Multicomponent flow injection analysis and quantitative infrared emission spectroscopy: chemometric applications

Abstract

Chemometrics involves the use of mathematical and statistical methods to provide maximum information from chemical data. A number of methods and algorithms have been developed which have been found useful in the analysis of chemical data. The developments in this volume include the use of chemometric methods in two areas which have generally been overlooked in the past. The first is the use of flow injection analysis (FIA) methods for mixtures using non-specific detectors. The Self-Modeling Curve Resolution method of Lawton and Sylvestre is used to distinguish analytes from background absorber in the sample using a UV/Visible absorbance detector. The second is infrared emission spectroscopy which, although currently used in a few applications, should become a much more generally useful tool when combined with the power of multivariate calibration methods. The Partial Least Squares method is applied to emission spectra for estimation of sample composition, thickness and temperature.