Symposium ES6 : Scientific Basis for Nuclear Waste Management

Synroc, mineral-titanate based ceramics for immobilisation of high-level radioactive wastes were invented by Ringwood in the late 1970s. A demonstration plant utilising inactive materials was set up at ANSTO (then the AAEC) was soon set up, with the process route based on sol-gel precursors, drying + calcination in a rotary calciner, followed by graphite-die hot-pressing at ~1200oC. The materials had ~100-1000 times more aqueous durability than borosilicate glass in short-term laboratory tests, although borosilicate glass had been actively demonstrated at full scale at that stage. However in the late 1990s, a synroc derivative was chosen ahead of glass for immobilising impure US/Russian Pu. Although this was not significantly pursued, synroc gained considerable credibility. Since the early 2000s, synroc has morphed from a titanate-based wasteform for reprocessing waste to a hot isostatic pressing platform for high- and intermediate level wastes which are problematic for glass in terms of waste loading and processing temperatures. While at ANSTO considerable basic as well as commercial research is being undertaken on a wide range of ceramics, glass-ceramics and more latterly glasses, ANSTO is currently in the detailed design stages for dealing with waste from its own 99Mo radiopharmaceutical production.

A possible method for the reduction of civil Pu stockpiles is the reuse of Pu in mixed oxide fuel (MOX). During MOX fuel production, residues unsuitable for further recycle will be produced. Due to their high actinide content MOX residues require immobilization within a robust host matrix. Although it is possible to immobilize actinides in vitreous wasteforms; ceramics, such as brannerite (UTi₂O₆), are attractive due to their high waste loading capacity and relative insolubility. A range of uranium brannerites, formulated M_x(U_0.9Ce_0.1)_1-xTi₂O₆ (M = Gd³⁺, Ca²⁺), were prepared using a mixed oxide route. Charge compensation of divalent and trivalent cations was expected to occur via the oxidation of U⁴⁺ to higher valence states (U⁵⁺ or U⁶⁺). Ce⁴⁺ was added as an analogue for the Pu⁴⁺ fraction in mixed oxide fuel. Gd³⁺ was added to act as a neutron absorber in the final Pu bearing wasteform. X-ray powder diffraction of synthesized specimens found that phase distribution was strongly affected by processing atmosphere (air, Ar or H₂/N₂). In all cases prototypical brannerite was formed accompanied by different secondary phases dependent on processing atmosphere. Microstructural analysis (SEM) of the sintered samples confirmed the results of the X-ray powder diffraction. Analysis of Ce L-III edge and Ti K edge XANES found that irrespective of processing conditions, Ce⁴⁺ had been reduced to Ce³⁺ and Ti was present in the tetravalent oxidation state. Analysis of U L-III edge XANES confirmed that charge compensation was achieved by oxidation of U⁴⁺. The preliminary results presented here indicate that brannerite is a promising host matrix for mixed oxide fuel residues.

The UK has over 100 tonnes of separated PuO₂ stored at the Sellafield site. The UK’s policy for dealing with this plutonium material is to fabricate all usable material into MOx fuel.¹ Unfortunately, due to contamination by elements including; Cl, Fe, Cr and Am, not all of the material is suitable for reuse as fuel and has been classified as higher activity waste. These Pu-residues require immobilisation into stable wasteforms for long-term storage and eventual geological disposal. One proposed treatment plan for these wastes is to process them into glass-ceramic of full ceramic wasteforms by hot isostatically pressing the waste and precursor materials inside stainless steel canisters. Glass-ceramic materials are proposed for the low purity streams of these highly variable wastes, whereby the glass phase provides wasteform flexibility to accommodate impurities and variations in the waste feed composition. The plutonium partitions into the more durable ceramic phase, zirconolite (CaZrTi₂O₇). Zirconolite has excellent wasteform properties including durability and radiation tolerance, and readily accepts actinides and rare earths into its crystal structure.

In this work, the formation of zirconolite is shown to vary as a function of the glass fraction and composition, such that an Al rich glass promotes a higher yield of zirconolite.² The thermodynamic activity of Si in the system drives the crystalline phase assemblage, by determining whether it is consumed in the amorphous glass phase or unwanted crystalline phases sphene (CaTiSiO₅) and zircon (ZrSiO₄). After defining an optimised formulation that minimises the presence of unwanted phases, cerium was utilised as an actinide surrogate in waste incorporation experiments. The digestion of CeO₂ and Ce partitioning into the ceramic phase is studied by SEM, EDX and XRD whilst the oxidation state of the Ce is identified from Ce L3 edge XANES data. All samples were processed by hot isostatic pressing at 1250 °C, 103 MPa (15,000 psi) for 4 hr in 30 ml stainless steel canisters.

References
¹ Nuclear Decommissioning Authority (NDA), Progress on approaches to the management of separated plutonium - Position paper - v1.0. Nuclear Decomissioning Authority, 2014.
² E. Maddrell, S. Thornber, and N. Hyatt, “The influence of glass composition on crystalline phase stability in glass-ceramic wasteforms,” J. Nucl. Mater., (2014).

Microporous stannosilicates consisting of heteropolyhedral structure where the simplest units are SnO₆ octahedra and SiO₄ tetrahedra have been raising considerable interest.^{1, 2} Sn-umbite (K₂SnSi₃O_9^●H₂O) crystallising in orthorhombic system with the space group P2₁2₁2₁ (a = 10.101, b = 13.136, c = 7.157 Å) has shown its ion exchange to both Cs and Sr.³ It is investigated that the ion exchange can be significantly improved by modifying the crystallographic and electrostatic environment via framework doping. Partial (25%) incorporation of pentavalent elements for Sn⁴⁺ on the octahedral site was achieved via hydrothermal synthesis. The substitutions were confirmed by XRD and XRF and unit cell parameters calculated. The structural incorporations lead to a slight change of the unit cell volume, suggesting an isomorphous substitution can be achieved. As compared to Sn-umbite, the substituted Sn-umbites show remarkable increases in both Cs and Sr capacity. An increase in ion exchange properties can be explained in terms of their inherent tunnel sizes to accommodate counterions due to partial substitution and bond strengths associated with the charge-neutralising cations and framework oxygens.^{4, 5} The Cs- and Sr-exchanged umbites were thermally converted by hot isostatic pressing for evaluation of ceramic wasteforms for Cs and Sr immobilisation.

References

1. Lin, Z.; Rocha, J.; Valente, A., Chemical Communications 1999, 2489-2490.
2. Peixoto, M. A. R.; Ferdov, S., Journal of Porous Materials 2013, 20, 1171-1178.
3. Pertierra, P.; Salvado, M. A.; Garcia-Granda, S.; Khainakov, S. A.; Garcia, J. R., Thermochimica Acta 2004, 423, 113-119.
4. Cherry, B. R.; Nyman, M.; Alam, T. M., Journal of Solid State Chemistry 2004, 177, 2079-2093.
5. Chitra, S.; Sudha, R.; Kalavathi, S.; Mani, A.; Rao, S.; Sinha, P., Journal of Radioanalytical and Nuclear Chemistry 2012, 1-7.

Layered materials, like titanate and vanadate nanostructures, manganese oxides or metal sulphides, were recently found to be efficient in the selective capture of cesium from multi-component aqueous solutions. Even though the retention mechanism was not fully clarified, there were strong indications that the retention performance of such materials is mainly related to their lamellar structure.
In the present study, various vanadate nanostructures with different morphologies (nanotubes and nanosheets) were elaborated and characterized so as to determine their structural and textural properties. Structural characterizations were performed using such different techniques as XRD, TEM, SEM, XPS, elemental analysis, ICP. Special attention was paid to improve the wettability and dispersion ability of nanoparticles in aqueous media. Cesium and strontium adsorption onto such materials was investigated in ultrapure water multi-component aqueous solutions. The individual adsorption isotherms were measured with the aid of HPLC by varying the solution composition in order to quantify the rate and extent of the underlying sorption phenomena. In particular, vanadate nanosheets were found to present a very high selectivity toward cesium irrespective of the aqueous medium used.
The computer-assisted structure determination was combined with Monte Carlo simulations to elucidate the adsorption mechanism depending on the nature of the compensating cations.
The selectivity and reversibility of cesium sorption will be considered as the main criteria for the selection of materials for further shaping and sizing.

Environmental contamination with radionuclides has resulted from accidental releases of radionuclides from nuclear facilities. Approximately 28 million m³ of soil, for example, was exposed and contaminated with radionuclides from the Fukushima Daiichi nuclear disaster. Decommissioning of nuclear facilities also produces a large amount of contaminated soil waste. Unfortunately, the very strong and specific adsorption of cesium into clay interlayers hampers the remediation of soils with general treatment techniques. Although marked progress has been made in understanding the mechanism of the sorption of cesium by clay since the Fukushima accident, the desorption of cesium from clay has not been successfully investigated. For example, attempts at cesium desorption using cation exchange agents including potassium, ammonium, magnesium and hydrogen ions have yielded the removal of only a very small amount of cesium owing to the stong adsorption of cesium in clay. For this reason, an efficient cesium desorption technique must be developed for the treatment of cesium-contaminated soil waste.
In this presentation, we report the desorption behavior of cesium from clay minerals (i.e. montmorillonite, vermiculite, and illite) by various ion exchange agents and a decontamination process for cesium-contaminated clay. We hypothesized that polycations having a high charge density will enhance the ion exchange with cesium ions owing to the extremely high local concentration of cations resulting on their adsorption in clay. We demonstrated significantly improved cesium desorption using a polymer-type cation exchanger, and the cesium desorption behavior by polycation treatment was compared with the results using single cations and cationic surfactants under various reaction conditions. For example, cationic polyethyleneimine successfully removed most cesium ions from montmorillonite (~97%) and vermiculite (~91%) under acidic reaction conditions, even though a limited amount of cesium (~60%) could be desorbed from illite by the polyethyleneimine treatment. Nevertheless, polyethyleneimine desorbed a significantly larger amount of cesium from illite than did strong acids and surfactants. After the cesium desorption step, the polymeric cation-exchange agent was readily separated from the aqueous waste containing desorbed cesium ions by an ultrafiltration membrane, and the cesium ions could then be concentrated by cesium-adsorbents for a reduction of the waste volume.

Layered materials, like titanates and hydroxyapatites, manganese oxides or metal sulphides, were recently found to present a high selectivity for the capture of strontium from multi-component aqueous solutions. The lamellar structure of such materials is considered to strongly influence the retention performance.
In the present study, nanoflower-like manganate nanostructures were synthesized and their structural and textural properties were evaluated to understand the retention behavior of these innovative solids. Structural characterization was based on the use of such different techniques as XRD, TEM, SEM, XPS, elemental analysis, ICP. In view of the application envisaged, cesium and strontium adsorption isotherms were measured both from ultrapure water and multi-component aqueous solutions by means of HPLC. The impact of the solution composition on the individual adsorption of cesium and strontium was quantified. High selectivity performance of such materials towards strontium was clearly demonstrated in comparison with previously reported structures.
Molecular simulations were performed to rationalize the retention process. The selectivity and reversibility of strontium sorption will be considered as the main criteria for the selection of materials for further shaping and sizing.

Recently, some crystalline hybrid porous solids known as Metal Organic Frameworks (MOFs) emerged as promising systems for gas adsorption and drug encapsulation.¹ Indeed these hybrid solids, constituted by inorganic nodes (metal chains or clusters) linked each other by organic linkers, present a large specific surface area and pore volumes as well as a high chemical versatility allowing to modulate the chemical and physical properties indefinitely.
Even if few results are available in the literature, the adsorption by such solids possessing extra-framework ions could be evidenced as a plausible, economic and simple solution to eliminate toxic contaminants in water.² Various parameters can influence the ability for sorption of porous solids such as the charge of the framework, the adsorption interaction and the diffusion in the pores.³
For this study, we have followed a strategy combining experimental techniques (adsorption calorimetry and isotherm, X-ray Diffraction) with computational approach (Molecular Dynamics) to elucidate both the microscopic mechanisms in parallel of the macroscopic behaviors allowing us to rationalize the ion exchange process for both anion (I^-) and cation (Sr²⁺) in MOFs containing extra-framework ions. Further, one anionic and one cationic MOFs have been chosen for this study, in which the impact of the topology and the framework charge can be discussed on the ionic exchange properties, i.e. the adsorbed ion quantity as well as the adsorption energy.

[1] G. Férey, Chem. Soc. Rev., 2008, 37, 191
[2] (a) A. Sachse, A. Merceille, Y. Barrès, A. Grandjean, F. Fajula, A. Galarneau, Micro Meso Mater., 2012, 164, 251, (b) C. Delchet, A. Tokarev, X. Dumail et al., RSC Adv., 2012, 2, 5707, (c) S. Sen Gupta, K.G. Bhattacharyya, Phys. Chem. Chem. Phys., 2012, 14, 6698

The separation of Krypton (Kr) from Xenon (Xe) is an industrially relevant problem. Kr and Xe are widely used in fluorescent light bulbs. High-purity Xe, has been used in commercial lighting, medical imaging, anaesthesia and neuroprotection. During the reprocessing of used nuclear fuel, two of the gases of concern is radioactive 133Xe and 85Kr. By the time fuel is processed, Xe would decay down to stable isotope however Kr has long half-life as a result can not be released into atmosphere freely. Effectively separating Kr from Xe in nuclear reprocessing plants, would lead to a considerable reduction in storage costs, and in potential revenue generated from the sale of pure Xe. The conventional method to separate these two gases is fractional distillation at cryogenic temperatures, which is an energy intensive process. Furthermore, even after cryogenic distillation, trace levels of radioactive Kr in the Xe-rich phase are too high to permit further use. Alternative environmental friendly separation technologies therefore could save energy. In this respect, membrane technology could play a key role in making this separation less energy intensive and therefore economically feasible. Membrane separation processes have several advantages over conventional fractional distillation; for instance, it is a viable energy-saving method, since it does not involve any phase transformation, furthermore, the required membrane process equipment is simple, easy to operate, control and scale-up. In particular, if prepared in membrane form, metal organic frameworks combine highly desirable properties, such as uniform micropores, high surface areas, and exceptional thermal and chemical stability, making them ideal candidates for challenging molecular gas separations, such as Kr/Xe separation.

Herein, we demonstrate the feasibility of preparing continuous and reproducible ZIF-8 (a type of MOF) membranes for Kr/Xe separation. It has been demonstrated that the effective aperture size of ZIF-8 is in the range of 0.40 nm to 0.42 nm, which makes this particular MOF composition ideal candidate for molecular sieve Kr over Xe. In the ideal case scenario, Kr molecules would diffuse rapidly through the pores, while Xe at most will diffuse slowly meaning that high Kr selectivities could be potentially achieved based on molecular diffusion differences. The synthesized ZIF-8 membranes displayed separation selectivities for Kr/Xe gas mixtures as high as 16.2, and Kr permeances as high as 37.9 GPU at transmembrane pressures of 138 KPa. To our best knowledge this work represents the first example of any MOF membrane composition displaying effective separation for Kr/Xe gas mixtures.

The projected UK plutonium contaminated material (PCM) waste volume is >30000 m³ with 70% arising at Sellafield. The current baseline treatment is supercompaction with cement encapsulation. Thermal treatment, i.e. in-container or plasma vitrification has been identified as the main alternative waste treatment method. Key drivers for the application of thermal treatment processes include the reduced volume, improved passive safety, and superior long term stability of the vitrified wasteform products. These advantages have led to a renewed interest in thermally treating various UK ILW streams, including PCM waste. To support the increased investment in thermal treatment technologies, a fundamental understanding of the processes and the impact of waste inventory needs to be established. The research aims to provide the evidence necessary to support a major investment in the thermal treatment of plutonium contaminated materials.
Laboratory scale experiments using PCM waste simulants (using Ce as a Pu surrogate) and glass forming additives have been performed in order to understand the reactions and processes of waste digestion and incorporation during thermal treatment.

Characterisation of the vitrified product phase assemblage have been performed using techniques that include XRD, SEM/EDX and EXANES. Mossbauer spectroscopy was used to investigate the REDOX conditions of the melt. It was found that PuO₂ (CeO₂ surrogate) from the PCM is physically and chemically immobilised in the resulting materials, i.e no residual PuO₂ (CeO₂) remains after processing. All of the analysis indicated that Ce was incorporated into the vitreous phase in all samples. Estimated volume reductions of ca. 80–95% were demonstrated, against a baseline of un-compacted 200 L PCM waste drums.
The research also aims to gain a greater understanding of the vitrified product stability with respect to generic ILW disposal concepts, through accelerated dissolution experiments. The most likely disposal option is for the resulting vitrified ILW product to be placed in a geological disposal facility in a high – pH environment with cemented ILW and a cement-based backfill. Therefore, the potential effects of such a high pH (12.5), calcium rich cement-based environment on the dissolution behaviour of simulant ILW glasses have been studied using a modified version of the product consistency test (PCT).
The research will contribute to accelerating the acquisition of knowledge and experience required to support the NDA in deploying thermal technologies as a national asset for ILW treatment.

The US Department of Energy is evaluating the “modified-open” nuclear fuel cycle to increase the efficiency of nuclear power production and reduce the amount of high level waste (HLW). In the nuclear fuel cycle, part of the fission products generated during burn-up in a nuclear reactor are non-fissionable, and once separated from the fissionable material must be immobilized in stable waste forms. The majority of these products are in the following three waste streams generated by the projected transuranic extraction (TRUEX^plus) process: alkali/alkaline-earths (¹³⁷Cs and ⁹⁰Sr), lanthanides (Ln), and transition metals. A glass-ceramic, with targeted crystalline phase assemblage comprising: powellite AEMoO₄, oxyapatite, (A,AE)_xLn_(10-x)Si₆O₂₆ (where A is alkali and AE is alkaline-earth) and lanthanide borosilicate (e.g., Ln₅BSi₂O₁₃), is being developed to immobilize these non-fissionable products. The proposed multi-phase borosilicate glass-ceramic waste forms are expected to exhibit significantly higher chemical durability in comparison to the reference borosilicate glass along with higher waste loading (~50%) and higher thermal stability.
The major hurdle in the development of this glass-ceramic is the high MoO₃ (~14 mass%) and alkali (Rb₂O, Cs₂O ~12 mass%) content of the waste stream. The presence of these species can lead to liquid-liquid phase separation and the uncontrolled crystallization of alkali/alkaline-earth molybdates. Similar challenges are also being faced during vitrification of French and UK HLW.
A key barrier to maturation and exploitation of glass-ceramic technology is the gap in our fundamental understanding of the mechanisms of phase separation and crystallization which lead to the development of the desired phase assemblage and microstructure determining long-term product performance. The challenge is in predictably achieving the targeted phase assemblage and microstructure, requiring a detailed understanding of the transformation process as a function of both cooling rate and melt chemistry.
Accordingly, the present study aims at understanding the fundamental science controlling the solubility of molybdenum in nuclear waste glasses. The compositional dependence of MoO₃ solubility in four–to–six components simplified nuclear waste glass compositions in the system Na₂O-CaO-B₂O₃-Al₂O₃-Nd₂O₃-SiO₂ has been studied. The solubility limit of molybdenum in these glasses has been determined by inductively coupled plasma – optical emission spectroscopy. The molecular structure of glasses has been studied by various spectroscopic techniques, while phase separation and crystalline phase evolution in glasses and glass-ceramics has been followed by electron microscopy, and X-ray diffraction, respectively. The obtained results pertaining to solubility of molybdenum in glasses along with the discussion about structural mechanisms controlling the same will form the gist of the presentation.

Glassy materials for HLW immobilization were produced from HLW surrogate, quartz sand, datolite (CaBSiO₄OH), and bentonite clay at a temperature of up to 1200 °C. Waste loading ranged between 20 and 40 wt.%. The glasses were characterized by X-ray diffraction, scanning electron microscopy and Fourier-Transform infrared spectroscopy. Glasses with waste loading of up to 35 wt.% obtained by melt pouring onto a metal plate were found to be rather homogeneous but contained minor noble metal oxides and britholite (at high waste loadings) while those annealed in turned-off furnace were partly devitrified. Average chemical composition of britholite corresponded to formula Na_1.00Ca_4.02Y_0.33Ce_0.05Nd_3.64Gd_0.17Si_6.79O_24.39. The glass network is built from SiO₄ units with one or two bridging oxygens and complex borate groups with primarily ternary coordinated boron. Increase of waste loading resulted in shift of band’s maxima to lower wavenumbers exhibiting increasing the fraction of SiO₄ unit with lower number of bridging oxygen ions and thus reduction of glass network connectedness. Glasses with up to 30 wt.% waste loading kept their high hydrolytic durability making them suitable for HLW immobilization.

Spent Fuel Management at Sellafield includes reprocessing of spent nuclear fuel from stations across the UK and also from overseas. At Sellafield, methods have been developed for the processing of high level wastes, including highly active liquor (HAL), which is a by-product of reprocessing irradiated nuclear fuel
This Highly Active (HA) raffinate is concentrated in evaporators in the Highly Active Liquor Evaporation & Storage (HALES) facility before feeding to the Waste Vitrification Plant (WVP). Here, the resultant HAL feed is calcined and combined with glass before pouring into containers to produce an immobilised HA wasteform
Ruthenium is a fission product possessed of two relatively long lived isotopes: Ru-103 (t_1/2 = 39.8 days) and Ru-106 (t_1/2 = 1 year). Both isotopes form part of the inventory of HA waste during reprocessing of spent fuel. Volatilisation of fission products in nuclear waste generally occurs at high temperature – apart from ruthenium where volatilisation occurs at the lower temperature stages of the vitrification process
Given its volatile nature and high specific radioactivity, ruthenium presents a strong challenge to the nuclear industry in effectively managing its abatement. Part of the challenge is to fully understand the highly complex solution chemistry under conditions relevant to HA waste streams and associated abatement systems
Experimental work within the National Nuclear Laboratory (NNL), UK has demonstrated the presence of oxidising metal ions in HA waste (e.g. Ce(IV)) enhancing volatility of ruthenium through a chemical conversion of Ru(III) species to what is assumed to be RuO₄. A better understanding of these species, their electrochemical processes and reaction kinetics is required to underpin the empirical evidence gathered to date, in particular to develop gravimetric, electrochemical and spectroscopic analytical methods that will improve the understanding of ruthenium speciation in high nitric acid environments, establish the kinetics of inter-conversion between ruthenium species and establish the mechanism by which metal ions such as Ce(IV) may oxidise ruthenium.
We have studied the electrochemical behaviour of ruthenium and present here the thermodynamics of complexed and uncomplexed ruthenium. Electrochemical and spectroscopic methods have been used to determine as bought RuCl₃ to be a mixture of Ru(III) and Ru(IV). Subsequently, a method to electroreduce the mixture to a pure Ru(III) solution was developed. Complexed RuNO³⁺solutions show no sign of any Ru(IV) present, indicating NO stabilises against Ru(III) to (IV) oxidation. Once Ru(IV) has formed, tetroxide generation occurs, in both complexed and uncomplexed systems at 1.2V vs. Ag/AgCl. These results suggest the Ru(III) to (IV) transition is the key precursor process for volatilisation, implying nitrate complexation plays no role in promoting volatilisation and volatility of ruthenium is an intrinsic ruthenium problem coupled with nitric acid chemistry.
Sukhraaj K. Johal^a, Colin Boxall^a*, Colin Gregson^b,Carl J. Steele^c
aThe Lloyd’s Register Foundation Centre for Nuclear Engineering, Engineering Department, Lancaster University, Bailrigg, Lancashire, LA1 4YR, U.K.
^bNational Nuclear Laboratory, Sellafield, Seascale, Cumbria, CA20 1PG, U.K.
^cSellafield Ltd., Sellafield, Seascale, Cumbria, CA20 1PG, U.K.

The dissolution rate of spent nuclear fuel depends on intrinsic factors such as fuel structure and burn-up, as well as environmental factors, including groundwater composition. The burn-up of future spent fuel to be disposed of is expected to increase, causing actinide accumulation in the rim zone and an increase of the content of lanthanides and other fission products. The high burn-up structure at the fuel rim is characterised by much smaller fuel grains and a large number of submicron fission gas bubbles, which both increase the surface area. The increased actinide content in spent fuel at higher burn-ups leads to a higher a-dose rate in the surrounding water and the higher content of fission products will also contribute to a higher b- and g-dose rate initially. The higher dose rates together with the increased surface are expected to increase the dissolution rate. All available experimental results show that the presence of fission products like lanthanides and other dopants in the UO₂ matrix has an inhibiting effect on UO₂ dissolution. The increase of non-uranium cation concentration at high burn-up seems to counteract effectively the influence of higher surface area and higher dose rates.
The anoxic corrosion of massive iron containers considered in most deep disposal concepts produces large amounts of dissolved hydrogen in the groundwater. At the relatively low repository temperatures, hydrogen is expected to be inert in bulk solution. During the last decade, a large impact of dissolved hydrogen on the dissolution of the LWR or MOX fuel and UO₂(s) doped with ²³³U or ²³⁸Pu has been observed. For hydrogen concentrations above a certain limit, the dissolution rates of these highly radioactive materials drop to very low values. A discussion of the results obtained with spent fuel or α-doped UO₂ in the presence of a range of hydrogen concentrations is presented. Typical for all measurements under such conditions are the very low long term concentrations of uranium and other redox-sensitive radionuclides, such as Tc and the minor actinides. Measurements of the radiolytic oxygen after long leaching periods result in values below detection limit. The investigation of the surface of spent fuel or UO₂(s) pellets doped with ²³³U by XPS after long periods of testing shows absence of oxidation. The kinetics of the release of non-redox sensitive elements such as Sr and Cs, used to estimate fuel matrix dissolution rates, is also discussed. Potential mechanisms responsible for the observed behaviour are based mainly on data from studies on the interaction of water adsorbed on the surfaces of metal oxides or actinide oxides with radiation. A discussion of the relative importance for a repository of the oxidative versus the non-oxidative dissolution of the fuel matrix will also be given.

Recently, cement has been considered as a possible material present in the Deep Geological Disposal (DGD) (ENRESA, 2014), for instance as a sealing material. Therefore, it is considered necessary to determine the effect of cementitius water in the Spent Fuel (SF) near field. With this objective, a series of electrochemical experiments were performed to ascertain the influence of two important components of cementitious water: calcium and silicate. The electrode was prepared by using 3% at. SIMFUEL, as a chemical analogue of SF with 41 GWd/T_U of burn-up.
Test solutions were prepared at pH 12 with NaCl 0.1 mol.dm^-3, and various concentrations of both Na₂SiO₃ and/or CaCl₂. The corrosion process was studied by performing cyclo-voltammograms from -1200 mV to 400 mV at 10 mV.s^-1, potentiostatic experiments at 200 mV for 1 hour and, finally, corrosion potential experiments for 24 hours. After performing both potentiostatic and potential corrosion experiments, the SIMFUEL suface was analysed by means of X-Ray Photoelecton Spectroscopy (XPS).
The results showed that the experimentally used silicate concentrations had no significant effect on the cyclic-voltammograms obtained, although its presence decreased the SIMFUEL oxidation at potential values above -100 mV. When calcium was added to the dissolution, the whole oxidation process was shifted to higher potentials. The XPS results obtained after performing the potentiostatic experiments at 200 mV showed that in the absence of both silicate and calcium, the surface was highly oxidized, with 75% of uranium as U(VI). When silicate was added to the electrolyte, the XPS spectrum showed a decrease of the U(VI) amount on the surface and both U(VI) and U(V) were approximately 38%. After calcium was added to the electrolyte solution, the predominant component on the surface was identified as U(V). Finally, after the corrosion potential experiments, the electrode surface presented a similar composition with 45% of U(V) as the main oxidized state, either with or without silicate in solution. However, when calcium was added to the electrolyte, the SIMFUEL surface showed that the predominant oxidized state was U(IV).

References:
ENRESA, 2014, 7th Plan Nacional de I+D. 2014-2018.

The oxidation of uranium dioxide (UO₂) has been widely studied due to the potential risks that this process may cause in the event of shielding failure during the storage of such a nuclear fuel. In case of failure under dry interim storage conditions, the UO₂ matrix of the spent nuclear fuel (SNF) might be oxidized owing to its contact with the atmospheric oxygen and the high temperatures present due to the decay heat of the SNF. The transformation of UO₂ into U₃O₈ via the two-step reaction UO₂→U₄O₉/U₃O₇→U₃O₈ entails an increase in volume of around 36% and, consequently, it might cause the loss of integrity of the UO₂ matrix. Since this fuel matrix is responsible for retaining the fission products and transuranium elements formed by the irradiation process, release of radionuclides into the biosphere might occur.
In spite of the large number of studies that have been carried out on this matter, a more specific characterization of the different uranium oxides involved in the conversion of UO₂ into U₃O₈ needs to be done for a better understanding of the structural and chemical evolution of the system. On this basis, the present study is focused on the first oxidation stage, from UO₂ to U₄O₉, with the aim of characterizing the UO_2+x (x < 0.25) hyperstoichiometric oxides in detail, as well as assessing the structural evolution taking place as oxidation proceeds. For this purpose, different UO_2+x powder samples, with controlled degree of non-stoichiometry, have been identified by thermogravimetric analysis and characterized by X-ray diffraction (XRD) and Raman spectroscopy. XRD analysis reflects that the commonly assumed Vegard’s law is not applicable over the whole hyperstoichiometry range, since a slight increase of the lattice constant is observed for 0.13 < x < 0.20. A quantitative Raman analysis of the UO_2+x spectra as a function of the oxidation degree is also shown. A new method to characterize any UO_2+x sample (for x < 0.20), based on the shift of the 630 cm^-1 band observed in the Raman spectrum, is proposed here for the first time. Moreover, three structure transitions have been detected at x = 0.05, 0.11 and 0.20, giving rise to four distinct regions associated with consecutive structural rearrangements over the hyperstoichiometry range: x < 0.05, 0.05 < x < 0.11, 0.11 < x < 0.20 and 0.20 < x < 0.25.

The fraction of fission products that dissolve faster than the Spent Nuclear Fuel (SNF) matrix and were segregated during irradiation are called Instant Release Fraction (IRF), and can be considered as the most important source of radiological risk in the performance assessment of a deep geologic repository. After the irradiation, IRF radionuclides (RN) like Cs and I are accumulated in the surrounding of the fuel pellet. Previous experiments with a BWR SNF with a Burn-Up (BU) of 42 GWd/t_HM and 215 W/cm of Linear Power Density (LPD) (42BWR), have shown that the RN trapped in the dishing (space between pellets) are the major contribution to the I and Cs IRF. Similar experiments were performed in this work with a PWR SNF with a BU of 60 GWd/t_HM and a LPD of 255 W/cm (60PWR). The contribution of the dishing in this case may be higher because of the higher LPD and BU of the PWR SNF but the morphological changes of the pellet during irradiation can also have an impact on the IRF RN accumulation in the dishing.
Static leaching experiments were performed using simulated granitic groundwater and three different cladded segments (CS). The CS were cut taking into account the position of the dishing, one in the top of the CS (TOP), another in the middle of the CS (MID) and the last one without dishing (FULL). Leaching solution was stirred continuously and completely replaced at each sampling time.
After the first analysis it is possible to observe that the dishing has a minor effect on the IRF corrosion experiments with the 60PWR SNF, especially when compared with the dishing effect on the 42BWR. In the 60PWR the fractures are much less abundant and narrower than in the 42BWR which makes more difficult for the water to reach the dishing in the MID CS, retarding the RN release from this region. Moreover, the dishing of the 42BWR TOP CS is completely open and slightly concave but the 60PWR one has still part of the following pellet attached to its surface and it is almost flat, decreasing the amount of IRF RN accumulated in it.
Since these two fuels had a different irradiation history, their pellets have also suffered from different morphological changes and the effect of the dishing on the IRF corrosion tests has also different magnitudes. Therefore, based on the first analysis, it is possible to consider that the dishing effect on the IRF corrosion tests depends on the irradiation history and the morphological characteristics of the SNF.

Nuclear power is of great importance to the future of low carbon energy production and the ability to separate and recover the actinide elements from spent fuel is a key requirement for a sustainable nuclear fuel cycle. While the extraction of U and Pu for the fabrication of new fuel is well established with the PUREX process, recovery of the actinides, and their separation from the chemically similar lanthanides, remains challenging.

A range of new organic extractant molecules, such as N,N,N’,N’’ tetraoctyl diglycolamide (TODGA), have been developed for the recovery of trivalent actinides through solvent extraction processes and it is important that they be well characterised with new understanding required for the associated chemical extraction mechanisms and kinetics.

Consequently, a study of the interfacial and mass transport kinetics of cerium extraction by TODGA has been conducted using a rotating diffusion cell (RDC) apparatus. The RDC comprises two solution phases which are separated by a defined area membrane interface and subjected constant rotation. This rotation establishes controlled hydrodynamic flow and well characterised boundary / diffusion layer conditions within each solution phase, facilitating the study of both diffusion and kinetic contributions to the rate of mass-transfer and the interrogation of the mechanism of extraction.

Studies to date have revealed significant insights into the Ce(III) / TODGA extraction system, indicating an interesting dependency on local hydrodynamics at the solution phase boundary with the key complexation reaction occurring in the aqueous phase. The extraction rate of Ce(III) has been shown to correlate with aqueous [Ce(III)] while the simultaneous extraction of HNO₃ by TODGA is also demonstrated. The use of HNO₃-pre-contacted TODGA indicates that the extraction of the acid may be inhibitive towards the continued extraction of metal ions and warrants further investigation.

A theoretical description of the Ce(III) / TODGA RDC system has been developed and combined with spectrometric quantification of the interfacial flux allowing for the determination of several key rate parameters including both the forward / complexation and back / decomplexation reaction rates, the aqueous decomplexation length and the interfacial rate constant.

The Swedish spent nuclear fuel canister design KBS-3 consists of a copper cylinder surrounding an iron insert that holds the spent fuel. Like in most other canister designs the mass of iron constitutes the majority of the canister weight. In order for groundwater to access the spent fuel in a future repository the outer canister must fail and iron corrosion occur. Spent nuclear fuel dissolution will therefor likely proceed under conditions of simultaneous anoxic iron corrosion. The iron corrosion can likely supress the spent fuel release by creation of strongly reducing conditions from Fe(II) formation and the generation of large quantities of hydrogen. Redox sensitive radionuclides may either be reductively precipitated by dissolved Fe(II) or from interaction with iron corrosion products such a magnetite or green rusts. The generated hydrogen (up to several MPa) may also inhibit the spent nuclear fuel dissolution at the surface of the fuel via the so called hydrogen effect. In order to probe these effects an autoclave experiment was performed in which a basket with PWR spent nuclear fuel (burnup ~43 MWd/kgU) was suspended in an autoclave containing a simplified groundwater (10 mM NaCl, 2 mM NaHCO3) together with iron powder. The autoclave was sparged and pressurised with argon. Following the expected initial rise in radionuclide concentrations from dissolution of pre-oxidised phases and the so called instant release fraction the U concentration dropped to 3x10^-9 M within 76 days, in-line with the expected solubility of amorphous UO₂, expected to form under reducing conditions. Any measurable Cs and Sr release also ceased within 223 days indicating a complete transition from dissolution of instant release fractions to conditions with inhibition of the dissolution of the fuel matrix. Gas phase and pressure monitoring showed a steady build-up of hydrogen at a rate higher than what could be attributed to radiolysis, reaching hydrogen partial pressures of serval hundred kPa. The results indicate no passivation of the iron corrosion, with magnetite as the likely major iron corrosion product.

Silver-functionalized silica aerogel (Ag⁰-aerogel) is being developed for the removal and sequestration of iodine compounds from the off-gas of a nuclear fuel reprocessing plant. This material shows promise as a potential replacement for silver mordenite because of its high selectivity and sorption capacity for iodine, and its feasible sequestration to a durable SiO₂-based waste form. The iodine-loaded Ag⁰-silica aerogel can be rapidly consolidated with hot isostatic pressing (HIP) and spark plasma sintering (SPS) at moderate temperatures and pressures into a waste form consisting of AgI particles encapsulated in the fused silica matrix. Highly iodine-loaded Ag⁰-aerogel was successfully consolidated with HIP at 1200°C with a 30-min hold and under 207 MPa. The fully densified sample had a bulk density of 3300 kg/m³ and contained ~39 mass% of iodine, The promising preliminary results were also obtained for samples consolidated with SPS, which offers the advantage of high densification rates at a lower processing temperature. The presentation will summarize the results from a series of consolidation studies.

The local chemistry of Tc-99 in oxide glasses is important for understanding the incorporation and long-term release of Tc from nuclear waste glasses, both those for legacy defense wastes and fuel reprocessing wastes. It is known that Tc preferentially forms Tc(VII), Tc(IV), or Tc(0) in glass, depending on the level of reduction of the melt. Tc(VII) in oxide glasses is normally assumed to be isolated pertechnetate TcO₄^- anions surrounded by alkali, but can occasionally precipitate alkali pertechnetate salts such as KTcO₄ and NaTcO₄ when Tc concentration is high. In all these cases of Tc(VII), Tc is 4-coordinated with oxygen. A reinvestigation of the chemistry of alkali-technetium-oxides formed under oxidizing conditions and at temperatures similar to those used in the melting of nuclear waste glasses showed that higher coordinated alkali Tc(VII) oxide species have been reported, including those with the TcO₅^- and TcO₆^- anions. The chemistry of alkali Tc(VII) and other alkali-Tc-oxides is reviewed, along with relevant synthesis conditions.
Additionally, we report the attempts to make alkali compounds of K, Na, and Li technetates as TcO₅^- and TcO₆^-. It was found that higher coordinated species are very sensitive to water, and easily decompose into their respective pertechnetates. It was difficult to get pure compounds, but mixtures of the pertechnetate and another phase(s) were frequently found, as evidenced by x-ray absorption spectroscopy (XAS), neutron diffraction (ND), and Raman spectroscopy. In addition, low temperature electron paramagnetic resonance (EPR) measurements showed the possibility of Tc(IV) and Tc(VI) in Na₃TcO₅ and Na₅TcO₆ compounds.
It was suspected that smaller counter cations would result in more stable technetates. To confirm the synthesis method, LiReO₄ and Li₅ReO₆ were created, and their Raman spectra match those in the literature. Subsequently, the Tc versions LiTcO₄ and Li₅TcO₆ were synthesized and characterized by ND, Raman spectroscopy, XANES, EXAFS, and nuclear magnetic resonance (NMR). The Li₅TcO₆ was a stable compound which appears to have the same structure as that known for Li₅ReO₆. Analysis of LiTcO₄ is still underway.
Some implications of the experimental work on stability of alkali technetate compounds and possible role in the volatilization of Tc are discussed.

The capture of radioactive iodine (I2) remains a important concern for safe nuclear wastes, since a more efficient technology related to the retention of radioelements is still needed. While various solids have been already tested such as zeolites, clays, MOFs,... it is still required to rationalize the impact of chemical composition of solids on the adsorption of this vapor. To reach this aim, we propose to study by Monte Carlo simulations using classical force fields various Metal Organic Frameworks (or MOFs). These solids are porous frameworks containing both organic and inorganic parts which can be modulated as will. Here, we focus on a series of UiO-66 (a MOF stable to the water) with various functions (Br, Cl, CF3, COOH,...) carried by the phenyl rings to investigate the influence of chemical functions on the adsorption mechanism. Furthermore ionic MOFs containing anions (such as Cl-, NO3-,...) or alkali/alkali-earth cations as compensating ions such as MIL-141, MIL-127 and Zn-BTeC have been also studied to determine the impact of the electrostatic charges on the I2 adsorption.

Using molecular simulations, it is possible to calculate the impact of such chemical modifications both on the affinity of the solids at low loading and the saturation at high loading. Furthermore, microscopic models allow us to elucidate the adsorption sites as well as to determine the adsorption isotherms, which can lead to propose some recommendations for the design of new MOFs in view of the capture of iodine vapor.

The immobilisation of radioiodine produced in the nuclear fuel cycle is a growing priority for nuclear wasteform research and development. In particular, 129I is of concern for used nuclear fuel reprocessing facilities due to its very long half-life (1.6 x 10⁷ years) and its high mobility in most geological environments. CuI in water has a very low solubility product and unlike AgI does not decompose when exposed to water containing Fe metal. We have investigated various methods for its production and have investigated its consolidation by sintering in argon and/or hot isostatic pressing in stainless steel cans. Ag4Al3Si3O12I has been formed by sintering in air or hot isostatic pressing in Ni or Cu cans at temperatures in the 750-900^oC range. HiPed samples were investigated by SEM to check the stoichiometry of the AgI sodalite phase, the general microstructure and the reaction between the sodalite and the can material. PCT and MCC-1 leach data will be reported on material HIPed in Cu cans.

A vital aspect of any sustainable nuclear fuel cycle is the utilization of a viable waste form for the immobilization of radioisotopes and fission products. One isotope of particular concern within the radioactive waste community is Iodine-129. The current proposed technology for the removal of I-129 from the radioactive waste stream typically involves the use of caustic scrubbing and Ag solid sorbents to capture iodine off-gas. Following the capture of iodine gas, the Ag solid sorbents are either dissolved and stored in waste tanks as a liquid (as is done with caustic solution) or stored in waste tanks as a solid. This storage of capture medium, however, is not a viable long-term solution of immobilization and a waste form remains to be thoroughly developed for various reasons. Conventional borosilicate vitrification, for instance, is not a viable option due to the low solubility of iodine in many glass chemistries. Most importantly, iodine is highly volatile at typical glass processing temperatures (1000-1100 °C). Alternative waste forms have been explored, but sufficient maturity and satisfactory properties have yet to be achieved to be considered viable.

Apatites have long been considered a possible candidate for the immobilization of radioactive iodine. Due to its ability to accommodate halides into its crystal structure along with its acceptable durability against radiation, apatites make a very viable option. The challenge involving apatite synthesis, however, has always been incorporating the large iodide ion into the crystal structure. The most promising, applicable composition is the iodoapatite, Pb₁₀(VO₄)₆I₂. The synthesis of the iodoapatite, however, has mainly consisted of melting at elevated temperatures (500-800 °C) in controlled, sealed reaction environments due to the volatility of iodine. This method would prove challenging for a large-scale production. Other novel methods have included spark plasma sintering, hot isostatic pressing, microwave heating, and high-energy ball milling, but those methods also continue to have flaws limiting their adoption.

Our research presents the first reported instance of low temperature wet-chemical synthesis of this iodoapatite. By lowering the synthesis parameters to ambient temperature and ambient atmosphere along with the incorporation of a solution-based method, the process can easily be incorporated onto a large scale. Along with the synthesis method, solid solution studies involving the gradual substitution of phosphate for vanadate and calcium for lead, Pb_(10-x)Ca_x(VO₄)_(6-y)(PO₄)_yI_2, were also performed with numerous characterization techniques for the purpose of investigating chemical durability improvement. Sintering studies were performed as well for the purpose of minimization of surface area and, hence, an improvement in chemical durability and leaching resistance. The results pertaining to synthesis and characterization of these minerals will be discussed at the symposium.

The International Simple Glass (ISG) is considered as reference benchmark glass and it has been developed in the frame of an international collaboration for the study of the dissolution mechanisms of high-level vitrified nuclear waste.
In this work the forward dissolution rate of the ISG has been determined in different alkaline solutions, as a simulation of the disposal conditions foreseen by the Belgian concept for geological diposal of vitrified nuclear waste. The determination of the forward dissolution rate has been carried out at 30 °C in four different KOH solutions with pH varying from 9 to 14 and in artificial cementitious water at pH 13.7 ± 0.2.
The values determined in this study have been compared with the rates measured in the same conditions for SON68 glass in a previous work [1]. The results obtained for the two glasses are comparable in both alteration media. However, ISG glass shows a smaller forward dissolution rate with respect to SON68 in KOH at pH 14 (0.204 ± 0.126 g/m²d for ISG and 0.355 ± 0.265 g/m²d for SON68), while in artificial cementitious water the large uncertainties make the comparison of the results more difficult. The forward dissolution rates calculated for the ISG in KOH solutions, moreover, are in good agreement with the initial dissolution rates presented by Inagaki et al. and obtained for a lower pH range [2].

References:
1. Ferrand, K. and K. Lemmens, Determination of the forward rate of dissolution for SON68 and PAMELA glasses in contact with alkaline solutions, in Scientific Basis for Nuclear Waste Management Xxxi, W.E. Lee, et al., Editors. 2008, Materials Research Society: Warrendale. p. 287-294.
2. Inagaki, Y., et al., Initial Dissolution Rate of the International Simple Glass as a Function of pH and Temperature Measured Using Microchannel Flow-Through Test Method. International Journal of Applied Glass Science, 2013. 4(4): p. 317-327.

Within the United Kingdom (UK) , it is proposed that nuclear waste will be disposed in a geological disposal facility, 200 m to 1 km underground¹. This facility will incorporate an engineered barrier system that will be optimised to physically and chemically impede the transport of radionuclides to the biosphere. The facility will house a large volume of cemented Intermediate Level Waste (ILW), in addition to vitrified ILW. A significant volume of concrete will be used in its construction. Interaction of groundwater with the cementitious components of the facility (both the waste and construction materials) will lead to the presence of high pH conditions within a repository. The effect of cement leachates on vitrified wasteforms is not well understood.

We present results from a glass durability study using idealised cement leachates to develop our understanding of glass durability mechanisms in these complex repository like environments. Simulant ILW glasses relevant to the UK disposal program have been utilised. We also investigated a simulant UK high level waste glass (MW-25%) and the International Simple Glass² (ISG), a 6 component borosilicate glass, with components that are common to most borosilicate nuclear glasses. Glass powders were exposed to idealised cement leachates of “intermediate” and “old” ages, approximately representative of GDF conditions at ~1000 and ~10,000 years of operation, according to the product consistency test B³. Analysis of the normalised mass loss and normalised leaching rate of these glasses as a function of cement leachate composition was achieved through analysis of solution concentrations. Simultaneously we present analysis of monolith sample alteration layers by SEM/EDX and GA-XRD. Collectively, these data support a mechanistic understanding of glass dissolution in the context of a complex geological disposal environment for vitrified UK waste.

References
1 Department of Energy & Climate Change, Implementing Geological Disposal, 2014.
2 S. Gin et al, An international initiative on long-term behavior of high-level nuclear waste glass, Materials Today, 2013, vol. 16.
3 Standard Test Methods for Determining Chemical Durability of Nuclear , Hazardous , and Mixed Waste Glasses and Multiphase Glass Ceramics, The Product Consistency Test (PCT) ASTM C1285-14, 2002, vol. 15.

At the Hanford site, USA, low activity tank wastes will be immobilised by vitrification to create 150-350,000m³ of Immobilised Low Activity Waste (ILAW) destined for disposal in a shallow subsurface Integrated Disposal Facility (IDF). During reprocessing, tank wastes have a separable low activity waste fraction that will report to the LAW facility for treatment. Radionuclides of concern in LAW glasses will include ⁶⁰Co, ¹³⁷Cs, ¹⁵⁴Eu, ⁹⁹Tc and ⁹⁰Sr alongside toxic metal contaminants. Conditions in the IDF are expected to differ from those within a deep geological disposal facility for high level waste with temperatures expected to be ~ 15^o C, an arid climate, variable pH and groundwater flow rates.

Project GLAD (Glass Leaching Assessment for Disposability) investigates newly developed leaching technologies for assessing the durability of ILAW glasses. Four new methodologies developed by the U.S. Environmental Protection Agency (EPA) for application to the accelerated ageing of ILAW glass are compared to established leaching tests accepted for evaluation of high level waste glasses, including PCT and MCC-1 protocols. The GLAD project studies the process of glass dissolution and constituent leaching as a function of temperature, pH, groundwater composition and flow rate.

Three candidate glasses, LAW A44, ORP LB2 and LAW A23 were analysed by US EPA leaching methods ‘1313’ and ‘1315’ in both deionised water and synthetic groundwater and results were compared to those obtained by standard PCT-B and MCC-1 tests. In addition, longer term leaching tests were performed to determine the effects of alteration layer formation on the rate of contaminant leaching.

Furthermore, simplified glasses, representative of candidate glasses, were designed and analysed in parallel to improve understanding of how glass composition affects dissolution rates.

Funding for this work was provided by William F. Hamel, Jr., Assistant Manager, of the U.S. Department of Energy Office of River Protection Waste Treatment & Immobilization Plant Project.

Project collaborators: Pacific Northwest National Laboratory (PNNL) and the Consortium for Risk Evaluation with Stakeholder Participation (CRESP).

HLW waste glasses are to be stored in a deep geologic repository. Some potential repository geologies have oxidizing ground waters while some have reducing or anoxic ground waters. The differences in the oxidizing potential of the groundwater, expressed as groundwater Eh, causes different rates of dissolution of the major glass components such as B, alkali, and silica. Moreover, the groundwater Eh has a profound impact on the release of multivalent species such as iron and uranium from the glass. ASTM-C1220 experiments at 90°C were performed in an Ar glovebox under anoxic conditions with iron present in both deionized water and in simulated basaltic groundwater. The deionized water and basalt groundwater were sparged of oxygen by bubbling argon gas through the solutions for >48 hours. The basaltic groundwater was then pre-equilibrated with ground basalt under these deoxygenated conditions. After argon sparging, the iron and or ground basalt served as the low Eh buffer. A companion set of experiments was done where the leach vessels were filled in the Ar glovebox but the durability was performed at 90°C in an oven on a bench top. Since Teflon vessels were used for the leach testing, the bench top experiments slowly oxidized. The same uranium doped HLW glass was leached in the anoxic glovebox environment and in the oxic bench top environments.

The leachate solutions were split into two aliquots and one was filtered and one was not. The releases of B, Na, Li, Fe and U were monitored. All oxic tests gave higher releases of U and higher releases of all glass components than the anoxic tests. Since all the tests included an iron bar representative of a potential waste package component, the iron pumping mechanism described in the 1980’s and shown to rapidly deteriorate HLW waste glass did not occur under anoxic conditions as it does under oxic conditions. The filtered samples gave the same releases of soluble B and Li but much lower concentrations of uranium. Geochemical modeling of the measured Eh-pH conditions from the oxic and anoxic experiments at 90°C, using Geochemist’s Workbench software, demonstrated that this is caused by the precipitation of the uranium as UO_2.6667 and/or UO₂(OH)₂.

Microporous inorganic solids composed of networks of octahedrally coordinated zirconium and tetrahedrally coordinated silicon atoms occur extensively in nature, and include petarasite (Na₅Zr₂Si₆O₁₈(Cl,OH)^.2H₂O) and umbite (K₂ZrSi₃O₉^.2H₂O). These are chemically and structurally related to titanosilicates such as CST (Na₂TiSi₂O₇^.2H₂O), the active ingredient in IONSIV, which is known to be an excellent material for the selective removal of radioactive Cs and Sr from aqueous solutions via an ion exchange process. We are investigating these and other related materials for use in the nuclear decommissioning activities in the UK and post-Fukushima clean-up activities in Japan. Results on the parent materials, germanium analogues and transition metal doped analogues will be presented covering synthesis, ion exchange testing and thermal conversion into dense ceramics.

The barium titanate hollandite is a promising crystalline host for Cs-immobilization, particularly for long term disposal of used fuel processed in a combined waste stream. Hollandite has been shown to sequester Cs during the formation of multiphase ceramics; however, Cs incorporation is compositionally dependent. That is, the M-site dopant of hollandite with the form (Ba_xCs_y)(M_zTi_8-z)O₁₆ can impact the formation of the Cs-doped hollandite phase. In this work, we undergo a detailed evaluation of the atomic structure across the Ba-Cs hollandite binary for Zn-, Ga-, and Al-doped hollandite. Quantum-mechanical calculations were used to observe the impact of A-site and B-site ordering on the structural stability of hollandite. The enthalpy of formation was quantified and agrees with calorimetric measurements of related hollandite phases. Ground state geometry optimizations show that, for intermediate compositions (i.e., Cs₂Ba₂Ga₆Ti₁₈O₄₈), mixing on the A-site is not energetically favored. That is, configurations with Cs and Ba mixed within a channel resulted in higher total energy (~ 0.04 eV) as compared with configurations with Cs and Ba segregated into separate channels; however, the energetics of Cs and Ba mixing may be overcome by the decay heat associated with the β-decay of ¹³⁷Cs to ¹³⁷Ba. The B-site dopants (i.e. Zn, Ga, or Al) prefer ordering with high symmetry and tend to arrange within a single tunnel layer; however, the six B-site dopants in the intermediate Ga- and Al-doped systems cannot arrange symmetrically about the eight B-sites that comprise each tunnel layer, and instead align along the tunnel direction.

The Ga-doped hollandite compositions were synthesized and characterized, showing agreement between DFT, XRD, EXAFS, and neutron diffraction measurements of the atomic structure. The lattice parameter associated with the tunnel dimension was found to increase with Cs concentration. A trend of decreasing thermodynamic stability with smaller tunnel cations was ascribed to the increasing structural distortion observed in the system. The interatomic distances and arrangement of tunnel cations reveals that the hollandite structure can strongly stabilize the A-site cations in the tunnel, even at elevated temperatures up to 500K. A direct investigation of cation mobility in tunnels using electrochemical impedance spectroscopy was conducted to demonstrate the ability of the hollandite structure to immobilize cations over a wide compositional range. The pure Cs-hollandite, with the largest tunnel size and longest rod-like microstructural features, exhibited the highest ionic conductivity. Thus, control of grain size and optimized Cs concentration are essential to limit cation motion and propensity for elemental release.

A previous study revealed Cs₂TiNb₆O₁₈ to be the major Cs-containing phase after hot isostatic pressing (HIPing) Cs-loaded IONSIV (a commercial exchanger). This material has demonstrated excellent waste form properties including aqueous durability. Both experimental and theoretical studies have been carried out in order to access if Cs₂TiNb₆O₁₈ is able to retain ¹³⁷Ba²⁺, the transmutation product of ¹³⁷Cs⁺. A series of samples with different charge compensation mechanisms have been synthesised including Cs_2-xBa_xTi^(III)_xTi^(IV)_1-xNb₆O₁₈ (Ti^(IV) reduction to Ti^(III)) and Cs_2-xBa_xTiNb^(IV)_xNb^(V)_6-xO₁₈ (Nb^(V) Reduction to Nb^(IV)). However due to difficult sample preparation, the majority of samples synthesised instead follow the formula Cs_2-xBa_xTi^(IV)_1+xNb^(V)_6-xO_18. Though not what would form in real conditions, these samples could still give good indication whether Ba can be retained in the structure. X-ray diffraction (XRD), X-ray fluorescence (XRF) and microscopy studies have proved inconclusive and therefore a series of calculations have been carried out using the GULP (General Utility Lattice Program) code in order to see if Ba incorporation is energetically favourable.

Phosphates compounds attract attention of specialists for the development of a ceramic-based immobilization product of long-lived actinides and fission products. A successful solution to this problem requires fundamental knowledge in chemistry and crystal chemistry of the chemical systems, containing important components of the waste streams.

Phosphates of the general composition M'_xM''₂(PO₄)₃ with M'=Na/Eu, M''=Zr, and M'=Fe, Ga, Y, In, lanthanides, M''=Hf, belonging to the NaZr₂(PO₄)₃-type structure family (NZP), were synthesized by high temperature treatments of precursors obtained by precipitation. The higher neutron cross section of Hf can potentially be used on demand in some applications in which neutron absorption properties are required. The scientific basis for such «tailoring» of properties of the NZP phosphates is the high isomorphic capacity of the structure. Isomorphism is an important property that must be taken into account for the immobilization of mixed radioactive wastes in ceramic materials.

The structures of selected representatives were refined by the Rietveld method from the X-ray powder diffraction data. New compounds were characterized by several techniques such as SEM-EDX, DTA/TG, Mössbauer and Raman spectroscopy, solid state NMR and high temperature drop-calorimetry.

The solid solution Na_xEu_(1-x)/3Zr₂(PO₄)₃ with Eu mimicking chemical behavior of a 3-valent actinide showed complex structural features, involving a morphotropic transition in the series as a result of ordering of Na and Eu cations. This conclusion was supported by the results of Rietveld refinement, Raman and solid state NMR spectroscopy. Phosphates M'_0.33Hf₂(PO₄)₃ with M' = Fe, Ga, Y, In and lanthanides were found to be structural analogues to the corresponding zirconium phosphates, however higher reaction temperatures were required compared to the latter. Exceptionally, the compound La_0.33Hf₂(PO₄)₃ crystallizes in two coexisting crystallographic modifications with space groups Pc and R. The enthalpy increments of the studied phosphates were obtained by the high-temperature drop-calorimetry and compared with the analogous Zr-phosphates.

A promising class of materials for the immobilization of alkaline and alkaline-earth fraction of nuclear waste is formed by the phosphates with the structure of mineral pollucite CsAlSi₂O₆. Structural information and the possibilities of chemical substitutions in phosphates with this structure are not very well explored. At the same time, phosphates appear to be advantageous compared to silicates with respect to preparation and processing. In the present work new phosphates of Cs, Cs-Sr, in combination with other elements (Fe, Co, Ni, B and Al) with the pollucite structure are synthesized through the solution route, their structural features and physico-chemical properties are analyzed.

The safe management of radioactive waste and ultimately its deep geological disposal requires a sound scientific understanding in particular with respect to its long-term behavior. Many properties of the waste materials are determined by atomic scale processes, which often are challenging to obtain by experimental methods. On the other hand, facing the continuously increasing performance of supercomputing resources we can use ab initio methods of computational quantum chemistry for simulating even complex materials that are of interest for nuclear waste management research. In this contribution we show various examples of complementary atomistic modeling and experimental studies, which resulted in more complete characterization of materials of interest. The materials discussed include monazite- and pyrochlore-ceramics as prospective waste forms, and uranium-bearing minerals such as coffinite, studtite ([(UO₂)O₂(H₂O)₂]2(H₂O)), metastudtite ([(UO₂)O₂(H₂O)₂]), diuranium pentoxide (U₂O₅) and strontium uranate (SrUO₄), considered as potential secondary phases but also as materials of interesting oxidation chemistry and technologically important properties. The investigated properties include the structural, elastic, thermochemical and thermodynamical parameters such as the reaction enthalpies, the elastic moduli, the heat capacities, the excess enthalpies of mixing as well as the defect formation and incorporation energies, and relationships between them. One important aspect leading to reliable atomistic modeling is careful selection of accurate computational methodology, which represents a challenge for materials containing strongly correlated f-electrons. We will show that DFT+Uapproach with Hubbard U parameter derived ab initio is a method that can provide results of sufficient accuracy and using reasonable amounts of computing time, which is usually a challenge for higher order, beyond DFT, methods of computation chemistry. Last but not least, we will show that complementary atomistic modeling and experimental studies result is superior characterization and enhanced understanding of nuclear materials.

In disposal of high-level radioactive waste, a carbon steel overpack will be corroded after closure of repository then the environment in the vicinity of repository will be changed into reducing and low pH condition. Plutonium diffusion experiment was carried out by using Kunigel V1, which is a typical Japanese bentonite and contains c.a. 50 % of montmorillonite, contacted with iron coupon for 15 years. Ten microliters of tracer solution that contained 1 kBq of ²³⁸Pu was spiked on the interface between iron coupon and compacted bentonite saturated with deionized water. After diffusion period, plutonium, iron and sodium distributions in the bentonite specimen were obtained by measuring with alpha scintillation counter for Pu and ICP-MS for Fe and Na, respectively. Plutonium penetrated into the bentonite to depth 2 mm. More than 90 % of plutonium wasn't moving from the interface. The color of the bentonite around the interface was black green like a magnetite according to watch observation. Iron was detected by the whole bentonite and existed with high concentration from the interface to 2 mm in particular. The concentration of sodium fell a little from the interface to 2 mm. It was supposed that ferrous ion was diffused into bentonite with corrosion of iron, and then precipitated as magnetite. The precipitation of magnetite could make pH in the bentonite fall, then dissolution and migration of plutonium. A small crack was also observed at 6 mm from the interface and could be made by migration of hydrogen occurred by corrosion.

Swelling clays are considered for nuclear applications since they present a multi-scale structure and therefore a multi-scale porosity hierarchy: macroporosity between aggregates / inter-particle mesoporosity / microporosity in the interlayer spacing, as described by mercury intrusion porosimetry, nitrogen adsorption and XRD experiments respectively. In this work, XRD measurements, water adsorption and calorimetry¹, thermoporometry² and electrical conductivity results are coupled with electrostatic calculations to investigate (i) the hydration sequence occurring in alkali/alkali-earth cations exchanged montmorillonites, and (ii) the diffusion properties of the interlayer cations.³
The comparison between montmorillonites saturated with homoionic alkali or alkali-earth cations ⁴ and montmorillonite saturated with both alkali and alkali-earth cations leads us to elucidate the impact of the nature of cations and the possible interactions existing upon hydration.
Furthermore the determination of the driving forces for hydration in swelling clays as a function of the nature of the interlayer cations allows us to better capture the ability of the swelling clays for ionic exchange.

1. F. Salles, J.M. Douillard, R. Denoyel, O. Bildstein, M. Jullien, I. Beurroies, H. Van Damme, Journal of Colloid and Interface Science, 2009, 333(2), 510-522
2. F. Salles, I. Beurroies, O.Bildstein, M. Jullien, J. Raynal, R. Denoyel, H. Van Damme, Applied Clay Science, 39(3-4), 186-201, 2008.
3. F. Salles, O. Bildstein, J.M. Douillard, B. Prélot, J. Zajac, H. Van Damme, Journal of Physical Chemistry C, 2015
4. F. Salles, O. Bildstein, J.M. Douillard, M. Jullien, J. Raynal, H. Van Damme, Langmuir, 2010, 26(7), 5028-5037

Compacted clays as engineered barriers are used for nuclear waste containment. However, the thermo-chemo-structural properties of clays which determine the integrity of these materials at elevated temperatures in the event of exothermic nuclear reactions are not well understood. In this study, we investigate the structure-property relationships of clays over a wide spatial (or “q”) range. The clays of interest include Na- and Ca-montmorillonite, and illite. We provide a coupled understanding of the microstructural changes in clays using ultra-small-angle X-ray scattering (USAXS), changes in the interlayer spacing using small-angle X-ray scattering (SAXS), and wide-angle X-ray scattering (WAXS), as the temperature is increased from 25 ^°C to 1150 ^°C. The key thermal events are the removal of interlayer water which occurs in the range of 25 °C to 300 ^°C, the dehydroxylation of clays between 200 °C and 800 ^°C, and the changes in the quasi-crystalline phases in clays to well defined crystalline phases on heating from 800 ^°C to 1150 ^°C. Changes in the X-ray scattering contrast that correspond to these thermal events are considered in determining the dynamic changes in the volume fractions, surface areas, porosities, and size distributions, which are related to the phase changes of the materials on heating. The insights from the quantification of the fundamental and dynamic structure-property changes in clays could be applied to develop more robust engineered barriers for nuclear waste containment.

Smectite clays are widely found on the Earth surface, and they are used as nuclear barrier systems¹. They are porous materials possessing connected mesopores in the micrometer range, in-between mineral grains, and nanopores inside the grains. These grains are stacks of individual 1 nm-thick clay particles (the layers) and have the ability to swell by incorporating H₂O molecules (or other molecules such as CO₂) in-between the layers, depending on the ambient temperature and on the humidity present in the mesopores surrounding the grain. Imposing a gradient of relative humidity RH along a temperature- controlled dry sample of smectite clay (synthetic fluorohectorite in the present case), we investigate the diffusive transport of water molecules in vapor phase through the material. As water molecules diffuse through the mesopores, some of them intercalate into the nanopores, causing the grains to swell and causes a decrease in the mesoporous volume available for vapor diffusion. We monitor this process using space- and time-resolved X-ray diffraction. We map the interlayer repetition distance (d-spacing) of the stacks in space and time we obtain humidity RH(x) profiles along the direction of the initial humidity gradient, at different times t. To model the data we consider a one-dimensional effective diffusion process described by a fractional time diffusion equation with a diffusion coefficient that depends on humidity, and we show that it is possible to rescale the humidity profiles onto a single master curve as a function of the parameter (x/t)^γ/2, where γ is an exponent characterizing the diffusion process as normal (γ=2), subdiffusive (γ<2), or superdiffusive (γ>2). We observe that γ strongly depends on the type of intercalated cation in the clay stacks, and we suggest that this is linked to the difference in time scales observed for the water adsorption dynamics by individual stacks for the two cation cases. The effective mechanism is reminiscent of retardation mechanisms known in other subsurface media, with a nanoscale trapping mechanism and a feedback effect of the mesoporous humidity on the local porosity of the medium.

1. Patrik Sellin, Olivier X. Leupin (2013) The Use Of Clay As An Engineered Barrier In Radioactive-Waste Management – A Review, Clays and Clay Minerals 61 (6) 477-498
2. Hemmen, H., Alme, L. R., Fossum, J. O., & Méheust, Y. (2010). X-ray studies of interlayer water absorption and mesoporous water transport in a weakly hydrated clay. Physical Review E, 82(3), 036315.
3. Michels, L., Méheust, Y., Altoé, M. A. S., dos Santos, E. C., Hemmen, H., Droppa Jr, R., Fossum J. O., da Silva, G. J., (2016) Water vapor transport in porous swelling clays: Control of normal vs. anomalous diffusion. Under review

Two isotopes of cesium, ¹³⁵Cs and ¹³⁷Cs, are among the most important contaminants associated with the nuclear fuel cycle because of their high fission yield, long half lives (2.3 10⁶ and 30 a, respectively), high solubility in water, and metabolic similarity to potassium. These radioisotopes are key concerns in the geologic storage of radioactive waste, the management of legacy sites contaminated during the development of nuclear power, and the remediation of soils contaminated by catastrophic releases such as those that occurred at the Chernobyl and Fukushima Daiichi nuclear power plants. Cesium has a well-known affinity for clay minerals and is likely to be primarily adsorbed on 2:1 clay minerals in soils, sediments, and sedimentary rocks. Experimental data indicate that illite, in particular, carries a small density of surface sites with a very strong affinity for cesium. These sites are widely thought to be located on the “frayed edges” of nanoparticles, but they are difficult to characterize at molecular scales. Here we discuss the methodological challenges associated with atomistic simulations of illite edge surfaces and present both large-scale molecular dynamics (MD) and thermodynamic integration calculations of cesium adsorption by illite nanoparticles contacting liquid water. New insights into the identities of different illite surface sites and their respective affinities for cesium are provided by our results.

One of the major concerns associated with the geologic storage of high-level radioactive waste (HLRW) is the production of hydrogen from the thermal and radiolytic breakdown of nuclear waste components. The accumulation of hydrogen in HLRW storage facilities has the potential to damage the integrity of these facilities by creating leakage pathways through engineered and natural barriers. Clay minerals are key constituents of these barriers in several countries, either as engineered clay barriers or clay-rich sedimentary rocks in the near- and far-field of potential HLRW repositories. At present, there remains considerable uncertainty regarding the behavior of H₂ in water-saturated argillaceous media. In this study, we use molecular dynamics (MD) simulations to predict the partitioning of H₂ between clay (specifically, Na-montmorillonite) interlayer nanopores and bulk liquid water. Our results reveal the impact of the clay surfaces on the coordination structure and diffusion dynamics of H₂ in liquid water. Our simulations of other gases (noble gases, methane, CO₂) provide additional insights into the behavior of H₂. Ultimately, the fundamental insights provided by our atomistic simulations of H₂-water-clay systems could be used to engineer more robust barriers for nuclear waste containment.

In case of container failure in a deep geological repository for spent nuclear fuel, intruding water possibly gets into contact with the waste. Besides anaerobic corrosion of the container material and consecutive production of hydrogen, radionuclides will be released into the groundwater. Initially a fast radionuclide release is expected to occur, which will include the release of of fission gases and volatile radionuclides segregated at the gap between fuel and cladding occur, at fuel fractures and at grain boundaries. Simultaneously to the fast leaching of the so-called instant release fraction, a relatively slow dissolution of the fuel matrix will begin, resulting in the release of matrix-bound radionuclides. The matrix dissolution is a long-term process, which will continue after the fast leaching processes will have been ceased. Dissolution of the fuel matrix will be affected both by the presence of molecular hydrogen and by water radiolysis. The later will result in the production of oxidizing radiolysis products (e.g. OH radicals and H₂O₂) in the micrometer scale vicinity of the fuel surface that might enhance the matrix dissolution, even if reducing conditions are expected to prevail within the near-field of the deep geological repository.
This paper focuses on the impact of H₂ overpressure on the release of uranium and other matrix-bound radionuclides, in particular actinides. Therefore, the dissolution behavior of an irradiated UO₂ fuel with an average burn-up of 50.4 MWd/kg_HM was studied under deep geological repository conditions with a simulated near neutral pH groundwater and H₂ overpressure. Two batch experiments were performed in autoclaves with a cladded pellet and fragments of the spent nuclear fuel. The samples were leached in bicarbonate water, which composition was 19 mM NaCl and 1 mM NaHCO_3, under (40 ± 1) bar of Ar/H₂ mixture, with a H₂ partial pressure of (3.2 ± 0.1) bar. The solution was completely changed after one day in order to reduce the amount of Cs in solution as well as to remove any potential U(VI) likely present as a pre-oxidized layer on the spent nuclear surface. Later on, the experiments continued for up to350-446 days in batch mode taking periodically 10 mL of liquid aliquots to analyze the amount of uranium / actinides released.
The concentration of uranium in solution was found to be virtually constant with time at (1 ± 0.1)×10⁻⁸ M in the cladded pellet experiment and at (1 ± 0.1)×10⁻⁷ M in the fragments experiment. In both experiments, aqueous concentrations of Am, Cm, Np and Pu remained below 10⁻⁹ M after 200 days. These observations indicate that the matrix dissolution of both types of samples, cladded pellet and fragments, is significantly inhibited under a partial pressure of H₂ of 3.2 bar in bicarbonate water.

Cement conditioning is a common solidification-stabilization (S/S) technique prior to near-surface or deep geological disposal of low- and short-lived intermediate level radioactive waste. The currently proposed disposal concepts further foresee the use of cementitious materials for the construction of the engineered barrier system (container, backfill and liners). In performance assessment it is considered that hardened cement paste (HCP) retards the release of radionuclides from the near field into the far field. In the long term HCP is subject to chemical alteration processes caused by the interaction with groundwater from the host rock. As a consequence of cement alteration, the mineral composition of HCP changes with time. Therefore, identification of the cement phases of HCP responsible for the uptake of radionuclides, i.e. identification of the so-called uptake-controlling cement phase, is indispensable with the aim of assessing the influence of the chemical evolution of the cementitious near field on radionuclide immobilization and the development of predictive modelling tools coupling cement alteration and radionuclide retention in the cementitious near field. The present trend in research is to advance molecular-level understanding of radionuclide interaction with cementitious matrices and thereby improve long-term predictions of radionuclide immobilization in the near field.
The core focus of the presentation will be on the uptake processes of actinides (e.g. U, Np) and dose-determining anions (e.g. I, Se) by cement phases and on the identification of the uptake-controlling cement phase in HCP. Recent advances in developing a mechanistic understanding and thermodynamic models of the interaction of actinides and dose-determining anions with cementitious materials will be presented. The combined information from macroscopic batch-type sorption studies and structural investigations allows the development of thermodynamic models of aqueous-solid solution systems predicting radionuclide uptake by cement phases. Determination of the end-member stochiometries of solid solutions requires structural information, such as on the long-range order of crystalline cement phases obtained from X-ray diffraction (XRD) measurements or on the local coordination environment of radionuclides in amorphous cement phases obtained from synchrotron- and laser-based spectroscopic techniques (e.g. X-ray absorption spectroscopy (bulk/micro-XAS) and time-resolved laser fluorescence spectroscopy (TRLFS)).

The main advantages of using cement materials for nuclear waste immobilisation are: i) the high internal pH, provided through hydration of cement minerals, which can supress the solubility of radioactive species in the waste and; ii) the cement minerals and their hydrates can act as selective binders for radioactive elements. Both of these immobilisation processes become increasingly important at longer time scales, when the wastes begin to break down within the cement, and radioactive elements are released, requiring uptake by the cement itself. The mineralogy of the cement, and particularly of the cement hydrate minerals that form is, therefore, of critical importance to the immobilisation of radioactive waste. Additionally, the evolution of the cement mineralogy with time is crucial in governing both structural and immobilisation efficiency of the cement.

Current understanding of the mineralogy of cement hydration, a thermodynamically controlled process by which cement minerals dissolve and new cement hydrate minerals are formed, in the long-term is limited in an experimental sense to a small number of common cement materials. Detailed kinetic information is particularly important in the context of UK nuclear waste cements, which contain supplementary cementitious materials, such as blast furnace slag and limestone powder. We present the results of a two-year study of the hydration kinetics of a number of nuclear waste cements, performed in the world’s first long duration synchrotron experiment, at the I11 x-ray diffraction beamline at Diamond Light Source. Using this technique, we have identified, at high temporal resolution, the presence of transient, trace and poorly crystalline cement hydrate minerals, which are key to identifying the cement hydration mechanism and predicting long-term performance within a geological disposal facility. Additionally, we describe the hydration products of a novel cement backfill material, developed with enhanced radionuclide (⁹⁹Tc) sorption capacity.

At the Savannah River Site (SRS) low activity salt solution is stabilized via combination with a mixture of blast furnace slag (BFS), fly ash (FA), and ordinary Portland cement (OPC) to form a grout referred to as saltstone. The flowable grout is emplaced into large concrete tanks, termed Saltstone Disposal Units (SDUs), where it cures to encapsulate the waste. Technetium-99 (⁹⁹Tc) is a long-lived radionuclide contained in the low activity salt waste and subsequently incorporated into the grout waste form: it is considered a significant contributor with respect to the long-term transport of radioactive material to the environment surrounding the SDUs. In the reducing, high pH cementitious environment within the grout, ⁹⁹Tc is expected to be relatively immobile since it exists in a reduced Tc(IV) oxidation state in the form of sparingly soluble sulfides (TcS_x) or hydrated oxides (TcO₂.xH₂O). It is believed that sulfide (S^2-) and ferrous iron (Fe²⁺), both present in the BFS component of saltstone, serve as Tc reductants. However, in the presence of O₂ (associated with the future infiltration of air or oxygenated ground waters into the saltstone monolith) it is possible for redox-sensitive Tc(IV) to transition into highly soluble (and mobile) Tc(VII) species, such as pertechnetate (TcO₄^-).

Traditional approaches to quantifying the leaching behavior of ⁹⁹Tc from cementitious matrices have involved partitioning experiments using ground saltstone samples, and measurement of adsorbed or desorbed ⁹⁹Tc. Such experiments create artificially high solid-solution contact areas that give rise to higher ⁹⁹Tc leachate concentrations than would be expected for an intact sample. In the current study, different techniques have been used to investigate the ⁹⁹Tc leaching behavior of intact (not ground) monolithic saltstone samples. These studies are intended to inform the SRS Saltstone Disposal Facility (SDF) Performance Assessment (PA) which models the long-term transport of radionuclides from the SDUs to the surrounding environment. The techniques employed include EPA Method 1315, a recently adopted method for evaluating contaminant leaching from intact monolithic materials, and a novel Dynamic Leaching Method (DLM). The latter technique utilizes a flexible-wall permeameter to achieve saturated leaching under an elevated hydraulic gradient in an effort to evaluate the persistence of reductive capacity, and subsequent changes in contaminant partitioning within intact cementitious monoliths. Simulant saltstone samples spiked with ⁹⁹Tc, and actual saltstone samples extracted from an SDU, have been assessed utilizing the aforementioned methods. The effect of dissolved oxygen (DO) in the permeant solutions has also been evaluated. Initial findings for both techniques indicate that the ⁹⁹Tc leaching behavior is likely controlled by the solubility of TcO₂.xH₂O compounds irrespective of DO in the permeant.

Numerous elements such as uranium, plutonium, minor actinides and fission products are present in the spent nuclear fuel. When the fuel is dissolved in nitric acid a precipitation that involves the fission products molybdenum and zirconium can occur, leading to the zirconium molybdate compound, Zr^IVMo₂O₇(OH)₂(H₂O)₂¹. Depending on the chemical conditions, the concomitant presence of tetravalent elements in solution like tetravalent plutonium or cerium can alter the composition and/or nature of the precipitated solid either by substituting Zr^IV in Zr^IVMo₂O₇(OH)₂(H₂O)₂ or by precipitating new crystal structures.

This work deals with a better understanding of the precipitation mechanisms involved in the Pu^IV/Zr^IV/Mo system using first Ce^IV as a chemical analog of Pu^IV then directly Pu^IV in targeted experiments. Precipitations were carried out in nitric acid 3M using various M^IV/(M^IV+Zr^IV) (M=Ce or Pu) ratio and two aging times: 3 and 20 hours. Depending on the aging time and the M^IV/(M^IV+Zr^IV) ratio different phase compositions were highlighted by using powder and single crystal X-Ray diffraction and by refining the unit cell parameters.
In the Ce^IV/Zr^IV/Mo system, after 3 hours, three domains were identified:

- a solid solution Zr_1-xCe_xMo₂O₇(OH)₂(H₂O)₂ for the lowest cerium molar fractions,
- a definite compound Zr_0,5Ce_0,5Mo₂O₇(OH)₂(H₂O)₂ for the intermediate cerium fractions,
- an unknown phase mixed with Ce₃Mo₆O₂₄(H₂O)₄² for the highest cerium molar fractions.

Increasing the aging time to 20 hours affects the phase composition in that the unknown phase is not found anymore and a solid solution, Ce_3-xZr_xMo₆O₂₄(H₂O)₄, appears in place of the definite compound, Ce₃Mo₆O₂₄(H₂O)₄, in the [90%≤X_Ce≤100%] domain. Zr_0,5Ce_0,5Mo₂O₇(OH)₂(H₂O)₂ and
Ce_2.7-Zr_0.3Mo₆O₂₄(H₂O)₄, ending members of the solid solutions, Zr_1-xCe_xMo₂O₇(OH)₂(H₂O)₂ and Ce_3-xZr_xMo₆O₂₄(H₂O)₄, are found in a large domain.

Beyond its role on the phase composition, the Ce^IV substitution rate influences the microstructure since a low Ce^IV content tends to form cubic particles whereas a high Ce^IV one leads to cauliflower type agglomerates.
Some insights into the influence of other concomitant elements in solution, such as tellurium which add effects on the zirconium molybdate precipitation mechanisms³, and preliminary transposition to tetravalent plutonium based systems are then finally considered.

1.Clearfield, A.and Blessing, R. H. The preparation and crystal structure of a basic zirconium molybdate and its relationship to ion exchange gels. J. Inorg. Nucl. Chem. 34, 2643–2663 (1972).
2.Cross, J. N. and al. From yellow to black: Dramatic changes between cerium(IV) and plutonium(IV) molybdates. J. Am. Chem. Soc. 135, 2769–2775 (2013).
3.Magnaldo, A. Use of certain chemical elements for inhibiting the formation of precipitates containing zirconium molybdate in an aqueous solution containing the element molybdenum ans the element zirconium. PATENT EP2010/066212 (2010).

Thermal treatment of legacy sludge waste arising from the United Kingdom’s Magnox nuclear fuel is a novel technique. Vitrification technology offers higher waste loading, volume reduction and greater durability compared to the current baseline plan; storage within an engineered facility whilst awaiting cementation. Cost effective in the short term, cementation increases the volume of the waste to be disposed by over 300% increasing the cost of final disposal in the UK’s planned geological disposal facility and hence durable glass wasteforms are potentially more effective at clearing up this legacy waste.
Two extremes for the waste found within the First Generation Magnox Storage Ponds were proposed, a highly corroded waste and a metallic waste. Borosilicate and aluminosilicate glass samples were batched based off MgO-B₂O₃-SiO₂ (MBS) and MgO-Al₂O₃-SiO₂ (MAS) phase diagrams and doped with the two waste streams. Non-radioactive neodymium and mischmetal were run alongside uranium for a surrogate compatibility study, with the same processing and analysis techniques used. Multiscale characterisation was accomplished with a variety of techniques to determine what crystalline phases are present, uranium oxidation states, melting temperatures of the glass and long term aqueous durability.
MBS glass samples melted at 1250 °C and MAS samples melted at 1500 °C forming fluid glass demonstrating passive oxidation and digestion of metallic waste into the glass network. XRD and SEM-EDX of the highly metallic waste glass showed phase separation and a high degree of crystallisation whereas glass from the corroded waste stream appeared single phase with no crystallisation. Various crystal phases evident from XRD were confirmed by SEM-EDX imaging identifying fused and dendritic crystals such as UO₂, NdB₆ & Nd_0.3Ce_0.7O_1.85 that are likely to have formed during the melting process and allowed to grow during casting and annealing. U L_III edge XANES proved uranium had been oxidised higher in MAS samples due to higher processing temperatures. DTA observed glass transition between 617 °C and 672 °C with at least two distinct crystallisation points observable whilst liquidus temperature was reached at 1420 °C in MAS samples and between 1135 °C and 1153 °C for MBS samples. Dissolution of glass samples over 28 days was observed to be very low with little release of uranium or neodymium into solution. A pH buffer from 8.5 to 12 in MAS samples saw higher rates of dissolution with boron release rates over 3x higher compared to MBS samples. Uranium was detected leaching significantly from MAS samples during the first 7 days until reaching a steady rate due to the formation of alteration layers visually confirmed by SEM-EDX analysis of 28 day altered samples. Volume reduction achieved by vitrification of Magnox sludge could be as great as 80% compared to the current baseline plan with a cost saving of approximately £83 million ($120 million) for long term storage.

Borosilicate glassy materials for immobilization of HLW from Russian WWER (PWR) spent nuclear fuel reprocessing were designed, synthesized in a resistive furnace, and characterized by XRD, SEM/EDS, and FTIR spectroscopy. Hydrolytic durability was determined by PCT-A procedure and compared to EPA glass and reference data. The glasses with 20 and 25 wt.% waste loading were found to be X-ray amorphous, homogeneous and hydrolytically durable. Glass network has formally relatively low degree of connectedness but it is increased due to embedding of different structural groups thus improving hydrolytical durability. Boron is present primarily in trigonal oxygen coordination. The glasses with 30-35 wt.% waste loading contained minor britholite phase concentrating rare earth elements and as expected trivalent actinides. Glassy product with up to 30 wt.% waste loading was also produced by cold crucible inductive melting at the IPCE RAS lab-scale unit equipped with 56 mm inner diameter copper cold crucible and energized from a 10 kW/5.28 MHz generator. The product was composed of vitreous phase and minor britholite with average composition K_0,39Sr_1,99Fe_0,16Nd_5,50Si_7,96O_26,60.

Cementation is often used as a treatment strategy for materials contaminated with fission products. A potential problem with cementing metallic waste is internal metal corrosion within the grout, which has been observed for cemented Magnox fuel cladding, leading to expansion of the wasteform and hydrogen gas production. Reworking such problematic cemented wasteforms can be achieved using thermal treatment to produce a passively-safe vitreous wasteform.

In this study representative cemented Magnox wasteforms were thermally treated at 1200 and 1300^oC with the addition of readily-available glass additives (sand, boric oxide and sodium carbonate). Four different waste loadings were used- 89%, 72%, 46% and 36%. At higher waste loadings a porous slag-like wasteform is produced, with mineral phases present, whereas at lower waste loadings a homogenous glass is produced. Spinel phases were present in most of the glasses, determined via SEM/EDX and XRD, and the formation of the spinels was linked to oxidation of stainless steel (from the original cemented waste). Also present in the waste is a small amount of uranium (~0.5wt%). This was added to one of the glass melts and found to be incorporated into the glass structure, with the average oxidation state of uranium, in the glass, found to be 4.06-4.11 as determined by XANES.

Crystallization of aluminosilicates during the conversion of Hanford high-level waste (HLW) to glass is a function of the composition of the glass-forming melt. In high-sodium high-aluminum waste streams, the crystallization of nepheline (NaAlSiO₄) removes chemically durable glass-formers from the melt, leaving behind a residual glass that is enriched in less durable components, such as sodium, boron, and iron. We seek to further understand the effect of boron and iron addition on the crystallization of sodium aluminosilicate glasses as analogues for the complex waste glass. In these systems, boron and iron behave as glass intermediates which allow for crystallization when present in low additions but frustrate crystallization at high additions. Furthermore, boron and iron addition promotes the formation of nepheline over carnegieite (the high-temperature polymorph). In this work, we seek to understand the crystallization behavior of model alkali aluminosilicates as a function of boron or iron addition. Crystallization behavior will be measured through X-ray diffraction and vibrating sample magnetometry for iron bearing samples. We will then seek to compare the average structures of quenched and heat treated glasses through Raman spectroscopy and X-ray scattering. Finally, we seek to compare the crystallization behavior and structures of the feldspathoid endmember glasses of this study (NaAlSiO₄, NaBSiO₄, and NaFeSiO₄) to other chemically related endmember glasses in the feldspar-like (NaAlSi₃O₈, NaBSi₃O₈, NaFeSi₃O₈) and pyroxene-like (NaAlSi₂O₆, NaBSi₂O₆, NaFeSi₂O₆) families. Such a comparison will provide further insight on the complex relationship between the average chemical ordering and topology of glass on crystallization.

To turn nuclear waste to glass via vitrification, the waste is mixed with glass-forming and glass-modifying additives and charged into an electric melter. Feed-to-glass conversion progresses in response to heating. In the melter, conversion occurs in a cold cap composed of a reaction layer and a foam layer. The conversion heat comes from molten glass on which the cold cap floats. The foam layer at cold cap bottom hinders the heat flow and thus has a negative impact on the rate of melting. The response of melter feed to heating is affected by the choice of feed materials, including the particle size of chemicals and minerals. This study focuses on the effects of alumina sources on melting behavior of a high-alumina high-level waste melter feed. The feeds were prepared with three different sources of alumina: gibbsite [Al(OH)₃], boehmite [AlO(OH)], and corundum [Al₂O₃]. We observed the volume expansion of these feeds in response to heating and determined their conversion heat using the differential scanning calorimetry (DSC). For the volume expansion study, dry feeds were pressed into pellets, heated in a furnace from room temperature to 1100°C at 5 and 10 K/min, and monitored with a camera through a quartz-glass window. The volume of pellets was calculated from the profile area of the images by numerical integration. The reaction heat and heat capacity were obtained using the thermogravimetric analyzer equipped with the DSC. Dry feeds were heated from room temperature to 600°C at 10 K/min using the run/rerun method. Foaming was more extensive and started at a lower temperature in the melter feed with corundum than in those with gibbsite and boehmite. The total mass change of feed with gibbsite, ~19%, was greater than that of feeds with boehmite, ~17%, and corundum, ~4%. The mass loss within the temperature interval from 200°C to 600°C was caused by the evolution of H₂O and CO₂. As expected, the reaction heat of feeds with gibbsite and boehmite was higher than that of feed with corundum. The high-alumina high-level waste melter feed with gibbsite and boehmite are known to melt faster than a feed with corundum. This indicates that foaming tends to affect the melting behavior of high-level waste feeds to a greater extent that the conversion heat.

The technical specification of a containment glass composition is based on a trade-off between three requirements:
●the glass microstructure must be homogenous, demonstrating a complete incorporation of the chemical elements of the waste,
●the glass must achieve long-term durability,
●the glass elaboration process must be achievable at an industrial scale.
These three objectives are considered at the early stage of glass formulation studies. These objectives can be translated in terms of constraints on physical and chemical properties of glass melt and final glass. Research performed at Marcoule site within the CEA-AREVA Joint Vitrification Laboratory is based on both accumulated knowledge since the sixties and industrial lessons from experience. In borosilicate glass some components of the waste streams are not easily incorporated into the glass structure. Among them, molybdenum and rare-earth elements have limited solubility in the glass melt that can induce liquid-liquid phase separation and/or crystallisation issues. This presentation points out the different stages of the glass elaboration process that can be concerned by both these phenomena and key parameters that consequently have to be well controlled.

After 30 years since the meltdown of reactor 4 at the Chernobyl nuclear power plant in Ukraine, scientists are no closer to fully understanding the behaviour of the lava-like radioactive materials formed during the explosion. Vital to the long term remediation of the site, the removal of these materials from the reactor and the sub-reactor buildings is a key step and can only be completed once a thorough understanding of the composition and microstructure of the lavas has been accomplished. The Chernobyl lava-like fuel contaminated materials (LFCM) were formed from the melting of fuel in the reactor core during the meltdown, combining the fuel with the core itself and surrounding structural materials such as concrete and steel, forming a highly crystalline glass like material that subsequently spread like molten lava through the building beneath the core. Simulant LFCM is needed to understand the durability of these materials and to determine the rate of dissolution of these materials into groundwater which could assist in the development of glass wasteforms for the long term disposal of civil nuclear waste in engineered facilities.
Synthesis of simulant LFCM batched from analysed compositions of actual LFCM from Chernobyl has been successful in producing material that is microstructurally similar without the radiological hazard. Melting of the batch materials in a tube furnace at high temperature, and under reducing atmosphere, with controlled cooling to room temperature was used to simulate conditions of lava formation. Post synthesis heat treatment was required to grow the crystalline phase assemblage for detection under an electron microscope. Characterisation using XRD and SEM-EDX has identified several crystal phases including high uranium zircon, uranium oxides and zirconium, as well as un-melted metal particles all of which are surrounded by a glassy phase. The UO_X + Zr phase morphology was very diverse from fused crystals to dendritic crystallites from re-crystallisation of uranium initially dissolved in the glass phase. Stainless steel particles were also detected. High uranium zircon was often found with the stainless steel and not found in isolation from other zircon crystals.

Once a severe nuclear accident is over and cooling established, a thorough assessment of the remediation of the nuclear power plant (NPP) is needed. The degree of the accident will have an important impact. At Fukushima, melting was pronounced and in at least one core the vessel was breached, while the spent fuel ponds were also affected, with potential physical damage to the fuel, without melting. During the accident, spent nuclear fuel (SNF) assemblies were exposed to emergency cooling aqueous media, especially borated and sea water, all of which can have different effects on the stability of damaged SNF. It is therefore important to provide basic data on stability of damaged SNF in these aqueous media to offer science-based advice and indication for potential radionuclide releases and their environmental impact.
In order to assess the stability of damaged SNF in aqueous cooling media used in Fukushima, experiments were carried out at ITU in collaboration with CRIEPI using a BWR and a PWR SNF with a burn-up of 54 GWd/tHM and 60 GWd/tHM, respectively. Static experiments were performed in sea water (SW), either from the Japanese Pacific Coast or simulated, and borated water (BW) saturated with air and at hot-cell temperature. In all cases the solution was continuously stirred and it was renewed several times during the experiments. The results were compared with similar experiments carried out in deionized water (DW) and in simplified groundwater (GW, 1 mM NaHCO3 and 19 mM NaCl).
After two weeks results show uranium dissolution rates in SW to be one order of magnitude faster than in GW and two orders of magnitude faster than in BW and DW, while Pu dissolution rates are similar in all the studied media. Instant release fraction elements like Cs do not differ significantly in all the cases studied and, as expected, presents dissolution rates orders of magnitude higher than those for uranium

Coffinite (USiO₄) is expected to play an important role in the field of direct storage of spent nuclear fuels in underground repository since it could control the uranium concentration in groundwaters. However, the thermodynamic properties associated with coffinite, especially the solubility product, remain poorly defined.
Coffinite precipitation (from a mixture of U(IV)-containing acidic solution and sodium metasilicate) then purification step were optimized to prepare pure synthetic samples suitable for solubility experiments [1]. Thus, several assemblages were submitted to under-saturated dissolution experiments performed in 0.1 mol L^-1 HCl in air or under Ar atmosphere. The solubility constant of coffinite was determined (log *K_S°(USiO₄, cr) = -5.25 ± 0.05), as well as the standard free energy of formation of coffinite (Δ_fG°(298 K) = -1867.6 ± 3.2 kJ mol^-1), which enables one to infer the relative stability of coffinite and uraninite as a function of groundwater composition. The standard free energy associated with the formation of the coffinite from a mixture of uraninite and quartz was found to 20.6 ± 5.2 kJ mol^-1, in agreement with data obtained from calorimetry measurements [2], that indicates unambiguously that coffinite is less stable than the quartz + uraninite mixture at 298 K.
Coffinite is thus formed through a precipitation mechanism involving the dissolution of uraninite in silica rich solutions. Geochemical simulations using PHREEQC 2 software indicate that this phase precipitates in solutions supersaturated with respect to UO₂(cr), but undersaturated with respect to UO₂(am) in aqueous solutions with silica concentrations typical of groundwater [2]. These favorable conditions during the formation of sedimentary uranium ore deposits, as well as slow dissolution kinetics, explain the common occurrence of coffinite.

[1] A. Mesbah et al., Inorg. Chem., 54 (2015) 6687-6696
[2] S. Szenknect et al., Geochimica Cosmochimica Acta, 181 (2016) 36-53

For the safe disposal of high-level nuclear waste, such as spent nuclear fuel (SNF), a mechanistic understanding of possible corrosion processes and the release of radionuclides in a deep geological repository is indispensable. SNF from light water reactors is very heterogeneous and one of the most complex materials. During the operation inside a nuclear reactor the composition and microstructure of the UOX-fuel is changed. Besides UO₂, SNF consists of: (1) fission gas bubbles, (2) oxide precipitates of e.g. Rb, Zr which are referred to as “grey phase”, (3) metallic particles of e.g. Mo, Tc, Pd mainly known as epsilon-particles and (4) solid solutions of rare-earth elements (REE) as well as actinides which are formed with the UO₂. Due to the chemical and microstructural complexity of SNF and its high beta- and gamma radiation field, fresh SNF is not suitable to gain straightforward mechanistic insight into the corrosion of aged SNF. One consequence of the complexity is that corrosion is driven by various independent and simultaneous reactions and it is difficult to link certain dissolution phenomena to individual reactions.
Here, we present a new approach to derive a mechanistic understanding of important SNF long-term matrix corrosion processes. The long-term matrix corrosion is controlled by the electrochemical properties of the system, which are influenced by the concentration and distribution of REE and epsilon-particles in the UO₂ matrix. Simplified UO₂-based model systems are prepared and characterized with respect to their structure and their corrosion kinetics to complement studies on irradiated "real" SNF. To address the relationship between the REE concentrations, their structural environment and potential synergies with epsilon-particles and the long-term matrix corrosion, we pursue a bottom-up approach, from simple to more complex systems. As a first step UO₂ pellets containing 1-2 wt% Nd are synthesized. Nd serves as fission product surrogate of the complete REE-series and REE concentrations of SNF are mimicked. To go one step beyond the pure doping of UO₂ ceramics with REE, metallic particles will be introduced into the matrix. A key aspect is to adjust the microstructure and grain size of these UO₂-based model systems to represent the microstructure of the central region of a SNF pellet. We have applied and compared various wet chemical approaches such as the weak acid resin process, sol-gel routes as well as coprecipitation routes. Characterization of the pellets by XRD, SEM and EDX confirmed the formation of a homogeneous solid solution and appropriate microstructures to mimic some aspects of the microstructure of SNF. In depth characterization by neutron total scattering provides insight into the oxygen sublattice. This allows to link structural properties of these UO₂-based model systems to their corrosion behavior and to understand highly relevant long-term matrix corrosion mechanisms of SNF.

Coffinite (USiO₄) is an abundant form of U(IV) found in both sandstone and hydrothermal deposits. Characterization of the thermodynamics and kinetics of coffinite precipitation and dissolution will enable scientists to understand the long-term stability of used nuclear fuel in a geologic repository. Coffinite readily forms in natural systems; however, laboratory synthesis has proved challenging. Even with recent achievements in the synthesis and characterization of coffinite, the complex process of coffinitization of uraninite remains unclear. Quantifying the energetics of different coffinite surfaces and the relationship between potential uraninite and coffinite interfaces can elucidate the role of uraninite as a template for coffinitization. In this work, density functional theory calculations are used to predict the surface energies of coffinite, which are compared with calculated and known surface energies of uranium dioxide (UO₂). Thus far, surface energy calculations were performed for the (100), (110), and (001) surfaces of coffinite. The (100) surface is terminated with singly-coordinated Si and U atoms (i.e., each surface cation is bonded to one surface O atom), while the (110) surface is terminated with doubly-coordinated Si and U atoms. For both terminations, the surface O atoms are bonded to one U and one Si atom. The (001) surface is more complex with the doubly-coordinated U atoms and uncoordinated Si atoms. Future work will include necessary hydration of the surfaces. Preliminary calculations were performed to ensure appropriate slab thickness (i.e., three to eight layers each containing two formula units), surface termination, and spin configurations were used in the final computation of the surface energies. The bulk coffinite spin configuration is antiferromagnetic with the spin alternating every two uranium atoms along the [001]. The bulk spin configuration was maintained in the formation of the slab geometries and throughout the slab optimizations for (100) and (110) coffinite surfaces; however, spin contamination was observed for the (001) surface. As expected, the surface energy trends for the (100) and (110) system decreased with increasing slab thickness. The surface energy for the (100) and (110) surfaces converge around six layers thick, with associated surface energies of 0.64 Jm^-2 and 1.02 Jm^-2, respectively, indicating that the favorable surface for coffinite is the (100) as compared with (110). As a comparison, the quantum-mechanically determined surface energies for the (100), (110), and (111) surfaces of uraninite are 1.04 Jm^-2, 0.73 Jm^-2, and 0.27 Jm^-2, respectively (Skomurski et al. Am. Mineral. 2006). Quantification of additional coffinite surface energies, comparison of coffinite surface energy trends with zircon surface energy trends, and structural comparison of dominant coffinite surfaces with the uraninite (111) surface will lead to a greater understanding of the coffinitization of uraninite.

Aiming to mitigate greenhouse gas emissions, nuclear has become more and more attractive comparing to the traditional fossil-fuel based energy sources. However, the management of radioactive wastes has been a grand challenge that hinders the further development of nuclear energy industry. The short-lived and low-level radioactive wastes are routinely disposed in qualified sites; while the disposal of high level radioactive waste (HLW) is still facing a number of major problems. With more than decades of study, many potential nuclear waste forms have been proposed for disposing HLW and these forms can be classified as glass, ceramic and glass-ceramic types. The industry commonly use glassy form to handle radioactive wastes due to its good chemical durability and the flexibility of accommodating a wider range of waste compositions. However, the glassy form often has very low solubility of actinides (U, Pu, Np, Am, Cm) which are rich in HLW from modern nuclear reactors. Ceramic waste form is primarily designed to immobilize the separated radionuclides from HLW. Disadvantages of ceramic waste form are: (1) easy to form secondary phase that is water soluble; (2) may undergo a transformation of crystalline to amorphous content which leads to the decrease of chemical durability and inferior physical properties. Glass-ceramic waste form aims to avoid the drawbacks of both glassy and ceramic waste forms but take the advantages of them through the double-barrier containment concept to immobilize radionuclides. In our studies, zirconolite-based glass-ceramics, in which zirconolite is designed as the hosted crystalline phase of actinides, have been successfully synthesized with the SiO₂-Al₂O₃-CaO-TiO₂-ZrO₂-Na₂O-Ln-Oxide (Ln=Ce, Nd, Gd, Yb) system via a two-step thermal treatment scheme. The partitioning ratios (PR) of actinide surrogate (Ln) between zirconolite and the glass were studied by Rietveld quantitative X-ray diffraction analysis (QXRD) and transmission electron microscopy-energy dispersive X-ray spectroscopy (TEM-EDX). The effects of crystallization temperature, surrogate concentration and surrogate radius on the PR were revealed. Up to 41% of Nd³⁺ was incorporated into zirconolite after crystallized at 1000 °C for 2h while a further increase of crystallization temperature led to a decrease of PR. Higher Nd₂O₃ concentration in the system decreased the PR, and the larger ionic radius of lanthanides decreased the PR. This study provides an effective method to quantify the partitioning of actinide surrogates in crystalline and glass phases within the glass-ceramic matrix. It can largely contribute to the future designs of reliable glass-ceramic systems optimized for incorporating different radionuclides into specific hosting phases.

Different pyromettalurgical processes (“pyroprocesses”) have been developed for recovering the actinide elements in spent nuclear fuel by electrotransport through molten salt eutectic ( typically LiCl-KCl) to solid and liquid metal cathodes [1]. This waste salt contains alkali-metals, alkaline-earth, rare earth and minor actinides elements and can be considered as a high level waste containing chloride salts. Such chlorides constitute an issue in the production of vitreous waste due to their low solubility and negative impact on the chemical durability [2]. This is why salt occluded zeolites have been identified as potential immobilisation matrices for waste salts. In this study, we will mainly focus on the capacity of retention of the zeolite using a model salt. The wasteform is then sintered using HIP (High Isostatic Pressing) and the leach resistance of the wasteform is then assessed. The aims are to find an easy way of recycling the molten salt for further use in pyroprocessing and to consolidate the final powder using HIPing, in order to produce the most durable waste achievable.


[1] I. Taylor; M. Thompson; T. Johnson In Proc. Int. Conf. on Future Nuclear Systems: Emerging Fuel Cycles and Waste Disposal Options (GLOBAL 93) 1993; Vol. 1, p 690.

[2] W. E. Lee; M. Gilbert; S. T. Murphy; R. W. Grimes, Opportunities for advanced ceramics and composites in the nuclear sector Journal of the American Ceramic Society, 2013, 96, 2005.

Pyrochlores, a class of complex oxides with formula A₂B₂O₇, are one of the candidates for nuclear waste encapsulation, due to the natural occurrence of actinide-bearing pyrochlore minerals and laboratory observations of high radiation tolerance. In this work, we use atomistic simulations to determine the role of surfaces, chemical interfaces, and cation disorder on the plutonium immobilization properties of pyrochlores as a function of pyrochlore chemistry. We find that both Pu³⁺ and Pu⁴⁺ segregate to the surface for the four low-index pyrochlore surfaces considered, and that the segregation energy varies with the chemistry of the compound. We also find that pyrochlore/pyrochlore bicrystals A₂B₂O₇/A’₂B’₂O₇ can be used to immobilize Pu³⁺ and Pu⁴⁺ either in the same or separate phases of the compound, with the solute always segregating to the phase containing the larger ion of the same valence. Finally, we find that Pu⁴⁺ segregates to the disordered phase of an order/disorder bicrystal, driven by the occurence of local oxygen-rich environments. However, Pu³⁺ is weakly sensitive to the oxygen environment, and therefore only slightly favors the disordered phase. Together, these results provide new insight into the ability of pyrochlore compounds to encapsulate Pu and suggest new considerations in the development of waste forms based on pyrochlores.

High radiation and chemical attacks on the extractant used at spent nuclear fuel reprocessing and high-level waste IHLW) partitioning lead to the formation and accumulation of impurities, which can deteriorate its fire and explosion safety characteristics. During the operation, the degraded extractant may find a way into the high-temperature stages of nitrate solutions evaporation, molten uranyl nitrate production, etc. In many countries, including Russia, there were accidents at radiochemical plants as a result of the thermal explosion due to the interaction of oxidants with an extraction system subjected to radiation and thermal impact. That is reason why a study of the thermal stability of the considered extraction system under ionizing radiation is an important task
A study of thermal stability of a 30% solution of tri-n-butyl phosphate (TBP) in an Isopar-M diluent, saturated with 4.3, 8.2 and 12.0 M nitric acid, and the characterization of its gas release under atmospheric pressure depending on the temperature, and the concentration of nitric acid and doses of pre-irradiation to determine the safe operating conditions was performed. Experiments on thermal oxidation were conducted witin the range of the thermostat temperatures from 70 to 170 °C.
Pre-irradiation of TBP solutions in Isopar-M was carried out using a linear electron accelerator UELV-10-10-C-70 (energy is 8 MeV, pulse duration is 6 ms, pulse frequency is 300 Hz, the average beam current is ≤800 mA, scan width is 245 mm, scan frequency is 1 Hz). The organic solutions were exposed to doses of 0.5; 1 and 2 MGy.
The results of this study showed that heating the single-phase extraction systems in open vessels, despite significant content of oxidant, does not create conditions for the development of intense autocatalytic oxidation. In open vessels high heat loss due to evaporation and boiling of system components is realized due to intermixing of the phases by releasing gaseous products preventing the run of autocatalytic exothermic processes with the self-heating of the mixture in the organic phase. In addition, the accumulation of significant content of oxidants in the organic phase at heating in open vessels does not occur.
Infrared studies showed that among radiolytic products nitro compounds and carboxylic acids dominate. The concentration of carboxylic acids particularly sharply increases with the increase of absorbed irradiation dose. Traces of esters are also present. NMR studies have shown that the relative reduction in TBP concentration at the absorption dose of 500 kGy was ~11%.
This study was financially supported by the Ministry of Education and Science of the Russian Federation in the framework of the Agreement with IPCE RAS, the unique identifier RFMEFI60414X0153.

Fire and explosion safety of extraction systems used at the facilities of the nuclear fuel cycle reprocessing is largely determined by their thermal and radiation stability under the conditions of the technological process. At the same time ensuring the radiation resistance is related to the control and limitation of the amount of radiolytic products which depends on the irradiation dose. This is also true to the extraction process of high-level waste (HLW) partitioning, including the UNEX-process.
As follows from reference data [1-3] phenyltrifluoromethylsulfone (FS-13) has a high flash point and sufficient radiation and thermal stability. The aim of this study was to investigate the explosion and radiation resistance of extraction mixtures based on CMPO in FS-13. Ionizing radiation of waste nuclides was simulated by irradiation of samples using a linear electron accelerator UELV -10-10 C70 with vertically scanning electron beam up to absorbed doses of 0.5; 1 and 2 MGy. Experiments were carried out in open and closed vessels. Thermostat temperature during these experiments was 150 °C; autoclave temperature - 170-200 °C.
The results of experiments to determine the nature of the influence of irradiation on the gas release from 0.2 mole/L CMPO in the FS-13 with an aqueous phase of 14 mole/L nitric acid showed that at external heating temperature of 200 °C gas release in this mixture increases from a temperature of about 120 °C and proceeds at a constant rate up to the temperature of the mixture above ~ 160 °C, at which a self-heating occurs (exothermic process). However, the exothermic process was relatively weak and short-lived followed by temperature and pressure stabilization. The data obtained conclude HLW partitioning process using the extractant based on CMPO in the FS-13 diluent is fire- and explosion safe at the conditions investigated.
This study was supported by the Russian Science Fund (project 16-19-00191).
1. V.N. Romanovsky. Ph.D. Thesis: St-Petersburg, 2001. P. 46-56.
2. P.K. Sinha, K. Shekhar, U.K. Mudali, R. Natarajan // J Radioanal. Nucl Chem. 2011. V.289. P. 899-901.
3. B.J. Mincher, R.S. Herbst, R.D. Tillotson, S.P. Mezyk // Solvent Extraction and Ion Exchange. 2007. V.25. P 747-755.

Physical and chemical properties of fuel debris generated at the accident in the Fukushima Daiichi Nuclear Power Station have not been characterized sufficiently by investigation inside the primary containment vessel, and a concept of fuel debris disposal has not been yet decided. In order to understand the outline of barrier system required for the fuel debris disposal, sensitivity analysis of radionuclides migration on the properties of fuel debris and safety functions of engineered and natural barriers was carried out under the assumption of a similar disposal concept of high-level vitrified radioactive wastes (HLW) in Japan. Fuel debris was assumed to be directly contained in metal containers without vitrification, while parameters on their safety functions were set up the same as HLW disposal for the analysis in a basic case. Considering uncertainties on the properties of fuel debris by reference to previous studies, we evaluated migration fluxes of radionuclides with conservative values of parameters such as dissolution rate, instant release fraction (IRF) of the fuel debris using the radioactive inventory of the fuel debris. The dissolution rate and the IRF were set to values as a standard based on data for fuel matrix in geological disposal for spent fuels [1]. In addition, we evaluated the migration fluxes changing lifetime of the overpack and buffer material's function, of which parameter was expressed by combination of permeability coefficient, diffusion coefficient and distribution coefficient were 0.1, 0.1 and 10 times as the standard value, respectively in the case of promoting migration.
As a result, the peaks of migration fluxes at the outlet of the natural barrier in terms of 4n+2 series nuclides including the main radionuclide of fuel debris, U-238, showed much the same as those in HLW disposal system, even though the uncertainties of properties of the fuel debris and the engineered barrier were considered. And the fluxes of the other series nuclides and fission product radionuclides such as Se-79 and Cs-135 were smaller than those in HLW disposal. On the other hand, the peaks of migration fluxes of C-14 and I-129 which were not contained in HLW have increased the sensitivity for the IRF of the fuel debris, because their distribution coefficients were relatively lower. In addition, the migration flux of C-14 of which half-life was 5,730 years was sensitive to lifetime of the overpack which was 40,000 years. The results indicated that it is important to understand the radioactive inventory of C-14 and I-129 in fuel debris for disposal and the physical and chemical properties at release of the nuclides from waste form of fuel debris.
[1] Nagra (2002): Project Opalinus Clay Safety Report, NTB 02-05

In Finland the spent nuclear fuel from nuclear power plants is planned to be deposited to a repository at a depth of more than 400 meters in the bedrock of Olkiluoto. The repository system of multiple release barriers consists of the sparingly soluble nuclear fuel, a copper canister with a cast iron insert, a bentonite buffer around the canister, backfilling of the tunnels and the surrounding bedrock. Safe disposal of spent nuclear fuel requires information on the radionuclide transport (i.e., advection and matrix diffusion) and retention properties (i.e., diffusion and distribution coefficients) within the porous and water saturated rock matrix along the water conducting flow paths.
To this end, various laboratory and in-situ experiments have been performed as a part of a project rock matrix REtention PROperties (REPRO). The research site is located in ONKALO, an underground rock characterization facility in Olkiluoto, at a depth of 420 meters close to the planned repository site. The aim of the REPRO project is to study matrix diffusion and sorption of radionuclides in the surrounding bedrock in laboratory and in-situ conditions and demonstrate how the conditions affect the results. More importantly, the results are utilized to investigate if the assumptions applied in the safety case are in line with the site evidence. Furthermore, the laboratory experiments are used to produce parameters for analyzing in-situ experiments and to test analysis tools used for in-situ experiments.
In currently running Water Phase Diffusion Experiment in laboratory (WPDElab) a short concentrated pulse of H-3 and Cl-36 was injected to a synthetic ground water flow through an artificial fracture in contact with a veined gneiss sample from the REPRO site. The 2 mm thick fracture is formed between a 0.8 m long plastic tube (inner diameter 46 mm) and a drill core sample (outer diameter 42 mm) placed in the center of the tube. It is assumed here that due to diffusion radionuclides migrate at a lower speed than that of the advective flow of water. In this experiment synthetic ground water was pumped through the fracture with a constant flow rate of 20 µl/min and a mixture of H-3 and Cl-36 tracers was injected into the advective flow. Breakthrough curves of the radionuclides were measured from collected water samples with a liquid scintillation counter (Tri-Carb 2910 TR). The activities were determined using double labelling and manual quench correction. Preliminary results of the experiment show that the heterogeneity of the advection field affects the early part of the breakthrough curve. Late part of the curve shows that the transport of Cl-36 is retarded less than that of H-3 due to anion exclusion. Currently the experiment is still running and more detailed results with numerical analyses will be provided in the conference presentation. A similar experiment has been performed in in-situ conditions also and a discussion on possible differences will be included.

IONSIV is a commercial ion exchanger developed and engineered by UOP. The material is based on crystalline silicotitinate (CST) and has been employed extensively around the globe to remove radioactive Cs from nuclear waste streams. Despite mostly being used for Cs, IONSIV is also selective for Sr. Along with ¹³⁷Cs, ⁹⁰Sr is one of the more problematic radionuclides produced from fission. It has a half life of around 29 years and is one of the primary heat generators in nuclear waste and it is often one of the main radionuclides accidently released into the biosphere. This work is concerned with the conversion of Sr-loaded IONSIV into robust ceramic waste forms suitable for final geological disposal. A number of samples have been thermally converted via traditional sintering methods in air and also via hot isostatic pressing (HIPing). Initial studies indicate both process methods produce crystalline waste forms which are currently undergoing analysis to fully describe the phase assembly using XRD, SEM and TEM. Furthermore, aqueous durability tests are currently taking place in order to access their suitability as final waste forms. As well as Sr-loaded IONSIV, mixed Cs/Sr loaded IONSIV has been studied.

V. Khachatryan, H. Mezhlumyan, H.Hovsepyan
The purpose of development is to enhance the radiation safety of radioactive waste by treatment and disposal using a new type of matrix spin ceramics. Disposal of radioactive wastes is carried out by immobilizing it’s by the intermolecular micro encapsulating with magnetically structured nano composites, developed on the basis of natural iron oxide magnetite minerals .by the introduction of radionuclide’s into stable new generation of matrix materials, dipole crystalline-amorphous materials – spin ceramics. As a result of the above processing is formed structured spin ceramic having radiation-resistant and radiation-absorbing properties. Go up to protect and neutralizing properties of the immobilizing matrix and hence the degree of safety of the treated radioactive waste.

The optimization of the conventional borosilicate glasses to improve the solubility of key nuclear waste products with limited solubility such as MoO₃ is of paramount importance for the overall nuclear waste management system. In this contribution, this work is dedicated to the formulation of improved borosilicate glass-ceramic system of SiO₂-Nd₂O₃-CaO-Na₂O-B₂O₃-Al₂O₃-ZrO₂-MoO₃ with the aim to incorporate higher quantity of Molybdenum ions in the glassy matrix. The concentration of MoO₃ in the system was increased from 0 to 6.7 mol%. High amounts of [MoO₃] ≥ 5 mol% (9 mass%) in the system were resulted in liquid-liquid phase separation and that was followed by the crystallization of the structure. Differential thermal analysis was implemented to study the thermal behavior of the compositions. The modifications of the glass matrix due to increase in MoO₃ at microstructure level and their morphology were investigated by implementing X-ray diffraction and Scanning Electron Microscopy.

The migration of radionuclides in the geosphere is to a large extent controlled by sorption processes onto minerals and colloids. On a molecular level, sorption phenomena involve surface complexation, ion exchange as well as solid solution formation. The formation of solid solutions leads to the structural incorporation of radionuclides in a host structure. Such solid solutions are ubiquitous in natural systems – most minerals in nature are atomistic mixtures of elements rather than pure compounds because their formation leads to a thermodynamically more stable situation compared to the formation of pure compounds.
In some scenarios describing the evolution of a geological waste repository system for spent nuclear fuel in crystalline rocks ²²⁶Ra dominates the radiological impact to the environment associated with the potential release of radionuclides from the repository in the future. The solubility of Ra in equilibrium with a Ba_xRa_1-xSO₄ solid solution is much lower than the one calculated with RaSO₄ as solubility limiting phase. Due to the expected conditions in the repository near field, a likely scenario will be a release of Ra from the spent fuel matrix into a solution in equilibrium with pre-existing barite. Batch-type laboratory experiments mimicking this scenario were carried out and indicate the uptake of Ra, leading to a reduction of more than 99% of the initial Ra concentration. The grain size and morphology of the barite grains are very similar before and after the Ra uptake, although ToF-SIMS analyses indicate that Ra was taken up into the particle volumes. In order to follow the uptake of Ra into barite and identify the possible pathways into the particle volume, we applied a new approach for the detailed characterization of Ra-barites obtained at different stages of the recrystallization experiments utilizing a combination of electron microscopy and atom probe tomography (APT). The preparation of the barite samples was done by adapting focused ion beam milling procedures to the material.
A layered structure caused by size and density variations of nano-scaled pores was observed by electron microscopy in Ra-free reference samples. The APT reconstruction showed chemical inhomogeneities of H₂O, Na and Cl present in layers of similar length scale. In conclusion, both findings indicated that the layered structure consisted of nano-scaled pores filled with NaCl-bearing fluids, providing a fast pathway for Ra into the barite particles. Subsequently, elemental maps of Ra-barites were obtained with energy-dispersive X-ray spectroscopy (EDX) enabling the analysis of the evolution of the Ra distribution within the solid with time. The maps showed an intermediate heterogeneous Ra distribution which becomes homogeneous at equilibrium state.

To assess the performance of the hypothetical future Deep Geological Repository (DGR) that would store the Spent Nuclear Fuel (SNF), several studies have been centered in the corrosion/alteration processes of the SNF matrix. SNF contains all the fission products, transuranium elements and radioactive daughters. Although some of them are segregated, the highest fraction remains into the UO₂ matrix, and their release is highly dependent on the alteration processes that may affect this matrix. In addition, UO₂ is highly sensible to redox conditions and it can be oxidized to the more soluble state U(VI)^1–3. Therefore, it is mandatory to design new in situ methodologies to identify the possible formation of oxidized uranium secondary phases. Moreover, it is necessary to screen and to detect the possible reactions in which SNF could be involved under different near-field conditions.
In this communication, we present the application of Ag-Nanoparticles (NPs) to enhance the detection of uranium phases (such as UO₂ or UO₂₍(OH)₂) by Raman Spectroscopy. The synthesis of Ag-NPs was obtained by the Seed Mediated Growth Approach method, as a simple and reproducible methodology for NPs preparation^4,5. These NPs were properly characterized by Electron Microscopy and UV-Visible spectrophotometry.
The experimental results showed that the use of Ag-NPs in Raman analysis resulted in the enhancement of the Raman signal of a fixed amount of UO₂ due to the Surface Enhanced Raman Spectroscopy (SERS) effect^6,7.
Raman-SERS Spectroscopy was used to follow up the reaction between UO₂ and H₂O₂. As a result, it was observed the appearance of two peaks at 818 and 875 cm^-1 of Raman shift, which were identified as corresponding to studtite. At the same time, the UO₂ signal decreased through time, proving the feasibility of Raman spectroscopy to in-situ screen the oxidation of UO₂ under the experimental conditions.



References:

1 A. Martínez-Torrents, S. Meca, N. Baumann, V. Martí, J. Giménez, J. De Pablo and I. Casas, Polyhedron, 2013, 55, 92–101.
2 J. Giménez, F. Clarens, I. Casas, M. Rovira, J. De Pablo and J. Bruno, J. Nucl. Mater., 2005, 345, 232–238.
3 A. Rey, S. Utsunomiya, J. Gimenez, I. Casas, J. de Pablo and R. C. Ewing, Am. Mineral., 2009, 94, 229–235.
4 N. R. Jana, L. Gearheart and C. J. Murphy, Adv. Mater., 2001, 13, 1389–1393.
5 Z.-W. Lin, Y.-C. Tsao, M.-Y. Yang and M. H. Huang, Chem. - A Eur. J., 2016, n/a–n/a.
6 W. J. Plieth, J. Phys. Chem., 1982, 86, 3166–3170.
7 A. F. Chrimes, K. Khoshmanesh, P. R. Stoddart, A. a Kayani, A. Mitchell, H. Daima, V. Bansal and K. Kalantar-zadeh, Anal. Chem., 2012, 84, 4029–35.

As a fluorite related superstructure, pyrochlore have attracted widespread attention over the years owing to their distinctive chemical and structural flexibility, opening doors into applications such as superconductivity, magnetism, ion exchange, solid oxide fuel cells, transmutation targets and host matrices for nuclear waste immobilisation [1,2]. The latter is of particular interest as pyrochlore are able to accommodate radionuclide species whilst satisfying charge balance requirements. Plus, the cationic homogeneity and its influence on the chemical durability of the nuclear wastes are key parameters in assessing the performance of the wasteform in long term disposal. In this study three methods of production have been chosen; solid state synthesis (SSS), molten salt synthesis (MSS) and oxalate precipitation synthesis (OPS). We are investigating the impact of the production method on the cationic homogeneity of Y₂Sn₂O₇, this parameter can be observed at different scales on this particular pyrochlore through EDS and NMR [3,4]. At the same time, the chemical durability of the samples is evaluated by dissolution experiments in different conditions.

[1] G. R. Lumpkin, Ceramic waste forms for actinides Elements, 2006, 2, 365.
[2] Y. Zhang; M. Stewart; H. Li; M. Carter; E. Vance; S. Moricca, Zirconolite-rich titanate ceramics for immobilisation of actinides–Waste form/HIP can interactions and chemical durability Journal of Nuclear Materials, 2009, 395, 69.
[3] M. de los Reyes; K. R. Whittle; Z. Zhang; S. E. Ashbrook; M. R. Mitchell; L.-Y. Jang; G. R. Lumpkin, The pyrochlore to defect fluorite phase transition in Y₂Sn_{2− x}Zr_xO₇ RSC Advances, 2013, 3, 5090.
[4] S. W. Reader; M. R. Mitchell; K. E. Johnston; C. J. Pickard; K. R. Whittle; S. E. Ashbrook, Cation disorder in pyrochlore ceramics: 89Y MAS NMR and first-principles calculations The Journal of Physical Chemistry C, 2009, 113, 18874.

Concrete and clay play an important role in the deep geological disposal of nuclear waste. In the Netherlands, the Boom clay formations have been identified as a potential host rock for low and intermediate level waste and high level waste [1]. Concrete has several functions in the underground shafts in the Dutch disposal concept, such as encapsulation of waste canisters, backfilling and lining of the galleries [2].

In the conditions of a repository, concrete and clay will be in direct contact. Concrete is exposed to infiltrating water which is not in equilibrium with the cement mineralogy. This causes different chemical detrimental reactions such as carbonation, decalcification/ leaching amongst others. During concrete degradation, highly alkaline waters from the concrete will diffuse through the clay barrier. Pore water of concrete has high pH (12.5 to 13.4) compared to the pH of pore water in a clay (pH ~8.5). This may produce structural transformations in the clayey materials, lead to the dissolution/precipitation of minerals in the clay/concrete interface and, consequently, result in changes of the clay characteristics. Therefore, investigations of the processes at the interface are of high importance for the geological storage of radioactive waste.

The present study is focused on the interface between high pH cementitious materials (CEM-III/B 42.5 LH HS (concrete); CEM-I (foamed concrete); CEM-III/B 42.5 LH HS (foamed concrete)) and the Boom clay, excavated from the depth of 68-88 m in the Netherlands. Recent studies [3] have shown that the composition of the pore water in this clay differs significantly from the one measured for Belgian Boom clay with respect to saline content. We are aiming at quantification the relevant alteration processes and their impact on physical properties, especially on the diffusive and advective transport characteristics for pore water and the dissolved species. The characterization of porosity and permeability is carried out using a combination of techniques: X-ray CT-scanning and neutron imaging, spin-echo small-angle neutron scattering (SESANS) and positron annihilation. The impact of the related microstructural characteristics on the transport parameters is being investigated. The goal of our experimental work is providing laboratory observations of the temporal and spatial extent of these processes, and notably of porosity changes.

References

1. E. Verhoef, E. Neeft, J. Grupa, A. Poley, Outline of a disposal concept in clay. Report OPERA-PG-COV008, 2011. 17 p.
2. E.V. Verhoef, A.M.G. de Bruin, R.B. Wiegers et al., Cementitious materials in OPERA disposal concept in Boom Clay. Report OPERA-PG-COV020, 2014. 17 p.
3. T. Behrends, I. van der Veen, A. Hoving, J. Griffoen. Geochemical characterization of Rupel (Boom) Clay material: pore water composition, reactive minerals and cation exchange capacity. Report OPERA-PU-UTR521, 2015. 43 p.

The radiation damage studies of phosphates Ca_0.25Sr_0.25Zr₂(PO₄)₃ and Sm_0.33Zr₂(PO₄)₃, belonging to the NaZr₂(PO₄)₃-family were carried out by using Xe²⁶⁺ ion implantation and by doping with 2% Cm (75.5% ²⁴⁴Cm), respectively. Both compounds are described by the general formula M'_xM''₂(PO₄)₃, where M' cations occupy larger cavities (interstitial position of the structure) and M'' are in the framework positions.

The precursor powder of the phosphate Ca_0.25Sr_0.25Zr₂(PO₄)₃ was synthesized by a sol–gel process using citric acid and ethylene glycol (the Pechini method). From this powder ceramic samples were prepared by spark plasma sintering (SPS), their relative densities were found to be 99.5 ± 0.3% after the isothermal treatment at 860°C for 3 min.

Sintered disc-shaped ceramic samples (d = 10 mm, h = 4 mm) were bombarded at 300 K by 167 MeV Xe²⁶⁺ ions with fluences ranging from 6●10¹⁰ to 1●10¹³ ions/cm². It was found that the exposure to the highest fluence of 10¹³ ion/cm² led to complete amorphization of the irradiated layer. The observed phase transition is ascribed to the formation of amorphous latent tracks via dense electronic excitations. Post-radiation heat treatment revealed that the transformation from the metamict to crystalline form took place after successive annealing at T = 200, 300, 400, 500, 600 and 800°C and t = 3, 13, 11, 5, 17 and 15 h, respectively.

Cm-doped phosphate Sm_0.33Zr₂(PO₄)₃ was synthesized by precipitation from the solutions of ZrOCl₂, NH₄H₂PO₄ and samarium and curium nitrates. The study of radiation stability of this compound implied observation of changes of the unit cell parameters depending on the absorbed dose, and characterization of its chemical durability. After the disappearance of all XRD reflections the amorphization doze was estimated: 2.0●10¹⁸ α-decay/g after the 13 months from the moment of synthesis. The leaching rate of Cm at the 14^th day of contact of the pelletized sample with water at 90°C (MCC-1 test) was estimated to be 6.0●10^-7 g●cm^-2●d^-1, and that of Sm - 3.6●10^-6 g●cm^-2●d^-1. This allowed us to characterize the obtained phosphate as chemically resistant.

Decommissioning and clean up of nuclear facilities requires the development of new technologies to condition radioactive wastes, producing passively safe waste packages of minimal volume, to reduce storage and disposal costs. We have applied hot isostatic pressing to demonstrate conceptual wasteforms for ion exchange materials and sludges present on the Sellafield site, UK, and Fukushima site, Japan. These radioactive wastes pose several challenges which demand early conditioning to produce a passively safe wasteform, including: i) the materials are of a wet and granular nature, and hence dispersible as a result of loss of containment; ii) the materials are characterised by very high dose rates as a result of their selectivity for short lived radionuclides (e.g. Sr-90, Cs-137, Co-60); iii) the materials exhibit radiogenic self heating, as a result of the concentration of such lived radio nuclides; iv) the materials exhibit hydrogen production as a result of radiolysis of entrained water.
In this presentation we demonstrate the conversion of several commercial and natural inorganic ion exchange materials into multiphase ceramic wasteforms, achieving a waste loading of 100 wt% and density in excess of 97% theoretical. High resolution thermogravimetric analysis coupled with mass spectroscopy was utilised to characterise the evolution of water and volatiles during the in-can bake out step, prior to the HIP cycle. This allowed optimisation of the bake out parameters (temperature, time and vacuum) to enable complete removal of water and volatiles, affording ceramic bodies with minimal residual porosity by hot isostatic pressing at 1250^oC for 4h in stainless steel cans. Characterisation of the ceramic wasteforms by SEM /EDX and X-ray diffraction revealed the nuclides of concern to be incorporated within well known natural mineral and synthetic phases, with Sr partitioning into the SrTiO₃ pervoskite phase and Co, Fe, Mn and Cr partitioning into a spinel phase. Dynamic alteration experiments revealed matrix dissolution rates of less than 10^-4 g m^-2 d^-1 under forward rate conditions at 90 ^oC.
Overall, hot isostatic pressing of inorganic ion exchange materials yields durable glass, glass-ceramic, and ceramic wasteforms, with minimal voidage and porosity, in which the radionuclide partitioning between glass and ceramic phases can be controlled by wasteform composition and processing parameters. The resulting ceramic wasteforms are considered to meet the disposability requirements of a UK Geological Disposal Facility for radioactive wastes.

Long half-life biologically active fission products, such as technetium-99, present particular problems for the disposal of spent nuclear fuel (SNF). Technetium is present in relatively high concentrations in fuel (approx. 1kg tonne^-1 SNF) and has very high mobility in oxidizing environments. Technetium is therefore generally removed from SNF either by solvent extraction and reduction, during the PUREX process, or by sorption via ion exchange processes. Historically technetium has been disposed of via dilution and dispersion in the sea but stringent regulations now mean that the preferred long term option is immobilisation in a highly stable and durable matrix. In this contribution we have looked at the synthesis of various crystalline host phases using Mo as an inactive surrogate for Tc. The solubility of Mo has been investigated using a range of characterisation techniques including X-ray diffraction (XRD), scanning electron microscopy (SEM), and X-ray absorption spectroscopy (XAS). Synthesis has been performed in the solid state under different processing conditions, to aid down-selection of suitable candidate phases, and hot isostatic pressing (HIP) has been utilized in an effort to minimize volatisation associated with high temperature processing of species such as technetium.

A new theoretical approach for determining the γ-ray full-energy peak efficiency at positions close to two HPGe detectors and at the well bottom of a well-type detector was developed for measuring environmental volume samples containing ¹³⁷Cs, ¹³⁴Cs and ⁴⁰K. The full-energy peak efficiency was estimated by considering coincidence-summing, attenuation factors and self-absorption corrections. The derived coincidence-summing correction factors were compared with those of experimental and Monte Carlo simulation methods and good agreements were obtained. The correction factor was derived as a function of the densities of several matrix materials. The present theoretical method was applied to the measurement of environmental samples and also low-level radioactivity measurements of water samples using the well-type detector.

Carbon-14 is a key radionuclide in the safety assessment of the low-level waste (LLW) disposal system in Japan. However, the chemical form of carbon-14 released from irradiated stainless steels (i.e., LLW) have not been sufficiently classified. The purpose of this study is to classify the chemical form of stable carbon released from unirradiated stainless steel under highly alkaline and low-oxygen conditions to develop an understanding of the release of carbon-14 from irradiated stainless steels. In this experiment, types 304 and 316 stainless-steel powders were immersed in 0.005M NaOH solution for about 14 days. The preparation was carried out in a glove box with an Ar atmosphere as much as possible. Gas and liquid samples were analyzed to identify the chemical form of carbon released from the stainless steel. The liquid samples were divided into unfiltered and filtered samples using an ultrafiltration membrane (10,000 MWCO) prior to the analysis. We found no significant difference between the analysis results for the SUS304 and SUS316 systems. In the gaseous phase, hydrocarbons such as methane and ethane were not detected. In the liquid phase, carboxylic acids (formic and acetic acids) were identified and quantified. However, the sum of the carbon concentrations of the carboxylic acids was significantly lower than the total organic carbon (TOC) concentration in the unfiltered samples. In the filtered samples, the TOC concentration was closer to the sum of the carbon concentrations than that for the unfiltered samples. Dynamic light scattering measurement showed that there were particles with a size of 10 - 100 nm order in the unfiltered samples. These results indicate that the chemical form of most of the carbon released from the stainless steels is likely to be colloidal. In addition, the concentrations of the metallic elements (Fe, Cr and Ni), which are main constituents of the stainless steels, tended to decrease upon ultrafiltration. This suggests that the sorption of carbon on metallic compounds (e.g., colloidal iron hydroxide) may have occurred.

Zeolites, especially Zeolite A, Zeolite P and Chabazite, are interesting for a wide range of water treatment applications including the removal of radioactive Sr and Cs in the nuclear industry. The magnetic modification of zeolites offers an advantage of efficient, easy and fast ion separation by applying a magnetic field after ion exchange.^[1] In the last few years, non-aqueous sol-gel routes have proven to be versatile for the synthesis of several pure inorganic metal oxide nanoparticles which were previously impossible to synthesize. Recently we proved that these routes are not only suitable for the formation of hybrid materials but also highly crystalline metal and metal oxide nanoparticles showing interesting structural, optical, electric and magnetic properties. A novel, facile method based on a non-aqueous sol–gel solvothermal process has been developed to synthesize spherical magnetic Fe₃O₄ nanoparticles in one pot at low temperature. Detailed XRD and HR-TEM studies prove that the pure phases of Fe₃O₄ and other oxides were formed with particle sizes in the range of between 3-15 nm.^[2] We are now making composites of zeolite particles with these nanoparticles to introduce magnetism and are testing these for the selective removal of Cs⁺ and Sr²⁺. The unique properties and reactivity of nanocrystalline zeolites and the potential for future applications of these nanocomposites will also be discussed.


References

[1] D. A. Fungaro, M. Yamaura, G. R. Craesmeyer, International Review of Chemical Engineering (I.RE.CH.E.) 2012, 4.
[2] a) M. Karmaoui, N. J. O. Silva, V. S. Amaral, A. Ibarra, A. Millan, F. Palacio, Nanoscale 2013, 5, 4277; b) M. Niederberger, G. Garnweitner, Chemistry - A European Journal 2006, 12, 7282.

Mixed actinide dioxides are currently used as fuels in Pressurized Water Reactors (PWR) (including Gen III, EPR) and also stand as potential candidates for several Gen IV concepts including Sodium-cooled Fast Reactor (SFR) or Gas-cooled Fast Reactor (GFR)¹.
Thorium-based dioxides are usually associated to very good performances in the perspective of the direct disposal for long-term storage (very high resistance to aqueous alteration or corrosion). This property is mainly associated to very low normalized dissolution rates as well as to the rapid establishment of saturation processes². While the multiparametric study of the Th_1-xU_xO₂ dissolution was already examined by varying conventional parameters such as temperature, chemical composition or leachate acidity^2,3 , that of Th_1-xCe_xO₂⁴ and Th_1-xPu_xO₂ remained poorly examined.
In this study, we describe dissolution experiments on Th_0.87Pu_0.13O₂ solid solutions which started in 2002 and were continued until 2011 in order to focus this study on the long term behavior of such samples after 9 years in several nitric acid concentrations.
The normalized weight losses determined after 9 years of leaching time are in good agreement with the linear regression proposed by Hubert et al. in 2008. Indeed the normalized dissolution rates estimated at 9 years, R_L,t^*(Pu), are almost similar to the leaching rates determined during the first 500 days of leaching.
Despite the very long duration of the dissolution tests, the amount of plutonium released remains very low even in very corrosive media : 1M HNO₃ (0.9% of dissolved solid) and 5M HNO₃ (2.1% of dissolved solid) that confirms the high chemical durability of such Th_0.87Pu_0.13O₂ solid solution. However, such slow normalized dissolution rates are large enough to cause microstructural modifications inducing a potential increase of the reactive surface area. In order to underline such modifications, leached powders were characterized by scanning electron microscopy.
1. M.S., Y.T. Ichimiya, , Am. Nucl. Soc., 2004. 90, 46-47.
2. S. Hubert, K. Barthelet, B. Fourest, G. Lagarde, N. Dacheux, N. Baglan, J. Nucl. Mater., 2001. 297, 206-213.
3. L. Claparede, F. Tocino, S. Szenknect, A. Mesbah, N. Clavier, P. Moisy and N. Dacheux, J. Nucl. Mater. 2015, 457, 304-316.
4. L. Claparede, N. Clavier, N. Dacheux, A. Mesbah, J. Martinez, S. Szenknect, P. Moisy, Inorg. Chem. 2011, 50, 11702–11714.
5. S. Hubert, G. Heisbourg, N. Dacheux and P. Moisy, Inorg. Chem. 2008, 47, 2064-2073.

Pellets of uranium-based mixed oxides, as potential fuels of the Gen(III) and Gen(IV) nuclear reactors, Th/U, U/Ce, U/Ln, and Th/Ln mixed dioxide were prepared from oxalate precursors, sintered then finally submitted to dissolution tests.
The macroscopic study of U1-xThxO2 solid solutions dissolution in nitric acid showed that uranium mole loading clearly impact their chemical durability. It also underlined significant modification of the preponderant dissolution mechanism when increasing the uranium incorporation rate in the ceramics. Indeed, for the lower uranium incorporation rates (i.e. xU < 0.5), the dissolution is mainly controlled by surface-controlling reactions at the solid/solution interface involving the adsorption of protons on reactive sites. On the contrary, for uranium enriched materials (i.e. xU > 0.5), oxidation of uranium (IV) into uranyl at the solid/solution interface becomes clearly preponderant (leading to the decrease of the chemical durability).
These strong modifications of the preponderant reaction controlling dissolution were confirmed at the microscopic level with the help of operando ESEM experiments. The evolution of solid/solution interface (reactive surface area, composition) during dissolution was followed as a function of the dissolution time. Preferential dissolution zones (triple junctions, grain boundaries, inter- and intra-granular porosities) were clearly observed for the lower uranium incorporation rates (xU < 0.5). They induced the rapid and significant increase of the reactive surface area even for short progress of the dissolution reaction. On the contrary, the dissolution appeared to be more generalized for uranium-enriched samples (xU > 0.5) due to the existence of the preponderant reaction of oxidation of uranium (IV) into uranium (VI) at the interface. For the latter, the increase of the reactive surface area with the progress of the reaction of dissolution was found to be more moderate.

Actinide oxalate crystals play an important role in nuclear fuel production and nuclear waste management. Metal oxalates are an attractive precursor material in metal oxide synthesis, expanding the range of accessible metal oxide morphologies for nanostructured nuclear fuel development. Equally important in nuclear waste management are the ability to prevent undesired nucleation and crystal growth and the need to selectively precipitate solids during waste processing. The challenge of controlling metal oxalate formation still requires a better understanding of the fundamentals of nucleation and crystal growth.

Characterization of crystal nucleation and growth presents many technical challenges. Nuclei are extremely small and growth kinetics are often rapid. Techniques well-suited for analysis of small particles, such as high resolution transmission electron microscopy (TEM) are often performed in high vacuum and are inherently incompatible with in situ analysis of suspensions or solution-based samples. However, the presence of drying artifacts in traditional TEM samples can lead to erroneous conclusions about the mechanism of crystal growth. Both cryogenic and fluid cell TEM and scanning electron microscopy (SEM) provide in situ routes for the detailed analysis of crystal growth kinetics. Cryogenic TEM is used to examine structures preserved in their in situ conformation by vitrification during specimen preparation. In fluid cell TEM and SEM, a fluid sample is placed in an airtight cell with electron transparent windows, isolating the specimen from vacuum while still allowing imaging to take place. We use cryogenic and fluid cell in situ TEM and SEM and in situ Fourier transform infrared spectroscopy to probe the mechanisms of lanthanide and actinide oxalate crystal growth. These techniques are complemented with computational experiments to probe the nucleation process at the molecular level.

About 100 tonnes of Pu are stored at the UK Sellafield site alone, the product of approximately 50 year’s civil nuclear fuel reprocessing. Interim storage of actinide waste is generally as calcined actinide oxide powders contained within a series of nested stainless steel cans with the outer can welded to maintain storage integrity. Under certain circumstances, high-level actinide wasteforms have been observed to cause gas generation within the cans and consequent pressurisation of the storage package. This comprises one of the most serious fault scenarios that must be considered in the safety cases for long term actinide storage and avoided in practice. Water adsorption on PuO₂ has previously been investigated by measuring headspace pressure as a function of temperature within a closed system containing in the presence of varying amounts of added water. There currently exists a gap in the knowledge regarding the exact mass of water which adsorbs on AnO₂ powders in the closed, heated conditions within a storage container.

We have coated thin films of ceria (CeO₂), thoria (ThO₂) and urania (UO₂, U₃O₈) onto piezo-active crystals and used QCM methodology to directly measure any mass changes under a range of temperatures and humidities. The mass changes, combined with accurate measurements of surface area, can then be used to calculate the amount of water adsorbed onto the ceria surface and the thermodynamic requirements for its desorption.

The films were deposited onto quartz or gallium phosphate wafers through spin-coating with a precursor actinide nitrate solution or oxalate dispersion, followed by calcination. SEM, XRF and AFM were used to determine the thickness and porosity of the layers produced. The coated crystals were mounted within the pressure vessel using a crystal holder which exposed one face of the crystal to the pressurized environment of controlled water vapour composition. The resonant frequency dependence of the crystals was first measured in the absence of moisture, to allow for compensation of purely temperature induced changes. Thereafter, a number of studies were undertaken that allowed the adsorption of water to be measured under a variety of real-world applicable conditions, through variation of temperature, relative humidity, and pressure.

As expected, it was found that increasing relative humidity of the environment by addition of water resulted in mass gain due to increased amounts of water absorbed onto the ceria layer. Analysis of the resulting absorption isotherms allowed the surface area of the AnO₂ films and the enthalpy of absorption of water to be calculated. Varying the temperature of the system from ambient to approximately 400^oC was found to decrease the mass of water absorbed onto the actinide oxide layers, due to desorption and evaporation from the surface. In a closed system, this effects appears to be primarily driven by the increased moisture carrying capacity of the heated atmosphere.

Pyrochlore ceramics (A2B2O7) are considered to be promising host phases for the immobilization of actinides and in particular Pu, due to their chemical flexibility and durability and ability of structural actinide uptake. Here, we present an overview of our recent studies on these materials.
Chemical flexibility: A combined experimental and theoretical approach was used to follow the pyrochlore – defect fluorite phase transition in the Nd2xZr1-xO2+x system. Disorder leading to the structural transition was introduced via partial substitution of the A site cation with excess Zr, resulting in (Nd,Zr)2Zr2O7+x. The transition occurs at around ~19 mol% Nd2O3 at 1600 °C with an associated transition enthalpy of ~30 kJ/mol (HT oxide melt solution calorimetry). Ab initio calculations were performed for quasi-random structures, assuming a random distribution of the excess Zr and oxygen on the A-site and oxygen vacancies, respectively and complete disordering for defect fluorite. The structural model was evaluated by predicting the formation enthalpies which are in excellent agreement with our experimental results.
Chemical durability: Dissolution experiments were carried out using a dynamic setup to ensure far from thermodynamic equilibrium conditions. The influence of temperature and pH on the dissolution rate was studied for the defect fluorite and the pyrochlore ceramics. An initial incongruent elemental release for Nd and Zr was observed. At steady state conditions and pH = 1, the dissolution rates of pyrochlore and defect fluorite are congruent and in the range of 1E-5 to 1E-6 g / m2 d. These macroscopic studies were complemented by electron microscopy and vertical scanning laser microscopy studies.
Actinide uptake: A wet chemical coprecipitation synthesis route of Nd2Zr2O7 was adapted to produce pyrochlores containing 5 and 10 mol% Pu. XANES spectra were recorded to obtain the oxidation state of the Pu. EXAFS spectra were obtained at the Pu L3-edge to unravel the adoption of the A or B site by Pu in the pyrochlore structure. Complementary spectra of the Zr K-edge were recorded for a complete characterisation of the synthesised ceramics. A further insight into the structural environment of Cm (and Eu) was obtained by time resolved laser fluorescence spectroscopy (TRLFS). Fluorescence spectra were obtained by UV and direct excitation of Cm and Eu doped La2Zr2O7 with the pyrochlore and the defect fluorite crystal structure, respectively. The observed emission spectra indicate the presence of Eu and Cm at the A position within the pyrochlore crystal structure.
In summary, a refined mechanistic understanding of key properties of pyrochlore-related zirconates with respect to their application as a nuclear waste form could be obtained during the last 5 years complementing earlier studies.

Nuclear waste-forms will be exposed to radiation and thermal effects during interim storage and permanent disposal. Alpha-decay processes occur for hundreds of thousands of years, depending on actinide or plutonium waste loadings, producing varying levels of alpha-recoil damage and alpha-particle accumulation. Swelling associated with He and defect accumulation, as well as possibly He bubble formation, in the waste-form may lead to cracking, resulting in increased leaching of radioactive material. This work examines the concomitant damage and He accumulation processes in pyrochlores Gd₂Zr₂O₇ and Gd₂Ti₂O₇, which have long been considered as candidates for crystalline radionuclide immobilization. Gd₂Zr₂O₇ undergoes a pyrochlore to defect-fluorite phase transformation at ~0.4 dpa and Gd₂Ti₂O₇ undergoes a pyrochlore to amorphous phase transformation at ~0.2 dpa, both within several hundred years of storage. Gd₂Zr₂O₇ and Gd₂Ti₂O₇ were implanted to several different He concentrations corresponding to various timescales for long-term disposal. Samples were pre-damaged by Au irradiation, prior to He implantation, to induce either the pyrochlore to defect-fluorite phase transformation or amorphization. Both pristine and amorphous Gd₂Ti₂O₇ samples were implanted with He in order to study the effect of initial He build-up in the crystalline vs. amorphous structure. The critical dose for He bubble formation, as well as bubble morphology, were investigated using transmission electron microscopy (TEM). Unit cell swelling and residual stresses were measured in crystalline Gd₂Zr₂O₇ as a function of He irradiation fluence using grazing-incidence x-ray diffraction (GIXRD) and thin film residual stress measurements.

Understanding of long-term behaviour of actinide-containing host-matrices under self-irradiation is crucial in ensuring safety of nuclear waste disposal. Radiation-induced damage of crystalline structure is usually, but not always accompanied by reduced chemical durability, swelling and crack formation in poly- and single-phase crystalline ceramics as well as in devitrified glasses containing large amounts of crystalline inclusions. Although radiation damage effects have been extensively examined in analogue minerals that naturally contain radionuclides and accumulate high doses of self-irradiation the mechanisms of damage that causes alterations such as the transition from the crystalline to disordered state are not well understood.
Under irradiation at room temperature (depending on cumulative dose and energy) the original crystalline structure may: (1) Be retained; (2) Be converted into another type of crystalline structure; or (3) become amorphous or metamict. The long-term stability and chemical durability of natural solid U-bearing Zr-silicate gel is assumed to be caused by two competing processes: (1) Crystallisation of the gel into U-doped zircon assisted by self-irradiation and (2) Metamictisation of the crystallised zircon back to a gel-like state. However recent extended experiments with actinide doped crystalline wasteforms have shown significant mechanical damage with crystal cracking within a decade of storage (see B.E. Burakov, M.I Ojovan, W.E. Lee. Crystalline Materials for Actinide Immobilisation, Imperial College Press, London (2010)). E.g. polycrystalline Pu-monazite underwent metamictisation, cracked into separate pieces. Zircon/zirconia ceramic did not crack under self-irradiation and preserved its initial light grey colour despite becoming completely amorphous. However, 238Pu-doped zircon single crystal examined over a 7-year period developed matrix cracking. Formation of tiny particles near the crystals was also observed, possibly arising from fracture of the crystal surface under self-irradiation. It has been suggested that this process could be due to inhomogeneous Pu distribution in the wasteforms. Indeed it has been earlier observed that inhomogeneous distribution of actinides in non-conducting host materials may lead to mechanical destruction due to accumulation of electric charge (M.I. Ojovan, P.P. Poluektov. Mat. Res. Symp. Proc., 648, P.3.1.1-6. (2001)).
This report aims to analyse the mechanism of mechanical damage of non-metallic nuclear wasteforms caused by electrical field induced in the insulating materials by decaying on non-homogeneously distributed radionuclides. The most important parameters that control the matrix integrity are the radioactive grain (inhomogeneity) size, specific radioactivity and effective electrical conductivity of matrix. Assessments show that this mechanism can really lead to some nuclear wasteform destruction in conditions of non-uniform radionuclide distribution.

In a scenario where ground water enters a canister for used nuclear fuel, the presence of trace elements in the water could possibly influence the fuel dissolution due to their effects on radiolysis yields. One species of particular interest in this context is bromide, which has a well-known ability to scavenge hydroxyl radicals much faster than molecular hydrogen, thereby preventing the beneficial effect of hydrogen in homogeneous γ-radiolysis. However, after a few hundred years, α-radiation dominates in the spent fuel. In the present work, radiolysis experiments have been performed to determine the effect of bromide ions on the yield of hydrogen peroxide by mass spectrometric measurement of its decomposition product oxygen. The use of high activity ²³⁸Pu solutions has made it possible to study this effect during pure α-radiolysis from a homogeneously distributed radiation field. To simulate repository conditions, and to minimize the influence of in-leaking oxygen from air, the studies were performed using graphite sealed stainless steel autoclaves with an initial atmosphere of 10 bar hydrogen overpressure. The results show that addition of 1 mM bromide to the solution gives no significant effect on the oxygen yield for α-radiation doses up to 2 MGy. The lack of effect is most likely explained by the limited radical escape yields from radiation tracks in pure α-radiolysis.

The ability of a material to withstand damage due to radioactive decay is a key property for nuclear wasteforms. Crystalline and amorphous inorganic materials are generally accepted as the best immobilisation options for separated actinides although there is limited qualitative understanding of the effect of radiation damage on the structure and the properties of these materials. The work presented here utilises complimentary surface sensitive spectroscopic techniques to probe the changes in local structure of key elements in a range of wasteforms which have undergone radiation damage. In these materials radiation damage has been simulated by heavy ion implantation which accurately mimics the damaging effect of alpha-decay recoil events in actinides such as plutonium.

In natural groundwaters such as brines associated with salt formations, there may be relatively high concentrations of borate. In addition, when borosilicate glass for high level nuclear waste (HLW) is corroded, borate is also released into the groundwater. In the field of nuclear waste management, actinides in +III state represented by Am(III) are present in nuclear waste in geological repositories. Therefore, the interactions between Am(III) and borate are important for performance assessment (PA) for geological repositories for nuclear wastes. As an example, in the Waste Isolation Pilot Plant (WIPP), a U.S. DOE geological repository for defense-related transuranic (TRU) waste in the bedded salt formations in New Mexico, USA, the inventory of Am(III) in waste was 143 kg for the WIPP Compliance Application Re-Certification in 2009 (CRA-2009). The borate concentrations in the two WIPP brines that are important for performance assessment (PA), i.e., Generic Weep Brine (GWB), and Energy Research and Development Administration (WIPP Well) 6 (ERDA-6), are 0.178 mol●kg^–1 (or 0.0445 mol●kg^–1 if it is expressed as B₄O₇^2-) and 0.0704 mol●kg^–1 (or 0.0176 mol●kg^–1 if it is expressed as B₄O₇^2-), respectively.
In this work, we present our evaluation of the equilibrium constant for AmHB₄O₇²⁺ and its Pitzer interaction parameters at 25^oC.

4B(OH)₄^– + 3H⁺ + Am³⁺ = AmHB₄O₇²⁺ + 9H₂O(l) (1)

Using Nd(III) as an analog to Am(III), solubility data of Nd(OH)₃(s) in NaCl solutions in the presence of borate ion, from the literature, is used to determine Am(III) interactions with borate suggested by Eq. (1). This evaluation is in accordance with the WIPP thermodynamic model in which the borate species include B(OH)₃(aq), B(OH)₄^–, B₃O₃(OH)₄^–, B₄O₅(OH)₃^2–, and NaB(OH)₄(aq). Also based on the above speciation scheme for borate and newly introduced AmHB₄O₇²⁺, we evaluated the equilibrium constant at 25^oC for dissolution of AmB₉O₁₃(OH)₄(cr).

AmB₉O₁₃(OH)₄(cr) + 19H₂O(l) = Am³⁺ + 6H⁺ + 9B(OH)₄^– (2)

According to the solubility data of NdB₉O₁₃(OH)₄(cr) in NaCl solutions in the presence of borate ion, from the literature, using Nd(III) as an analog to Am(III), the solubility constant for Reaction (2) is evaluated.
The equilibrium constants and Pitzer parameters evaluated by this study will be important to describe the chemical behavior of Am(III) in the presence of borate in geological repositories.

^A This research is funded by the WIPP programs administered by the Office of Environmental Management (EM) of the U.S. Department of Energy.
^B Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000. SAND2016-5266A

Siderite solubility in the Fe²⁺-Na⁺-Cl^--CO₃^2- system was explored at room temperature and high pH (>10). Experiments were carried out in a controlled anoxic glove box in which concentrations of O₂ were < 1ppmv. With increasing aging time, the pH of the solution decreased, signalling formation of a hydroxyl-bearing mineral phase. X-ray diffraction analyses identified the precipitates siderite and chukanovite. The evolution of the solution chemistry over time indicates that chukanovite was the reaction controlling phase. Accordingly, we sought to quantify the thermodynamic stability of chukanovite at high ionic strength (1 – 7.5m) and high pH (> 10). The formation free energy of chukanovite was estimated from the equilibrium constant of chukanovite dissolution reaction, log K. The aqueous speciation code EQ3/6 was used to model the experimental ferrous iron solubility data. The fitting process was repeated by adjusting log K for chukanovite and Pitzer interaction parameters until the sum of squared ferrous iron solubility differences between the aqueous model and the experimental data reached a minimum. Pure chukanovite syntheses were then carried out at pH range between 4 and 11. The synthesized chukanovite was aged in Na₂CO₃ + NaCl brines to ensure crystalline products. To further assess the equilibrium between siderite and chukanovite, we have investigated the formation of siderite phase in chukanovite-added Na₂CO₃ + NaCl brines. The data will be discussed in the context of waste disposal in carbonate-bearing brines under anoxic conditions. This research is funded by WIPP programs administered by the Office of Environmental Management (EM) of the U.S Department of Energy. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000. SAND2016-5578A

The dissolution and precipitation of gibbsite [α-Al(OH)₃] and boehmite [γ-AlOOH] is of prime importance to the final disposition of high-level nuclear waste (HLW) stored at the Hanford Site. The high aluminum content in the Hanford Tank waste stream is incompatible with current durable glass formulations and needs to be minimized prior to vitrification. The aluminum oxide that formed from the precipitation of waste streams from plutonium weapons processing is present as colloidal-sized particles in a highly alkaline and high ionic strength solution. Gibbsite and sodium aluminate, which are common in the tanks, are easily dissolved by heating under caustic conditions. However, the nano-particles (<100 nm) of boehmite in the Hanford Waste Tanks have been shown to be a much more intractable problem, as the concentration of Al(aq) in the tank waste does not fully account for the documented drastic decrease in boehmite dissolution rate and extent. In this study we focus on the extrinsic factors that could have a possible effect on the rate of dissolution including aggregation and attachment/adsorption of other phases/metal impurities to reactive bohemite surfaces. In this study we utilize cryo-transmission electron microscopy (TEM) and in-situ liquid TEM to achieve fundamental insights into the mechanisms of nano-particle aggregation and dissolution. Our initial results show that the particles aggregate along a preferred crystal orientation (the main flat (010)), irrespective of pH and solution content, forming large stacked agglomerates. When the same particles are synthesized with metal impurities (Fe, Cr, Mn) highly corrosive treatment at elevated temperatures has little to no effect on the morphology, unlike the pure boehmite particles. Scanning electron microscopy (SEM) of the dried material reveals the smooth and sharp edges of the boehmite even after several days under caustic leach, while the pure boehmite has rounded and uneven edges.

The alteration of the spent nuclear fuel (SNF) under repository conditions is one of the key issues in the performance assessment. During the SNF leaching dissolution, some of the radionuclides (RN) segregated from the matrix are known to have a faster release than the matrix itself. However, since all the RN dissolve simultaneously while performing the leaching experiments, it is difficult to separate the ones that contribute to the Instant Release (IR) from the other ones.

In order to identify the RN in the IR, a semi-empirical mathematic approach is proposed. The experimental release data obtained in the FIRST-Nuclides European Project were used to fit the dissolution parameters of this model. Two different SNF with different burn-ups were used to perform this study: one with 42 GWd/T_U and the other one with 54 GWd/T_U. From each SNF, two powder samples were used: one that came from the center part of the pellet and the other that came from the outer part of the pellet. The leaching dissolution had 1 mM HCO₃^- and 19 mM NaCl and the experiments were performed under oxidizing conditions. The concentration of the RN in the leaching was determined as a function of time.

The model assumes homogeneous RN distribution. The uranium release was fitted by using a dissolution model with two different components (both first order kinetics): a rapid component that was considered to come from previously oxidized phases and fines and a slow component that was considered to come from the UO₂ matrix. The dissolution rate constants and molar fraction obtained for the uranium release were corrected by the inventory and used to predict the rest of RN releases. Some RN, i.e. Am, were well estimated by using the matrix release parameters and therefore, their release was considered congruent with the SNF matrix. However, other elements were overestimated, i.e. Rh and Ru, because they partially belong to more insoluble phases than the matrix (ε-phases) which were not considered in the model. Finally, some RN, especially Cs, were underestimated and a third component (also first order kinetics) was needed in the model to fit the experimental results. This component was attributed to those RN more soluble than the matrix and segregated at the grain boundaries during irradiation, the so called IR Fraction.

Fine-grained sedimentary rocks (shale, mudstone) play important roles in global CO₂ abatement efforts through their importance in high-level radioactive waste (HLRW) storage, carbon capture and storage (CCS), and shale gas extraction. These different technologies rely on seemingly conflicting premises regarding the sealing properties of shale and mudstone, suggesting that fine-grained rocks that lend themselves to hydrocarbon extraction may not be optimal seals for HLRW storage or CCS, and vice versa. In this paper, a compilation of experimental data on the properties of well-characterized shale and mudstone formations is used to demonstrate that clay mineral mass fraction, X_clay, is a master variable that controls key material properties of these formations and that a remarkably sharp threshold at X_clay ~ 1/3 separates fine-grained rocks with very different properties. This threshold coincides with the predictions of a simple conceptual model of the microstructure of sedimentary rocks and is reflected in the applications of shale and mudstone formations for HLRW storage, CCS, and shale gas extraction.

Nuclear fuel reprocessing is important in maintaining a sustainable nuclear fuel cycle and involves the chemical separation and recovery of fissionable plutonium from irradiated nuclear fuel. However, the original Plutonium URanium EXtraction (PUREX) flowsheet still in use in the UK has a number of shortcomings, in particular the creation of a pure plutonium product and minor actinide heavy uranium product. As such both the UK and Europe are seeking to develop advanced spent nuclear fuel separation flowsheets that not only produce a proliferation resistant plutonium product, but also remove many of the long life transuranium (TRU) elements from the potentially re-useable uranium fuel.
One of the advanced spent nuclear fuel separation flowsheets currently under development is the European Grouped Actinide Extraction (EURO-GANEX) process. This flowsheet is broadly similar to the PUREX/Advanced-PUREX flowsheet, which is currently employed at the Thermal Oxide Reprocessing Plant (THORP) in the UK for the reprocessing of spent nuclear fuel. Nonetheless, two key differences exist between the EURO-GANEX and PUREX process:

1) DEHiBA and Odourless Kerosene (OK) are deployed in the primary separation cycle and TODGA and DMDOHEMA in the 2nd cycle TRU extraction in the EURO-GANEX process to produce an impure, proliferation resistant, plutonium stream.

2) Acetohydroxamic acid (AHA), while common to both EURO-GANEX and Advanced-PUREX processes, is supplemented by the hydrophillic, aqueous phase, An(III) selective stripping ligand SO₃-Ph-BTP, in the 2nd cycle TRU back extraction of EURO-GANEX, in order to improve minor An separation.

However, little is known about the influence of the above ligands on the corrosion behaviour of the steels that typically make up pipework, tanks and centrifugal contactors in each extraction step. In particular, the corrosion behaviour of AHA, already used in the Advanced-PUREX process and known to have high affinities for e.g. Fe³⁺ (present in all steels), has seen little study at concentrations >5 mmol dm^-3.

Thus, electrochemical corrosion measurements have been performed on the process steels, 304L and 316L stainless (SS) in the presence of EURO-GANEX equivalent AHA and SO₃-Ph-BTP concentrations. Such studies have also been performed in the presence of new An(III) selective stripping ligands SO₃-Ph-BTBP and SO₃-Ph-BTPhen. Using voltammetric and microgravimetric techniques AHA has been shown to greatly increase the rate of transpassive dissolution of process steels in 1.13 mol dm^-3 HNO₃ at E > 1.1V. SS316L is affected more than SS304L due to its decreased silica content, which protects against transpassive dissolution in acidic, oxidative media. In the presence of SO₃-Ph-BTBP, AHA driven transpassive dissolution is impeded. However, when SO₃-Ph-BTPhen or SO₃-Ph-BTP is added to an acidified AHA solution, higher currents and associated mass loss are observed in the transpassive region than those seen in the presence of AHA only.

Hydrogen embrittlement of zirconium alloys is a phenomenon of interest in the United States due to the lack of a long-term solution for disposal of spent nuclear fuel (SNF). Normal operation of nuclear fuel in a reactor results in the formation of a waterside corrosion layer and the introduction of hydrogen into the zirconium cladding. The processes used during the drying of SNF as it is transferred from wet storage to dry storage can expose the SNF cladding to temperatures and pressure-induced tensile hoop stresses high enough that radial hydrides can precipitate during subsequent cooling. These radial hydrides could provide an additional embrittlement mechanism as the cladding temperature decreases below the ductile-to-brittle transition temperature. To simulate this behavior, unirradiated Zircaloy-4 samples (hydrided by a gas charging method) and defueled irradiated cladding has been tested under high pressures at high temperatures to generate radial hydrides at several laboratories. In the ORNL experiments, we expanded on the earlier work by performing hydride reorientation tests using fueled high burnup irradiated SNF rod segments in the Irradiation Fuel Examination Laboratory. The high-burnup specimens were sectioned from PWR rods taken from a 15x15 assembly of the H.B. Robinson (HBR) Unit 2 reactor. This fuel had operated for seven cycles and reached a rod-average burnup of 67 GWd/MTU. The fuel enrichment is 2.90%. The nominal fuel pellet dimensions are 9.06 mm dia. x 9.93 mm height. The cladding is cold-worked/stress-relieved Zricaloy-4, 10.77 mm OD x 9.25 mm ID, with a nominal tin content of 1.42%. After out-of cell benchmark tests using unirradiated hydriding Zircaloy-4 specimens were conducted to determine the appropriate temperature, pressure, cooling times and number of cooling cycles needed to achieve significant hydride reorientation, the in-cell hydride reorientation tests were performed with high burnup fueled specimens under the hoop stress »140 MPa at 400°C. The specimens were heated to target temperature, held for 3 hours, cooled at 1°C/min to 170°C, and then heated at 1°C/min to target temperature 400°C again for five cycles. Post test metallographic examinations showed that a significant amount of radial hydrides were induced in the HBR fuel rods. The length of radial hydride was up to 60 mm. Limited out-of cell ring compression tests (RCT) were conducted to evaluate the sample’s ductility for unirradiated Zircolay-4 specimens. For unirradiated materials, the ductility of radial hydride treated specimens is significantly reduced as compared to the as-hydrided specimens having the same hydrogen concentration (»300 wppm in this work). The mechanical testing on irradiated fueled samples with and without hydride reorientation experiments are under way, and will be reported in the near future.

During operation of a nuclear reactor, the long-lived activation product C-14 is produced by neutron capture reactions mainly from N-14 impurities present in Zircaloy cladding and other metallic components of fuel elements.
For the waste management of irradiated Zircaloy or stainless steel, C-14 is a key radionuclide, which must be considered in safety assessments of deep geological repositories for nuclear waste. Corrosion of the emplaced waste possibly releases C-14 bearing volatile and/or dissolved compounds. Organic C-14 compounds reveal a high mobility either in the aqueous or in the gaseous phase and, once released, are potentially transported into the biosphere. On the contrary, volatile/dissolved inorganic C-14 bearing compounds are affected by various retention processes in the near field of a repository and the geosphere.
In this study, the inventory of C-14 and other radionuclides present in irradiated Zircaloy-4 cladding and a stainless steel plenum spring is determined. Furthermore, the chemical form of C-14 released from these materials is analysed.
Experimentally measured radionuclide contents are compared to theoretically predicted inventories of the irradiated Zircaloy-4 and stainless steel, obtained by means of Monte Carlo N-Particle calculations (MCNP-X).
The studied materials were sampled from the plenum of a fuel rod segment, which achieved an average burn-up of 50.4 GWd/t_HM in the Swiss Gösgen pressurized water reactor.
The Zircaloy-4 cladding (dose rate (DR) ≤ 20 mSv/h) as well as the stainless steel spring (X7 CrNiAl 17.7, 10.4 g, DR ≤ 1600 mSv/h) were dry cut and small subsamples (120–300 mg, DR ≤ 120 mSv/h) were digested in dilute acid solutions using an autoclave equipped with a gas collecting cylinder.
C-14 is separated from other radionuclides in aqueous and gaseous aliquots by stepwise extraction of the inorganic and organic carbon fractions by conversion into CO₂, which is then trapped in various alkaline washing bottles. Finally, the content of C-14 is analysed by liquid scintillation counting.
The measured C-14 inventory in Zircaloy-4, (3.7 ± 0.4)×10⁴ Bq/g, is in good agreement with the calculated one, (3.2 ± 0.3)×10⁴ Bq/g. The vast majority of C-14 (88 ± 10)% is released as gaseous organic compounds during dissolution of Zircaloy into the gas phase. Moreover, about (11 ± 10)% remains as dissolved organic C-14 bearing compounds in the acidic digestion liquor. Almost no inorganic C-14 bearing compounds (< 1%) are found in all experiments performed with Zircaloy, neither in the gaseous nor in the aqueous phase.
Preliminary experimental results obtained for the C-14 inventory present in stainless steel, (2.7 ± 0.3)×10⁵ Bq/g, agree within a factor of ~3 with the activation calculations performed for the material, (8.5 ± 0.9)×10⁴ Bq/g. In addition, the partitioning of C-14 released from steel between organic (~99%) and inorganic (< 1%) compounds is similar to that found in Zircaloy.

Spent nuclear fuel (SNF) from Advance Gas-Cooled (AGRs) in the UK is currently reprocessed at Thermal Oxide Reprocessing Plant (THORP) at Sellafield, this facility will cease operation within the next 5 years. The future plan for un-reprocessed AGR SNF and SNF that will be discharged from AGRs is to send it to a national GDF (geological disposal facility). The GDF is expected to start taking intermediate level waste in the late 2030s, with fuel to follow after the bulk of the accumulated ILW has been disposed of. The GDF may not be open for receipt of spent fuel until ~2075 and until then AGR SNF will be kept in interim storage ponds at Sellafield. These ponds are dosed with NaOH (to pH≈11.4) which acts as a corrosion inhibitor. Current storage periods are typically less than 10 years, although this may extend up to 100 years.
A new racking system for the THORP receipt and storage pond has been developed to accommodate all future SNF arisings. This will cause a rise in the temperature of the storage ponds giving an expected peak operating temperature of ~60°C and an average temperature of 45°C. Hence the evolution of fuel cladding, UO₂ and SIMFUEL surfaces upon exposure to pond water as a function of temperature are studied using electrochemical and imaging methods.
The corrosion of 20/25/Nb stainless steel (AGR cladding) and 304H stainless steel (Nb-free cladding analogue) was studied as a function of pH, temperature and chloride concentrations, both in the absence and presence of peroxide (as a simulant for water radiolysis) at concentrations typical of those found in pond water.
The National Nuclear Laboratory has been developing simulants for radiation sensitised fuel cladding by use of heat treatment methods on both stainless steel types. Thus heat treated 20/25/Nb and 304H samples were examined in an attempt to determine the extent and origin of any increased susceptibility to corrosion.
The results demonstrate the following, with respect to the initiation of corrosion:
It is advantageous in terms of minimising corrosion to dose the ponds to pH≈11.4. In most cases, at pH lower than ~7 the initiation of pitting is observed ~0.2V, pits are considered to be initiators of stress corrosion cracking (SCC). The initiation of pits is not seen above pH~7 for any temperature.
In the absence of peroxide, higher chloride concentrations,~200ppm, than those expected in pond water,~1ppm, are necessary to cause significant cathodic movement in breakdown potential and thus the onset of intergranular attack.
There generally appears to be no localised corrosion threat to fuel cladding as the electrolyte temperature is increased in the range 24°C-90°C, in the absence of peroxide,assuming that the fuel has not undergone SSC or intergranular attack before submersion in the ponds.
In the presence of peroxide, initiation of pitting is evident for a peroxide concentration of 0.5ppm for temperatures greater than 60°C for unsensitised samples and 45°C for sensitised samples.

Performance assessment calculations show that carbon-14 (C-14) is one of the main contributors to the annual dose from a cement-based deep geological repository for low- and intermediate-level radioactive waste (L/ILW) in Switzerland. For current performance assessment it is assumed that C-14 mainly contributes to the dose in its organic form, e.g. as C-14 bearing organic compounds, which are only weakly retarded in the cementitious near field. A compilation of activity inventories reveals that C-14 in L/ILW in Switzerland is mainly associated with activated steel (~85 %) which is produced due to the activation of nitrogen impurities in stainless steel exposed to thermal neutron flux, for example in nuclear reactors, according to the reaction 14N(n,p)14C. Release of C-14 occurs during the slow corrosion of the activated steel in the deep geological repository. Although the inventory of C-14 is well known, the chemical speciation of C-14 in the cementitious environment upon release from activated steel is still only poorly understood.
Identification and quantification of C-14 bearing organic compounds is a particularly challenging task because the inventory of C-14 in activated steel is low and the corrosion rate of steel under the alkaline conditions anticipated in a cement-based repository is very slow (only a few nm/year). As a consequence very low concentrations of C-14 bearing organic compounds are expected under these conditions. The only analytical technique suitable to determine these compounds with very low C-14 concentrations is compound-specific C-14 accelerator mass spectrometry (C-14 AMS).
The compound-specific C-14 AMS method was developed in several steps. At first, potentially C-14 containing organic target compounds were identified in batch-type corrosion experiments with non-activated carbonyl iron powders under anoxic alkaline conditions. Dissolved and volatile organic compounds were detected in the supernatant solution using high-performance ion exchange chromatography (HPIEC) and gas chromatography (GC) coupled to mass spectrometry (MS) detection. The main dissolved organic carbon species identified were carboxylic acids with a maximum of four carbon atoms. Then, chromatographic separation was combined with C-14 AMS for the detection of the individual C-14 containing carboxylic acids. To demonstrate the feasibility of the analytical strategy, measurements with C-14 labelled carboxylic acids were performed. Finally, the method was tested by identifying and quantifying the C-14 bearing carboxylic acids in the alkaline solution of a leach test with an activated steel sample.
The study shows results obtained in the course of the development of the C-14 AMS method along with results from the first experiments with activated steel samples. Furthermore, an ongoing corrosion study with activated steel will be outlined, including in the future also the analysis of volatile corrosion products using compound-specific C-14 AMS.

Monitoring and understanding of corrosion on nuclear sites plays a key role in safe asset management (predicting plant life, assessing efficacy of corrosion inhibitors for plant lifetime extension) and supporting informed choice of decontamination methods for steels due for decommissioning. Commonly used techniques for monitoring corrosion include: the measurement of bulk macroscopic mass changes on corrosion coupons; the measurement of material electrical resistance; and the measurement of material linear polarization resistance during sample immersion in the putatively corrosive environment. However the former is unsuitable for real time measurements, the medial and latter are not suitable for measuring in real time corrosion rates or pitting corrosion.
Recent advances in Quartz Crystal Nanobalance (QCN) technology offer a means to monitor corrosion in situ in radiologically harsh environments, in real time and with high sensitivity. The QCN measures minute changes in frequency of a quartz crystal resonator with weight gain/loss. Using the Sauerbrey equation, the drop in frequency observed during corrosion testing can be converted to an instantaneous corrosion rate with nanogram sensitivity. QCN technology is suitable for measuring uniform and pitting corrosion rates.
Experiments have concentrated on determining corrosion rates in acids and complexants used in chemical decontamination processes, particularly methods involving the commonly used cleaning agents nitric acid and oxalic acid (e.g. the CORD-UV process). Oxalic acid is currently being studied as an Enhanced Chemical Cleaning decontamination agent in the decommissioning of high level waste (HLW) storage tanks at the Hanford and Savannah River Sites. Oxalic acid has the ability to act as a rust remover and as a corrosion inhibitor. Therefore it is preferred over bulk nitric acid. The tanks are comprised of low carbon steel; thus the decontamination process must be carefully monitored to avoid over-aggressive decontamination that may result in a loss of asset structural integrity. To avoid this, the oxalic acid concentration being used has been reduced to 1 wt% from the 4-8 wt% range typically used during decontamination campaigns.
Using the QCN we have measured the corrosion rates of mild carbon steel surrogates in oxalic acid-based decontamination solutions, and have directly observed both the removal by oxalic acid of iron oxide corrosion product and the formation of a protective layer of ferrous oxalate at the metal surface. Studies have recently extended to newer decontamination solution formulations including basic and acid permanganate solutions and oxalic acid/nitric acid blends. In the case of the former, preliminary results indicate that, counter intuitively, base concentrations of>10mol dm^-3 will lead to a loss of HLW storage tank structural integrity. This is currently the subject of further study.