Multicomponent Determination of Chlorinated Hydrocarbons Using a Reaction-based Chemical Sensor. Part 2. Chemical Speciation Using Multivariate Curve Resolution

Abstract

A new multivariate curve resolution method that can extract analytical information from UV/visible spectroscopic data collected from a reaction-based chemical sensor is proposed. The method is demonstrated with the determination of mixtures of chlorinated hydrocarbons by estimating the kinetic and spectral profiles of the chemical species formed in the Fujiwara reaction. The three key aspects of the proposed method are (1) the initial estimation of the kinetic concentration profiles from evolving factor analysis; (2) the implementation of an alternating and constrained least squares method to optimize the determination of both the spectral and concentration profiles of the species present in the reaction, and (3) the development of a quantitative approach based on the simultaneous analysis of standards and unknowns for the determination of the initial concentration of the analytes in the mixtures