Investigations on swelling and fission gas behaviour in uranium dioxide

UO₂ irradiated at temperatures between 1000 and 2100 K was investigated with respect to fission gas behaviour and swelling. The amount of fission gas was measured in three steps as released fission gas, fission gas retained in bubbles and pores, and fission gas in the fuel matrix. The retained fission gas reaches concentrations up to $\text{1.6 × 10}{}^{\mathrm{?2}}$ gas atoms per uranium atom at temperatures below 1250 K and decreases with increasing temperature. The swelling was evaluated by measuring the volume changes and by immersion density measurements. The maximum fission gas swelling without extensive bubble migration is about 20% at 2000 K. It diminishes to about 5% at 1250 K.     

A microstructure-dependent model for fission product gas release and swelling in UO2 fuel

A model for the release of fission gas from irradiated UO₂ fuel is presented. It incorporates the relevant physical processes: fission gas diffusion, bubble and grain boundary movement, intergranular bubble formation and interlinkage. In addition, the model allows estimates of the extent of structural change and fuel swelling. In the latter, contributions of thermal expansion, densification, solid fission products, and gas bubbles are considered. When included in the ELESIM fuel performance code, the model yields predictions which are in good agreement with data from UO₂ fuel elements irradiated over a range of water-cooled reactor conditions: linear power outputs between 40 and 120 kW m⁻¹, burnups between 10 and 300 MW h(kg U)⁻¹, and power histories including constant, high-to-low and low-to-high power periods.The predictions of the model are shown to be most sensitive to fuel power (temperature), the choice of diffusion coefficient for fission gas in UO₂, and burnup. The predictions are less sensitive to variables such as fuel restraint, initial grain size and the rate of grain growth.     

The characteristics of fission gas release from monocrystalline uranium dioxide during irradiation

Release rates for ^85mKr, ⁸⁷Kr, ⁸⁸Kr, ¹³³Xe, ¹³⁵xe and ¹³⁸Xe were measured in the temperature range 700–1550°C. The data were analysed in terms of diffusion of the rare gases and their halogen precursors. The diffusion coefficients for xenon and iodine were found to be similar whilst krypton also had a similar mobility at ~1200°C but otherwise diffused more slowly. Bromine had a high mobility compared with the rare gases (X 200).     

Predicting the probability for fission gas resolution into uranium dioxide

Classical molecular dynamics simulations, using a set of previously established pair potentials, have been used to predict the minimum energy needed for krypton and xenon atoms to be resolved into uranium dioxide across a perfect (1 1 1) surface. The absolute minimum energy, E_min, is 53 eV for krypton and 56 eV for xenon atoms, significantly less than the 300 eV value often assumed in fuel modelling as the minimum energy required for gas resolution. The present values are, however, still sufficient to preclude thermal resolution at normal reactor temperatures. The discrepancies between the present and previous resolution energies are due to the significant variation in probabilities of absorption at different impact points on the crystal surface; we have mapped out the probability distribution for various impact sites across the crystal surface. The value of 300 eV corresponds to an 85% chance of resolution.     

Effect of non-equilibrium fission gas and fuel creep on swelling and release in irradiated carbide fuels

The code UCSWELL was developed to simulate fission gas behavior in carbide fuels. In the present work, one of the limiting assumptions in UCSWELL - that matrix gas bubbles are in equilibrium with gas atom concentration - is removed and non-equilibrium matrix fission gas bubbles are allowed, but with relaxation to equilibrium by means of vacancy diffusion and thermal and radiation-induced creep of the fuel. For a given grain size, the difference in swelling between equilibrium and non-equilibrium with relaxation bubble fission gas treatment increases with decreasing irradiation temperature. At a given temperature, the non-equilibrium effect is more pronounced for larger grain fuel. This is to be expected because the creep rate (and hence the rate at which bubbles grow to an equilibrium size) decreases as temperature decreases and/or as grain size increases. At temperatures, where the creep rate is grain size insensitive, grain size remains important to the equilibrium process in so far as the grain boundary is a source of vacancies to the non-equilibrium bubbles. While the difference in these quantities is at the most on the order of 20% for the steady operating conditions considered, it is anticipated that the non-equilibrium effects become more pronounced during reactor overpower and undercooling transients.     

Radiation re-solution of fission gas in uranium dioxide and carbide

Calculations have been performed to estimate the removal rate of fission gas atoms from bubbles due to collisions with energetic fission fragments and recoil cascades. The efficiency of this process was found to be higher than estimated earlier, but is still too low to be responsible for the experimental observations of fission gas bubble destruction during irradiation of oxide fuel. An irradiation experiment to investigate the interaction of fission spikes with free surfaces has enabled a simple theory to be developed which can explain the shrinkage of bubbles and pores by the surface relaxation of a shock wave produced by the passage of a fission fragment. This mechanism occurs in oxides but not carbides because of the faster dispersion of the fission fragment energy and provides the major reason for the difference in gas bubble distributions in oxide and carbide fuel. This process, however, does not remove gas atoms from the bubbles. Since high levels of apparently diffusive fission gas release are observed in oxides, the “effective solubility” of the fission gases required for this release must be sought in phenomena other than the fission spike.     

A comparison between fission gas release data and FEMAXI-IV code calculations

The FEMAXI-IV code is an extension of the earlier version FEMAXI-III. The primary improvement in the new version is the provision for treating the fuel rod behavior during an operational transient. For this purpose, the time-dependent models are used for heat conduction, fission gas release, and mixing of the released gas with the plenum gas.In FEMAXI-IV, the fission gas release model was thoroughly revised from the previous version. It is based on the fission gas release model presented by White and Tucker. The model takes into account the following mechanisms:

• - diffusion of gas atoms to the grain boundary;

• - sweeping of gas atoms by grain growth;

• - precipitation of gas atoms into intragranular gas bubbles;

• - resolution of gas atoms from intragranular and grain boundary gas bubbles;

• - fission gas release due to bubble interconnection.

The model was incorporated into FEMAXI-IV and code calculations were compared with the fission gas release data obtained in the Inter-Ramp and Over-Ramp experiments.This paper describes the fission gas release model involved and results of calculations.     

Crystal structure stability and fission gas swelling in intermetallic uranium compounds

A correlation between fission-induced amorphization and the behavior of fission gasses in intermetallic uranium compounds is proposed. Changes in fission-gas mobility and in the plastic flow rate of the fuel, that result from such a transformation, are likely to be responsible for the large fission-gas-driven swelling-rate increases that have been observed in certain fuels. This study attempts to assess the propensity for amorphization of intermetallic compounds as potential test-reactor fuels by means of their thermodynamic properties. The compound U₆Mn was thus identified as a probable candidate for a stable high-density intermetallic fuel to be used in dispersion-fuel elements.     

Modeling the diffusion yield of radioactive fission products from uranium dioxide fuel

The numerical aspects of a systematic solution of the problem of diffusion and yield of radioactive fission products from a homogeneous sphere, simulating a uranium dioxide fuel kernel, with realistic boundary conditions are discussed. The numerical scheme is based on a one-group method of calculating the production and radioactive mutual transmutations of fission products in combination with the standard expansion of the radial dependence of the concentration in terms of the eigenfunctions of the Laplace operator. It is demonstrated on illustrative examples that the approach is highly economical.     

The migration of intragranular fission gas bubbles in irradiated uranium dioxide

The mobility of intragranular fission gas bubbles in uranium dioxide, irradiated at 1600–1800°C, has been studied following isothermal annealing at temperatures below 1600°C. The intragranular fission gas bubbles, average diameter approximately 2 nm, are virtually immobile at temperatures below 1500°C. The bubbles have clean surfaces with no solid fission product contamination and are faceted to the highest observed irradiation temperature of 1800°C. This bubble faceting is believed to be a major cause of bubble immobility. In fuel operating below 1500°C the predominant mechanism allowing the growth of intergranular bubbles and the subsequent gas release must be the diffusion of dissolved gas atoms rather than the movement of entire intragranular bubbles.     

Dynamics of fission gas release from molten nuclear fuel

A model of gas release from molten nuclear fuel has been developed taking into account motion of bubbles and different physical processes leading to the coagulation of bubbles. It is shown that the fuel swelling and gas release dynamics are governed by the external pressure, geometry of the melt, initial gas concentration and properties of the liquid materials.     

Simulation of high-temperature swelling of uranium dioxide and deformation behavior of a fuel element

Models and computer codes, developed based on them, for simulating the swelling of uranium dioxide (BARS) and the stress-deformation state of a fuel element (SDS) under high-temperature irradiation are presented. It is shown that when developing a design for high-temperature fuel elements and validating their serviceability the quantitative indicator required for the swelling of uranium dioxide in the range ≥1400°C is the change in the external dimensions of the fuel caused by constant formation and growth of bubbles containing gaseous fission products during irradiation. The results of computational investigations using the models indicated are examined. These results eliminate the inconsistency of the data on the effect of the main operating parameters — the temperature and burnup — on the radiation characteristics and service life behavior of a fuel element. It is shown that the central channel in the fuel kernel and strengthening of the cladding improve the dimensional stability fuel elements. __________ Translated from Atomnaya énergiya, Vol. 103, No. 3, pp. 172–179, September, 2007.     

Modeling of fission gas swelling in LWR UO2 fuel under irradiation

An understanding of the behavior of fission gas in uranium dioxide (UO₂) fuel is necessary for the prediction of the performance of fuel rods under irradiation. A mechanistic model for matrix swelling by the fission gas in LWR UO₂ fuel is presented. The model takes into account intragranular and intergranular fission gas bubbles behavior as a function of irradiation time, temperature, fission rate and burn-up. The intragranular bubbles are assumed to be nucleated along the track of fission fragments, which play the dual role of creator and destroyer of intragranular bubbles. The intergranular bubble nuclei is produced until such time that a gas atom is more likely to be captured by an existing nucleus than to meet another gas atom and form a new nucleus. The capability of this model was validated by a comparison with the measured data of fission gas behavior such as intragranular bubble size, bubble density and total fuel swelling. It was found that the calculated intragranular bubble size and density are in reasonable agreement with the measured results in a broad range of average fuel burn-ups 6–83 GW d/tU. Especially, the model correctly predicts the fuel swelling up to a burn-up of about 70 GW d/tU.     

Rapid aqueous release of fission products from high burn-up LWR fuel: Experimental results and correlations with fission gas release

Studies of the rapid aqueous release of fission products from UO₂ and MOX fuel are of interest for the assessment of the safety of geological disposal of spent fuel, because of the associated potential contribution to dose in radiological safety assessment. Studies have shown that correlations between fission gas release (FGR) and the fraction rapidly leached of various long-lived fission products can provide a useful method to obtain some of this information. Previously, these studies have been limited largely to fuel with burn-up values below 50 MWd/kg U. Collaborative studies involving SKB, Studsvik, Nagra and PSI have provided new data on short-term release of ¹³⁷Cs and ¹²⁹I for a number of fuels irradiated to burn-ups of 50–75 MWd/kgU. In addition a method for analysis of leaching solutions for ⁷⁹Se was developed. The results of the studies show that the fractional release of ¹³⁷Cs is usually much lower than the FGR covering the entire range of burn-ups studied. Fractional ¹²⁹I releases are somewhat larger, but only in cases in which the fuel was forcibly extracted from the cladding. Despite the expected high degree of segregation of fission gas (and by association ¹³⁷Cs and ¹²⁹I) in the high burn-up rim, no evidence was found for a significant contribution to release from the rim region. The method for ⁷⁹Se analysis developed did not permit its detection. Nonetheless, based on the detection limit, the results suggest that ⁷⁹Se is not preferentially leached from spent fuel.     

A methodology to predict a fission gas release ratio of MOX fuel with heterogeneous microstructure

The objective of this study is to formulate a methodology to predict a fission gas release ratio of MIMAS MOX. An irradiated MIMAS MOX fuel with plutonium rich agglomerates was subjected to elemental analyses by electron probe micro analysis and secondary ion mass spectrometry in order to investigate xenon distribution. The results of the elemental analyses showed that the plutonium rich agglomerates at the periphery of the fuel pellet sample retained a high concentration of xenon as gas bubbles. Then, the results were used as reference data for modification of models in a fuel rod analysis code, FEMAXI-7. Using the modified FEMAXI-7, we applied an approach to prediction of fission gas release ratio of MOX fuel with plutonium rich agglomerates. In the approach, two separated analyses using FEMAXI-7 were performed for the plutonium rich agglomerates and the matrix. Fission gas release ratios obtained from the two analyses were processed through weighted-average with burnup ratios of the plutonium rich agglomerates and the matrix. Finally, the fission gas release ratios were compared with results of rod puncture tests. As a result of the comparison, it was confirmed that the proposed approach could well predict fission gas release ratio of MOX fuel with plutonium rich agglomerates.     

THETRIS: A micro-scale temperature and gas release model for TRISO fuel

The dominating mechanism in the passive safety of gas-cooled, graphite-moderated, high-temperature reactors (HTRs) is the Doppler feedback effect. These reactor designs are fueled with submillimeter-sized kernels formed into tristructural-isotropic (TRISO) particles that are imbedded in a graphite matrix. The best spatial and temporal representation of the feedback effect is obtained from an accurate approximation of the fuel temperature. Micro-scale models of TRISO particles are necessary in order to obtain accurate predictions during fast transients or when parameters internal to the TRISO are needed. Most accident scenarios in HTRs are characterized by large time constants and slow changes in the fuel and moderator temperature fields. In these situations, a meso-scale, or pebble- and compact-scale, solution provides a good approximation of the fuel temperature as the fission thermal energy transports out of the kernel and into the surrounding matrix with a much shorter time constant. Therefore, in most cases, the matrix can be assumed to be in quasi-static equilibrium with the kernels. These models, however, fail to provide accurate information on the state of the various components of the TRISO during the early stages of transients. Since the coated particles constitute one of the fundamental design barriers for the release of fission products, it becomes important to understand the transient behavior inside this containment system. An explicit TRISO fuel temperature model named THETRIS has been developed and incorporated into the CYNOD–THERMIX-KONVEK suite of coupled codes. The code includes gas-release models that provide a simple predictive capability of the internal pressure during transients. The new model yields similar results to those obtained with other micro-scale fuel models of TRISO particles, but with the added capability to analyze gas release, internal pressure buildup, and effects of a gap in the TRISO. Analysis of bounding benchmark transients yield good agreement with other codes in which the TRISO particles are modeled explicitly. In addition, a sensitivity study of the potential effects on the transient behavior of high-temperature reactors due to the presence of an inter-layer gap is included. Although the formation of a gap occurs under special conditions, its consequences on the dynamic behavior of the reactor can yield responses during fast transients that depart significantly from those in which no gap is present in the model. The new model was applied to an extreme (beyond design basis) scenario in order to observe the behavior of the fuel during a large prompt critical reactivity insertion. Although a large amount of fission energy was deposited rapidly into the fuel, the kernel temperature is shown to stay well below the melting point and the silicon carbide layer remained well below the temperature above which failure is expected to occur. The explicit treatment of the TRISO particle geometry leads to much lower estimations of power peaking during the transient and a greater degree of negative Doppler feedback.     

On the use of grain boundary loss terms in fission gas release and swelling models

An important feature of all physical models of fission gas release and swelling is the calculation of the gas deposition rate at the boundaries of the fuel grains. To accomplish this for the diffusional flow of gas to the boundaries, within the rate theory model of the lossy continuum, it is necessary to derive and apply a suitable grain boundary loss term. In the present study we investigate the use of several such terms derived by applying various levels of approximation. It is concluded that a grain boundary loss term originally derived by Speight gives reasonable accuracy under steady-state irradiation conditions. However, if, as must be the case during reactor operation, these conditions change during irradiation, a numerical solution to the diffusion problem must be sought.