Method for selection of optimum wavelengths for multicomponent spectrophotometric determinations

Abstract

A method for selecting optimum wavelengths (OWL) in multicomponent spectrophotometric determinations has been developed. The algorithm and criteria proposed allow selection of OWL, the value of which is equal to the component number. The method was applied to the Al3+-Ox, F@+-Ox and Ti02+-H202-Ox (where Ox is 8-hydroxyquinoline) chemical systems. About 5% relative prediction error was obtained for concentrations from 0.009 to 0.036 mmol dm-3.

Detection and Correction of Biased Results of Individual Analytes in Multicomponent Spectroscopic Analysis

“…The selection of the best sensors for quantification has been the subject of much recent interest. Criteria based on the condition number of S T S (superscript T means “transposed”, and S is the calibration matrix), the determinant 2 of S T S , the trace − of ( S T S ) -1 , and the accuracy and selectivity, among others, , have been optimized with different methods that include genetic algorithms 5 and simulated annealing . Recently, Xu and Schechter showed that the sensors that give more noise than signal may spoil the prediction results, and they developed a new criterion for wavelength selection.…”

Ultraviolet and Light Absorption Spectrometry

“…Validation of ultraviolet and visible methods using computer optimization techniques has been discussed (247). Two papers address the subject of computer-optimized wavelength selection for multicomponent determinations (248,249). The solving of ultraviolet spectra vectors via stepwise regression has permitted the simultaneous determination of multiple-component systems with indefinite components (250).…”

Aplicação de algoritmos genéticos na seleção de variáveis em espectroscopia no infravermelho médio: determinação simultânea de glicose, maltose e frutose

“…Recentemente tem-se observado uma tendência na aplicação do algoritmo genético na seleção de variáveis 17 , entretanto, vale ressaltar que existem outros métodos de seleção de variáveis [18][19][20][21] . A prefe-rência pelo algoritmo genético deve-se a sua eficiência, versatilidade e robustez 17 .…”