Determination of the metal complexing capacity of aqueous solutions containing ligands by titration in the presence of complexing resins

Abstract

The feasibility of ligand titrations with metal ions using complexing resins as sorbing solids is evaluated. The resins considered were the iminodiacetic resin Chelex 100, the carboxilic resin Amberlite CG50, and the anion exchange resin AG1X8, whose sorbing properties for aluminium(III) and copper(II) were known from previous investigations. Synthetical solutions containing known concentrations of ligands, EDTA and IDA were titrated separately with aluminium(III) and copper(II), by adding known concentration of titrant metal, and measuring the concentration of metal ion sorbed on the resin, c. The concentration of free metal ion was evaluated from c by the relationship: [M] = cV/ K* w. K* is the ratio of the concentration of metal ion sorbed on the resin to the free metal ion in solution, and strongly depends on the conditions. In the case of the resins used in this investigation it can be calculated since the sorption equilibria of the metal ions are known. The total concentration of the ligands and the conditional stability constants are evaluated by a linearization procedure, previously proposed and widely used for ligand titrations, based on the hypothesis that only a 1 to 1 metal to ligand complex is formed. Accurate results are always obtained if the titration is properly carried out, i.e. using the appropriate resin, and adding the required excess of titrant. Some criteria are suggested to evaluate the results obtained, which can be useful when solutions of unknown composition are titrated. For these evaluations the sorption equilibria of metal ions on the resin at the considered conditions must be exactly known. In this paper they were evaluated from the sorbing properties of the resins, previously investigated according to the Gibbs–Donnan model. This is very convenient, since it avoids the experimental determination at each particular set of conditions, which is often even impossible, yet giving acceptable results.