UK Advanced Gas-cooled Reactors Research Articles

Simulated advanced gas-cooled reactor spent nuclear fuels: Determination of the O/U ratio - an XRD, XPS and Raman study

This paper reports a detailed chemical and materials characterization study of novel Simulated Spent Nuclear Fuels (SIMFuel) which replicate the chemical and microstructural state of Spent Nuclear Fuel (SNF) discharged from a UK Advanced Gas-cooled Reactor (AGR). Recent advances in the analysis of pure UO2 samples by XPS, micro-Raman spectroscopy and XRD, have been assessed for deployment in the characterisation of SIMFuels generally and these AGR SIMFuels in particular. All three analytical methods reveal the extent that the UO2 bulk matrix is defected in the SIMFuels by the presence of lanthanide dopants. XPS, by inspection of the U4f peaks and their satellites, indicates the presence of U(V) in the UO2 matrix as a means to charge compensate for the incorporation of Ln(III) states and presence of a slight hyperstoichiometry in the UO2 solid solution. This was corroborated by detailed Raman analysis of the UO2 matrix of the SIMFuels – wherein the (Ba,Sr)ZrO3 and metallic particle phases, that simulate the precipitated grey phases and ε-particles formed in real spent fuel, were avoided – which indicated the presence of both interstitial oxygens and oxygen vacancies, the latter as a parallel means of charge compensation for the presence of Ln(III). Further confirmation was provided by XRD measurements through observation of a lattice parameter contraction arising from the presence of the smaller U(V) ion in the U(IV)O2 matrix. XPS and Raman allow for the reporting of O/U ratios in the SIMFuels. These XPS and Raman-derived O/U ratios are in good agreement, indicating: (i) that the SIMFuels are near stoichiometric/slightly hyperstoichiometric, concurring with previous analogous studies on real and simulated light water reactor fuels; and (ii) that either technique may find application in the analysis of real fuel analysis, both pre- and post-irradiation.

Creep deformation phenomena in near-surface carburised layers of 316H stainless steels

The local creep-induced phenomena occurring within the near-surface carburised layer in two pre-carburised type 316H austenitic stainless steels, containing 0.98 or 1.52 wt% Mn, was assessed at 550 °C. The pre-carburisation exposed both steel grades at either 550 °C (8000 h) or 600 °C (3000 h) to simulated CO2-rich gas environment of UK Advanced Gas-cooled Reactors. The local properties within the carburised layer causes a reduction in minimum creep rate as compared to the non-carburised condition. In both steels carburised at 600 °C, ferrite is observed to form in the most heavily carburised regions, including slip bands, during the initial plastic deformation to the target applied stress. The ferrite area fraction further increases with creep strain. Within the ferritic areas, new fine grains are observed in the creep tested specimens and the fraction of high angle grain boundaries is >60%. Those fine grains have a lower degree of lattice strain compared to the surrounding ferrite area and especially to the austenite matrix. The lattice strain localises within the carburised layer, and the formation of ferrite from the strained austenite, and potentially a dynamic recrystallization within the ferrite regions, pinpoints to a dominant strain relief-driven mechanism. Lattice strain localises primarily in the pre-carburised austenite grains and its carbide-decorated boundaries, and induces the continuous formation and refinement of ferrite grains, and the development of intergranular cracks that evolve further into the bulk structure. These results emphasise the necessity to account for environmental-creep interactions at elevated temperatures in structural assessment procedures, in the current gas-cooled fission reactor fleet and beyond into (very) high-temperature Gen IV and advanced modular reactor designs.

Large-Scale Modeling of Damage and Failure of Nuclear Graphite Moderated Reactor

Abstract The UK Advanced Gas-Cooled reactors (AGRs) have cores made of graphite bricks with dual functions: as structural elements of the core, providing space for and separating fuel and control rods; and as moderator of the nuclear reaction. Nuclear graphite is a quasi-brittle material, where the dominant mechanism for failure is cracking. While cracking of isolated bricks is expected due to operation-induced changes in graphite microstructure and stress fields, these could be tolerated as far as the overall structural function of the core is maintained. Assessment of the whole core behavior has been previously done with whole scale models where bricks have been considered as rigid body elements connected by elastic-brittle springs. This approach does not allow for the realistic assessment of the stresses in the bricks and associated brick cracking. Reported here are results from an ongoing project, which addresses this shortcoming. The proposed model uses deformable bricks with appropriate interactions, allowing for physically realistic whole core analysis. The results are focused on the damage that a graphite moderated reactor develops during a life cycle, how this affects the behavior of the whole core, and how changes in bricks' behavior impacts the core integrity. The proposed methodology is a major step toward high-fidelity assessment of AGRs' fitness for service, required for supporting continuous safe operation and life-extension decisions.

Microstructural Deformation and Modulus Changes of Nuclear Graphite due to Hydrostatic Pressure Loading

For the first time, the influence of pore closure on elastic modulus in the UK's Magnox and Advanced Gas Cooled Reactors (AGRs) graphite was determined by subjecting the graphite samples to high hydrostatic pressure and dynamic loading. Two grades of nuclear graphite were tested; anisotropic Pile Grade A (PGA) which was used in the Magnox reactors, and semi-isotropic Gilsocarbon as used in the AGRs [1] . The volumetric strain was measured in two directions during compression and separately the velocity of sound through the samples was also determined as a function of confinement pressure. Under hydrostatic loading, the stiffness of PGA graphite was reduced after a few percent of volume strain before increasing again after about ∼20% volumetric strain. Gilsocarbon showed similar behaviour to PGA at lower volumetric strain (∼10 to 13%) however due to Gilsocarbon having a higher density and lower porosity compared to PGA, the response was generally stiffer than PGA. During unloading, significant hysteresis was observed in both graphite grades, with the sample volume almost fully recovering when the pressure was completely removed. Micromechanical models are used to explain the poro-elastic response of graphite during hydrostatic compaction, and to make comparisons and extend data on the influence of pore change due to oxidation on modulus.

Creep performance of carburized 316H stainless steel at 550 °C

The creep behaviour of two ex-service type 316H austenitic stainless steel casts with different Mn contents and austenite grain size, that were pre-carburized for either 8000 h at 550 °C or 3000 h at 600 °C in simulated CO2-rich gas environment of UK Advanced Gas-cooled Reactors, was evaluated by uniaxial creep testing at 550 °C in air. The low Mn (0.98 wt.%) steel grade with an average austenite grain size of 142 ± 110 µm presents a higher creep deformation rate and creep ductility compared to the high Mn (1.52 wt.%) steel cast with an austenite grain size of 81 ± 67 µm. Below an applied stress of 318 MPa, the presence of the sub-surface carburized layer reduces the minimum creep rate, likely due to the significantly modified material properties of the carburized layer and the presence of local compressive stresses. The stress exponent determined for the high Mn steel carburized at 600 °C (n = 9.6), and its reference non-carburized microstructure (n= 9.8), identifies the dominant creep mechanism to be dislocation climb & glide. However, crack evolution through M23C6-decorated grain boundaries within the carburized layer was observed in all the tested material's conditions at creep strains of 0.50%. The average crack length increases with creep strain and attains values close to the carburized layer thickness at creep strains of ε ≥ 1.00%. The prevalence and size of cracking increases with the local hardness carburized layer depth. At applied stresses σ ≥ 318 MPa, cracking on-loading prior to forward creep already occurs in the carburized layer, leading to a sharp increase in minimum creep rates of carburized specimens. These results demonstrate the impact a relatively thin layer of modified material can have on the materials creep performance; emphasising the requirement to consider the effects of the service environment in structural assessment procedures, particularly for high temperature applications including Generation IV and advanced modular reactor designs.

Design of a direct-cycle supercritical CO2 nuclear reactor with heavy water moderation

A new reactor concept is described that directly couples a supercritical CO2 (sCO2) power cycle with a CO2-cooled, heavy water moderated pressure tube core. This configuration attains the simplification and economic potential of past direct-cycle sCO2 concepts, while also providing safety and power density benefits by using the moderator as a heat sink for decay heat removal. A 200 MWe design is described that heavily leverages existing commercial nuclear technologies, including reactor and moderator systems from Canadian CANDU reactors and fuels and materials from UK Advanced Gas-cooled Reactors (AGRs). Descriptions are provided of the power cycle, nuclear island systems, reactor core, and safety systems, and the results of safety analyses are shown illustrating the ability of the design to withstand large-break loss of coolant accidents. The resulting design attains high efficiency while employing considerably fewer systems than current light water reactors and advanced reactor technologies, illustrating its economic promise. Prospects for the design are discussed, including the ability to demonstrate its technologies in a small (∼20 MWe) initial system, and avenues for further improvement of the design using advanced technologies.

Reconstructing the conductivity profile of a graphite block using inductance spectroscopy with data-driven techniques

The radial depth profile of the electrical conductivity of the graphite channels in the UK's advanced gas-cooled reactors (AGRs) can be reconstructed and estimated by solving a non-linear optimisation problem using the mutual inductance spectra of a set of coils. This process is slow, as it requires many iterations of a forward solver. Alternatively, a data-driven approach can be used to provide an initial estimate for the optimisation algorithm, reducing the amount of time it takes to solve the ill-posed inverse problem. Two data-driven approaches are compared: multi-variable polynomial regression (MVPR) and a convolutional neural network (CNN). The training data are generated using a finite element (FE) model and superimposed on a noise floor in the interval [20, 60] dB of the weakest amplitude point in the corresponding spectrum. A total of 5000 simulated datasets are generated for training. The results on smoothed test data show that the two models have a comparable mean percentage error norm of 17.8% for the convolutional neural network and 17.3% for multivariable polynomial regression. A further 500 unsmoothed profiles are tested in order to assess the performance of each algorithm on conductivity distributions where the conductivity of each layer is independent of another. The performance of both algorithms is then assessed on reactor-type test data. The results show that the two data-driven algorithms have a comparable performance when estimating the electrical conductivity depth profile of a typical reactor-type distribution, as well as vast deviations. More generally, it is thought that data-driven approaches for depth profiling of some electromagnetic quantity have the potential to be applied to other ill-posed inverse problems where speed is a priority.

Advanced Gas-cooled Reactor SIMFuel Fabricated by Hot Isostatic Pressing: a Feasibility Investigation

The manufacture of a simulant UK Advanced Gas Cooled Reactor (AGR) spent nuclear fuel (SIMFuel) was achieved by Hot Isostatic Pressing (HIP). Characterisation of HIP AGR SIMFuels, tailored to burn ups of 25 GWd/t U and 43 GWd/t U (after 100 years cooling) demonstrated fission product partitioning, phase assemblage, microstructure and porosity in good agreement with spent nuclear fuels and SIMFuels, and AGR fuels in particular. A pivotal advantage of the application of the HIP manufacturing method is the retention of volatile fission products within the resultant SIMFuel as the result of using a hermetically-sealed container. This new approach to SIMFuel manufacture should enable the production of more accurate spent nuclear fuel surrogates to support research on spent fuel management, recycle, and disposal, and the thermal treatment of fuel residues and debris.

The effect of irradiation and radiolytic oxidation on the porous space of Gilsocarbon nuclear graphite measured with mercury porosimetry and helium pycnometry

Gas pycnometry and mercury porosimetry are used to investigate the porous network of Gilsocarbon nuclear graphite samples that are representative of the material present in the cores of UK Advanced Gas-Cooled reactors at different stages of the reactors’ operational lifetimes. Irradiation and radiolytic oxidation change the pore volume of nuclear graphite and the relative ratios of open (coolant gas accessible) and closed pore volume. Particular focus has been paid to the deformation of the Gilsocarbon graphite observed during mercury intrusion at high pressure, which has previously marred the use of porosimetry to characterise this material. The results show clear trends in the evolution of the Gilsocarbon graphite porous space. Semi-quantitative deductions are made that will assist the modelling of the evolution of the pore space in the context of the safe extension of the reactors’ working lifetimes.

Analysis of dynamic fracture and fragmentation of graphite bricks by combined XFEM and cohesive zone approach

It has been anticipated and recently observed that axial cracks can occur at keyway corners of graphite bricks in the UK Advanced Gas-cooled Reactors (AGR). The crack initiation is triggered by internal stress reversal during reactor operation and may potentially lead to brick more cracks through a mechanism known as Prompt Secondary Cracking (PSC). To assess the likelihood of this occurring, a method for modelling dynamic fracture, known as eXtended Finite Element Method with cohesive elements (XCZM), is used to analyse whole 3D component fragmentation. XCZM is used here to model primary crack interactions with a methane hole, the effect of graphite heterogeneity, and secondary crack profiles in a 3D brick slice under external hoop stress loading. The results show that methane holes in bricks can be beneficial for brick integrity, as these can lead to crack path deflection with associated increase of dissipated energy. Further, the results suggest that graphite heterogeneity does not affect significantly the crack profile and therefore the integrity assessment. Finally, it is demonstrated that PSC may readily occur and the secondary crack propagates towards regions of high kinetic energy.

The effect of ion irradiation on the dissolution of UO2 and UO2-based simulant fuel

The aim of this work was to study the separate effect of fission fragment damage on the dissolution of simulant UK advanced gas-cooled reactor nuclear fuel in water. Plain UO2 and UO2 samples, doped with inactive fission products to simulate 43 GWd/tU of burn-up, were fabricated. A set of these samples were then irradiated with 92 MeV 129Xe23+ ions to a fluence of 4.8 × 1015 ions/cm2 to simulate the fission damage that occurs within nuclear fuels. The primary effect of the irradiation on the UO2 samples, observed by scanning electron microscopy, was to induce a smoothening of the surface features and formation of hollow blisters, which was attributed to multiple overlap of ion tracks. Dissolution experiments were conducted in single-pass flow-through (SPFT) mode under anoxic conditions (<0.1 O2 ppm in Ar) to study the effect of the induced irradiation damage on the dissolution of the UO2 matrix with data collection capturing six minute intervals for several hours. These time-resolved data showed that the irradiated samples showed a higher initial release of uranium than unirradiated samples, but that the uranium concentrations converged towards ∼10−9 mol/l after a few hours. Apart from the initial spike in uranium concentration, attributed to irradiation induced surficial micro-structural changes, no noticeable difference in uranium chemistry as measured by X-ray electron spectroscopy or ‘effective solubility’ was observed between the irradiated, doped and undoped samples in this work. Some secondary phase formation was observed on the surface of UO2 samples after the dissolution experiment.

A multi-scale three-dimensional Cellular Automata fracture model of radiolytically oxidised nuclear graphite

A multi-scale approach for fracture simulation, based on the Cellular Automata technique, has been developed and then applied to a nuclear graphite that is used in structural components of the UK Advanced Gas-cooled Reactors (AGR). High resolution X-ray computed tomographs of Gilsocarbon grade graphite, with up to 68% weight loss by radiolytic oxidation, provide quantitative descriptions of the porosity within its constitutive filler particles and their surrounding matrix. The statistical distributions for tensile strength and elastic modulus obtained from small models of the filler and matrix are introduced to a large scale model of the heterogeneous microstructure. These microstructure-derived simulations achieve a good agreement with experimental data. The computationally efficient analysis is then used to investigate the stochastic effects on mechanical properties of possible variations in the microstructure between individual small test specimens. As these may represent material of nominally the same microstructure, there is an apparent increase in the variability of strength and modulus at high weight loss.

Kinetics of Duplex Oxide Growth on 9Cr Steels Exposed in CO2: Application of Dimensional Metrology

Investigations into potentially extending the lives of UK advanced gas-cooled reactors have highlighted the need for improved understanding of the long-term oxidation and carburisation of 9Cr ferritic steels. These steels were used in evaporators and primary superheaters and as these are to be used beyond their original design lives, it is necessary to ensure that these degradation routes remain within acceptable levels. A dimensional metrology technique has been applied to archived autoclave samples to measure such damage. These samples had previously been exposed to a range of temperatures, pressures and gas chemistries representative of those experienced by 9Cr steels in CO2-rich AGR gases. Earlier sample assessments had focused on weight change measurements, but the dimensional technique enables measurement of duplex oxide thicknesses around samples and the extraction of related data from longer exposure times. These data also support estimation of the extent of both carbon and oxygen uptake.

Corrosion of AGR Fuel Pin Steel Under Conditions Relevant to Permanent Disposal

Fuel pins from the UK's Advanced Gas-cooled Reactors (AGR) consist of ceramic UO2 fuel encased in a 20/25/Nb stainless steel cladding. Spent AGR fuel is currently reprocessed, but the option of direct disposal of spent fuel in a sealed, underground Geological Disposal Facility (GDF) is now under examination. It is assumed that over several thousand years groundwater from the environment will penetrate these barriers and come into contact with the fuel surface and steel cladding.Electrochemical studies on unsensitised samples of 20/25/Nb steel in simulant groundwater electrolytes have been performed, and show low corrosion currents, typically of the order μA/cm2, at the oxidative potential stresses found in such repositories. Whilst the cladding may therefore be considered to be passive short time periods, the very long timescales involved in a GDF project mean that these currents are sufficient to cause corrosion of the full thickness of the clad in a matter of decades.Furthermore, an increase in electrochemical potential from that expected of less than 50mV is sufficient to initiate severe pitting corrosion in a matter of hours. It can therefore be assumed that there is a risk of cladding corrosion in a repository environment, and the corrosion products created may have implications for the chemistry of the spent fuel ceramic.

Storage of thermal reactor fuels – Implications for the back end of the fuel cycle in the UK

Fuel from UK's Advanced Gas-Cooled Reactors (AGRs) is being reprocessed, however reprocessing will cease in 2018 and the strategy for fuel that has not been reprocessed is for it to be placed into wet storage until it can be consigned to a geological disposal facility in around 2080. Although reprocessing of LWR fuel has been undertaken in the UK, and this option is not precluded for current and future LWRs, all utilities planning to operate LWRs are intending to use At-Reactor storage pending geological disposal. This strategy will result in a substantial change in the management of spent fuel that could affect the back end of the fuel cycle for over a century. This paper presents potential fuel storage scenarios for two options: the current nuclear power replacement strategy, which will see 16 GWe of new capacity installed by 2030 and a median strategy, intended to ensure implementation of the UK's carbon reduction target, involving 48 GWe of nuclear capacity installed by 2040. The potential scale, distribution and timing of fuel storage and disposal operations have been assessed and changes to the current industrial activity are highlighted to indicate potential effects on public acceptance of back end activities.

Processing and microstructural characterisation of a UO2-based ceramic for disposal studies on spent AGR fuel

Preparation and characterisation of a Simulated Spent Nuclear Fuel (SIMFuel), which replicates the chemical state and microstructure of Spent Nuclear Fuel (SNF) discharged from a UK Advanced Gas-cooled Reactor (AGR) after a cooling time of 100years is described. Given the relatively small differences in radionuclide inventory expected over longer time periods, the SIMFuel studied in this work is expected to be also representative of spent fuel after significantly longer periods (e.g. 1000years). Thirteen stable elements were added to depleted UO2 and sintered to simulate the composition of fuel pellets after burn-ups of 25 and 43 GWd/tU and, as a reference, pure UO2 pellets were also investigated. The fission product distribution was calculated using the FISPIN code provided by the UK National Nuclear Laboratory. SIMFuel pellets were up to 92% dense and during the sintering process in H2 atmosphere Mo–Ru–Rh–Pd metallic precipitates and grey-phase ((Ba, Sr)(Zr, RE) O3 oxide precipitates) formed within the UO2 matrix. These secondary phases are present in real PWR and AGR SNF. Metallic precipitates are generally spherical and have submicron particle size (0.8±0.7μm). Spherical oxide precipitates in SIMFuel measured up to 30μm in diameter, but no data were available in the public domain to compare this to AGR SNF. The grain size of actual AGR SNF (∼3–30μm) is larger than that measured in AGR SIMFuel (∼2–5μm).

Estimating the occurrence of foreign material in Advanced Gas-cooled Reactors: A Bayesian Monte Carlo approach

The current occurrence of a particular sort of foreign material in eight UK Advanced Gas-cooled Reactors has been estimated by means of a parametric approach. The study includes both variability, treated in analytic fashion via the combination of standard probability distributions, and the uncertainty in the parameters of the model of choice, whose posterior distribution was inferred in Bayesian fashion by means of a Monte Carlo route consisting in the conditional acceptance of sets of model parameters drawn from a prior distribution based on engineering judgement.The model underlying the present study specifically refers to the re-loading and inspection routines of UK Advanced Gas-cooled Reactors. The approach to inference here presented, however, is of general validity and can be applied to the outcome of any inspection campaign on any plant system, and indeed to any situation in which the outcome of a stochastic process is more easily simulated than described by a probability density or mass function.

Preparation and characterization of UO2-based AGR SIMFuel

ABSTRACTPreparation and characterization of a Simulated Spent Nuclear Fuel (SIMFuel), which replicates the chemical state and microstructure of Spent Nuclear Fuel (SNF) discharged from UK Advanced Gas-cooled Reactor (AGR) after a cooling time of 100 years is described. Thirteen stable elements were added to depleted UO2 and sintered to simulate the composition of fuel pellets after burn-ups of 25 and 43 GWd/tU and, as a reference, pure UO2 pellets were also investigated. The fission product distribution was calculated using the Fispin code provided by NNL. SIMFuel pellets exhibit a microstructure up to 92% TD. During the sintering process in H2 atmosphere Mo-Ru-Rh-Pd metallic precipitates and grey-phase ((Ba, Sr)(Zr, RE)O3 oxide precipitates) formed within the UO2 matrix. These secondary phases are present in real PWR and AGR SNF, although they are smaller in size than those examined in this study. The grain size of the produced SIMFuel is in good agreement with literature references.

Effect of horizontal cross flow on the heat transfer form the moderator bricks in the Advanced Gas-cooled Reactor: A CFD study

The paper reports an investigation of the effect of the horizontal cross flow on the temperature of the moderator brick in the UK Advanced Gas-cooled Reactor (AGR) using computational fluid dynamics (CFD) with a conjugate heat transfer model for the solid and fluid. The commercial software package ANSYS Fluent is used to study the correlation between the flow structure and heat transfer which ultimately determine the temperature within the system. The CFD model comprises the full axial length of one-half of a typical fuel channel (assuming symmetry) and part of the neighbouring channels on either side. Two sets of simulations have been carried out, namely, one with cross flow and one without cross flow. The effect of cross flow has subsequently been derived by comparing the results from these two groups of simulations. The study shows that a small cross flow can have a significant effect on the cooling of the graphite brick, causing the peak temperature of the brick to reduce significantly. Two mechanisms are identified to be responsible for this. Firstly, the small cross flow causes a significant redistribution of the main axial downward flow and this leads to an enhancement of heat transfer in some of the small clearances, and an impairment in others although overall, the enhancement is dominant leading to a better cooling. Secondly, the cross flow makes effective use of the small clearances between the key/keyway connections which increases the effective heat transfer area, hence increasing the cooling. Under the conditions of no cross flow, these areas remain largely inactive in heat transfer. The study shows that the cooling of the moderator is significantly enhanced by the cross flow perpendicular to the main cooling flow.

A Three-Dimensional Methodology for the Assessment of Neutron Damage and Nuclear Energy Deposition in Graphite Components of Advanced Gas-Cooled Reactors

Abstract This paper describes the development of a three-dimensional methodology for the assessment of neutron damage and nuclear energy deposition (or nuclear heating) throughout the graphite cores of the UK’s Advanced Gas-cooled Reactors. Advances in the development of the Monte Carlo radiation transport code MCBEND have enabled the efficient production of detailed fully three-dimensional models that utilise three-dimensional source distributions obtained from Core Follow data supplied by the reactor physics code PANTHER. The calculational approach can be simplified to reduce both the requisite number of intensive radiation transport calculations, as well as the quantity of data output. These simplifications have been qualified by comparison with explicit calculations and they have been shown not to introduce significant systematic uncertainties. Simple calculational approaches are described that allow users of the data to address the effects on neutron damage and nuclear energy deposition predictions of the feedback resulting from the mutual dependencies of graphite weight loss and nuclear energy deposition.