Modeling of Complexometric Titration Data: Applications and Implications of New Computational Tools and Thermodynamic Data

Complexometric titration of natural ligands in seawater using the competitive ligand equilibration-adsorptive cathodic stripping voltammetry method (CLE-AdCSV) is the method of choice for characterizing the organic complexation of Cu and Fe in seawater. Interpreting such titration data is made difficult by the complexity of the modeling process, which arises from the need to estimate non-linear model equations, the potential for artifacts, and the use of reference equilibrium constants that have been subject of only limited study.

Due to the need to model multi-component equilibrium systems when these titration data, a variety of approximations have been made in order to allow standard linear and non-linear regression tools to be applied. Two software tools, KINETEQL and ProMCC, solve the model equations exactly and allow users to estimate complexation model parameters accurately. ProMCC excels in visualization and ease-of-use, while KINETEQL provides the user with flexibility in the definition of equilibrium models and has the additional capability of solving reaction kinetics problems.

A detailed example of the application of KINETQL to simulating the kinetics of Cu(II) complexation by EDTA in seawater will be illustrated. The implications of kinetics for experimental determination of the stability constants of natural Cu- and Fe-binding ligands will be addressed.

These modeling tools make it feasible to design experiments and analyze datasets using new, complex approaches to data analysis, i.e., data from multiple CLE-AdCSV titrations obtained in different analytical windows. This approach can help solve to the problem of internal calibration in waters that contain mixtures of weak and strong ligands. Because it attempts to model data that span a much wider range in chemistries, the “multiwindow” approach is especially vulnerable to bias in the reference complex stability constants. The difficulty of obtaining coherent models of multiwindow CLE-AdCSV datasets obtained using Cu(II) and salicylaldoxime raises the possibility that the accepted reference constants are in need of revision. New constants for the Cu-SA system will be presented, together with a model that explains the bias in previously published values and revisions required in previous ambient Cu speciation discussed.