Quantifying metallic components in aerosol filter samples using micro-synchrotron radiation X-ray fluorescence: With quartz filter as an example

Abstract

Metals in atmospheric fine particulate matter (PM2.5) are presumed to have toxic effects. For environmental epidemiological studies, measuring the metallic components of PM2.5 collected by personal samplers and quartz fiber filters (QFF) has two key advantages: reducing exposure misclassification and enabling analyses of multiple (organic and inorganic) components on a single filter. We developed a sensitive, low-sample-consumption method to quantify the metallic components of PM2.5 collected on QFFs using micro-synchrotron radiation X-ray fluorescence (μ-SRXRF). Teflon filter is used here as a reference to test the substrate applicability of QFF. This method requires only a 5-mm-diameter portion of a QFF and can simultaneously quantify 15 metals with atomic numbers Z = 19–34 (i.e., K–Se, excluding Sc). The analysis time for each sample was 120–360 s, and the method detection limits were 0.04–36.6 ng/m3. We validated this method by performing cross-comparisons with two external methods: offline inductively coupled plasma mass spectrometry (ICP-MS) and online X-ray fluorescence (XRF). The slopes of the linear fits between the μ-SRXRF and offline ICP-MS methods varied from 0.40 to 1, with R2 values ranging from 0.62 to 0.98. The slopes of the linear fits between the μ-SRXRF and online XRF methods were between 0.29 and 0.89, with R2 values ranging from 0.39 to 0.80. We applied the proposed method to analyze 391 personal PM2.5 samples collected in an epidemiological study conducted in urban Beijing. By comparing the results with the ambient concentrations from a fixed monitoring site, we found that the concentrations of some metallic components in personal PM2.5 samples were significantly different from those in the corresponding fixed-site samples, with up to an order of magnitude difference. Because personal sampling can reduce the potential for exposure misclassifications, the high-resolution temporal analysis of metallic components in personal samples using the μ-SRXRF method could be a useful approach for exploring the health risks and identifying the sources of metallic components of PM2.5.