Date of Award

12-2010

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Legacy Department

Environmental Engineering and Science

Advisor

DeVol, Timothy A.

Committee Member

Elzerman , Alan W.

Committee Member

Fjeld , Robert A.

Committee Member

Powell , Brian A.

Abstract

Uranium is an important natural resource used in production of nuclear reactor fuel and nuclear weapons. The mining and processing of uranium has left a legacy of environmental contamination that remains to be addressed. There has been considerable interest in manipulating the oxidation state of uranium in order to put it into an environmentally immobile form. Uranium forms two stable oxidation states in nature, uranium (IV) being much less soluble than uranium (VI), and therefore the preferred state with regard to limiting the potential exposure of man. One of the tools used in evaluating potential effects of manipulating the natural systems to prefer U(IV) formation is chemical equilibrium modeling. These models show the thermodynamically preferred chemical species that are formed in a particular defined system. Chemical equilibrium modeling is very dependent on the data that is used to support the calculations, the initial definition of the system, and the other constraints placed on the reactions. This research is designed to show the relative effects of inorganic chemical perturbations in the particular systems modeled, as well as the effects of humic/fulvic acids, mineral and bacterial sorption and temperature (to a limited extent), with a focus on uranium solubility, by performing a sensitivity analysis using Geochemist's Workbench, a commercially available chemical equilibrium software package.
Four groundwater systems of interest were selected upon which to perform the sensitivity analysis: the Yucca Mountain (YM) J-13 well water system (Harrar, et al., 1990), the mean well water system from the Simpsonville, SC (SSC) area (Woodruff, 2002), the Savannah River National Laboratory (SRNL) F-Area groundwater system, and Idaho National Engineering and Environmental Laboratory (INEEL) Snake River aquifer system (Ayaz, et al, 2000). The results of the sensitivity analysis are presented in terms of metrics called the uranium ratio (UR) and the sensitivity index (SIUR), defined as follows:
UR = URaq(mol)/URtot(mol)
and

SIUR = (URmin-URmax)/URbaseline
Where:
URmin = UR at the minimum constraint concentration evaluated
URmax = UR at the maximum constraint concentration evaluated
URbaseline = UR at the baseline for the system
The discussion of the results of the sensitivity analysis focuses on the relative effect varying a given constraint has on the system, defined as follows: significant or major effects (SIUR >1), moderate effects (0.1 < SIUR < 1), minor effects (0.01 < SIUR < 0.1%), no apparent effects (SIUR <0.01%).
The sensitivity analysis of the YM solubility controlled system indicated that aqueous carbonate concentrations dictated UR to a large extent. Phosphate and strontium, both of which were not included in the basis, were indicated to be of potential significance to uranium solubility in this system. Temperature also seems to have a strong role in uranium solubility in this system. Interestingly, pH seemed to have little effect on the UR. The presence, concentration and composition of the organic acid simulant appeared to be of little concern.
The solubility controlled SSC system precipitated Soddyite and Quartz in the baseline case at equilibrium indicating oversaturation of silicon and uranium in the natural groundwater. Similarly to the YM system, carbonate seemed to dominate the uranium solubility of the system, but silicon and temperature also had significant effects. The presence, concentration and composition of organic acids in this system could effect aqueous uranium concentrations by as much as 20-30%.
The SRNL system was modeled under both solubility and sorption controlled assumptions. The solubility controlled SRNL system was impressively unresponsive to physical, inorganic and organic constraint variations across the tested range, giving strong evidence that this system is indeed sorption controlled. Both pH and total uranium concentration exerted strong effects in the sorption controlled system, illustrating the importance of availability of sorption sites relative to the amount of uranium in solution. Other constraints, surprisingly including carbonate, had relatively little effect. The presence, concentration and composition of organic acids had little effect. Bacterial sorption was shown to have a significant potential to affect the aqueous uranium concentration in this system, especially as the total uranium in the system was reduced.
The sorption controlled INEEL system showed much more sensitivity than did the SRNL system, and seemed to be significantly (> 10%) effected by the variation of HCO3-, pH, pe, Na+, Ca2+, SO42-, Cl-, and Mg2+in both the mineral-only and mineral-bacterial cases. The UR in this system was not shown to be sensitive to the presence, concentration or composition of organic acids. The mineral-only and mineral-bacterial sorption controlled systems behaved nearly identically.
There are obvious limitations in the models and supporting data. Further, the sensitivity analysis did not evaluate synergistic or antagonistic effects of multiple constraint variability, except in a very rudimentary way. More research in these areas is suggested. Given these limitations, the sensitivity analysis can identify potentially unrecognized factors that may have a strong impact, and those that have little effect, on uranium mobility in the environment. This gives environmental scientists another tool to evaluate resource allocation and narrow down many potential avenues of research to those that have the greatest potential to bear fruit, ultimately aimed at defining and perhaps controlling uranium mobility in the environment.

Share

COinS