Please download to get full document.

View again

of 36
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.

Fan Fiction

Publish on:

Views: 24 | Pages: 36

Extension: PDF | Download: 0

10 Y AECL-8260 ATOMIC ENERGY»2J5a L'ENERGIEATOMIQUE OF CANADA LIMITED V j ^ ^ j f DU CANADA LIMITEE IRRADIATION OF A CANDU UO 2 FUEL ELEMENT WITH TWENTY-THREE MACHINED SLITS CUT THROUGH THE ZIRCALOY SHEATH Irradiation dun element combustible CANDU (UO 2 ) dont la gaine de Zircaloy est transpersee de 23 fentes usinees R.L.daSILVA Chalk River Nuclear Laboratories Laboratoires nucleates de Chalk River Chalk River, Ontario Septem ber 1984 septem bre ATOMIC ENERGY OF CANADA LIMITED Irradiation of a CANOD DO2 Fuel Element With Twenty-Three Machined Slits Cut Through the Zlrcaloy Sheath by R.L. da Silva Fuels and Materials Division Chalk River Nuclear Laboratories Chalk River, Ontario, KOJ 1J September AECL-8260 L'ENERGIE ATOMIQUE DU CANADA, LIMITEE Irradiation d'un élément combustible CANDU (UO*,) dont la gaine de Zircaloy est transpersée de 23 fentes usinées. par R.L. da Silva Résumé On a délibérément endommagé un élément combustible CANDU, en exposant une surface minimale de 272 mm du combustible U0 empilé au moyen de 23 fentes longitudinales usinées dans la gaine de Zircaloy 4. L'élément a ensuite été irradié dans la boucle X-2 du réacteur NRX durant une période de 14,64 jours à pleine puissance, à un taux linéaire thermique de 48 kw/m, pour que l'on puisse étudier la relation existant entre le dégagement de produits de fission et le comportement de l'oxydation de l'uo_ dans un élément où les gaz de fission font l'objet d'une capture minimale entre le combustible et le dégagement. La libération des produits de fission, telle que mesurée par la spectroscopie gamma en direct, a montré que les gaz nobles et les iodes radioactifs provenant de la matrice du combustible UO^ vont directement dans le caloporteur au moyen d'une simple cinétique ae diffusion et que leurs diffusibilités dans l'uo. hyperstoichiométrique sont approximativement égales. L'oxydation de U0_ jusqu'aux états supérieurs UC 2 +, U 0 et U 3 0 s'est accompagnée d'un important gonflement du combustible et d'une deformation de la gaine, localisés de préférence dans l'extrémité de l'élément ayant la plus faible puissance. Le comportement de l'écaillement et de l'érosion dans les pastilles de combustible a été mis en corrélation avec le taux d'oxydation du combustible. Division des matériaux et des combustibles Laboratoires nucléaires de Chalk River Chalk River, Ontario KOJ 1J0 Septembre 1984 AECL-8260 ATOMIC ENERGY OF CANADA LIMITED Irradiation of a CANDD DO2 Fuel Elenent Kith Twenty-Three Machined Slits Cut Through the Zircaloy Sheath by R.L. da Silva ABSTRACT A CANDU fuel element was purposely defected, exposing a minimum UO2 fuel stack area of 272 mm 2, by machining 23 longitudinal slits through the Z-ircaloy-4 sheathing. The element was then irradiated in the X-2 loop of the NRX reactor for a period of effective full power days at a linear heat rating of 48 kw/m to investigate the relationship between fission product release and D0 oxidation behaviour in an element with minimal fuel-to-gap fission gas trapping. The fission product releases, as measured by on-line gamma-ray spectroscopy, revealed that the noble gases and radioiodines are both released from the UO2 fuel matrix directly to the coolant via simple diffusion kinetics, and that their diffusivities in hyperstoichiometric UO2 are approximately equal. The oxidation of UO2 to the higher states U0 2+x' U 4 9 an * U 3^8» was accompanied by substantial fuel swelling and sheath deformation preferentially located in the lower powered end of the element. The spalling and erosion behaviour of the fuel pellets was correlated to the rate of fuel oxidation. Fuels and Materials Division Chalk River Nuclear Laboratories Chalk River, Ontario, KOJ 1J September AECL-8260 Irradiation of a CANBU DO2 Fuel Element With Twenty-Three Machined Slits Cut Through the Zircaloy Sheath 1. INTRODUCTION The uniqueness of the on-power refuelling capability of the CANDU* reactor system has provided impetus to improve our understanding of fission product release behaviour from in-core failed fuel during normal operating conditions. Small sheath defects may be tolerated when weighed against burnup penalties and core physics considerations following from premature refuelling; however, early detection and prompt removal of large sheath defects is necessary to mitigate the health hazards and economic penalties associated with high radiation fields, arising from the decay of circulating and deposited fission products within the reactor's primary heat transport system. Operating decisions regarding failed fuel removal must be based on a good understanding of secondary sheath deterioration and fission product release behaviour from a wide variety of fuel defect conditions. Pursuant to this, a series of irradiation experiments in the pressurized light water coolant loops of the NRX reactor at the Chalk River Nuclear Laboratories (CRNL) on deliberately and naturally defective UO2 fuel elements, have been performed (1-5). This report documents and describes one of the more recent experiments, Exp-FFO-103, on a purposely defected element, A3N, containing 23 slits cut into its sheath and irradiated at a midplane linear power of 48 kw/m for a period of 15 effective full power days. The experiment was designed to furnish information on IK^/coolant contact behaviour in a simulated power reactor environment, and to provide data from a fuel element with minimum impediment to fission product release, i.e. where the diffusion path from the whole of the UO2 to the coolant was made as small as possible. The analysis includes a discussion of both dissolved and deposited fission products release behaviour and contains comparisons to other experimental data from previous defect tests. 2. EXPERIMENTAL DESCRIPTION 2.1 Fuel Element Details The defect test element A3N was of the standard design currently used in CANDU 600 MW(e) and 750 MW(e) power reactors modified with support fittings for use in the X-2 loop of the NRX reactor. The Zircaloy-4 fuel sheath, with a mm outside diameter and 0.43 mm wall thickness, was fabricated with 23 machined slits of average length 35.7 mm and width 0.33 mm cut longitudinally in a helical pattern about the sheaths circumference, extending down the full length of the element. The circumferential * CANadian Deuterium Uranium separation between adjacent silts was typically 12.7 ran or 111 degrees. The element contained sintered U0 2 pellets enriched to 5.02 wt% U-235 in U with a nominal density of Mg/nr. The total exposed IK 2 area was approximately 272 mm^ or 1.5% of the total fuel surface area. A detailed drawing of element A3N is given in Figure 1 while complete design details are listed in Table Loop Description The X-2 loop of the NRX reactor is designed to study activity releases from defective fuel elements. At power, the loop operates with a nominal axial averaged cell thermal flux of 5.5 x 10 n/cm.s and has a pressurized light water coolant at 10.5 MPa (260 C) with a mass flow of 1.0 kg/s. Activity releases to the coolant are measured by both a stationary gamma-ray spectrometer at the test section inlet (position Ml) and by a mobile spectrometer at the test section outlet (position M2). The coolant piping at site position M2 has four different surface trap materials which were regularly scanned throughout the experiment. A schematic of the X-2 loop facility and a simplified flow diagram are shown in Figures 2a and 2b. Loop operating conditions are listed in Table Data Acquisition System The detectors for each gamma-ray spectrometer were fabricated from a single piece of high purity germanium containing one large diode and one small diode with active volumes of ~ 2.88 x 10~* m 3 and ~ 7.50 x 10~ 7 m 3 respectively. The excellent resolving characteristics of germanium detectors for gamma-ray spectrometry analysis are described in detail elsewhere (6). Coolant activity was measured on-line using a multi-channel pulse height analyzer. During steady state reactor operation, fission product spectra were collected every 2500 seconds using 1000 second count times; during power transients counting times were shortened to 200 seconds. The spectrometer data was recorded on magnetic tape for subsequent processing. Samples of typical spectral plots measured by the two spectrometers upstream and downstream of the test section, as processed by the GRAAS code (7), are given in Figures 3a and 3b. 3. IRRADIATION POWER HISTORY During experiment Exp-FFO-103 element A3N operated at an average midplane linear power of 48 kw/m. The axial flux gradient along the element length resulted in a linear power profile ranging from 37 kw/m at the top end to 57 kw/m at the bottom end of the element as shown in Figure 4. A plot of element A3N's average midplane linear power versus irradiation time is shown in Figure 5. The nominal calculated midplane burnup at the end of the irradiation was 17.5 MW.h/kg U. A complete summary of the power history of element A3N is given in Table 3. 4. POST-IRRADIATION EXAMINATION 4.1 Visual Inspection and Neutron Radiography Element A3N was visually examined and neutron radiographed prior to destructive examination. Neutron radiographs of the defect test element, A3N, and an intact filler element, A2D, are shown together in Figure 6*. The contrast in radiographs clearly reveals the extent of UO2 fuel erosion and oxidation experienced by element A3N over the 15-day irradiation. Erosion was greatest in the upper and lower regions of the fuel stack while indications of oxidation sintering is observed throughout. Visual studies of the element revealad a consistent pattern of localized plastic sheath deformation resulting from UO2 oxidation swelling that was axially dependent (refer to Figure 7b). Slit widths at the low powered top end of the element had increased by about a factor of ten over the as-measured pre-irradiation dimensions (0.3- -3.0 mm), while slits in the higher powered middle and bottom of the element increased by about factors of six and three respectively. In total, the defect area increased to an estimated cm^; a factor of 5.5 greater than the initial defect size. Unlike the observed axial dependence of U0 2 oxidation swelling, a uniformly distributed ZrO2 film 2-4 ym thick was observed on the sheath outer surface. Although parts of the sheath were excessively deformed there was no evidence of localized sheath hydriding on the outer surface of the element. 4.2 Metallographic Examination Destructive examination of element A3N included four transverse metallographic cross-sectional cuts through the fuel; one each from the top and bottom end of the element and two from the middle. A transverse sectional cut from the bottom end plug, three measurements of the hydrogen concentration in the fuel sheath, and a gravimetric analysis of the oxygento-metal ratio (0/U) of a fuel specimen from below the fuel centreline were also made. The photomacrographs of three of the four transverse sections of A3N are shown in Figure 7 along with a post-irradiation photograph of the element. All three photomacrographs reveal the presence of columnar grain growth, pellet cracking, significant oxidation and fuel erosion. In addition, the sections below the fuel midplane show evidence of void formation and central melting. The heaviest oxidation and fuel erosion (maximum sheath deformation) was found in pellets from the top half of the fuel. Composites of several oxide phases were present but most were indistinguishable. Some evidence of t^og was seen in the fuel from the upper half while the characteristic Widmanstatten pattern of U4O9 precipitate was observed only in fuel from the bottom half of the element. * Two intact filler elements were irradiated in a trefoil assembly along with element A3N to increase the efficiency of the loop calorimetric measurements. The 0/U measurement taken just below the fuel midplane gave a value of Grain sizes measured ranged from 6-7 {jo at the outer fuel surfaces to ym 0.5 mm in from the fuel periphery. Most of the peripheral oxide phases extended around 70% of the pellets circumference and penetrated as far as 1 on along radial crack surfaces. Intergranular corrosion and U0 pitting was heaviest in the pellet volumes opposite sheath slits. The amount of fuel loss due to coolant erosion, was estimated to be ~65 g (about 12Z of initial U0 2 weight) based on the neutron radiographs and metallographic cross sections. All sheath sections examined were lightly hydrided ( yg/g) with a slight concentration gradient extending down the length of the element. No regions of localized hydride concentrations) were found on any of the interior surfaces of the sheath. 4.3 Discussion of Post-Irradiation Examinations One of the more interesting observations arising from the metallographic examinations of element A3N is the difference between the extent of UO2 oxidation swelling, element deformation and fuel erosion in the top half of the element to that observed in the higher powered bottom half. The exact reason(s) for this are unknown but recent U0 oxidation experiments (in air) by McCracken and Woo (8) provide a possible explanation for this temperature-oxidation dependence. Their results show that as U0 2 oxidizes to the higher states (U4O9, U3O7 and U3O3) at temperatures 600 C the resulting volume expansion to U3O3 generates sufficient stress to produce fuel cracking along grain boundaries resulting in the spalling of grain sized U3O3 particles. Spalled material could easily be removed by the action of the coolant passing over the fuel in the vicinity of the slits. At higher temperatures ( C) the U 3 Og started to sinter and formed a thin surface layer over a much thicker layer of U4O9. It is postulated that this surface layer of U3O3 delays the onset of boundary oxidation and thus slows down the rate of oxidation. They also noted that the oxide phase of U3O3 formed at temperatures 800. C, was more adherent to the UO2 substrate than formed at temperatures of C. Any spalling at the higher temperatures was associated with greater particle size. Note ^3 8 was tentatively identified only in pellets above the fuel midplane while U4O9 was only observed in fuel in the bottom half. Although formation is not generally predicted for the coolant conditions existing during this test, Imoto et al. have also recently reported observing formation in fuel from defective UO2 fuel elements irradiated in the Mihama-1 power reactor (9). They calculated it was possible to form if temperatures were between 1080 and 1290 C given the proper oxygen potential. From the metallographic examination, it is difficult to provide estimates of fuel surface temperatures, especially along crack faces, due to the widely varying UO2 stoichiometry throughout the fuel stack. However, in view of the results of McCracken and Woo, the general trend of increasing temperatures and decreasing oxidation and fuel loss with element length must be ascribed to the faster formation of higher oxides in the cooler half, relative to the hotter half. The lack of any secondary sheath damage due to hydriding is consistent with previous experiments on artifically defected fuel irradiated at CRNL (2-4). This hole size dependence on secondary sheath hydriding has been previously attributed to a critical H2/O2 ratio required to promote hydrogen diffusion into the Zircaloy sheath surface (10). The critical H2/O2 ratio is dependent on the initial defect size; if the hole size is large enough to enable a replenishing of the oxidant needed to provide a protective oxide barrier on the sheath's inner surface, localized hydride formation is inhibited. Apparently in this test the size and number of defects did not allow oxygen starvation to occur. 5. FISSION PRODUCT RELEASE: RESULTS AND DISCUSSION During Exp-FFO-103, element A3N operated at constant power between 03:42 hours June 01 through to 14:00 hours 1981 June 14. The initial reactor transients (see Figure 5) occurred too early in the experiment to allow for any rigorous treatment of the data. Fission products released from the element were transported by the loop coolant, past the spectrometer locations (M2 and Ml) either in dissolved form or as particulates. In addition, certain individual fission product species plated out along the piping surfaces. Any particulates within the coolant were filtered out by graphite membrane filters downstream of the spectrometer at position M2. The filter efficiency for particle sizes 5 yra is estimated to be ~ 97%. Coolant concentrations in Bq/kg.J^O* for the dissolved species and concentrations in Bq/cm^ for the depositing species were calculated using the SUMRT program. The program has previously been described in reference 3. The analyses of the individual dissolved and depositing fission product species are treated separately in this report; each will be described and discussed in the following sections under different headings. 5.1 Noble Gases and Radioiodines The activity concentrations of the nobxe gases (Kr-85m, Kr-87, Kr-88, Xe-133, Xe-135 and Xe-138) and radioiodines (1-131, 1-132, 1-133, and 1-135) measured during the experiment are presented in Figures 8, 9 and 10. Loop degassing and ion-exchange purification during the periods June 09 (10:00-18:00 hours) and June 14 (06:00-24:00 hours) had a noticeable effect on the measured concentrations. Due to the large amount of UO2 surface area exposed to the loop coolant, traditional models for calculating isotopic fractional releases (3-5) which assumed fission product trapping in the fuel-to-sheath gap were considered inappropriate and a new generalized approach was developed. The method is described in detail below. * 1.0 Bq = 2.7 x 10-5 5.1.1 Release Model Description The governing mass balance equation for the production and release to coolant of a given isotope can be written as, or dn f (t) -B(t) -R(t) - XjL(t) [1] dt~ * dn f (t) + ta t) = B(t) [ 1 - % .\ [2] dt~ r where Nf(t) = total number of atoms in fuel at time t, (atoms) A s decay constant of isotope (s~^) R(t) = release rate of isotope from fuel (atoms/s) B(t) = average birth rate of isotope in fuel (atoms/s) R(t).. = isotope fractional release from fuel at time t t = irradiation time (s) The birth rate of an isotope exposed to a constant neutron flux may be written as*, B(t) = F R yc (1 - (1-0) e Yt ) [3] where F R = average fission rate in fuel (fissions/s) + yc = cumulative isotope fission yield (atoms/fission) 0 = ratio of the isotopes direct fission yield to its cumulative fission yield Y = precursor growth constant (s l) For the isotopes of interest in this report J«1, thus equation 3 simplifies to, B(t) = F R yc (1 - e Yt ) [4] Values for Y were calculated numerically using data generated with the CRNL physics code FISSPROD (11). The method is described in detail in Appendix A and the results for a few of the longer lived isotopes and/or those with long lived parents are given below: Isotope Xe Decay Constant A (s 1 ) 1.53 x 10~ x 10~ x 10 5 Precursor Constant V (s 1 ) 8,40 x 10~ x 10~ x 10 6 * See discussion in Appendix A. + Assuming negligible fuel loss from element The above values of Yreflect the direct dependence of the isotope's growth rate on parent(s) half life and branching fractions. For example, the isotope has a single parent Te-132 ( A* 2.46 x 10~ 6 s 1 ), the value of Y calculated for is within 1% of the Te-132 decay constant. The isotope Xe-133 on the other hand has a complex parent decay process involving branching fractions from three different isotopes; ( X= 9.21 x 10~ 6 s 1 ), I-133m ( A = 7.70 x 1(T 2 s 1 ) and Xe-133m ( *= 3.66 x 10 6 s 1 ). The Y calculated for Xe-133 (8.40 x 10 s 1 ) thus represents an effective single parent decay constant which accounts for the various fission yield contributions from the ~ decay chains of the 3 parent isotopes. Substitution
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks