Source identification and apportionment of ambient particulate matter in Beijing using an advanced...

The use of electron microscopy to analyze the morphology, composition, and sources of atmospheric particles has been studied extensively worldwide. However, in China, there are few studies on single-particle source analysis based on computer-controlled scanning electron microscopy (CCSEM) technology for a large number of particles, and the related technical methods need to be established and improved. In this study, ambient particulate matter (PM) was collected simultaneously from urban, suburban, and background areas of Beijing in spring 2018 and subsequently characterized using the IntelliSEM-EPAS™ technology (an advanced CCSEM software). The deposition velocity model was used to deduce the size distribution and calculate the concentration of ambient PM. Based on the k-means algorithm and empirical rules, all particles investigated were quantitatively apportioned to nine major sources, including soil/road dust, carbonates-silicates, carbonates, irregular carbonaceous particles, irregular iron oxides, combustion/industry, calcium sulfate, secondary particles, and salt-related particles. The size-resolved contributions (mass and number) of different sources were calculated. For example, soil/road dust (65.1 %), carbonate-silicate (16.1 %), and carbonate (7.1 %) were the top three sources contributing to PM₁₀ mass. This study was the first localized application of IntelliSEM-EPAS technology in China, demonstrating its great promise in PM source apportionment. For further accurate and refined source apportionment, it is essential to build localized individual particle source profiles.