RESUMEN
The aim of this paper is to determine the concentration of three dyes throughout the tanning process of leather using vegetable tanning agents with a sequential injection analyser with second-order data treatment. As the vegetable tanning agents used are highly absorbent species, we focus on three aspects: (i) difficulties with the resolution (ii) the reduction in the working concentration range; and (iii) matrix effects. Ideally, second-order instruments provide "second-order advantage"; i.e. calibration is possible in the presence of uncalibrated interfering species. However, if the interfering species change the instrumental response of the analyte (in scale or shape), standard additions must be used to ensure the accuracy of the estimated analyte concentration. Here we study the presence of matrix effects for three dyes in several samples in order to significantly improve the accuracy of predictions in the presence of such effects. We found that there were matrix effects in at least 80% of the samples with an alpha risk of 5%. We used this method to study the exhaustion of dyes in the dyeing process.
RESUMEN
This paper describes how artificial neural networks can be used to classify multivariate data. Two types of neural networks were applied: a counter propagation neural network (CP-ANN) and a radial basis function network (RBFN). These strategies were used to classify soil samples from different geographical regions in Brazil by means of their near-infrared (diffuse reflectance) spectra. The results were better with CP-ANN (classification error 8.6%) than with RBFN (classification error 11.0%).
