Copper and nickel in ultrafiltered brackish water: Labile or non-labile?

Copper and nickel were sampled at three stations in the Baltic Sea using diffusive gradients in thin film (DGT) passive samplers and ultrafiltration (< 1 kDa). Two versions of DGT devices were used, the normal open pore (OP) and a restricted pore (RP). The OP DGT and RP DGT concentrations closely followed each other both in depth profiles and time series. The lack of significant difference between OP and RP DGT suggests that the labile complexes were smaller than the pore size of the RP gel (approximately 1 nm). These data, together with OP DGT measurements at the same location in two different years, clearly demonstrate that the DGT method is robust and indicates reproducible results during routine field conditions.Between 50 and 80% of the ultrafiltered fractions for Ni and Cu could not be detected by the DGT method, using standard procedures. This suggests the presence of complexing ligands for Cu and Ni. Assuming 100% complexation of Ni to fulvic acid ligand gave DGT concentrations similar to ultrafiltered Ni concentrations. The equivalent calculation for Cu indicates that up to 75% of the ultrafiltered Cu fraction is non-labile. The non-labile Cu complexes are proposed to be produced at sea since the fraction increases with decreasing terrestrial influence.

► Ni complexes in the ultra-filtered fraction (< 1 kDa) are suggested to be completely DGT-labile.
► Up to 75% of Cu complexes in the ultra-filtered fraction were non-labile.
► Non-labile Cu complexes are proposed to be produced at sea since the fraction increases with decreasing terrestrial influence.
► Open pore and restricted pore DGT passive samplers gave similar results regarding both Cu and Ni measurements.