Speciation of copper and zinc in natural freshwater: comparison of voltammetric measurements, diffusive gradients in thin films (DGT) and chemical equilibrium models

In situ measurements of copper and zinc using diffusive gradients in thin films (DGT) in two distinct natural water systems were compared to metal speciation assessed by competitive ligand exchange (CLE) and voltammetric measurements. In a dynamic river system, where dissolved metal concentrations vary significantly over short-time periods, DGT technique provided averaged values of the metal concentrations over time. In microcosms, at different total dissolved concentrations of copper and zinc, DGT technique measured a similar fraction as measurements of labile metal performed by voltammetry. The proportion of DGT and voltammetric-labile zinc to dissolved zinc was 61(plus-minus)4% and, respectively, 76(plus-minus)9%. DGT technique was measuring 81(plus-minus)8% of exchangeable copper (by exchange with catechol). These two fractions were similarly influenced by the addition of NTA. In the absence of NTA, copper measured by DGT represented 34(plus-minus)4% of dissolved copper whereas in the presence of NTA, this proportion raised to 57(plus-minus)2%. These measurements were compared to calculations performed with speciation programs using several models for the complexation by humic and fulvic substances, namely Model VI (WHAM), NICA-Donnan and SHM. The predicted speciation by these three models was similar. The prediction of free zinc ion and labile zinc concentrations were in agreement with experimental data. Calculated concentrations of free copper ion were overestimated because these models are not considering strong specific copper-binding ligands probably present in natural water.