The objectives of this study are to develop analytical methods for determination of isotopes in fine particulate matters (PM2.5), identify potential sources of PM2.5 in the central-southern Taiwan with the isotopic fingerprints, and compare the difference between isotopes and PMF model in pollution source identification. δ13C, 14C, and lead isotopes were analyzed by Inductively Coupled Plasms Mass Spectrometer (IRMS) with Gas Chromatograph (GC), Accelerator Mass Spectrometer (AMS), and Inductively Plasma Mass Spectrometer (ICP-MS), respectively. PM2.5 samples in the atmospheres were sampled in Changhua, Erlin, Zhushan, Douliu, Taixi, Mailiao, Lunbei, Chiayi, and Xingang stations in the fall of 2016 and winter and summer of 2017. Moreover, in order to establish isotope fingerprints of specific sources, industrial (coal-fired power plant and oil plant), traffic, and biomass burning emissions were collected and analyzed for water-soluble ions, organic carbon, elemental carbon, metals, and isotopic compositions. Our data showed that δ13C, 14C, and lead isotopes were detected with high precision and accuracy by IRMS with GC, AMS, ICP-MS instruments, respectively. Chemical compositions of PM2.5 in the atmospheres displayed that nitrate, sulfate, ammonium, and total carbon, which accounted for 15%, 17%, 10%, and 20% of the PM2.5 mass, respectively, were the predominant species. Moreover, secondary aerosol (sulfate, nitrate, and ammonium) and modern carbon have higher contributions in high PM2.5 concentration event days. These results implied that photochemical reactions and biomass burning were important contributors of PM2.5 in the central-southern Taiwan. The sulfates were predominant composition of PM2.5 from coal-fired power and oil plants, the fossil carbon was an important carbon source (pMC < 50%), and the 206Pb/207Pb (208Pb/207Pb) ratios were ranged from coal-fired plant and oil plant on 1.1604-1.12229 (2.4141-2.4806) and 1.1528-1.2021 (2.4145-2.4425), respectively. In the part of biomass burning, organic carbon was a predominant component in PM2.5 and the ranges of δ13C and percent Modern Carbon (pMC) were -35 to -30.1‰ and 97 to 98%, respectively. Elemental carbon was a major component of PM2.5 from traffic emission and the ranges of δ13C and pMC were -28.6 to -26.3‰ and 7 to 21%, respectively, and the 206Pb/207Pb (208Pb/207Pb) ratios were ranged on 1.1329-1.1508 (2.4078-2.4294). The average δ13C values in PM2.5 were -27.8‰ (-30.4 to -25.6‰), -28.7‰ (-34.7 to -26.6‰), and -32.0‰ (-38.5 to -27.8‰) in the fall, winter, and summer, respectively. The results of δ13C suggested that industrial and traffic emissions were important carbon sources in the fall and winter, and biomass burning also contributed the carbon level in winter. The secondary aerosol was an important carbon source in summer. Moreover, the pMC results indicated that modern and fossil carbon were nearly equivalent on average in three seasons. The 206Pb/207Pb ratios were 1.1457 (1.1247 to 1.1728), 1.1534 (1.1407 to 1.1710), and 1.1465 (1.1194 to 1.1635) in the fall, winter and summer, respectively, and 208Pb/207Pb ratios were 2.4303 (2.4025 to 2.4577), 2.4323 (2.4168 to 2.4513), and 2.4208 (2.3936 to 2.4599). However, we need more data from pollution sources for more accurately identifying the lead sources in the atmospheres. Finally, this study used both isotopic compositions and PMF model to identify the pollution sources of PM2.5. Our data showed that the carbon sources of PM2.5 were similar from both source identification methods, however, the lead sources were different. These results suggested that more data are necessary from different pollution sources for source identification.

PM2.5, Isotope, Pollution Source Identification