Subsurface Transport of Plutonium in Organic and Aqueous Acidic Processing Wastes at the Hanford Site, USA.

Over 4 million liters of mixed acidic (∼pH 2.5), high ionic strength (∼5 M nitrate) plutonium (Pu) processing waste were released into the 216-Z-9 (Z-9) trench at the Hanford Site, USA, and trace Pu has migrated 37 m below the trench. In this study, we used flowthrough columns to investigate Pu transport in simplified processing waste through uncontaminated Hanford sediments to determine the conditions that led to Pu migration. In low pH aqueous fluids, some Pu breakthrough is observed at pH < 4, and increased Pu transport (14% total Pu breakthrough) is observed at pH < 2. However, Pu migrates in organic processing solvents through low pH sediments virtually uninhibited with approximately 94 and 86% total Pu breakthrough observed at pH 1 and pH 3, respectively. This study demonstrates that Pu migration can occur both with and without organic solvents at pH < 4, but significantly more Pu can be transported when partitioned into organic processing solvents. Our data suggest that under acidic conditions (pH < 4) in the vadose zone beneath the Z-9 trench, Pu present in organic processing solvents moved relatively unhindered and may explain the historical downward migration of Pu tens of meters below the Z-9 trench.

Plutonium Desorption from Nuclear Melt Glass-Derived Colloids and Implications for Migration at the Nevada National Security Site, USA.

The migration of low levels of plutonium has been observed at the Nevada National Security Site (NNSS) and attributed to colloids. To better understand the mechanism(s) of colloid-facilitated transport at this site, we performed flow cell desorption experiments with mineral colloid suspensions produced by hydrothermal alteration of NNSS nuclear melt glass, residual material left behind from nuclear testing. Three different colloid suspensions were used: (1) colloidal material from hydrothermal alteration of nuclear melt glass at 140 °C; (2) at 200 °C; and (3) plutonium sorbed to SWy-1 montmorillonite at room temperature. The 140 °C sample contained only montmorillonite, while zeolite and other phases were present in the 200 °C sample. Overall, more plutonium was desorbed from the 140 °C colloids (ca. 9-16%) than from the 200 °C colloids (ca. 4-8%). Furthermore, at the end of the 4.5 day flow cell experiments, the desorption rates for the 140 °C colloids and the Pu-montmorillonite colloids were similar while the desorption rates from the 200 °C colloids were up to an order of magnitude lower. We posit that the formation of zeolites and clays hydrothermally altered at 200 °C may lead to a more stable association of plutonium with colloids, resulting in lower desorption rates. This may give rise to more extensive colloid-facilitated transport and help explain why trace levels of plutonium are found downgradient from their original source decades after a nuclear detonation. Interestingly, in the case of cesium (a co-contaminant of plutonium), no difference was observed between the 140 and 200 °C colloids. This reflects intrinsic differences between cesium and plutonium sorption/desorption behavior (charge, cation size) and suggests that the Cs sorption mechanism (cation exchange) is not similarly affected by colloid formation temperature.

Kinetics of neptunium sorption and desorption in the presence of aluminum (hydr)oxide minerals: Evidence for multi-step desorption at low pH.

Kinetics analyses of sorption and desorption provide important insight into reaction mechanisms occurring at the mineral-water interface. They are also needed to determine when equilibrium is achieved, identify intermediate chemical species, and inform models describing neptunium mobility. Neptunium sorption to and desorption from four different aluminum (hydr)oxides - bayerite (α-Al(OH)3), gibbsite (γ-Al(OH)3), corundum (α-Al2O3), and γ-alumina (γ-Al2O3) - were investigated as a function of mineral concentration (5 - 170 m2 L-1), neptunium concentration (10-9 - 10-7 M), and pH (5.5 - 10.5). Neptunium sorption was characterized by a two-step reaction with an initial fast sorption step occurring within minutes followed by a slower equilibrium process, which was attributed to initial sorption of neptunium to a small number of strong sorption sites followed by sorption of neptunium to a larger number of weak sorption sites. The kinetics data were modeled using the linear and non-linear forms of the pseudo-first and pseudo-second order rate equations and the goodness of fit parameters were compared. Non-linear pseudo-second order rate constants described neptunium sorption to aluminum (hydr)oxides most accurately and were used to determine the reaction orders with respect to mineral concentration and [H+]. Neptunium desorption experiments demonstrated that the desorption mechanism changed as a function of pH and that the forward and reverse reactions were not equivalent. At pH ≥ 7.5, desorption reached steady-state within an hour and was accurately described by the non-linear pseudo-second order rate equations. A desorption plateau was observed at pH 5.5 that could not be described by either pseudo-first or -second order kinetics, suggesting the possibility of a multi-step desorption reaction. The comparatively slow desorption kinetics observed here suggests that sorbed neptunium could be slowly released back into the aqueous phase and act as a continuous source of contamination to the environment.

Kinetics of europium sorption to four different aluminum (hydr)oxides: Corundum, γ-alumina, bayerite, and gibbsite.

Kinetic studies can improve our understanding of the environmental mobility of radionuclides by providing insight into sorption mechanisms occurring at the mineral-water interface. While nearly identical equilibrium europium sorption edges are obtained for gibbsite (γ-Al(OH)3), bayerite (α-Al(OH)3), corundum (α-Al2O3), and γ-Al2O3, differences in surface acidity and crystal structure result in distinct reaction rates. Batch kinetic experiments were performed to investigate europium sorption to four aluminum (hydr)oxide minerals as a function of time (10-120 min), mineral concentration (0.5-50 m2 L-1), pH (5-7), and europium concentration (10-8-10-6 M). Linear and non-linear pseudo-first and pseudo-second order rate equations were used to fit the data. The overall reaction rates and reaction orders with respect to proton concentration, mineral concentration, and europium concentration were determined, leading to a governing rate law for each mineral system. Europium sorption is characterized by a two-step reaction with a fast initial sorption step followed by a slower sorption or aging process. The pseudo-second order equations resulted in better model fits of the data. Using the non-linear pseudo-second order rate constants, it was observed that the reaction orders for each mineral system were similar but resulted in different overall reaction rates following the trend bayerite ∼ γ-Al2O3 > corundum > gibbsite. However, the calculated reaction orders are similar to those previously reported for plutonium(IV/V) interactions with iron (oxyhydr)oxide minerals, suggesting that these binary systems share similar elementary reactions.