Premium
Center of Gravity (COG) Method as a Tool in Processing of Voltammetric Signals
Author(s) -
Jović Ozren,
Omanović Dario,
Zelić Marina,
Pižeta Ivanka
Publication year - 2015
Publication title -
electroanalysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.574
H-Index - 128
eISSN - 1521-4109
pISSN - 1040-0397
DOI - 10.1002/elan.201500065
Subject(s) - chemistry , ionic strength , stability (learning theory) , center of gravity , smoothing , analytical chemistry (journal) , algorithm , computer science , mathematics , chromatography , statistics , management , machine learning , aqueous solution , economics
The center of gravity method (COG) was applied (for the first time) in voltammetry (polarography) as a tool for very precise determination of peak potentials, and signal shifts. Basically, the adjustment of the method consists of finding the optimal fraction of the peak that contains information about peak position, either of original signal or of its 1 st or 2 nd derivatives, along with optimal selection of parameters for SavitzkyGolay smoothing of original curves and elimination of baseline influence. The principle of the method and its validation were demonstrated and checked on simulated differential pulse polarograms (DPP) representing a series of curves for the determination of stability constants of labile metal complexes (DeFord‐Hume method). It was shown that COG outperforms the classical “one‐point” method (OPM) in precision and accuracy, providing excellent results even if a very large step potential (e.g. 10 mV) was used. The problems of reduced accuracy and precision in case of curved (non‐ideal) baseline and asymmetric peaks were successfully overcome by applying COG on curves transformed by 1 st and/or 2 nd derivatives. The method was additionally examined in details on a simple experimental dataset of cadmium chloride complexes in 4 mol dm −3 ionic strength solution and on a more complex dataset of uranyl selenate complexes in 3 mol dm −3 ionic strength solution. Stability constants obtained by COG agree well with those in the literature with much better precision than the classical one‐point method (checked by standard error of the fit).