Multivariate calibration standardization used for the reduction of standards needed for multivariate calibration in inductively coupled plasma mass spectrometry

Multivariate calibration methods such as partial least squares regression (PLS) are capable of eliminating spectroscopic interferences in inductively coupled plasma mass spectrometry. These methods often require a large number of standards for the calibration, which can be a drawback if a full recalibration is needed very often. This problem can be circumvented by using multivariate calibration standardization to adjust either the sample or the calibration model to match a given state of the instrument. Four standardization methods (two univariate methods, direct standardization, and inverse model standardization) have been investigated and compared with full and partial calibration. Direct standardization gave the best results. It allowed for up to 78% reduction of the number of standards needed in the daily routine. Inverse model standardization needed more standards to accomplish similar results, and the two univariate methods could not be used for the determination of Gd. (C) 1997 Elsevier Science B.V.