Modelling the matrix interference of iron in the multivariate determination of copper by stripping voltammetry - Instrument standardization

Two different standardization procedures have been applied in order to mini mize the experimental effort necessary to determine copper in presence of i ron by differential pulse anodic stripping voltammetry. The significant mat rix interference caused by iron in the voltammetric determination of copper has been successfully solved by using partial least squares (PLS) regressi on. The use of this multivariate regression implies a large number of train ing set samples, so a standardization method was required. In fact, the sta ndardization methods used have reduced the calibration samples needed in fu ture determinations by up to 75%. Moreover, PLS regression allowed both met als to be determined simultaneously by using an adequate experimental desig n, without any limitation to their respective concentrations and without th e need to know the level of interference. (C) 1999 Elsevier Science B.V. Al l rights reserved.