Rapid Analysis of Oil Species Using Excitation-Emission Matrix and PARAFAC Analysis

Abstract

For oils (heavy oil A, light oil, kerosene) with the highest number of spill accidents in the Yodo River basin, we attempted rapid discrimination and simple quantification using the Excitation-Emission Matrix (EEM) and Parallel Factor Analysis (PARAFAC). However, it was considered that these oils could be separated into three fluorescent components and could be distinguished by the component score ratio. In addition, a high correlation was observed between the oil addition concentration and the component score, and it was considered that the oil spilled during a water quality accident could be easily quantified using these regression equations. Furthermore, it was suggested that oil that has flowed into raw water and has changed over time may be able to be quickly and easily discriminated by using the components separated by PARAFAC analysis.