Ambient carbonyl compounds play an important role in tropospheric atmospheric chemistry. Primary emissions and photochemical formation are both sources of carbonyls, and therefore it is challenging work to analyze their sources. In this study, carbonyl sources were apportioned using the source tracer ratio method (STR) and positive matrix factorization model (PMF) based on offline carbonyls observations at a site in Nanjing during March 2017. Eleven carbonyl compounds were detected, and their total concentrations were in the range of 2.57×10-9-22.83×10-9. Formaldehyde, acetaldehyde, and acetone were the main components, accounting for 36.8%, 21.6%, and 18.5% of the average concentration of eleven carbonyl compounds, respectively. The influences of tracer selection and background concentrations on the results of source apportionment using the STR method based on comparing the results of acetylene and toluene as tracers and the 5th and 10th percentages as background concentrations are presented. Five sources were resolved by PMF, including traffic emission, the petrochemical & chemical industry, paint & solvent use, secondary formation & background, and the chemical industry. Secondary formation & background sources were the largest contributors of carbonyl compounds, contributing 56.4%, 48.2%, and 58.3% to formaldehyde, acetaldehyde, and acetone, respectively. By comparing the carbonyl source apportionment results by STR and PMF, it was found that the STR depends on the selection of tracers. When the STR is applied in the areas with complex sources, it is difficult to use a tracer to indicate anthropogenic source emissions, and therefore it is not a suitable method for carbonyl source apportionment.
Keywords: Nanjing; carbonyl compounds; positive matrix factorization model (PMF); source apportionment; source tracer ratio method (STR).