Mitigation of Rayleigh and Raman Spectral Interferences in Multiway Calibration of Excitation−Emission Matrix Fluorescence Spectra

Abstract

A weighted parallel factor analysis (W-PARAFAC) model is applied to excitation-emission matrix (EEM) fluorescence spectra of carbamate pesticides to aid with calibration in the presence of Raman scattering. Traditional PARAFAC inefficiently models the Raman scattering, resulting in prediction and calibration errors when a significant background is present. Four different weighting strategies were investigated and compared with subtraction of the appropriate sample background. Using a binary weighting strategy produced superior results, compared with a continuous distribution of weights. Further choice of weighting strategies, which are optimized to include either maximum analyte signal or to exclude a maximum amount of background scattering, is dependent on the degree of overlap and relative signal intensity.

A fully robust PARAFAC method for analyzing fluorescence data

“…Parallel factor analysis (PARAFAC) [3,4] is a commonly used method for modeling fluorescence excitation-emission data (EEM) [5][6][7][8][9][10]. PARAFAC decomposes the fluorescence signals X into F tri-linear components according to the number of fluorophores present in the samples [11]:…”

“…Scatter can be considered as outlying elements or outliers of the second type, because this scatter does not contain any chemically relevant information and does not follow the tri-linear model. Several ways to handle scatter in relation to PARAFAC modeling can be found in the literature [7,11,[15][16][17][18][19][20][21] but treating scatter as outlying elements leads in Engelen et al [2] to the construction of the automated scatter identification method. The approach can handle both first and second order Rayleigh scatter as well as Raman scatter and require no a priori information about the scatter itself, like for example the placement or the width of the scatter ridges.…”

An investigation into the relationship between the formation of thermal cracked components and PM reduction during diesel combustion using water emulsified fuel

“…The expansion of the fluorescence emission area suggests an increase of PAHs. In this case, the addition of chrysene, anthracene and pyrene are predicted (JiJi andBooksh, 2000, Beltrán et al, 1998). W/O20 causes extreme thermal decomposition among these fuels so that more PAHs via LHCs and other thermal decomposition substances might be produced.…”

Calibration, standardization, and quantitative analysis of multidimensional fluorescence (MDF) measurements on complex mixtures (IUPAC Technical Report)

“…For PARAFAC modelling, one can insert NaN values in the scatter regions combined with a non-negativity constraint [81]. Jiji and Booksh [82] used data point weighting to make the scattering band insignificant during trilinear decomposition, whereas Wentzell et al [83] chose to eliminate scatter by using weighted PCA on the unfolded EEM matrix before refolding and subsequent decomposition of the EEM. There are various other strategies available, including interpolation [84] and modelling [85,86].…”