3-D URANIUM TRANSPORT EXPERIMENTS AT THE INTERMEDIATE SCALE: DOES ‘BIGGER' EQUATE WITH ‘MORE COMPLEX'?

To understand uranium transport in groundwater systems, model systems are often created at the bench scale. Working at this scale allows for the dissemination of fundamental system characteristics controlling uranium migration behavior. Conceptual models are built at the bench scale for field scale implementation. However, much of this small scale characterization cannot be included in conceptual models of larger scale systems due to computational and field characterization limitations. Upscaling of the conceptual model to the field scale is often done in an ad-hoc manner with significant data fitting. The limitation of our knowledge to these two extreme scales of inquiry limits our ability to determine upscaling methodologies. In this presentation we report on experiments that took place at the intermediate scale examining the desorption and transport of uranium. The central hypothesis was that the more complex (i.e. larger) experimental system would exhibit simple uranium transport behavior.

The experimental system consisted of a 3-D tank with dimensions of 2.4m x 0.61m x 0.61m (LxHxW). This tank was filled with uranium contaminated sediment collected from the Naturita field site, a former uranium mill in southwestern Colorado. 1157kg of sediment was heterogeneously packed into the tank using five distinct size fractions: <2mm composite, <0.250mm, >0.250mm, 0.125-0.250mm, and 4-12mm. Groundwater flow was established with constant head boundaries using a uranium free artificial ground water; steady state flow continued for 9 months. A bromide tracer test was used to characterize the flow field. Small groundwater wells were installed to allow for aqueous characterization. The results show simplified behavior in the sense that both the uranium and bromide breakthrough curves (BTC) were largely indistinguishable from typical column scale BTC, and several geochemical relationships appear to be at equilibrium or at least steady state. The implication of this work is that upscaled reactive transport models may not need to explicitly include all of the micro-scale nuances to give a reliable predictor of uranium migration behavior. From a mining perspective upscaled models would allow for more predictable production, better regulatory controls, and more reliable remediation strategies.