Quantitative Study On Fluorescence Spectroscopy Of Polycylic Aromatic Hydrocarbons In Soils

Polycyclic aromatic hydrocarbons(PAHs)are a group of chemicals containing two or more single or fused aromatic rings,which can cause cancer and transgenation.Rapid detection of PAHs is of great practical significance for the investigation of pollution distribution,the treatment of contaminated sites,and the emergency treatment of pollution accidents in soils.Traditional analytical methods for PAHs in soils,such as GC/MS,HPLC,usually need pretreatment and chromatographic separation.The pretreatment of samples(including extraction,concentration,and purification)is complicated costly,some steps being possibly harmful to the operator's health.It is not suitable for rapid in-site detection.Fluorescence spectroscopy has become a powerful analytical technique for organic contaminants for their high sensitivity,rapid and multicomponent analysis without sample preparation.However,this technique still has the problems of quantitative analysis caused by the weak fluorescence intensity,strong interference of unknown coexisting components and heavy overlap of spectra of the components.Therefore,quantitative study on PAHs in soils based on fluorescence spectral data has great theory significance and utility value for in-site monitoring.The main results obtained from this paper are as follows:(1)A laser-induced fluorescence spectra detection system for PAHs is built,a method for enhancing PAHs fluorescence signal based on surfactants is studied.The LODs of optimized system are 19.6mg/kg(anthracene),3.53mg/kg(pyrene)and 15.9mg/kg(phenanthrene),which can meet the requirements of the current screening and control values of PAHs in soils,and are also lower than the reported values in foreign literatures.Surfactant SDS can enhance the intensity of lasers-induced fluorescence of PAHs in soils.With the addition of 1%SDS,LODs of anthrance,pyrene,and phenanthrene are decreased to 16.5mg/kg,1.47mg/kg and 10.4mg/kg,respectively.The correlation coefficients between fluorescence intensity and concentration are above 0.99.(2)Quantification methods based on weighted non-negative least squares multivariate linear regression(MRL)and support vector regression(SVR)are proposed for quantifying multicomponent PAHs in soils.The results show that weighted non-negative least squares multivariate linear regression can significantly improve the accuracy of prediction results of bicomponent PAHs compared with univariate regression.The mean relative errors of phenanthrene and anthracene decrease from 24.7%and 23.5%to 11.8%and 11.7%,respectively.However,the mean relative errors of phenanthrene and anthracene are still 20.8%and 23.1%in multicomponent PAHs.In the SVR model,PSO is improved based on the nonlinear decreasing inertia weight and the asynchronous linear change of the acceleration factor,and GWO is optimized by combining with differential evolution(DE).The improved parameter optimization processes have a faster convergence rate,and results obtained by the SVR model are better than the linear regression analysis.Among them,the average relative errors of anthracene,pyrene and phenanthrene in the GWO-DE-SVR model do not exceed 6.9%.(3)Quantitative algorithm of PAHs in the complicated soil environment based on 3DEEM is studied.Identification and quantitative analysis methods of PAHs based on nonsmooth nonnegative matrix factorization and U-PLS/RBL are proposed.In order to identify the components of PAHs in soils,nonsmooth nonnegative matrix factorization is used to interpret 3DEEM data.The results show that nsNMF could extract spectral components effectively in the presence of unknown interference,the coefficients of similarity are greater than 0.90.Quantitative analysis of multicomponent PAHs in complex background soils based on U-PLS/RBL algorithm show that 3DEEM combined with U-PLS/RBL algorithm have a second-order advantage and could effectively predict the contents of multicomponent PAHs in soils with different regions.The average relative errors of PAHs do not exceed 13.3%,and the root mean square errors are less than 53.4mg/kg.The 2D correlation spectra constructed based on 3 DEEM data can further reduce background interference.The average relative errors do not exceed 12.0%,and the root mean square errors are less than 49.6mg/kg.(4)The fluorescence emission characteristics of PAHs under different physical and chemical parameters(humidity,minerals,organic matters)are analyzed.The quantitative effects of soil moisture and organic matter on PAHs and their correction methods are studied.The results show that the soil moisture and organic matters have a greater influence on the fluorescence emission intensity of PAHs.Using the humidity as the external disturbance,a laser-induced fluorescence synchronous and asynchronous 2D correlation spectra model of soil PAHs is constructed,and a correction model for quantitative influence of soil moisture is established based on the reflected light at 532nm.The accuracy of quantitative analysis of soil PAHs under different humidity has been improved.The rapid quantitative analysis of soil PAHs based on standard incremental extrapolation can reduce the impact of organic matters.PAHs can be rapidly detected in soil samples with different organic matter contents in the case of non-calibration.