Excitation emission matrix fluorescence spectroscopy and parallel factor framework-clustering analysis for oil pollutants identification

Serious ecological damage can be caused due to increased oil pollution. Identifying the source of oil can inform effective mitigation strategies and policies. A novel method for oil pollutants identification has been presented based on excitation emission matrix (EEM) fluorescence spectroscopy and parallel factor framework-clustering analysis (PFFCA). First, the EEM spectroscopy of the oil samples was measured by a FS920 steady-state fluorescence spectrometer. Second, EEM was analyzed and characterized by PFFCA. A total 90 EEM were decomposed into six components using excitation wavelengths from 260 to 400 nm and emission wavelengths from 280 to 450 nm. Finally, oil samples were classified and matched by using concentration vectors. The results were compared with those obtained by using linear discriminant analysis (LDA) employing parallel factor analysis (PARAFAC) scores, and partial least squares discriminant analysis (PLS-DA). The best classification result was obtained by using LDA employing concentration vectors with 96.7% accuracy. The results indicate that PFFCA-LDA offers a robust approach for the oil samples, which is of great significance in discrimination of oil pollutants. (C) 2021 Elsevier B.V. All rights reserved.

Automated summary

A novel method for oil pollutants identification using fluorescence spectroscopy and clustering analysis was proposed, showing significant potential in identifying the source of oil pollution. By decomposing excitation emission matrix and comparing concentration vectors, accurate classification of oil samples was achieved, providing a robust approach for discrimination of oil pollutants.