Chemometric-assisted fast quantification and source apportionment of PAHs in PM10 using gas chromatography-mass spectrometry

ABSTRACT

In this study, PM10 samples at four locations of Loudi, China were collected from January 2017 to March 2017 and December 2017 to February 2018 at first. Then, a new analytical strategy of second-order calibration based on the alternating trilinear decomposition (ATLD) coupled with gas chromatography-mass spectrometry (GC-MS) was firstly developed for quantitative analysis of seven polycyclic aromatic hydrocarbons (PAHs) in these samples. This strategy is more sensitive and efficient than traditional chromatography not only, but also it can resolve the problems of baseline drift and peaks overlapping in chromatographic analysis. Without the need for intricate pretreatment process and any derivation reactions, useful qualitative and quantitative information of seven PAHs was quickly extracted from the GC-MS-Sample data array by replacing chemical or physical separation with mathematical separation. In validation samples, the average recoveries of seven PAHs ranged from (96.7 ± 4.8) % to (107.9 ± 4.2) %, and the root-mean-square errors of prediction (RMSEPs) were within the range of 0.03–0.35 μg mL⁻¹. The highest concentrations of individual PAHs in these samples were for phenanthrene (1.34 μg mL⁻¹ in monitoring station), acenaphthylene (0.79 μg mL⁻¹ in municipal government) and chrysene (0.53 μg mL⁻¹ in monitoring station). Identified by component analysis, coal combustion and vehicle exhaust were the most possible sources of Loudi, China. These results demonstrated that the proposed method is characterised as a more precise, simple and promising analytical strategy to identify and quantify PAHs in actual aerosol.

KEYWORDS:

• PM10
• Polycyclic aromatic hydrocarbons
• GC-MS
• second-order calibration

Acknowledgments

The authors gratefully acknowledge the National Natural Science Foundation of China (Grant No. 21707032 and 31701689), the Research Foundation of Education Bureau of Hunan Province, China (Grant No.16A109) and School-level Supports for National Foundation Cultivation of Hunan University of Humanities, Science and Technology (No.2016PY04) for financial supports.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [No. 21707032 and 31701689]; School-level Supports for National Foundation Cultivation of Hunan University of Humanities, Science and Technology [No.2016PY04]; Research Foundation of Education Bureau of Hunan Province, China [No.16A109].