Four-dimensional data coupled to alternating weighted residue constraint quadrilinear decomposition model applied to environmental analysis: Determination of polycyclic aromatic hydrocarbons

• Published in: Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy Vol. 193; pp. 507 - 517
• Main Authors: Liu, Tingting, Zhang, Ling, Wang, Shutao, Cui, Yaoyao, Wang, Yutian, Liu, Lingfei, Yang, Zhe
• Format: Journal Article
• Language: English
• Published: England Elsevier B.V 15.03.2018
• Subjects: Alternating weighted residue; Constraint quadrilinear; Decomposition; Four-dimensional spectral data; Polycyclic aromatic hydrocarbons; Recovery rate; Spectrum; The wavelet transform; Three-dimensional fluorescence; Alternating weighted residue; The wavelet transform; Polycyclic aromatic hydrocarbons; Four-dimensional spectral data; Constraint quadrilinear; Three-dimensional fluorescence; Recovery rate; Decomposition; Spectrum
• ISSN: 1386-1425; 1873-3557
• DOI: 10.1016/j.saa.2017.12.003

Qualitative and quantitative analysis of polycyclic aromatic hydrocarbons (PAHs) was carried out by three-dimensional fluorescence spectroscopy combining with Alternating Weighted Residue Constraint Quadrilinear Decomposition (AWRCQLD). The experimental subjects were acenaphthene (ANA) and naphthalene (NAP). Firstly, in order to solve the redundant information of the three-dimensional fluorescence spectral data, the wavelet transform was used to compress data in preprocessing. Then, the four-dimensional data was constructed by using the excitation-emission fluorescence spectra of different concentration PAHs. The sample data was obtained from three solvents that are methanol, ethanol and Ultra-pure water. The four-dimensional spectral data was analyzed by AWRCQLD, then the recovery rate of PAHs was obtained from the three solvents and compared respectively. On one hand, the results showed that PAHs can be measured more accurately by the high-order data, and the recovery rate was higher. On the other hand, the results presented that AWRCQLD can better reflect the superiority of four-dimensional algorithm than the second-order calibration and other third-order calibration algorithms. The recovery rate of ANA was 96.5%~103.3% and the root mean square error of prediction was 0.04μgL−1. The recovery rate of NAP was 96.7%~115.7% and the root mean square error of prediction was 0.06μgL−1. [Display omitted] •Fluorescence Spectrometer•The mixture of organic contaminants (PAHs)•Compressed fluorescence spectra, reduce redundant information•Classification of mixtures of organic contaminants (PAHs) by AWRCQLD•Quantitative analysisof mixtures of organic contaminants (PAHs) by AWRCQLD