This study investigated the chemical profiles of filterable particulate matter (FPM) and condensable particulate matter (CPM) from a coal-fired power plant and the sintering unit of a steel manufacturing plant, which were in turn applied to improve forensics for the pollution sources of PM2.5 in central Taiwan using the Positive Matrix Factorization (PMF) receptor model. In addition, ambient PM2.5 samples were collected from eight sites (Fengyuan, Dali, Xitun, Chungming, Shalu, Taichung EPB, Wang Gao Liao, and Changhua) during March 4 to 21, 2022. All PM2.5 samples were analyzed for water-soluble ions, organic carbon, elemental carbon, crustal elements, heavy metals, and organic markers (organic acids, PAEs, Polycyclic Aromatic Hydrocarbons (PAHs), and Levoglucosan).
The results showed that CPM was the predominant component in total PM2.5 emitted from these two stationary pollution sources, which accounted for >89%. Moreover, contribution of inorganic CPM (82%) was significantly larger than that of organic CPM (18%). The major chemical constituents in the FPM and CPM samples from these two stationary pollution sources were: (1) Coal-fired power plant: there were higher concentrations of water-soluble ions in CPM (5644 ± 1328 μg/m3) than that in FPM (186 ± 20 μg/m3), and the predominant specie was sulfate. The metal concentration in CPM (246 ± 46 μg/m3) was also significantly higher than that in FPM (95 ± 14 μg/m3). The concentrations of Zn and Na were significantly higher in CPM as compared to FPM, by 58 and 17 times, respectively. The concentrations of organic acids and PAEs were also higher in CPM than those in FPM, and the predominant species were oxalic acid, DBP, and DEHP. The major PAH compounds were naphthalene, pyrene, and benzo perylene; (2) Steel manufacturing plant: there were higher concentrations of water-soluble ions in CPM (16883 ± 2694 μg/m3) than that in FPM (1162 ± 164 μg/m3), and the predominant specie was sulfate. The metal concentration in CPM (245 ± 29 μg/m3) was lower than that in FPM (688 ± 196 μg/m3). However, the concentration of Zr was higher in CPM. The concentrations of organic markers were higher in CPM than those in FPM, and the predominant species were phthalic acid, oxalic acid, DBP, DEHP, naphthalene, and pyrene.
The average level of ambient PM2.5 was 24.5 ± 11.5 μg/m3 in the study area. The predominant chemical compounds of the ambient PM2.5 were organic carbon, nitrate, sulfate, and ammonium, which accounted for 26%, 20%, 16%, and 12%, respectively. Oxalic acid (40.8 ± 19.0 ng/m3) and Phthalic acid (39.5 ± 22.0 ng/m3) were predominant organic acids. The highest concentration of PAEs was Di(2-ethylhexyl)phthalate (DEHP) (93.9 ± 77.3 ng/m3), followed by Di-n-butyl phthalate (DBP) (42.2 ± 52.7 ng/m3). The Benzo[b]fluoranthene ＆ Benzo[j]fluoranthene (BbF＆BjF) (0.188 ± 0.129 ng/m3) and Benzo[g,h,i]perylene (BghiP) (0.137 ± 0.087 ng/m3) were predominant PAHs.
Seven pollution source factors in PM2.5 were extracted from the chemical composition dataset for the study area, where the contributions of these seven sources ranged between 12% and 17% at respective sampling sites. We found that plastic burning could also be a major contributor to PM2.5 in the study area, and further investigation is suggested. Moreover, our data showed that the dominant PM2.5 pollution sources are different on different sampling days, but the dominant pollution sources will contribute continuously for 2-3 days, which implies that the influence of these pollution sources to air quality are not a result of occational events.