Processing multi-way chromatographic data for analytical calibration, classification and discrimination: A successful marriage between separation science and chemometrics

Abstract

Recent research works on multi-way chromatographic data for analytical calibration, classification and discrimination are reviewed. Focus is directed towards measured data arrays in the form of matrices, three- and four-dimensional mathematical objects, depending on the number of elution time and instrumental detection modes. Chemometric models typically used to process these data and to obtain the maximum amount of information on the studied systems are discussed. The advantages in processing such data for complex samples are highlighted, both for quantitative and qualitative purposes. In the former case, the achievement of the second-order advantage which permits analyte quantitation in the presence of uncalibrated constituents is perhaps the most relevant contribution to this field. For classification and discrimination, the processing of multi-way chromatographic data provides highly compressed information which can then be submitted to appropriate algorithms. This represents an additional advantage, because individual analytes do not need to be fully resolved and quantitated.