Skip to main content

Association of Uranium with Organic Matter- and Iron-bearing Colloids in Wetland Environments

Periodic Reporting for period 1 - UMIC (Association of Uranium with Organic Matter- and Iron-bearing Colloids in Wetland Environments)

Periodo di rendicontazione: 2016-10-01 al 2018-09-30

In the context of decreasing water standards for uranium (U) concentrations in drinking water, the use of natural or constructed wetlands is often proposed for cost-effective small-scale water treatment. The underlying mechanism is immobilization of the dissolved form of uranium, U(VI), through its reduction to a less mobile form, U(IV). However, the long-term stability of U(IV) species formed is not well understood. Indeed, a significant amount of U is found labile, even in its reduced form, and contaminates the hydrologic network in several sites. Previous studies suggest that the colloidal phase is responsible for the dispersion of U(IV) in the watershed, but few occurrences of U(IV)-colloids were reported in wetland environments – and no study so far reported the occurrence of U(IV)-colloids in wetlands not impacted by anthropogenic activities. In addition, information is lacking on the processes leading to the formation of the present vector of U(IV) dissemination, i.e. Fe-NOM colloids. Understanding these processes is crucial for predicting the efficiency of wetlands as traps of U water contamination.

We proposed to (i) search for new occurrences of U(IV)-colloids in wetland environments (including wetlands not impacted by anthropogenic activities) and further characterize wetland colloids; (ii) test the effect of various geochemical conditions (including Fe-reducing conditions) on the formation of colloids in batch experiments; and (iii) identify the U(IV) binding mechanisms at molecular levels on Fe-NOM colloids.
Several wetlands have been sampled for analysis of (pore)water composition and colloids identification. For this, different size fractions were separated by ultrafiltration, and colloids were observed with transmission electron microscopy (TEM). These wetland sites include a wetland in the valley between Lausanne and Fribourg, which is influenced by agricultural activities; a high-altitude wetland close to the lake Cadagno in the Swiss canton of Tessin, and a wetland in Gola di Lago in the canton of Tessin.

The wetland of Gola di Lago was selected for further research, based on 1) the primary results of bulk uranium concentrations, 2) the fact that this is a pristine mountain wetland not impacted by anthropogenic activities, and 3) the fact that information was available about the composition of the sediment phase. Three piezometers that allow for the sampling of porewater under anoxic conditions were installed in April 2017. Porewater samples were collected in July 2017, October 2017, April 2018 and September 2018. Streams flowing in and out of the wetland were also collected. All samples were filtered immediately in a field glovebox and stored in anoxic, cold conditions. Dissolved oxygen and pH were analyzed on site. In the laboratory, the samples were analyzed for the geochemical composition, including non-purgeable dissolved organic carbon (DOC), S and S(-II), Fe and Fe(II), U and other major and trace elements.
Within a maximum of fourteen days after sampling, the association of the metals to the colloids was characterized by coupling Asymmetrical Flow Field-Flow Fractionation (AF4) with a UV detector, a fluorescence detector, and an Inductively Coupled Plasma Mass Spectrometer (ICP-MS). This characterization by AF4-ICP-MS was conducted at University of Geneva (UNIGE). The redox state of the samples was preserved during the AF4 analyses. The proportion of U(IV) and U(VI) was determined using strongly basic anion exchange resin. Selected porewater samples were prepared for TEM analysis, including size fractions that were collected downstream of the AF4 instrument. Colloids were observed and elemental maps were collected using scanning transmission electron microscopy and energy dispersive X-ray spectroscopy (STEM-EDS).

The main results of this study were as follows. U porewater concentrations ranged from less than a µg/L to tens of µg/L, challenging the available analytical approaches for U speciation in natural samples. AF4-ICP-MS results showed that most of the U was associated with three fractions that varied spatially and seasonally.

The most significant results of the study were submitted to Environmental Science and Technology in the form of a research article. They were also presented at the international conference on uranium biogeochemistry in October 2018. In addition, in the course of the research project, the outcomes of the project were used as a support for several actions of communication to the general public such as science days and a high school movie project.
Because we characterized U-colloids in a pristine wetland where U concentrations were low, we challenged the technical potential of analytical methods such as AF4 (for the composition of size-fractionnated colloids) and resin exchange (for the redox state of U). Also, this study is one of the first to use AF4-ICP-MS for the study of natural colloids sampled in anoxic conditions, and this should develop in the future because the redox state of natural waters influences the speciation of the colloidal elements and thus the sizes of colloids.
In addition, this study provided results of important implications. Small, organic-rich colloids can be expected to be mobile and quite stable over significant distances. Hence, their presence suggests that there is the potential for colloid-facilitated transport of U(IV) far afield. The results of this work show that U(IV) colloids can form in undisturbed environments, even though the porewater U concentrations are very low. This result is important in the scope of understanding the fundamental processes of U scavenging vs. potential release through wetland environments. In the case of an environmental disturbance of such pristine wetlands, for example changes in hydraulic conditions or geochemistry (due to climate change or to changes of land management, for example), the U(IV) colloids present in the wetland porewaters could be mobilized. Therefore, the possible occurrence and formation of U(IV) colloids and in particular very small organic-rich U(IV) colloids in wetland environments should be taken into account in predictive transport models, for instance when constructed wetlands are used to remove U(VI) from mine-impacted water.
sampling porewater in the field