Epsilon Archive for Student Projects

Abstract

This study investigates the advantages and disadvantages of several destructive and non-destructive analytical methods that can be employed to characterize airborne par-ticulate matter (PM) deposited in seasonal snow. To this end, a comparative study was conducted on PM obtained from snowpack surveys in two Arctic/subarctic regions of Northern Sweden and Interior Alaska, USA. This study measured and characterized carbonaceous particles (CPs) in snow-filtered PM using three different optical analytical methods: a Particle Soot Absorption Photometer (PSAP), a Thermo-Optical Transmittance (TOT) analyser, and a Light-Absorbing Heating Method (LAHM) instrument. The study additionally tested the applicability of a portable X-Ray Fluorescence (pXRF) analyser for quantifying the particulate trace metal content in the snow. The ratios of analytes from the measurements performed with the pXRF were also compared with the ratios of elements from bulk chemical analyses by In-ductively Coupled Plasma Mass Spectrometry (ICP-MS). Snow meltwater was filtered with two filtration set-ups (syringe and vacuum-pump) on two differently sized tissuquartz filters (25 mm and 50 mm). The analysis of CPs suggests a higher correlation between the measurements of PSAP/TOT in comparison to LAHM/TOT meas-urements. The lower association between LAHM and TOT were i.a. assumed to de-rive from heterogeneities of the loading of the filter material especially on the smaller 25 mm filters produced by syringe-filtration. Organic carbon (OC) and elemental carbon (EC) values derived from TOT measurements gave insight into possible emission sources in different geographic regions of Sweden and Alaska. High OC values in rural Alaska suggested the influence of wood smoke on the particle composition in the snow. The chemical analysis of the empty and blank tissuquartz filters used for filtration showed high base elemental concentrations. Despite these high initial values of the filter matrix, an analysis of the two inbuilt calibrations ‘GeoMining’ and ‘GeoExploration’ showed that both calibrations measure relatively precisely. pXRF calibration ‘GeoExploration’ further yielded higher count rates and thus is more suit-able for trace metal analysis filters than calibration ‘GeoMining.’ The correlation of ICP-MS vs. pXRF further showed no significant relationship between the measurements of the two methods. Yet, for filters with high filter loadings the relative concentration pattern (RCP) of the pXRF measurements reflected the chemical composition of the respective sample environment. This suggests that further investigations into using pXRF for snowpack analyses could be a valuable time- and cost-effective tool for the analysis of chemical composition of snow.