Spectral renormalization for multi-component fourier transform infrared absorbance data: Importance to multivariate calibration and quantitative exploratory chemometrics studies

Abstract

Although infrared spectroscopy is a very common analytical tool in the chemical sciences, quantitative infrared spectroscopic studies of multi-component solutions (particularly on-line or in-line studies) have been hampered by the lack of a concise and robust numerical approach. Using an inert chemical species as internal standard, the usual absorbance ratio analysis is generalized into a mathematical form that converts typical absorbance measurement A into its renormalized absorbance Arenorm, with the latter directly relating to the absolute moles rather than the concentration of the constituents. The renormalized absorbance has a number of very important properties, including (1) insensitivity to path length changes due to variations in temperature or pressure of the Fourier transform infrared (FT-IR) cell and (2) insensitivity to solution volume changes arising from reaction, chemical transport, or thermodynamic conditions. The methodology is first applied to a simple model test system to demonstrate its utility for multivariate calibration, and then re-applied to a real problem: the calibration of the rhodium carbonyl hydride HRh(CO)4, a long-sought species in organometallic chemistry and homogeneous catalysis (Angewandte Chemie Int. Ed., vol. 41, 3786 (2002)). The utility of the developed methodology for multivariate calibration and exploratory chemometric studies is demonstrated. Emphasis is placed on providing a numerical recipe for the practicing analytical chemist. © 2007 Society for Applied Spectroscopy.