Uranyl peroxide enhanced nuclear fuel corrosion in seawater Christopher R. Armstronga,1, May Nymanb, Tatiana Shvarevaa, Ginger E. Sigmonc, Peter C. Burnsc,d, and Alexandra Navrotskya,2 a Peter A. Rock Thermochemistry Laboratory and Nanomaterials in the Environment, Agriculture and Technology Organized Research Unit, University of California Davis, Davis, CA 95616; bSandia National Laboratories, Albuquerque, NM 87185; cDepartment of Civil Engineering and Geological Sciences, University of Notre Dame, Notre Dame, IN 46556; and dDepartment of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, IN 46556
Contributed by Alexandra Navrotsky, November 30, 2011 (sent for review October 31, 2011)
The Fukushima-Daiichi nuclear accident brought together compromised irradiated fuel and large amounts of seawater in a high radiation field. Based on newly acquired thermochemical data for a series of uranyl peroxide compounds containing charge-balancing alkali cations, here we show that nanoscale cage clusters containing as many as 60 uranyl ions, bonded through peroxide and hydroxide bridges, are likely to form in solution or as precipitates under such conditions. These species will enhance the corrosion of the damaged fuel and, being thermodynamically stable and kinetically persistent in the absence of peroxide, they can potentially transport uranium over long distances. calorimetry ∣ actinyl peroxide ∣ actinide ∣ uranium ∣ spent fuel
F
ailed cooling systems in the reactors and spent fuel cooling ponds at the Fukushima-Daiichi nuclear power plants resulted in compromised irradiated fuel and release of radionuclides. Copious amounts of seawater were subsequently used to cool the fuel. The collocation of large quantities of damaged fuel, an intense radiation field, and massive amounts of seawater has created a highly heterogeneous and presumably rapidly evolving system that has the potential to release vast quantities of radionuclides to the environment. Currently, large quantities of contaminated water remain onsite (1), and presumably some has been released to the subsurface as well as to the Pacific Ocean. The fuel matrix at the Fukushima-Daiichi site is mainly UO2 , whose behavior will largely dictate release of matrix-incorporated plutonium and various other radionuclides into water used as a coolant. The intense radiation field of the fuel will cause radiolysis of water and formation of peroxide (as well as other species) (2). Peroxide enhances the corrosion rate of UO2 by oxidizing U (IV) to the much more soluble U(VI) that exists as the linear dioxo uranyl cation, ðUO2 Þ2þ . Peroxide strongly complexes uranyl (3), which increases its aqueous solubility under alkaline conditions. Simple uranyl peroxide complexes contain a single uranyl ion coordinated by as many as three peroxide groups that, each being bidentate, define the equatorial edges of hexagonal bipyramidal coordination polyhedra (4) (Fig. 1). These small complexes associate with counterions locally to balance charge in solution and readily crystallize as alkali metal salts (4–6). When peroxide bridges uranyl ions, the configuration is bent (7–9) and nanoscale cage clusters containing as many as 60 uranyl ions self-assemble in aqueous systems (10) (Fig. 1). These soluble clusters carry negative charges, are associated with counterions in solution, and can be crystallized. Under acidic conditions in deionized water, the combination of uranyl and peroxide causes the precipitation of studtite, ½ðUO2 ÞðO2 ÞðH2 OÞ2 ðH2 OÞ2 (11). Some insight into the geochemical interactions of uranium and peroxide that may occur in the Fukushima-Daiichi systems emerge from cases where deionized water, UO2 , and ionizing radiation have been combined. Irradiated fuel was stored in the K-East Basins of the plutonium-production facility at the Hanford Site under 3.7 m of water maintained at 10 °C and continuously deionized by pumping through ion exchange columns. Studtite was a major 1874–1877 ∣ PNAS ∣ February 7, 2012 ∣ vol. 109 ∣ no. 6
alteration phase of fuel-element claddings, and also occurred on the basin floor and in canister sludge (12). Studtite also formed on spent fuel in deionized water under laboratory conditions (13), and on UO2 doped with alpha emitters or irradiated by an external source in water (2, 14–17). Where seawater flow is relatively stagnant in the FukushimaDaiichi systems, peroxide will accumulate, it will increase corrosion rates of exposed UO2 , and it will complex uranyl ions in solution. The seawater, with a pH of approximately 8, will provide abundant Na and lesser K to balance the charge of the resulting uranyl peroxide complexes that form in solution and in precipitates. We have measured enthalpies of formation enthalpies of formation of model uranyl peroxide compounds to gain insight into the formation and stability of simple and complex uranyl peroxide species in aqueous solution, as well as the solid phases that may form and persist (e.g., upon reduction of water volume due to evaporation or boiling, and eventual drying of the systems). We selected model compounds that could be synthesized with high purity and yield, as detailed in Materials and Methods. The salts M½ðUO2 ÞðO2 Þ3 ðH2 OÞ9–10 [M ¼ Lið10H2 OÞ, Nað9H2 OÞ, Kð9H2 OÞ] contain uranyl ions coordinated by three peroxide groups and linked through the corresponding counterions and H bonding networks (designated LiUT, NaUT, and KUT, respectively) (4–6). The water-soluble U60 nanoscale cage cluster contains 60 uranyl ions that are bridged through bidentate peroxide as well as through hydroxyl groups (Fig. 1) (18), and crystals have composition Li40 K20 ½UO2 ðO2 ÞðOHÞ60 ðH2 OÞ214 . U60 is used here as a model for the more than 30 known cage clusters (containing between 20 and 60 uranyl ions) that are built from uranyl ions bridged through peroxide and hydroxide groups (10). Given that the role of alkali counterions in templating uranyl peroxide cage clusters is poorly understood, it is difficult to predict which uranyl peroxide cluster(s) are most likely to form where Na is abundant; indeed we crystallized several with Na as a counterion. Results and Discussion Enthalpies of formation from oxides (normalized to 1 mol of U) as a function of alkali ionic radius are shown in Fig. 2. Insight can be gained by directly comparing formation reactions under environmentally relevant conditions, i.e., equilibria involving UO2 (c), aqueous systems, and pertinent secondary mineral phases. In contrast to studtite, which has limited thermoAuthor contributions: M.N., P.C.B., and A.N. designed research; M.N., C.R.A., and T.S. performed research; G.E.S. contributed new reagents/analytic tools; C.R.A. analyzed data; and C.R.A., M.N., T.S., P.C.B., and A.N. wrote the paper. The authors declare no conflict of interest. Data deposition: The data is deposited at the Inorganic Crystal Structure Database, www. fiz-karlsruhe.de/icsd.html (CSD-423478). 1
Present address: Savannah River National Laboratory, Aiken, SC 29808.
2
To whom correspondence should be addressed. E-mail:
[email protected].
This article contains supporting information online at www.pnas.org/lookup/suppl/ doi:10.1073/pnas.1119758109/-/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1119758109
A
C
D
B
dynamic stability (19), alkali uranyl peroxide formation from the oxides features highly exothermic ΔHo f ;ox values (Tables 1 and 2; Eq. 1): 3 UO3ðcÞ;25 °C þ 2M2 OðcÞ;25 °C þ 9H2 OðlÞ;25 °C þ O2ðgÞ;25 °C 2 ⇌ M4 ½UO2 ðO2 Þ3 ðH2 OÞ9ðcÞ;25 °C ðM ¼ Na;KÞ
[1]
Comparing the enthalpies of formation of these compounds offers insight into relative stability (Fig. 2), because each phase involves the same amount of oxygen in the reactants side of the ΔHo f ;ox reaction Eq. 1, differences in entropy of reaction (ΔSo ) should be small. Thus the free energy of the reactions, ΔGo (i.e., stability) parallels ΔHo f ;ox . Consistent with studtite (5), the formation of alkali uranyl peroxide species in solution, and the precipitation of M4 ½ðUO2 ÞðO2 Þ3 ðH2 OÞ9–10 solids, requires the corresponding alkali metal ion, water, and peroxide (Reactions A, B, and C), as reactions without peroxide are clearly endothermic (Reactions D, E, and F). An overall increase in stability is evidenced down the alkali series, with a possible small maximum in exothermicity observed with NaUT (Fig. 2). Analogous patterns of energetics as a function of ionic radius for alkali-bearing compounds are well established. For example in alkali uranates (Li-Cs), the maximum stability is observed in K 2 UO4 (20, 21). In contrast to the M4 ½ðUO2 ÞðO2 Þ3 ðH2 OÞ9–10 phases and studtite, the formation of crystals containing the U60 cluster in equilibrium with an aqueous phase and UO2 does not require excess peroxide (Reaction G). In addition, the formation of the U60 cluster from the pertinent M4 ½ðUO2 ÞðO2 Þ3 ðH2 OÞ9–10 com100 0
*
Studtite
∆Hf,ox(kJ/mol)
-100
U 60 cluster
-200 -300
LiUT
-400
KUT
-500
NaUT -600 0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Ionic radius (6-coord) angstrom Fig. 2. Comparison of enthalpies of formation from the oxides for LiUT, NaUT, KUT and the U60 nanocluster, normalized per mole of U. Note: Studtite data from Kubatko Hughes, et al. (19) are also included.
Armstrong et al.
pounds and studtite is also favorable (Reaction H), suggesting that M4 ½ðUO2 ÞðO2 Þ3 ðH2 OÞ9–10 with Li and K, as well as studtite, are metastable with respect to crystals containing U60. At present it is not possible to rigorously determine the stability of crystals containing the U60 cluster with respect to common uranyl oxyhydrates because (i) a U60 cluster with Na as the only counterion has not yet been synthesized, and (ii) no enthalpy data exist for Li uranyl oxyhydrates. However, because the exchange of Na for Li in the formation of U60 without peroxide (analogous to Reaction G) is also exothermic (−9;114 kJ∕mol), the formation of a Na-containing U60 cluster is probably thermodynamically favorable. The energetics of Na-bearing U60 formation and crystallization from an aqueous system in contact with UO2 has been estimated in Reaction L. This reaction is exothermic, indicating that crystals of a Na-K bearing U60 are metastable with respect to secondary U(VI) phases (Na compreignacite and K compreignacite). In the absence of the aqueous phase, uranyl peroxide compounds are expected to eventually decompose to a stable uranyl oxyhydrate; e.g., metaschoepite (Reaction I), clarkeite (Reaction J), Na compreignacite (Reaction K), or Na and K compreignacite (Reaction L). In any case, due to its redox properties and its high affinity for uranyl ion complexation (3), peroxide significantly affects U mobility by acting as a catalyst in U oxidation, by increasing the solubility of the uranyl ion, and by enhancing the production of relatively stable secondary uranyl oxyhydrate phases. Although it is not currently possible to measure the enthalpies of formation of aqueous alkali uranyl peroxide species (22), or complex cage clusters in solution, our studies of the corresponding solid phases provide useful insight pertaining to the interaction of irradiated fuel and seawater. Peroxide production will be the highest where water contacts fuel directly (owing to the importance of alpha radiolysis), and it will achieve the highest concentration locally in the water near the surface of the fuel, where the fuel to water ratio is high, and where water is relatively stagnant. Possible locations where peroxide buildup will be greatest at Fukushima-Daiichi include spent fuel cooling pools and areas in the reactor cores where water intimately interacts with UO2 fuel. We contend that simple uranyl peroxide species, as well as complex nanoscale cage clusters built from uranyl peroxide polyhedra, will form in these environments. The thermochemical data further demonstrate the importance of uranyl peroxides in systems combining intense radioactivity, water, and UO2 . Precipitation of compounds such as M4 ½ðUO2 ÞðO2 Þ3 ðH2 OÞ9–10 is expected in such systems, and especially the thermodynamically stable NaUT phase in the case of seawater. Self-assembly of nanoscale uranyl peroxide clusters in such systems is also likely, and, at least in the case of U60, these are thermodynamically stable in the absence of excess peroxide. As such, these clusters are expected to persist in water even after peroxide production stops (i.e., the water is no longer in contact with fuel). Where clusters become sufficiently concentrated, precipitation may occur to give thermodynamically stable cluster PNAS ∣
February 7, 2012 ∣
vol. 109 ∣
no. 6 ∣
1875
ENVIRONMENTAL SCIENCES
Fig. 1. Ball-and-stick and polyhedral representations of the uranyl ion coordinated by three bidentate peroxide groups as in LiUT, NaUT, and KUT (A, B); and two bidentate peroxide groups and two hydroxides as in U60 cage clusters (C, D). Uranium and O atoms are shown as yellow and red spheres, respectively.
Table 1. Summary of thermodynamic data for uranyl peroxide compounds in kJ∕mol. ΔHds : enthalpy of drop solution; ΔHf ,ox : enthalpy of formation from the oxides; ΔHf ,el : enthalpy of formation from the elements Compound Studtite (19) LiUT NaUT KUT U60
ΔHds
Formula ðUO2 ÞO2 ðH2 OÞ4 Li4 ½UO2 ðO2 Þ3 ðH2 OÞ10 Na4 ½UO2 ðO2 Þ3 ðH2 OÞ9 K4 ½UO2 ðO2 Þ3 ðH2 OÞ9 Li40 K20 ½UO2 ðO2 ÞðOHÞ60 ðH2 OÞ214
+273.2 +794.4 +725.8 +460.5 +350.0
(± (± (± (± (±
ΔHf ;ox 3.6) 12.6) 7.7) 7.9) 4.6)
+22.3 −260.9 −515.5 −453.4 −138.1
(± (± (± (± (±
ΔHf ;el 3.9) 13.2) 8.9) 9.3) 5.0)
−2,344.7 −5,540.2 −5,147.4 −4,975.8 −2,784.3
(± (± (± (± (±
4.0) 14.0) 9.9) 9.3) 5.1)
Data refer to one mole of U.
compounds that eventually convert to alkali uranyl oxyhydrates, carbonates, silicates, and other solid phases. These uranyl peroxide cluster compounds are thus an energetic intermediate between dissolved aqueous species and uranyl minerals. The thermodynamic stability of such clusters in the absence of excess peroxide also indicates they may disperse in the environment through transport in water. To verify the persistence of U60 clusters in aqueous solution in the absence of excess peroxide, crystallized U60 was dissolved in ultrapure water and the resulting solution was monitored using electrospray ionization mass spectroscopy (ESI-MS). The spectrum attributable to U60 persisted for the solution through at least 294 d without detectable change (Fig. S1). Materials and Methods Synthesis of Li4 UO2 ðO2 Þ3 · 10H2 O (LiUT). Aqueous lithium hydroxide solution (4 M, 4 mL) and 30% H2 O2 (3 mL) are combined in a 40 mL beaker with a stir bar and placed in an ice bath. Uranyl nitrate hexahydrate (0.5 g, ∼1 millimole) was dissolved in 6 mL deionized H2 O, that was previously cooled in the ice bath. While stirring, the uranyl nitrate solution was added to the LiOH∕H2 O2 mixture via pipette, and a clear, orange, bubbly solution formed. The beaker was then placed inside of a large jar containing a layer of ethanol on the bottom, and the jar was capped with a lid. The ethanol diffuses slowly into the aqueous-uranyl solution. After 1 d, small orangeyellow dodecahedral crystal have appeared; these are the prior reported ½UO2 ðO2 Þ3 12 ½ðUO2 ðOHÞ4 Li16 ðH2 OÞ28 3 ·Li6 ½H2 O26 (6). After 5 d, long yellow blade-like crystals grew out from the original crop of crystals. Crystals of LiUT were harvested by breaking of the dodecahedral crystals at the tips. Yield ∼0.32 g, 55% based on U. The blade-like crystals were characterized by single-crystal X-ray diffraction, and the bulk was characterized by powder X-ray diffraction (5–60° 2-theta, Cu-Kα radiation) to ensure that the single-crystals are representative of the bulk, and the bulk is pure-phase (Fig. S2). A representative figure of the structure is shown in Fig. S3. Synthesis of Na4 UO2 ðO2 Þ3 · 9H2 O (NaUT). Aqueous sodium hydroxide solution (4 M, 4 mL) and 30% H2 O2 (3 mL) were combined in an 80 mL beaker with a
stir bar and placed in an ice bath. Uranyl nitrate hexahydrate (0.5 g, ∼1 millimole) was dissolved in 6 mL deionized H2 O, that was previously cooled in the ice bath. While stirring, the uranyl nitrate solution was added to the NaOH∕H2 O2 mixture via pipette, and a clear, orange, bubbly solution formed. Upon rapid addition of ethanol, a heavy orange precipitate formed, and the liquid became colorless. The crystalline precipitate was isolated via vacuum filtration and washing with additional ethanol, and left to dry in air. Yield ¼ 0.55 grams, 88%. This product is characterized by powder X-ray diffraction (5–60° 2-theta, Cu-Kα radiation) and thermogravimetric analysis. Both characterization techniques confirm the sodium uranyl peroxide salt reported by Alcock was formed (4, 6). Synthesis of K4 UO2 ðO2 Þ3 · 9H2 O (KUT). The synthesis of a mixed Cs∕K or Rb∕K salt similar to KUT was described previously (22). The pure K-salt was synthesized in this study. Aqueous potassium hydroxide solution (4 M, 4 mL) and 30% H2 O2 (3 mL) was combined in a 40 mL beaker with a stir bar and placed in an ice bath. Uranyl nitrate hexahydrate (0.5 g, ∼1 millimole) was dissolved in 6 mL deionized H2 O, that was previously cooled in the ice bath. While stirring, the uranyl nitrate solution was added to the KOH∕H2 O2 mixture via pipette. Initially an orange solution formed followed by the rapid precipitation of a bright yellow crystalline salt. The salt was isolated by vacuum filtration and washing with ethanol. X-ray powder diffraction, an SEM image, and TG-DTA (thermogravimetric-differential thermal analysis) analysis of KUT are shown in Figs. S4, S5, and S6, respectively. Due to the high solubility of this salt, and the fact that LiUT and NaUT dominate under identical conditions in a freshly prepared solution, we reason that this salt is KUT. TGA analysis gave a formula consistent with nine water molecules (calculated, 31% wt loss; observed, 32% wt loss). Energy Dispersive Spectroscopy gave a K∶U ratio of 4∶1. Synthesis of U60 Nanoclusters. U60 crystals were synthesized using methods described by Sigmon, et al. (18). In summary, using standardized materials in a 20 mL scintillation vial: 1 mL (0.436 M uranyl nitrate hexahydrate), 0.25 mL (0.285 M potassium chloride), and 1 mL of 30% H2 O2 were combined and the pH was adjusted to 9.0 using 2.203 M lithium hydroxide aqueous solution. After 7 d, large, yellow crystals formed in solution. Identity of U60 was confirmed by single-crystal X-ray diffraction.
Table 2. Enthalpies of reactions involving uranyl peroxide compounds NaUT, LiUT, and KUT (and studtite) formation with peroxide: UO2 ðcÞ þ 4Naþ ðaqÞ þ 4H2 O2 ðlÞ þ 7H2 OðlÞ ⇌ Na4 ½UO2 ðO2 Þ3 9H2 OðcÞ þ 4Hþ ΔH ¼ −352.3 (A) UO2 ðcÞ þ 4Liþ ðaqÞ þ 4H2 O2 ðlÞ þ 8H2 OðlÞ ⇌ Li4 ½UO2 ðO2 Þ3 10H2 OðcÞ þ 4Hþ ΔH ¼ −307.7 (B) UO2 ðcÞ þ 4Kþ ðaqÞ þ 4H2 O2 ðlÞ þ 7H2 OðlÞ ⇌ K4 ½UO2 ðO2 Þ3 9H2 OðcÞ þ 4Hþ ΔHM ¼ −133.4 (C) NaUT, LiUT, and KUT formation without peroxide: UO2 ðcÞ þ 4Naþ ðaqÞ þ 11H2 OðlÞ þ 2O2 ðgÞ ⇌ Na4 ½UO2 ðO2 Þ3 9H2 OðcÞ þ 4Hþ ΔH ¼ 37.9 (D) UO2 ðcÞ þ 4Liþ ðaqÞ þ 12H2 OðlÞ þ 2O2 ðgÞ ⇌ Li4 ½UO2 ðO2 Þ3 10H2 OðcÞ þ 4Hþ ΔH ¼ 82.5 (E) UO2 ðcÞ þ 4Kþ ðaqÞ þ 11H2 OðlÞ þ 2O2 ðgÞ ⇌ K4 ½UO2 ðO2 Þ3 9H2 OðcÞ þ 4Hþ ΔH ¼ 256.8 (F) U60 formation without peroxide: 60UO2 ðcÞ þ 40Liþ ðaqÞ þ 20Kþ ðaqÞ þ 274H2 OðlÞ þ 60O2 ðgÞ ⇌ Li40 K20 ½UO2 ðO2 ÞðOHÞ60 214H2 OðcÞ þ 60Hþ ðaqÞΔH ¼ −7592.3 (G) U60 formation from LiUT, KUT, and studtite: 10Li4 ½UO2 ðO2 Þ3 10H2 OðcÞ þ 5K4 ½UO2 ðO2 Þ3 9H2 OðcÞ þ 45UO2 ðO2 Þ 4H2 OðcÞ ⇌ Li40 K20 ½UO2 ðO2 ÞðOHÞ60 214H2 OðcÞ þ 81H2 OðlÞ þ 15O2 ðgÞΔH ¼ −4374.3 (H) Uranyl oxyhydrate formation from uranyl peroxides: UO2 ðO2 Þ 4H2 OðcÞ ⇌ UO3 2H2 OðcÞ þ 2H2 OðlÞ þ 0.5O2 ðgÞΔH ¼ −17.0 (I) Na4 ½UO2 ðO2 Þ3 9H2 OðcÞ þ 3UO2 ðcÞ ⇌ 4NaðUO2 ÞOðOHÞðcÞ þ 7H2 OðlÞΔH ¼ −493.7 (J) Na4 ½UO2 ðO2 Þ3 9H2 OðcÞ þ 5UO2 ðcÞ þ O2 ðgÞ þ 2Hþ ðaqÞ ⇌ Na2 ½ðUO2 Þ3 O2 ðOHÞ3 2 7H2 OðcÞ þ 2Naþ ðaqÞΔH ¼ −844.7 (K) 1∕60Na40 K20 ½UO2 ðO2 ÞðOHÞ60 214H2 OðcÞ þ 2UO2 ðcÞ þ 14∕15H2 OðlÞ þ 1∕2O2 ðgÞ ¼ 1∕3Na2 ½ðUO2 Þ3 O2 ðOHÞ3 2 7H2 OðcÞ þ 1∕6K2 ½ðUO2 Þ3 O2 ðOHÞ3 2 7H2 OðcÞΔH ¼ −243.6 (L) (For clarity, reaction L has been normalized to one mole U) All values in kJ∕mol; c ¼ crystal, l ¼ liquid, g ¼ gas, aq ¼ aqueous solution. For clarity, the peroxo groups are shown in blue. 1876 ∣
www.pnas.org/cgi/doi/10.1073/pnas.1119758109
Armstrong et al.
Electrospray Ionization Mass Spectra (ESI-MS). ESI-MS were recorded on a Bruker microOTOF-Q II high resolution quadrapole time of flight instrument (Q-TOF) (3,200 V capillary voltage, 0.4 Bar nebulizer gas, 4 L∕ min dry gas, 180 °C dry gas temperature, negative-ion mode). The samples were introduced by direct infusion at 7 μL∕ min and scanned over 500–5;000 m∕z with data averaged over 2–5 min. Data was deconvoluted using the MaxEnt algorithm software.
ment and experimental conditions for study of uranyl-bearing materials are provided elsewhere (19, 27, 28). Small pressed pellets (∼5 mg) of powdered sample were dropped from room temperature into a melt of 20 g of sodium molybdate (3Na2 O · 4MoO3 ) at 976 K inside platinum crucibles. The calorimeter was calibrated prior to analysis with ∼5 mg pellets of a standard reference of α-Al2 O3 . To ensure an oxidizing environment, O2 was continuously flushed over the melt head space. Gas flushing also assisted in sweeping evolved H2 O out of the calorimeter. Upon rapid and complete dissolution of the sample, the enthalpy of drop solution ΔHds , was measured, and, using appropriate thermochemical cycles (see Tables S1, S2, and S3), enthalpies of formation from the oxides, ΔHf ;ox , and from the elements, ΔHf ;el , were calculated.
Calorimetry. High temperature oxide melt solution calorimetry was conducted using a Tian-Calvet twin microcalorimeter. The details of this instru-
ACKNOWLEDGMENTS. Jennifer Szymanowski conducted TG and ESI-MS analysis. This work was supported as part of the Materials Science of Actinides, an Energy Frontier Research Center funded by the Department of Energy, Office of Science, Office of Basic Energy Sciences under award number DE-SC0001089. Sandia National Laboratories is a multiprogram laboratory operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin company, for the Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000.
1. Tokyo Electric Power Co. (2011) Roadmap towards restoration from the accident at Fukushima Daiichi Nuclear Power Station revised version., http://www.meti.go.jp/ english/earthquake/nuclear/roadmap/pdf/110719_nps_03.pdf, (July 19) . 2. Christensen H, Sunder T (2000) Current state of knowledge of water radiolysis effects on spent nuclear fuel corrosion. Nucl Technol 131:102–123. 3. Meca S, et al. (2011) Determination of the equilibrium formation constants of two U (VI)-peroxide complexes at alkaline pH. Dalton Trans 40:7976–7982. 4. Alcock NW (1968) Crystal and molecular structure of sodium uranyl triperoxide. J Chem Soc A O:1588–1594. 5. Kubatko K-A, Forbes TZ, Klingensmith AL, Burns PC (2007) Expanding the crystal chemistry of uranyl peroxides: Synthesis and structures of Di- and triperoxodioxouranium (VI) complexes. Inorg Chem 46:3657–3662. 6. Nyman M, Rodriguez MA, Campana CF (2010) Self-assembly of alkali-uranyl-peroxide clusters. Inorg Chem 49:7748–7755. 7. Miro P, Pierrefixe S, Gicquel M, Gil A, Bo C (2010) On the origin of the cation templated self-assembly of uranyl-peroxide nanoclusters. J Am Chem Soc 132:17787–17794. 8. Sigmon GE, et al. (2009) Uranyl-peroxide interactions favor nanocluster self-assembly. J Amer Chem Soc 131:16648–16649. 9. Vlaisavljevich B, Gagliardi L, Burns PC (2010) Understanding the structure and formation of uranyl peroxide nanoclusters by quantum chemical calculations. J Amer Chem Soc 132:14503–14508. 10. Burns PC (2011) Nanoscale uranium-based cage clusters inspired by uranium mineralogy. Mineral Mag 75:1–25. 11. Burns PC, Hughes K-A (2003) Studtite, (UO2)(O2)(H2O)(2)(H2O)(2): the first structure of a peroxide mineral. Am Mineral 88:1165–1168. 12. Abrefah J, Marschmann S, Jenson ED (1998) Examination of the Surface Coatings Removed from K-East Basin Fuel Elements (Pacific Northwest National Laboratory, Richland, WA), PNNL-11806. 13. Hanson B, et al. (2005) Corrosion of commercial spent nuclear fuel. 1. Formation of studtite and metastudtite. Radiochim Acta 93:159–168.
14. Amme M, et al. (2002) Raman microspectrometric identification of corrosion products formed on UO2 nuclear fuel during leaching experiments. J Nucl Mater 306:202–212. 15. Clarens F, et al. (2004) Formation of studtite during the oxidative dissolution of UO2 by hydrogen peroxide: a SFM study. Environ Sci Technol 38:6656–6661. 16. Jegou C, et al. (2005) Effect of external gamma irradiation on dissolution of the spent UO2 fuel matrix. J Nucl Mater 341:62–82. 17. Sattonnay G, et al. (2001) Alpha-radiolysis effects on UO2 alteration in water. J Nucl Mater 288:11–19. 18. Sigmon GE, et al. (2009) Symmetry versus minimal pentagonal adjacencies in uraniumbased polyoxometalate fullerene topologies. Ange Chem Int Edit 48:2737–2740. 19. Hughes Kubatko KA, Helean KB, Navrotsky A, Burns PC (2003) Stability of peroxidecontaining uranyl minerals. Science 302:1191–1193. 20. O’Hare PAG, Hoekstra HR, Frederickson DR (1976) Thermochemistry of uranium compounds. VII. Solution calorimetry of alpha—and beta -sodium monouranate, standard enthalpy of formation of beta -sodium monouranate, and the enthalpy of the alpha to beta transition at 298.15 K. J Chem Thermodyn 8:255–258. 21. Cordfunke EHP, Ouweltjes W, Prins G (1986) Standard enthalpies of formation of uranium compounds. XIII. Cesium uranate (Cs2UO4). J Chem Thermodyn 18:503–509. 22. Nyman M, Rodriguez MA, Alam TM (2011) The U28 nanosphere: What’s inside? Euro J Inorg Chem 14:2197–2205. 23. Bruker AXS (2002) SMART (Version 5.054) (Bruker AXS, Madison, WI). 24. Bruker AXS (2000) SAINT (Version 6.02) (Bruker AXS, Madison, WI). 25. Altomare A, et al. (1997) SIR97. A package for crystal structure solution by direct methods and refinement (Bari, Rome, Italy). 26. Sheldrick GM (2008) A short history of SHELX. Acta Crystallogr A 64:112–122. 27. Navrotsky A (1977) Progress and new directions in high-temperature calorimetry. Phys Chem Miner 2:89–104. 28. Navrotsky A (1997) Progress and new directions in high temperature calorimetry revisited. Phys Chem Miner 24:222–241.
Armstrong et al.
PNAS ∣
February 7, 2012 ∣
vol. 109 ∣
no. 6 ∣
1877
ENVIRONMENTAL SCIENCES
Single-Crystal X-Ray Diffraction. Single-crystal X-ray diffraction of LiUT was performed at −85 °C on a Bruker AXS SMART-CCD diffractometer with graphite monochromated Mo-Kα (0.71073 Å) radiation. Data collection and reduction were carried out with SMART 5.054 (23) and SAINT 6.02 (24) software, respectively. A numerical absorption correction from face indexing was applied. The structure was solved by Direct Methods [program SIR97 (25)] and refined by full matrix least-squares on the basis of F 2 using SHELX97 (26). Formula: H20 O18 Li4 U; FW ¼ 575.97; Monoclinic, P21 ∕cð#14Þ, a ¼ 7.4761ð8Þ, b ¼ 13.7915ð14Þ, c ¼ 14.9976ð16Þ Å, β ¼ 98.659ð2Þ°, V ¼ 1;528.7ð3Þ Å3 , Z ¼ 4, ρcalcd ¼ 2.503 Mg·m−3 ; μðMoKαÞ ¼ 10.704 mm−1 ; 2.02 ≤ θ ≤ 26.40°; R1 ½I > 2σðIÞ ¼ 0.0168; wR2 ½I > 2σðIÞ ¼ 0.0391, GOF ¼ 1.03.
