Dounreay Fast Reactor Research Articles

THE DOUNREAY FAST REACTOR VAULT CHARACTERISATION USING THE MONTE CARLO N-PARTICLE CODE AS A TOOL TO SUPPORT DECOMMISSIONING PLANNING

A combined empirical and theoretical characterisation approach was taken in order to investigate the activity distribution and inventory in the Dounreay Fast Reactor (DFR) vault, so as to aid with the decommissioning and waste management strategy of the cooling circuits located within the shielded vault. The DFR vault is a no-man access, shielded void, which surrounds the reactor core and the graphite shield. Empirical data was collected through various Non-Destructive Assay methods, utilising all available accesses into the reactor vault. This collection of information was then used to make a first assumption about the activity distribution within the vault; this was then included into a MCNP model of the vault as the source terms. The model was run and a dose rate map of the vault obtained, and compared to the empirical dose rate data previously gathered in the accessible penetrations in the vault; the model was altered until a satisfactory agreement was obtained between theoretical and measured data. Once this was obtained then, the model was used to provide an estimate of the activity inventory within the vault, and specifically within the various reactor components.

Characterization of radioactive particles from the Dounreay nuclear reprocessing facility

Radioactive particles originating from nuclear fuel reprocessing at the United Kingdom Atomic Energy Authority's Dounreay Facility were inadvertently released to the environment in the late 1950s to 1970s and have subsequently been found on site grounds and local beaches. Previous assessments of risk associated with encountering a particle have been based on conservative assumptions related to particle composition and speciation. To reduce uncertainties associated with environmental impact assessments from Dounreay particles, further characterization is relevant.Results of particles available for this study showed variation between Dounreay Fast Reactor (DFR) and Materials Test Reactor (MTR) particles, reflecting differences in fuel design, release scenarios, and subsequent environmental influence. Analyses of DFR particles showed they are small (100–300 μm) and contain spatially correlated U and Nb. Molybdenum, part of the DFR fuel, was identified at atomic concentrations below 1%. Based on SR-based micrometer-scale X-ray Absorption Near Edge Structure spectroscopy (μ-XANES), U may be present as U(IV), and, based on a measured Nb/U atom ratio of ~2, stoichiometric considerations are commensurable with the presence of UNb2O7. The MTR particles were larger (740–2000 μm) and contained U and Al inhomogeneously distributed. Neodymium (Nd) was identified in atomic concentrations of around 1–2%, suggesting it was part of the fuel design. The presence of U(IV) in MTR particles, as indicated by μ-XANES analysis, may be related to oxidation of particle surfaces, as could be expected due to corrosion of UAlx fuel particles in air. High 235U/238U atom ratios in individual DFR (3.2 ± 0.8) and MTR (2.6 ± 0.4) particles reflected the presence of highly enriched uranium. The DFR particles featured lower 137Cs activity levels (2.00–9.58 kBq/particle) than the MTR (43.2–641 kBq 137Cs/particle) particles. The activities of the dose contributing radionuclides 90Sr/90Y were proportional to 137Cs (90Sr/137Cs activity ratio ≈ 0.8) and particle activities were roughly proportional to the size. Based on direct beta measurements, gamma spectrometry, and the VARSKIN6 model, contact dose rates were calculated to be approximately 74 mGy/h for the highest activity MTR particle, in agreement with previously published estimates.

Deterministic and Monte Carlo Neutron Transport Calculations of the Dounreay Fast Breeder Reactor

A homogenized whole-reactor cylindrical model of the Dounreay Fast Reactor has been constructed using both deterministic and Monte Carlo codes to determine neutron flux distributions inside the core and at various out-of-core components. The principal aim is to predict neutron-induced activation levels using both methods and make comparisons against the measured thermal reaction rates. Neutron transport calculations have been performed for a fixed source using a spatially lumped fission neutron distribution, which has been derived from measurements. The deterministic code used is based on the finite element approximation to the multigroup second-order even-parity neutron transport equation, which is implemented in the EVENT code. The Monte Carlo solutions were obtained using the MCNP4C code, in which neutron cross sections are represented in pointwise (or continuous) form. We have compared neutron spectra at various locations not only to show differences between using multigroup deterministic and continuous energy (point nuclear data) Monte Carlo methods but also to assess neutron-induced activation levels calculated using the spectra obtained from both methods. Results were also compared against experiments that were carried out to determine neutron-induced reaction rates. To determine activation levels, we employed the European Activation Code System FISPACT. We have found that the neutron spectra calculated at various in-core and out-of-core components show some differences, which mainly reflect the use of multigroup and point energy nuclear data libraries and methods employed, but these differences have not resulted in large errors on the calculated activation levels of materials that are important (such as steel components) for decommissioning studies of the reactor. The agreement of calculated reaction rates of thermal neutron detectors such as the 55Mn(n,γ)56Mn against measurements was satisfactory.

Modelling of the Dounreay Recovered Spent Nuclear Fuel Database

AbstractWe report the initial results of modelling studies performed on the Dounreay Database of Spent Nuclear fuel particles recovered from the marine environment and coastal foreshore adjacent to the Dounreay Fast Reactor Complex located in the North East coast of Scotland. We demonstrate how interrogation of the fuel-particle frequency versus 137Cs isotopic distribution allows us to generate particle behaviour models and residence lifetimes of various particle subgroups. Similarly we find the recovered fuel particle's spatial distribution and shape characteristics are determined by environmental selection rules governed by properties of adjacent sediment and sediment transport mechanisms.Analysis of the Spent Fuel particle database allows us to speculate on the original Site discharge pathways and fuel composition at the time of discharge. Further such studies correlate well with other sample morphology and compositional data and allow us to speculate upon the ultimate fate of different spent fuel types.

Observation of a high burnup rim-type structure in an advanced plutonium–uranium carbide fuel

The observation is reported of a `rim-type' structure with small subgrains in an advanced plutonium–uranium carbide (U 0.8Pu 0.2)C fuel pin, which had been irradiated in the Dounreay Fast Reactor to a burnup of 8.3% FIMA.

Irradiation creep and void swelling of austenitic stainless steels at low displacement rates in light water energy systems

The majority of the high fluence data on the void swelling and irradiation creep of austenitic steels was generated at relatively high displacement rates characteristic of fast reactors. The application of these data to other reactor systems which operate at lower displacement rates requires that some estimate be made of the dependence of swelling and irradiation creep on displacement rate. Austenitic components in typical light water reactors (LWRs) experience displacement rates that are an order of magnitude or more lower than that found in fast reactors. Since irradiation creep has a component that is proportional to swelling, it is anticipated that irradiation creep will exhibit a sensitivity to displacement rate that is a direct expression of the dependence of swelling on displacement rate. Until recently, however, the non-swelling-related component of creep was also thought to exhibit its own dependence on displacement rate, increasing at lower fluxes. This perception originally arose from the work of Lewthwaite and Mosedale on irradiation creep of cold-worked steels in the Dounreay Fast Reactor at temperatures in the 270–350°C range. It now appears that this interpretation is incorrect and, in fact, the steady-state value of the non-swelling component of irradiation creep is relatively insensitive to displacement rate. The earlier perceived flux dependence appears to have arisen from a failure to properly separate the transient and steady-state regimes of irradiation creep. As evidence begins to accumulate on void swelling of stainless steels at LWR-relevant displacement rates, voids are being observed at lower-than-expected temperatures and often at very low dpa levels, implying that the ‘temperature shift’ phenomenon may indeed be operating to increase swelling. It is demonstrated in this paper that if swelling exceeds ∼ 10% at PWR-relevant temperatures, stainless steels become very brittle.

Determination of trace impurities in sodium coolant from a Fast Breeder Reactor by inductively coupled plasma atomic-emission spectrometry

Trace impurities in samples of the liquid sodium coolant from the Dounreay Fast Reactor are determined by inductively coupled plasma (ICP) atomic-emission spectrometry of vacuum distillation residues. A 72-channel vacuum spectrometer with a 1-kW argon plasma source, fed by a coaxial capillary nebuliser, simultaneously measures 59 elements with a precision varying from <1% at concentrations 20 times the detection limit to 30% at detection levels. The ICP torch box is enclosed in a fume-cupboard fitted with a filtered extract system to allow safe handing of the radioactive solutions. Comparisons of some results have been made with spark-source mass spectrometry, d.c. arc emission spectrography, neutron-activation analysis and atomic-absorption spectrometry.

Void-swelling and precipitation in a neutron-irradiated, niobium-stabilised austenitic stainless steel

Unstressed specimens of FV548 niobium-stabilised austenitic steel in three heat-treatment conditions (1150°C solutiontreated, 20% cold-worked, and 850°C aged) have been neutron-irradiated in the Dounreay Fast Reactor in the temperature range 380°–738°C and to displacement doses of up to 30 dpa. The irradiated specimens have been examined using transmission electron microscopy and EDX techniques to establish the void-swelling and precipitation behaviour under neutron irradiation. Where possible, the observations are compared with results obtained on type 316 steel. The similarities and differences in the behaviour of the two steels are used in an attempt to understand the factors affecting the precipitation under irradiation and the void-swelling behaviour of FV548 and type 316 steels.Irradiation is shown to enhance the formation of some precipitate phases (e.g. G-phase and carbides of the M6C type) while having no effect on others (e.g. Sigma phase and M23C6). It is concluded that fine intragranular dispersions of niobium carbide particles play a major role in suppressing void-swelling in neutron-irradiated FV548 steel at relatively high irradiation temperatures.

The calibration of fast reactor irradiated silicon carbide temperature monitors using a length measurement technique

Silicon carbide temperature monitors have been calibrated against thermocouples in the Dounreay Fast Reactor. Statistical confidence levels have been determined for future temperature estimates based on as-irradiated length change and on the post-irradiation annealing intersection temperature. Possible dose limitations are discussed.

Pellet and Sphere-Pac (U, Pu)C Fuel Comparative Irradiation Tests

Mixed (U,Pu)C fuel pins of both pellet and sphere-pac designs have been irradiated in epithermal and fast fluxes. The results of postirradiation examples, which are reported in this paper, allow a limited comparison of the two designs to be made. Two experiments were performed, one in the Belgian Reactor 2 at Mol, Belgium and the other in the United Kingdom's Dounreay Fast Reactor. 10 refs.

The effects of temperature and dose-rate variations on the creep of austenitic stainless steels in the dounreay fast reactor

Loaded helices of cold-worked, austenitic stainless steels (En58B, E, J, FV548 and type 316) have been irradiated in DFR at damage rates ranging over an order of magnitude at temperatures between 513 and 633 K. Creep rates varied between steels but a common pattern emerged where strain per unit dose appeared to increase with decreasing dose-rate. Comparison with experiments at 773 K on three of the steels reveals creep rate increases by factors ranging from 1.2 to 5. The results are discussed in the context of other fast reactor data and possible explanations of the findings are considered.

Void swelling and irradiation-induced void shrinkage in neutron irradiated molybdenum and TZM

Molybdenum and TZM samples have been neutron-irradiated in the Dounreay Fast Reactor to doses up to 8 × 10 26n/m 2 ( E > 0.1 MeV) and at temperatures between 450 and 900°C. This paper presents and discusses the results of a transmission electron microscope study of the void parameters in the irradiated samples. In general, these parameters were not too dependent on either purity or irradiation temperature. Results showing the effects of dose were rather limited but at 450°C sufficient evidence was obtained to show convincingly that in both molybdenum and TZM the usual void growth processes were being replaced during the irradiation by a mechanism tending to produce void shrinkage. The shrinkage process is discussed in terms of the segregation of a transmutation impurity, possibly technetium, to the void surface where its accumulation gradually inhibits the flow of vacancies into the voids.

Void-swelling in irons and ferritic steels: II. An experimental survey of materials irradiated in a fast reactor

The void-swelling response of a wide range of ferritic alloys irradiated in the Dounreay Fast Reactor to displacement doses up to 30 dpa (N/2) and covering the temperature range 380–615°C have been compared. The materials selected included high purity irons, together with commercial mild and low alloy steels, high chromium (12–14%) ferritic and martensitic stainless steels and a range of high purity binary iron-chromium alloys containing chromium contents up to 15%. The pure irons and binary iron-chromium alloys exhibited measurable but relatively low swellings (<1%) whilst all the commercial ferritic steels appeared to be void-swelling resistant, with swellings below the experimental detection limit (0.1%). The pattern emerging is thus one of overall swelling resistance in ferritic materials as a general class. Void-swelling in the pure iron peaked at two irradiation temperatures (~420, ~510°C), and the low magnitude of the swelling was rationalized in terms of the operation of solute-controlled swelling suppression mechanisms involving residual interstitial impurities. The complex functional dependence of peak-swelling on chromium content in the binary iron-chromium alloys was explained in terms of void-swelling suppression based on the presence of weak interactions between chromium atoms in solution and vacancies, modified by depletion of chromium from solid solution by α' precipitation at chromium contents exceeding 10%. The validation of the high swelling resistance of the 12% Cr martensitic stainless steels in a fast reactor environment provides confidence in the selection of these alloys as alternative core component materials for commercial fast reactor systems.

Acoustic resonances as a means of blockage detection in sodium-cooled fast reactors

This paper outlines the basis of a system giving early warning of the presence and build up of a local blockage in sodium-cooled fast reactor sub-assemblies. The method uses acoustic resonances and reports on the analysis of signals recorded during special experiments on the Dounreay Fast Reactor. These tests proved the detectability of the standing waves and showed theoretical predictions of their frequency and shift in the presence of a blockage to be in good agreement with measurements.

Oxygen potential measurements on irradiated oxide nuclear fuel

A method has been developed whereby direct data on the radial variation of oxygen potential in an oxide fuel pin can be determined. The method uses a miniature solid state galvanic cell which enables emf measurements to be made on micro samples extracted from different radial positions in the fuel pin. Measurements have been made on a pin irradiated in the Dounreay Fast Reactor to ~6 % burn-up. The oxygen gradient deduced from the experimental results was smaller than that predicted theoretically although the mean O/M was within the manufacturing tolerances of the fuel pin. During the testing of the galvanic cell, new data were obtained for the relationship between oxygen to metal ratio and oxygen potential for hypostoichiometric uranium, plutonium oxide. Some anomalous results from these experiments, arising from inhomogeneities in the sample, have been interpreted as an explanation for discrepancies in the published data for this system.

The structure of fast-reactor irradiated solution-treated AISI type 316 steel

Transmission electron microscopy has been used to study the damage structure of solution-annealed AISI type 316 steel irradiated in the Dounreay Fast Reactor at measured temperatures in the range 430—740°C to displacement doses of 6 and 16 dpa ( N 2 ). The structures observed are complex and are more sensitive to temperature than to dose. The dislocation and void components have been characterised quantitatively. The maximum temperature for void stability is about 600° C and the peak swelling temperature, which cannot be defined precisely from these experiments, is thought to lie between 400 and 450°C. These findings agree with published data from other unstressed reactor irradiations of this composition of steel but not with observations made on fast-reactor fuel-pin cladding. The results are compared with the damage structures generated in steel of the same cast exposed to heavy ion and electron bombardment in accelerated voidage experiments.

The effect of the neutron energy spectrum on neutron displacement damage in niobium

This paper presents an analysis of displacement damage produced in niobium, which has potential use as the first wall material of a fusion reactor.Lindhard's formalism was used together with cross sections taken from the UK nuclear data file on niobium. Data sets describing damage energy created at various neutron incident energies were generated. It was calculated that: The ratio of damage density (ie mean damage energy × cross section) at 14 MeV to that at 1 MeV is 3·5 to 1·00.The damage energy data sets were integrated over two different neutron spectra in order to compare the total effects. The spectra were taken from: (a) the first wall of a model fusion reactor, with a total flux of 2·8 × 1015 neutrons/cm2 s; (b) the core centre of the Dounreay Fast Reactor (DFR), total flux 2·5 × 1015 neutrons/cm2 s.Analysis showed that the number of atomic displacements per atom per year is165 displacements for spectrum (a);78 displacements for spectrum (b).

Irradiation Creep in the Dounreay Fast Reactor

HELICAL springs, mostly from batches used in a thermal reactor irradiation1 and loaded in tension in strings of five, have been used to measure irradiation creep in three experiments in the Dounreay Fast Reactor (DFR). The specimens were immersed in sodium and creep deflexions were estimated from pre and post-irradiation X-radiographs of the rigs. The deflexion of each spring was limited by design to about 15 mm. One nickel specimen, the uppermost in the second rig (Table 1) and of larger coil radius than the rest, was cut from a spring supplied by Dr R. V. Hesketh from the batch he used in his thermal irradiation2. For this spring a strain of 10−4 corresponded to a deflexion of 2.54 mm, twice that for the others, the comparison being for twenty-five active turns.

The ( n, α) cross sections for natural nickel, iron, chromium and molybdenum in a fission neutron spectrum

A technique based on the absolute determination of helium by a combination of vacuum fusion extraction and mass spectrometry has been used to measure the ( n, α) cross sections for nickel, iron, chromium and molybdenum samples which had been irradiated in the core of the Dounreay Fast Reactor. Particular care was taken to determine the helium contribution due to impurities in the samples. The neutron flux was calculated from activation measurements on the threshold reactions 54Fe( n, p)Mn and 58Ni( n, p) 58Co in the samples. The values obtained for the cross sections averaged over a fission neutron spectrum ( σ F ) are: nickel, 4·7(±0·6) mb; iron, 0·18(±0·04) mb; chromium, 0·17(±0·06) mb and molybdenum, 0·13(±0·03) mb.

High temperature fast-neutron irradiation of pyrolytic-carbon-coated ThO 2 microspheres

We determined the effects of fast-neutron damage alone on four types of pyrolytic-carbon coatings with spherical geometry. The coatings, which ranged in density from 1.75 to 2.7 g/cm 3 and had Bacon anisotropy factors from 1.0 to 1.4, were irradiated in the Dounreay Fast Reactor to 4.5 × 10 21 neutrons/cm 2 (> 0.18 MeV) at about 1200°C. A porous buffer layer prevented interaction between the coating and the sol-gel ThO 2 kernels. Essentially no fissioning occurred in the ThO 2 in the fast-neutron spectrum. We found that only the coatings with the highest anisotropy factor failed. Coatings with an initial density of 1.8 g/cm 3 or less densified while coatings with initial density of 1.9 g/cm 3 or greater swelled. Metallographie examination revealed that the preferred orientation increased in the coatings that densified. These changes must be considered in mathematical models describing coated-particle behavior.