Evaluation Techniques for Two-Way Data from in Situ Fourier Transform Mid-Infrared Reaction Monitoring in Aqueous Solution

Abstract

A model chemical reaction was monitored with in situ Fourier transform mid-infrared spectroscopy using an attenuated total reflectance probe. The evaluation of the IR spectra is complicated by the fact that the reaction runs in nonisothermal aqueous solution with large variations in pH. Despite this, it was possible to extract large amounts of useful information on the reaction after suitable pretreatment of the spectra. Alternating least-squares (ALS) multivariate curve resolution is shown to be a useful technique for obtaining pure component spectra and concentrations if suitable spectral regions are analyzed. Rank mapping methods are used as the basis for this sectioning into smaller regions. Techniques for finding and analyzing selective spectral regions are also shown to be applicable to this type of data. Partial least-squares (PLS) regression models based on spectral data were used to verify the results where possible. The correlation between the concentrations predicted from PLS and ALS is excellent.