Multivariate data analysis as a tool for evaluating emission intensity, background equivalent concentration and detection limit obtained for different plasma positions in direct current plasma-atomic emission spectrometry

Abstract

Relative emission intensity, background emission concentration (BEC) and detection limit (DL) obtained for different analytes and different plasma positions are examples of multivariate data sets. The observations can be related to the emission distribution in the plasma for the different elements (the spatial profiles). Principal component analysis (PCA) as a tool for modelling, interpretation and visualisation of such data sets was applied (i) to elucidate the data structure caused by the profiles, (ii) to enhance structural information using replicate or similar data sets, (iii) to predict model results that are less prone to errors and random variations, and (iv) to compare data sets of different origin (e.g. directly observed results with those calculated from the profiles). The selection of a suitable optimisation element can be guided by visual procedures or rather simple calculations based on the PCA model.