaceous aerosols play an important role in the atmosphere as they are responsible for visibility reduction, radiative forcing, atmospheric chemical reactions, adverse human health effects, and degradation of cultural heritage. These aerosols consist mainly of organic carbon (OC) and elemental carbon (EC, also called black carbon BC). EC/BC is a by-product of incomplete combustion of carbonaceous fuels and it is one of the most efficient absorber of solar light amongst the aerosol components, while organic carbon is a light-scatterer. Carbon measurements, beyond problems due to sampling artefacts show also analytical uncertainties mainly ascribed to difficulties in accurately separating OC and EC/BC. A universal methodology for the determination of the two components EC/BC and OC is still not available. In fact, OC and EC/BC are operationally defined by the analytical method and, consequently, different methods can produce different results (Schmidt et al., 2001). In the frame of a national project and an INFN-experiment aiming at the study of the carbonaceous components of atmospheric aerosol, the authors set-up two different laboratory systems based on a polar photometer and on a multi-wavelengths transmission photometer to perform particle light absorption measurements on filter supports with different typology (i.e. quartz fibre filters, PTFE filters, Nuclepore filters). To validate these methodologies a set of laboratory standards were generated using mixtures of Black Carbon and (NH4)2SO4 in different proportions. Field measurements were performed in parallel using a Multi Angle Photometer (Thermo, mod. 5012) and an Aethalometer (Magee Scientific, mod. AE-42). Thermal Optical Transmittance (TOT, Sunset analyser) analysis was also carried out on quartz fibre filters using the NIOSH5040 protocol (Birch and Cary, 1996). The aim of this work was three-fold: 1) set up laboratory systems to assess aerosol light absorption properties directly on filters used in standard PM monitoring campaigns; 2) compare techniques for measuring EC and BC, and determine their uncertainties; 3) assess, for the first time in Milan, BC concentration with high time resolution. During two weeks in winter 2009, a measurement campaign was carried at the urban background station located in the campus of the University of Milan (Po valley, Italy). Parallel PM10 samplings were carried out on PTFE and quartz fibre filters during nighttime and daytime and continuous BC measurements were running during the same period (5 min averaged data). The PTFE and quartz fibre filters were analysed using the laboratory systems; the polar photometer results were obtained applying the same radiative transfer calculations as described in Petzold and Schönlinner (2003) and used by the MAAP. The inter-comparison on BC (using the same factor for conversion as done by MAAP) showed a good agreement between the laboratory systems and the correspondent on-line system. However, differences were observed between particle light absorption values obtained by the polar photometer on PTFE and quartz fibre filters highlighting a different role of the filter matrix in the model used for multiple scattering corrections.