Impact of low ionic strength on DGT sampling with standard APA gels: Effect of pH and analyte.

Only a limited and scattered knowledge is currently available on the conditions leading to the occurrence of sampling alteration at low ionic strength (<10-3 mol L-1) with DGT (diffusive gradients in thin films technique). In this study, the role of the pH and the charge of the analyte were comprehensively evaluated with DGT equipped with APA (polyacrylamide with agarose-derivative crosslinker) diffusive gels and ZrO or Chelex binding phases. The sampling of four cations (CdII, CuII, NiII and PbII) and two anions (AsV and CrVI) was compared for pH 4, 6 and 8 at common (10-2 mol L-1) and low (10-4 mol L-1) ionic strengths. Results showed that the sampling was modified at low ionic strength only in the most acidic condition (pH 4) for both anions and cations with an opposite incidence: cations' sampling was halved whereas anions' sampling was increased. Furthermore, cations sampling alteration was similarly reproduced using diffusion cell experiments, which requires only the APA gel, indicating that the binding layer does not participate in the low ionic strength effect. The intensity of DGT sampling modification was consistent with a prediction based on Donnan partitioning of analytes at gel/solution interface for several valences (from -I to + III). All these results strongly suggest that the APA diffusive gels carry positive charges that create a Donnan effect at low ionic strength. Since no ionic strength effect could be evidenced at pH 6 and 8, it can be reasonably assumed that this effect occurs only marginally for DGT deployments in most natural waters.

Evaluation of a mercapto-functionalized silica binding phase for the selective sampling of SeIV by Diffusive Gradients in Thin films.

This study evaluates binding discs based on 3-mercaptopropyl-functionalized silica gel for the selective sampling of selenite (SeIV) using Diffusive Gradients in Thin films sampler (DGT). SeIV accumulation was quantitative and selective over SeVI and followed the theoretical linear accumulation with the exposure time up to 0.7 µg. The sampling was not affected by ionic strength variations down to 10-2 mol L-1 (as NaNO3) but SeIV accumulation was found to decrease significantly for pH greater than 5 and was nearly zero at pH 9. Both the limited accumulation range and the pH dependence were unexpected because they have not been reported in the literature related to the SeIV trapping by thiol-based solid phases. Our experiments showed that after SeIV was bound to thiol functional groups, a further pH-dependent reaction occurred with free thiols, resulting in the reduction of SeIV into elemental selenium (Se0) followed by its release and back-diffusion through the DGT sampler. Unfortunately, such a reversible accumulation is incompatible with the implementation of the mercapto-functionalized silica binding phase in DGT devices for SeIV selective sampling.