login

Multivariate Calibration

Published 1 January 1984
Harald Martens, Tormod Næs
Citations597

Abstract

Determination of chemical concentrations can be made more rapid and reliable by combining information from several measurement variables, e.g. light absorbances at several wavelengths. Systematic errors can thereby be eliminated so that very unspecific measurement data can be used for quantitative determinations. Automatic warning of unexpected errors is also possible. In addition, multivariate calibration can give new information on how the chemical components behave in situ in the sample. Near InfraRed (NIR) reflectance analysis of wheat flour is used as an example.

Keywords

Chemistry