In situ speciation of uranium in treated acid mine drainage using the diffusion gradients in thin films technique (DGT).

The exchange membranes P81 and DE81 and Chelex-100 resin were used to perform in situ speciation of uranium in treated acid mine drainage at the Osamu Utsumi mining site, Poços de Caldas city, Southeast Brazil. To investigate possible chemical modifications in the samples during analysis, the three ligands were deployed in situ and in a laboratory (in lab). The results obtained in situ were also compared to a speciation performed using Visual MINTEQ software. Chelex-100 retained total labile U for a period of up to 48 h. The labile U fraction determined by Chelex 100 ranged from 107 ± 6% to 147 ± 44% in situ and from 115 ± 22% to 191 ± 5% in lab. DE81 retained anionic U species up to 8 h, with labile fractions ranging from 37 ± 2% to 76 ± 3% in situ and 34 ± 12% to 180 ± 17% in lab. P81 exhibited a lower efficiency in retaining U species, with concentrations ranging from 6± 2% to 19± 2% in situ and 3± 2% to 18± 2% in lab. The speciation obtained from MINTEQ suggests that the major U species were UO2OH+, UO2(OH)3-, UO2(OH)2(aq), Ca2UO2(CO3)3(aq), CaUO2(CO3)32-, UO2(CO3)22-, and UO2(CO3)34-. This result is in accordance with the results obtained in situ. Differences concerning speciation and the total and soluble U concentrations were observed between the deployments performed in situ and in the laboratory, indicating that U speciation must be performed in situ.

In situ determination of V(V) by diffusive gradients in thin films and inductively coupled plasma mass spectrometry techniques using amberlite IRA-410 resin as a binding layer.

Amberlite IRA-410 anionic exchange resin was evaluated as the binding layer for sampling V(V) by using Diffusive Gradients in Thin Films (DGT). V(V) was determined by inductively coupled plasma mass spectrometry (ICP-MS). Mass vs. time DGT deployments (ionic strength = 0.03 mol L-1 NaNO3, pH = 5.6 and T = 23.5 ± 0.5 °C) was characterized by excellent linear relationship (R2 = 0.9993) and a significant retention of V(V) by the binding layer. An exchange capacity of at least 40 µg V g-1 resin was achieved for the proposed binding layer. The diffusion coefficient obtained (7.13 ± 0.6 10-6 cm2 s-1) agrees with the literature. The accumulation rate of V(V) was not significantly affected by ionic strength of solutions up to 0.03 mol L-1 and for the entire studied pH range (from 3 to 9). Furthermore, when comparing the concentrations obtained using IRA-410-DGT and those obtained by direct measurement of the solution concentrations, the proposed approach provided a reduction of the 35Cl16O interference on V(V) determination by ICP-MS. Determination of V in normal mode (without collision cell) in solutions containing analyte:Cl- concentration ratio up to 1:500,000 was not affected by interference of 35Cl16O+ polyatomic ion even when normal mode ICP-MS was used. Potential interfering ions on sampling V(V) by DGT (PO43- and SO42-) showed no significant effects on the accumulation rate of V(V). Laboratory tests performed using synthetic samples, natural freshwater and acid drainage water showed an excellent performance (recoveries from 93% to 110%). For in situ deployment, measurements of V(V) by the proposed approach was not significantly different (95.5%) from the value of dissolved V concentration.