J.M. Díaz-Cruz

Multivariate curve resolution with alternating least squares optimisation: a soft-modelling approach to metal complexation studies by voltammetric techniques

Abstract Voltammetry of metal ions in the presence of ligands is usually interpreted by means of hard-modelling approaches. Soft-modelling approaches have been scarcely applied to electroanalytical data, especially in comparison with their use in spectroscopy. In recent years, the multivariate curve resolution–alternating least squares (MCR–ALS) method, based on factor analysis, has been shown to be a powerful tool in the analysis of metal complexation processes by a variety of voltammetric techniques. In this article, an overview of the method and its applications is presented. For this purpose, either numerically simulated or experimental data corresponding to different characteristic systems are analysed. A very interesting feature of MCR–ALS is that, by using different constraints, both the pure voltammograms of the different electrochemical processes involved and the corresponding concentration profiles are easily obtained. This type of information cannot be reached by means of classical univariate data analysis techniques. From calculated concentration profiles, the model of complexation and the equilibrium stability constants are estimated.

Application of multivariate curve resolution to voltammetric data. Part 1. Study of Zn(II) complexation with some polyelectrolytes

Multivariate curve resolution is applied to the study of Zn(II) complexation with anions of polymethacrylic acid and polyacrylic acid in aqueous solution containing 0.01 M KNO3. Data matrices obtained by different electrochemical techniques (normal pulse polarography, reverse pulse polarography, differential pulse polarography and differential pulse anodic stripping voltammetry) and for different ligand deprotonation degrees have been analysed individually and simultaneously by a procedure which consists of using several chemometrical techniques based on factor analysis: principal component analysis, evolving factor analysis and multivariate curve resolution with alternating least-squares optimization. These techniques were used to investigate the number of species present in the systems, their concentration profiles at increasing ligand concentration and their individual voltammograms together with the global (apparent) stability constants K and the ratio of diffusion coefficients of the individual species. The comparison of the results obtained by soft (model free) and hard (model postulated a priori) modelling for the systems investigated is discussed.

Chemometrics in Electroanalytical Chemistry

The use of chemometrics in electroanalytical chemistry is not as popular as in spectroscopy, although recently, application of these methods for mathematical resolution of overlapping signals, calibration and model identification have been increasing. Self-modelling curve resolution and multivariate analysis have been shown to be very powerful for in the analysis of electroanalytical data, especially for multianalyte calibration and modelling in multicomponent dynamic systems. In this paper, an overview on the application of chemometrics to electroanalytical data is presented, with special attention to the contributions of the last decade.

Voltammetry of metal ions in mixtures of ligands Part II: Application to successive labile complexes

Abstract In previous work, a model was proposed for the voltammetric study of metal ions in mixtures of macromolecular and simple ligands, with the assumption that the complexes formed are totally labile or totally inert. In this paper, the general equations of that model are applied to the study by reverse pulse polarography (RPP) of the system Cd(II) + polymethacrylic acid + iodide, which is taken as a model of labile complexation with the simple ligand forming successive 1: m complexes. In general terms, there is good agreement between the experimental set of current data and the theoretical predictions. In contrast, the influence of the electrodic adsorption on the potential data was too high to obtain results close to the theoretical ones. According to theoretical predictions, an experimental strategy was developed for the estimation of the stability constants of simple complexes which adsorb onto the electrode. The methodology is based on the use of RPP as a well-known tool for minimizing adsorption. However and in contrast with the DeFord-Hume method, it considers only current measurements and not potential values, since previous papers showed that RPP potentials can be much more affected by adsorption than RPP currents. As expected, the application of the method to the Cd(II) + I − complexes yielded results much closer to the potentiometric values reported in the literature than those obtained by the application of the DeFord-Hume method.

Voltammetry of metal ions in mixtures of ligands part I. Theoretical formulation and application to 1:1 labile complexes

A model is proposed for the voltammetry of metal ions in mixtures of macromolecular and simple ligands. Among other hypotheses, the model assumes excess of ligand and absence of both electrodic adsorption and kinetic effects (complexes are supposed to be totally labile or totally inert). The mathematical solution of the model provides general equations which are used for simulation purposes. Besides that, the limiting case when only 1:1 labile complexes exist is extensively discussed and the resulting equations applied to the polarographic study of the systems Zn(II) + polymethacrylic acid + phthalate and Cd(II) + polymethacrylic acid + bromide.

Comparison of the zinc–cadmium exchange properties of the metallothionein related peptide {Lys–Cys–Thr–Cys–Cys–Ala} and a zinc-containing metallothionein: study by voltammetry and multivariate curve resolution

Abstract The complexation of the hexapeptide Lys–Cys–Thr–Cys–Cys–Ala (FT), peptidic fragment {56–61} in the α-domain of mouse liver metallothionein I (MT), in the simultaneous presence of Cd(II) and Zn(II) is compared with that of a Zn-containing mammalian MT (Zn–MT II) in the presence of Cd(II). The influence of Cd(II) is studied by square wave voltammetry and differential pulse polarography, at different Cd:Zn:FT molar ratios, which yield a rough picture of the metal exchange processes taking place. The application of multivariate curve resolution with alternating least squares optimization (MCR-ALS) allows both the unitary voltammograms and the concentration profiles to be obtained. Moreover, MCR-ALS is applied for the first time to the study of the complexing properties of a complete MT. On the basis of these results, the electrochemical reduction processes and some structures are depicted for the chemical species considered in the mixed metal system Cd–Zn–FT. In the case of Cd–Zn–MT II, additional information is required before proposing similar models.

Comparison of Voltammetry Assisted by Multivariate Analysis with EXAFS as Applied to the Study of Cd- and Zn-Binding of Metallothionein Related Peptides

Zn- and Cd-complexes of glutathione (GSH) and the metallothionein related peptide Lys-Cys-Thr-Cys-Cys-Ala (FT) were investigated by EXAFS and by voltammetry, assisted by multivariate curve resolution (MCR). EXAFS data from the Zn-FT system seem to show the formation in low extent (under 50%) of a metal cluster with a coordination number of four. In the cases of Cd-FT, Zn-GSH and Cd-GSH no sound conclusions can be deduced from EXAFS. These results are in contrast with those found by voltammetric techniques. The use of MCR allows to calculate the voltammograms corresponding to each single process and its corresponding concentration profile. Then models of complexation are proposed.

Chemometrics in Electrochemistry

The use of chemometrics in electrochemistry is not as extended as in spectroscopy although during the past years several chemometric methods have been proposed and used for experimental design, data preparation and exploration, sample classification, resolution of overlapping signals, calibration, and model identification. Soft-modeling and multivariate analysis methods have been shown to be very powerful for the analysis of electrochemical data, especially for multianalyte calibration and modeling in multicomponent dynamic systems. In this chapter, an overview on the application of chemometrics to electrochemical data is presented, with special attention to recent contributions.

Polarography and differential pulse anodic stripping voltammetry of Pb(II)/polycarboxylate complexes

Abstract Complex formation of Pb(II) with polymethacrylic acid (PMA) and polyacrylic acid (PAA) is studied by sampled direct current, normal pulse, reverse pulse, and differential pulse polarography, by cyclic voltammetry and by differential pulse anodic stripping voltammetry. The two Pb(II) systems appear to be labile on the different time scales of several techniques, and at different ligand-to-metal ratios. Voltammetric titrations of Pb(II) with partially neutralized PMA or PAA allow the precise determination of the apparent formation constants K of the complexes present in both systems. Under conditions where adsorption of Pb(II) onto the cell walls is not negligible, i.e. at low concentrations and at pH > 4.5, the application of a previously described methodology also allows the precise determination of K by introducing a correction due to adsorption. In some cases anomalous behaviour is still observed, which cannot be explained by adsorption losses.