Reapportioning the sources of secondary components of PM2.5: combined application of positive matrix factorization and isotopic evidence

Secondary particles account fora considerable proportion of fine particles (PM2.5) and reasonable reapportioning them to primary sources is critical for designing effective strategies for air quality improvement. This study developed a method which can reapportion secondary sources of PM2.5 solved by positive matrix factorization (PMF) to primary sources based on the isotopic signals of nitrate, ammonium and sulfate. Actual PM2.5 data in Beijing were used as a case study to assess the feasibility and capacity of this method. In the case, 20 chemical components were used to apportion PM2.5 sources and source contributions of nitrate were applied to reapportion secondary source to primary sources. The model performance was also estimated by radiocarbon measurement (C-14) of organic (OC) and elemental (EC) carbons of eight samples. The PMF apportioned seven sources: the secondary source (36.1%), vehicle exhausts (18.7%), industrial sources (13.6%), biomass burning (11.4%), coal combustion (8.10%), construction dust (7.93%) and fuel oil combustion (424%). After the reapportionment of the secondary source, vehicle exhausts (28.7%) contributed the most to PM2.5, followed by biomass burning (25.1%) and industrial sources (18.9%). Fossil oil combustion and coal combustion increased to 8.00% and 11.4%, respectively, and construction dust contributed the least. The average gap between contributions of identified sources to OC and EC and the C-14 measurements decreased 2.5 +/- 12% after the reapportionment than 132 +/- 10.8%, indicating the good performance of the developed method. (C) 2020 Elsevier B.V. All rights reserved.

Automated summary

This study developed a method to reapportion secondary sources of PM2.5 based on isotopic signals of nitrate, ammonium, and sulfate, and validated it using actual PM2.5 data in Beijing. After reapportionment, major sources of PM2.5 included vehicle exhausts, biomass burning, and industrial sources, showing good performance of the method.