No abstract

Various approaches to infrared multicomponent quantitative analysis including K-matrix, multivariate least-squares, principal component regression (PCR), and partial least-squares (PLS) are compared. The advantages and disadvantages of each are discussed. A particular implementation of the PLS method is detailed, with emphasis on the methods provided for calibration optimization and evaluation.

No abstract

No abstract

An implementation of the PLS (partial least-squares) statistical approach to quantitative analysis was applied to a set of mid-infrared spectra obtained from a series of commercial detergent samples. The components analyzed included the base detergent, sodium benzoate, 2-propanol, 1,2-propanediol, polypropylene glycol, and glycerol. The samples were analyzed with the use of a horizontal attenuated total reflectance (ATR) accessory equipped with a zinc selenide crystal. The PLS model created in the calibration was found to provide excellent results for a set of validation samples.