Indirect source apportionment of methyl mercaptan using CMB and PMF models: a case study near a refining and petrochemical plant.

Identifying the sources of volatile organic compounds (VOCs) is key to air quality control and pollution prevention. Though receptor models have been widely used in source apportionment of VOCs, they are not applicable to identify the potential source of labile species. In this study, the potential source of methyl mercaptan (MeSH) near a large refining and petrochemical plant was identified using an indirect method. When wind directions were controlled, the study period was separated into two subperiods depending on the detection of MeSH. Relative contributions from potential sources were predicted by chemical mass balance model and positive matrix factorization model based on ambient concentrations of sulfur-free compounds. Both models predicted that petroleum refinery and petrochemical production were the dominant sources of VOCs in the study area. When MeSH was detected, the relative contribution from gasoline, liquefied petroleum gas, or crude oil increased by 7.4 to 26.8% point, depending on wind direction and the predictive model used, suggesting a close relationship between MeSH and the emission from petroleum refinery. Consistent with the indirect source apportionment, among the coexisting VOCs, MeSH was most highly correlated or associated with ethane, propane, isobutane, cis-2-pentente, and isoprene, which are major components of the products or by-products of petrochemical refining processes.