In-situ recovery (ISR) of uranium (U) from sandstone-type roll-front deposits is a technology that involves the injection of solutions that consist of ground water fortified with oxygen and carbonate to promote the oxidative dissolution of U, which is pumped to recovery facilities located at the surface that capture the dissolved U and recycle the treated water. The ISR process alters the geochemical conditions in the subsurface creating conditions that are more favorable to the migration of uranium and other metals associated with the uranium deposit. There is a lack of clear understanding of the impact of ISR mining on the aquifer and host rocks of the post-mined site and the fate of residual U and other metals within the mined ore zone. We performed detailed petrographic, mineralogical, and geochemical analyses of several samples taken from about 7 m of core of the formerly the ISR-mined Smith Ranch-Highland uranium deposit in Wyoming. We show that previously mined cores contain significant residual uranium (U) present as coatings on pyrite and carbonaceous fragments. Coffinite was identified in three samples. Core samples with higher organic (N 1 wt.%) and clay (N6-17 wt.%) contents yielded higher 234 U/ 238 U activity ratios (1.0-1.48) than those with lower organic and clay fractions. The ISR mining was inefficient in mobilizing U from the carbonaceous materials, which retained considerable U concentrations (374-11,534 ppm). This is in contrast with the deeper part of the ore zone, which was highly depleted in U and had very low 234 U/ 238 U activity ratios. This probably is due to greater contact with the lixiviant (leaching solution) during ISR mining. EXAFS analyses performed on grains with the highest U and Fe concentrations reveal that Fe is present in a reduced form as pyrite and U occurs mostly as U(IV) complexed by organic matter or as U(IV) phases of carbonate complexes. Moreover, U-O distances of~2.05 Å were noted, indicating the potential formation of other poorly defined U(IV/VI) species. We also noted a small contribution from U_O at 1.79 Å, which indicates that U is partially oxidized. There is no apparent U-S or U-Fe interaction in any of the U spectra analyzed. However, SEM analysis of thin sections prepared from the same core material reveals surficial U associated with pyrite which is probably a minor fraction of the total U present as thin coatings on the surface of pyrite. Our data show the presence of different structurally variable uranium forms associated with the mined cores. U associated with carbonaceous materials is probably from the original U mobilization that accumulated in the organic matter-rich areas under reducing conditions during shallow burial diagenesis. U associated with pyrite represents a small fraction of the total U and was likely deposited as a result of chemical reduction by pyrite. Our data suggest that areas rich in carbonaceous materials had limited exposure to the lixiviant solution, continue to be reducing, and still hold significant U resources. Because of t...

Subsurface transport of plutonium (Pu) may be facilitated by the formation of intrinsic Pu colloids. While this colloid-facilitated transport is largely governed by the electrokinetic properties and dispersion stability (resistance to aggregation) of the colloids, reported experimental data is scarce. Here, we quantify the dependence of ζ-potential of intrinsic Pu(IV) colloids on pH and their aggregation rate on ionic strength. Results indicate an isoelectric point of pH 8.6 and a critical coagulation concentration of 0.1 M of 1:1 electrolyte at pH 11.4. The ζ-potential/pH dependence of the Pu(IV) colloids is similar to that of goethite and hematite colloids. Colloid interaction energy calculations using these values reveal an effective Hamaker constant of the intrinsic Pu(IV) colloids in water of 1.85 × 10(-19) J, corresponding to a relative permittivity of 6.21 and refractive index of 2.33, in agreement with first principles calculations. This relatively high Hamaker constant combined with the positive charge of Pu(IV) colloids under typical groundwater aquifer conditions led to two contradicting hypotheses: (a) the Pu(IV) colloids will exhibit significant aggregation and deposition, leading to a negligible subsurface transport or (b) the Pu(IV) colloids will associate with the relatively stable native groundwater colloids, leading to a considerable subsurface transport. Packed column transport experiments supported the second hypothesis.

In alkaline carbonate solutions both technetium and uranium exist as anionic species, TcO 4 − and UO 2 (CO 3 ) 3 4− , which makes separating Tc and U from each other more challenging compared to acidic PUREX type solutions where uranium forms neutral or cationic complexes. We tested the batch separation of 99 Tc, 238 U, and 239 Pu from carbonate solutions using anion exchange resins with different properties, including Purolite A530E, Reillex HPQ, and Dowex Marathon WBA. Overall, the highest K d values for pertechnetate were achieved using the strong base resins in solutions of the lowest carbonate concentration. Pretreating the resins in higher carbonate concentrations before use was shown to increase the K d values for pertechnetate. The measured K d values for uranium generally were low, resulting in high selectivity for the separation of pertechnetate from uranium in carbonate solutions.

vii • Pu sorption/desorption experiments on colloid minerals identified in NTS groundwater, and • Transport of Pu-doped colloids through fractured rock core. Chapter 1 is a background review of our current understanding of colloids and their role in facilitating contaminant transport. Chapters 2, and 3 are field studies that focused on characterizing natural colloids at different hydrologic environments at the NTS and address Ryan and Elimelech's (1996) first criteria regarding the existence and stability of colloids. Chapters 4, 5 and 6 are laboratory experimental studies that investigate the sorption/desorption behavior of Pu and other low-solubility radionuclides on colloid minerals observed in NTS groundwater. These studies evaluate Ryan and Elimelech's (1996) second criteria that the affinity of Pu for colloids must be high. Chapters 7, 8, 9, and 10 are laboratory studies that focus on whether colloids can be transported through fractures. These transport studies address Ryan and Elimelech's (1996) third criteria that colloids must be transported. Chapter 11 discusses the implications of the fracture flow experimental results. Chapter 12 provides recommendations for future work that would help reduce uncertainties associated with the prediction of colloid-facilitated radionuclide transport at the NTS.

Understanding the parameters that control colloid-mediated transport of radionuclides is important for the safe disposal of used nuclear fuel. We report an experimental and reactive transport modeling examination of americium transport in a groundwater-bentonite-fracture fill material system. A series of batch sorption and column transport experiments were conducted to determine the role of desorption kinetics from bentonite colloids in the transport of americium through fracture materials. We used fracture fill material from a shear zone in altered granodiorite collected from the Grimsel Test Site (GTS) in Switzerland and colloidal suspensions generated from FEBEX bentonite, a potential repository backfill material. The colloidal suspension (100 mg L(-1)) was prepared in synthetic groundwater that matched the natural water chemistry at GTS and was spiked with 5.5 × 10(-10) M (241)Am. Batch characterizations indicated that 97% of the americium in the stock suspension was adsorbed to the colloids. Breakthrough experiments conducted by injecting the americium colloidal suspension through three identical columns in series, each with mean residence times of 6 h, show that more than 95% of the bentonite colloids were transported through each of the columns, with modeled colloid filtration rates (k(f)) of 0.01-0.02 h(-1). Am recoveries in each column were 55-60%, and Am desorption rate constants from the colloids, determined from 1-D transport modeling, were 0.96, 0.98, and 0.91 h(-1) in the three columns, respectively. The consistency in Am recoveries and desorption rate constants in each column indicates that the Am was not associated with binding sites of widely-varying strengths on the colloids, as one binding site with fast kinetics represented the system accurately for all three sequential columns. Our data suggest that colloid-mediated transport of Am in a bentonite-fracture fill material system is unlikely to result in transport over long distance scales because of the ability of the fracture materials to rapidly strip Am from the bentonite colloids and the apparent lack of a strong binding site that would keep a fraction of the Am strongly-associated with the colloids.