Date of Award

8-2019

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Environmental Engineering and Earth Sciences

Committee Member

Brian A Powell, Committee Chair

Committee Member

Lawrence C Murdoch

Committee Member

Ronald W Falta

Abstract

Several long-lived radioisotopes of cesium, such as 135Cs and 137Cs are produced as byproducts of the fuel and targets of production reactors at SRS, and the radioactive isotope of particular interest for hydrogeologic and geochemical studies is 137Cs due to a 30 year half-life, high fission yield, high solubility, high transferability, wide distribution, and rapid assimilation by living organisms. Due to the environmental releases and adverse health effects from Cs exposure, it is important to characterize Cs sorption to soils and sediments, which is a major factor in transport in the environment. Cs mobility is controlled by sorption to geologic materials, and sorption may be a rate limiting process controlling Cs migration in the environment. If the sorption rate is fast compared to the transport rate, relatively simple equilibrium based sorption models will be appropriate. If sorption is slow or involved with dynamic, competitive processes, then the sorption/desorption rates should be utilized in reactive transport models to simulate Cs transport. The objective of this research was to determine if Cs sorption rates must be utilized to model the transport, or if implementation of isotherms in transport models is sufficient.

Competition between Cs and other alkali and alkali earth ions for sorption sites can alter Cs mobility, and increasing the concentration of competing ions limits Cs sorption while also increasing mobility. In order to study these phenomena, two soil-packed flow-through and stop-flow column experiments with different competing ion concentrations were conducted in order to obtain Cs breakthrough curves. This was done to study transport through SRS (SRS) soil under different ionic strength conditions. Desorption experiments were also conducted to investigate the Cs desoprtion. Radioactive conservative and nonconservative tracers were used to estimate dispersivity and study transport of Cs sorbed to low capacity, high affinity binding sites.

The data were modeled analytically using finite-step and semi-infinite step groundwater contaminant transport equations, which simulate Cs release as a pulse and as a continuous release, respectively, in order to estimate the longitudinal dispersion and distribution coefficients, respectively. The effects of Cs diffusion from a high-capacity, low-affinity binding site to a low-capacity, high affinity binding site, which is known as aging, after stop-flow periods was also studied using these models. A numerical contaminant transport model that implemented two types of Cs binding sites, was developed in COMSOL Multiphysics, in order to estimate forward and reverse rate constants for Cs exchange reactions at both sites. Rate constants that apply to systems with both flowing and stagnant groundwater were estimated.

Due to the dynamics of the competition and ion exchange processes, equilibrium models can be fit to initial Cs breakthrough data but are insufficient for describing the behavior after periods of stop-flow. Nonequilibrium models must be used to fit both flow and stop-flow periods, particularly subsequent flow periods after stop-flow. The ion exchange reaction appears to follow two steps with an initial exchange on FESs followed by diffusion of Cs into interlayer sites on 2:1 clays. Distribution coefficients increase after stop-flow periods. as well as two step model is that Cs sorption appears to have an aging step (via diffusion into the interlayers) that leads to sorption hysteresis. Aqueous effluent Cs was measured in desorption experiments. It was observed from Cs gamma scans that strongly bound Cs transport is still affected by advection and dispersion.

Equilibrium models are sufficient before the effects of aging strongly influence Cs transport after periods of long groundwater residence times. As Cs loading continues after stop-flow periods, kinetic models are needed due aging and the system being perturbed from equilibrium. A 1-site nonequilibrium model cannot adequately fit the breakthrough data, therefore, a 2-site model was used, and this indicates the presence of 2 binding sites. The distribution coefficient increases after stop-flow periods as more Cs becomes strongly bound to the interlayer with time. Desorption experiments and radioactive nonconservative tracer mobility indicate Cs sorption to FESs interlayer sites is reversible.

Share

COinS