Three-way data analysis of pollutant degradation profiles monitored using liquid chromatography-diode array detection

Abstract

There are a wide variety of environmental contaminants that need to be better characterized. It is important to know the degradation pathways for these pollutants in order to better understand their effect on the environment. The hydrolysis of Glean® (chlorsulfuron), a sulfonylurea herbicide, is currently under study in this laboratory. Liquid chromatography coupled with diode array detection (LC–DAD) offers a means of studying these types of reactions. With this approach, three‐way data are obtained—absorbance measurements as a function of chromatographic retention time, UV wavelength and reaction time. The direct trilinear decomposition (DTD) algorithm was applied to these data, presuming an internal trilinear structure in the data set. Upon inspection of the results, however, deviations from trilinearity were found that caused chemically meaningless results to be obtained using this approach. This difficulty was addressed by using the three‐way multivariate curve resolution–alternating least squares approach (MCR–ALS). This technique allows for the optional use of the trilinearity constraint, and some additional constraints related to the features of the chromatographic and spectral profiles can be included. In the present work the application of this approach has been evaluated for resolving the overlapped responses that arise when measuring herbicide degradation reaction profiles. Copyright © 1999 John Wiley & Sons, Ltd.

Multiway Data Analysis: Eigenvector-Based Methods

“…193 However, it can still be applied if deviations from trilinearity are slight. 193 However, it can still be applied if deviations from trilinearity are slight.…”

Evaluation of the multivariate selectivity of multi‐way liquid chromatography methods

“…Models and methods for analyzing multi-way data abound in the literature [21][22][23][24]. In particular, the PARAFAC model is commonly used to evaluate three-way and higher order data [25].…”

Trends in chemometric analysis of comprehensive two-dimensional separations

“…Although the application of multivariate curve resolution techniques to 2D separations has yet to be realized, the success of these methods for the peak deconvolution of data from separation techniques hyphenated with spectroscopic techniques indicates that they will be powerful tools for the analysis of 2D separation data [24].…”