AECL
SIMULATING THE BEHAVIOUR OF ZIRCONIUM-ALLOY COMPONENTS IN NUCLEAR REACTORS
(Presented at the ASTM Symposium on Zirconium in the Nuclear Industry, Annecy, France, 2001 June, on receipt of the Kroll Medal for 1998)
bY
C.E. Coleman
Fuel Channels Division Chalk River Laboratories Chalk River, Ontario KOJ 1JO Canada
200 1 December
AECL-12132 EACL
SIMULATIONDUCOMPORTEMENTDESCOMPOSANTSD'ALLIAGE DE ZIRCONIUMDANSDE~R~ACTEURSNUCLI~AIRES
(prkentation faite dans le cadre du symposium de la ASTM sur l’utilisation du zirconium dans I’industrie nuchire tenu 5 Annecy, France, en juin 2001 lors de la remise de la mkdaille de Kroll pour 1998)
Par
C. E. Coleman
Resume
Pour empecher la defaillance de composants nucleaires, il est necessaire de comprendre les interactions entre les materiaux environnants et les changements de leurs proprietes physiques & toutes les &apes d’exploitation des reactems. Trois exemples relatifs a des reacteurs CANDU illustrent l’utilisation de simulations de situations compliqutes likes a des reacteurs, a savoir :
des essais de gonflement qui ont permis d’elaborer une methode pour augmenter la tolerance de la gaine de combustible en zircaloy sous des conditions de variation continue de puissance; des observations du comportement des fissurations donnant lieu a des fuites dans des tubes de force d’alliage Zr-Nb 2,5 % qui instaurent une confiance dans l’utilisation de la methode de fuite avant rupture comme moyen de defense contre l’apparition de defauts; - des essais d’echauffement par contact relatifs aux modifications des surfaces des tubes de cuve en zircaloy qui augmentent la capacite du moderateur a eau lourde d’agir comme une source froide a la suite d’un accident hypothetique de perte de caloporteur.
MOTS CL&3 : simulation, alliages de zirconium, gaine de combustible, tubes de force, tubes de cuve, essais de gonflement de la game de combustible, variation de puissance, fuite avant rupture, CRACLE, conductance par contact, modification de surface, source froide. Canaux de combustible Laboratoires de Chalk River Chalk River (Ontario) KOJ 1JO Canada
Decembre 200 1
AECG12132 AECL
SIMULATING THE BEHAVIOUR OF ZIRCONIUM-ALLOY COMPONENTS IN NUCLEAR REACTORS
(Presented at the ASTM Symposium on Zirconium in the Nuclear Industry, Annecy, France, 2001 June, on receipt of the Kroll Medal for 1998)
C.E. Coleman
Abstract
To prevent failure in nuclear components one needs to understand the interactions between adjacent materials and the changes in their physical properties during all phases of reactor operation. Three examples from CANDU reactors are described to illustrate the use of simulations that imitate complicated reactor situations. These are:
swelling tests that led to a method for increasing the tolerance of Zircaloy fuel cladding to power ramps, observations of the behaviour of leaking cracks in Zr-2SNb pressure tubes that provide confidence in the use of leak-before-break as part of the defence against flaw development, and contact boiling tests on modifications to the surfaces of Zircaloy calandria tubes that enhance the ability of the heavy water moderator to act as a heat sink after a postulated loss-of-coolant accident.
KEYWORDS: simulation, zirconium alloys, fuel cladding, pressure tubes, calandria tubes, fuel swelling test, power ramp, leak-before-break, CRACLE, contact conductance, surface modification, heat sink.
Fuel Channels Division Chalk River Laboratories Chalk River, Ontario KOJ 1JO Canada
200 1 December
AECL-12132 I. INTRODUCTION
The 1998 Kroll Medal was awarded for the “contribution to the evaluation of key properties for use of zirconium alloys in power reactors which have been used for lifetime assessments and improvements of fuel sheathing and pressure and calandria tubes. Included are studies of in-reactor deformation and ductility, delayed hydride cracking, pellet-cladding interaction, and principles for leak-before-break defence of CANDU fuel channels.” I am grateful to my nominators and the Kroll and ASTM B10.02 Committees for thinking me worthy to follow in the footsteps of the father of Zircaloy [l] and the mother of Zr-Nb alloys [2]. One of the traditions of this award is “to collect first-hand accounts on key aspects of the history of the application of zirconium and its alloys” and the request was made to outline “the pertinent aspects of lifetime assessments of zirconium alloy components in power reactors.” Rather than describe techniques [3] and the valuable results of post-irradiation examination of components removed from power and research reactors [4,5], three simulations have been chosen to represent the types of insights that can be obtained on three components of a CANDU reactor:
- to increase the tolerance of fuel cladding (sheathing) to power increases, - to defend leak-before-break in pressure tubes, and - to enhance safety margins through modifications to calandria tubes.
Simulations may take many forms, from mathematical and computer analyses to full- scale models. Irradiating in a high flux reactor [6,7] or with charged particles [8,9,10] simulate power reactor irradiations. The type of simulation we will consider here are small-scale imitations that capture many of the essential features of the conditions the component may encounter. Since the examples are all from the CANDU system, a brief description is given as background. 2. CANDU REACTOR
A CANDU nuclear reactor, (Fig. l), consists of a large tank, which holds heavy-water moderator at about 7O”C, penetrated horizontally by about 400 pairs of concentric tubes forming the fuel channel. The inner tube of the fuel channel, the Zr-2SNb pressure tube, contains 12 or 13 fuel bundles and the hot heat-transport fluid while the outer tube, the Zircaloy-2 calandria tube, is separated from the hot pressure tube by spacers. The calandria tube isolates the hot pressure tube from the cool moderator, insulation being provided by filling the annular space between the two tubes with dry CO2 at atmospheric pressure. The 500 mm long fuel bundle, (Fig. 2), comprises a number of elements, up to 37, held together by welding at end plates. A fuel element consists of a cylindrical Zircaloy tube (cladding or sheathing) containing UOZ. fuel pellets. Fuelling machines that can attach to the ends of the fuel channels allow refuelling on-power. 3
3. AVOIDING FUEL DEFECTS
In the late ‘60’s it was noticed that the failure rate of fuel bundles in Douglas Point, although very low (
During irradiation, radial cracks form in the outer annulus of the UOZ because of thermal stresses. The cladding and fuel touch from either expansion of the fuel pellet or collapse of the cladding. When the bundle power is increased, the fuel tends to expand radially thus separating the outer segments of fuel. If there is no slippage between the cladding and the fuel, the stress and the deformation in the cladding will be concentrated over the cracks in the fuel. The cladding will appear to have low ductility because only a small part of the tube has been deformed. If there is sliding between fuel and cladding, the strain in the sheath will be less concentrated since the stress will be spread over a greater area; the ductility will appear greater as more material is being deformed. Analysis [13] of this behaviour shows that the strain concentration, E,,,&zav, in the cladding at the point of plastic instability is given by:
where E),,,~~= maximum strain in the cladding E 53” = strain that would result if the stress was not concentrated over the cracks = coefficient of friction between the fuel and the cladding LJ = number of radial cracks in the fuel adjacent to the cladding, and m = work hardening coefficient of the cladding material.
To minimise the strain concentration, and therefore premature fracture, we deduce from’ equation (1) that there should be many cracks in the fuel, the work hardening coefficient of the cladding material should be high and the coefficient of friction between cladding and fuel should be low.
A simple swelling test was devised to test the implications of this analysis (Fig. 3). The early work was done by C.D. Williams (later of General Electric) and was extended to include stress corrosion by iodine and caesium [ 141. Similar techniques have been used on cladding for light-water reactor fuel [ 15,161. The core is aluminum, to represent the hot centre of the fuel, that is surrounded by an annulus of either UO2 or A1203, to 4
represent the outer part of the fuel, which fits inside a short section of fuel cladding. To imitate initial irradiation the soft aluminum is compressed; it expands radially and cracks the brittle ceramic annulus. Power increases are simulated by further compression that presses the cracked ceramic radially against the cladding, deforming it to rupture. To quantify the effect of the number of cracks in the fuel the ceramic annulus is replaced by segments of steel. The total circumferential strain is estimated from either changes in diameter or circumference and the strain concentration is estimated from measurements of local wall thickness variation and the average wall thickness change.
The model of fuel behaviour during a power increase was confirmed. Strain concentrated in the cladding over the “fuel” cracks (Fig. 4), and compared with a burst test the total elongation was much reduced; for example, with cold-worked cladding the total elongation was reduced from 9% to 4%. Increasing the work-hardening of the cladding by heat-treatment reduced the stress concentration whereas irradiation damage increased it. The total elongation of the cladding was increased as a consequence of the strain concentration being inversely related to the number of cracks in the outer annulus of the “fuel” (Fig.5). Application of graphite or siloxane to the inside surface of the cladding reduced the coefficient of friction between the cladding and the “fuel” and hence decreased the strain concentration in the cladding and increased the ductility, Table 1. The latter method of reducing strain concentration was thought to be a simple way of overcoming power ramp defects. This opinion was confirmed when full-size elements containing the graphite or siloxane layer survived large power increases after a period at low power in a research reactor whereas unprotected elements, made with the same batch of cladding, failed in the same irradiation. Further testing of the graphite modification, including post-irradiation examination, detected no adverse effects [ 171. This simple modification to the surface of the fuel cladding was called CANLUB (CANada LUBrication). The graphite modification was chosen as the standard for CANDU fuel in 1972. In the past 29 years over 1.5 million bundles containing graphite coatings have produced power in Canada with the failure rate being very low; for example, in the two years 1997 and 1998 the defect rate in fuel from all causes was about 0.01% with none being attributed to power ramping [ 181.
Although the test simulates a mechanical model of fuel behaviour, subsequent analysis suggests that the role of the graphite or siloxane layer may be more to prevent stress corrosion cracking than reduce strain concentration in the cladding [ 19,201. Nevertheless, the application of the results of the simulation has been most fortunate, even if done for the wrong reason. 5
4. DEFENDING LEAK-BEFORE-BREAK
Leak-before-break (LBB) is applied widely in the nuclear industry to protect against unstable crack propagation in pressurized components [21]: detecting leakage of water indicates a through-wall crack and the reactor is shutdown before the crack becomes unstable. Two general approaches are used:
For the main steel piping most reactor designers follow the philosophy of the US NRC using LBB as a licensing tool [22]. The material is not permitted to be susceptible to failure by degradation mechanisms during service. A through-wall flaw, that is stable and less than half the critical crack length (CCL), is postulated in the highest loaded region, Leakage of over 200 kg/h must be detectable before the pipe ruptures; in practice, leaks ten times smaller have been easily found.
Some construction materials are susceptible to degradation mechanisms and these must be taken into account when applying LBB, for example, zirconium alloys for pressure tubes [23] and nickel alloys for steam generators [24]. As a result the application of LBB is an economic issue rather than one of safety. For example, the simultaneous failure of a pressure tube and its calandria tube is a licensable event in CANDU.
The operating experience from the power reactors has led the way for Zr-2.5Nb pressure tubes in both Canadian [25] and Russian [26] reactors. In 1974, leakage was detected in the gas annulus between a pressure and calandria tube in Pickering Unit 3. The pathway for the leakage was cracking in the Zr-2.5Nb pressure tube close to the rolled joint. The modular construction of the reactor allowed the cracked pressure tube to be simply removed and replaced. Subsequently LBB was demonstrated 24 times and resulted in much research on the cracking and leak detection leading to formal operating procedures and Fitness-for-Service Guidelines [23, 271. The probability of crack initiation has been much reduced by lowering the residual stresses in the fuel channels and modifying operating procedures. The mechanism of cracking was delayed hydride cracking (DHC) and the basis for LBB is that leakage from a crack growing by DHC is detected long before CCL is reached. Thus one needs to know how much time, T, is required to detect the leak and place the reactor in a safe condition, and how much time, t, is available between the crack penetrating the tube wall and becoming unstable and then show that t>>T.
Experiments at Pickering A and Bruce A have shown that the annulus gas system is very sensitive to water leakage. Alarms based on dewpoint and rate of rise in dewpoint are triggered by a few g/h of moisture and the value of T has been determined to be less than 10 hours. To evaluate t one needs to know the crack length at first leakage, L, the rate of crack growth, V, and value of CCL, C, and in the simple case:
t = (C-L)/2V (2). 6
Fracture toughness controls C. Much research has yielded information on determining the effects of irradiation [4,5,7,28], specimen geometry [29], hydrogen [4, 30,311 and other trace elements on toughness [32,33], providing confidence in the knowledge of CCL and improving the ability to maintain high toughness after irradiation. Similarly, the mechanism and phenomenology of DHC are well understood and the effects on V of irradiation [5,7,34,], microstructure [35-371, temperature [5,7,38,39], temperature cycling [30,40-42] and temperature gradients [43] are known and methods for minimizing cracking have been developed [30,37,42,44,45]. The cracks found in the pressure tubes removed from power reactors grew in both the radial and axial directions and the shape of the cracks and the value of L was determined by the relative crack velocities in each direction [27]. Using conservative values for L, C and V yields t 2 18 h thus t 2 1.8T providing an acceptable margin. In assessmentsthe actual calculations are more involved than those suggested by the simple model of equation (2). They include the effect on the DHC growth rate and CCL of reducing the coolant pressure and temperature as the reactor cools. The analysis is used to show that during the cool-down the operating procedures maintain the CCL value above the current crack length so that the crack remains stable.
Leaking cracks may behave differently from non-leaking cracks because the metal near the crack is cooled to supply the latent heat needed as the escaping pressurized water flashes to steam on the crack face [46]. The leakage itself will not only depend on the crack opening but also on obstructions from oxidation and debris caught on the crack face. To verify the safety margin for detecting a leaking cracks, a simulation was devised in which a leaking crack was grown. The apparatus was called the Chalk River Active Crack Leak Evaluation, CRACLE [47].
The main features of the apparatus are depicted schematically in Fig. 6. A 450 mm long section of pressure tube and its rolled joint are attached to a source of flowing hot water and sealed at the other end. The water is pressurized up to 10.3 MPa and maintained at CANDU chemistry. Usually a deliberate flaw is spark-machined on the inside surface of the pressure tube. A water-cooled jacket to simulate the temperature conditions imposed by the calandria tube surrounds the pressure tube. To protect people from gamma irradiation and blast should CCL be exceeded, the whole assembly is contained in a lead- lined, water-cooled, registered pressure-vessel. Eddy current probes and ultrasonics measure the crack length. The test chamber is filled with dry nitrogen at 100 kPa and dew point of -30°C. Initial leakage is measured as an increase in dew point while large leakage, up to 200 kg/h, is collected and weighed in tanks below the test chamber. The temperature of the effluent from the crack is estimated by a thermocouple wedged between the pressure tube and the end-fitting.
Several rolled-joints removed from power reactors have been tested, one of which contained a crack formed during reactor operation. The spark-machined flaws were grown using temperature cycling and their penetration of the tube wall was easily detected by water leakage. During subsequent experiments the rolled joints were subjected to various pressure and temperature cycles and the resulting cracking and leakage was measured. Crack velocities were generally lower than would be predicted from tests on dry cracks (Fig. 7). This result was attributed mainly to the reduction in material temperature caused by the leakage since the temperature of the effluent at the mouth of the crack was close to 100°C even when the internal water temperature was 240°C. Crack growth was not stopped by approaching the test temperature by heating as is observed at temperatures above 170°C with dry cracks [30,40 - 42,441. This result is attributed mainly to the effect of the temperature gradient [43]. Cracks up to 70 mm long have sustained full system pressure at room temperature. Some stages of the experiments lasted up to 190 h, a consequence of the low crack velocities and large CCL, with accumulated test times of several hundred hours, thus providing confidence that the margins on t were very large.
The leak rate of water through the crack depended not on the absolute crack length but on the change of crack length after leakage (Fig. 8). In cracks that develop over several years before penetrating the tube wall, the oxide and debris on the crack faces, as well as crack tunnelling, impede the passage of water. Once wall penetration is achieved, a small increase in crack length allows a large increase in leak rate that should be easily detected in the dry annulus gas system, providing confidence in small values of T.
The main shortcomings of the simulation are that for operating reasons the apparatus was held vertically, whereas in CANDU the fuel channels are horizontal, and the internal flow of the loop water is not as great as in the power reactor thus the heat-transfer conditions were not the same as those in the power reactor. The leakages observed in the experiments were so large that the orientation of the apparatus would only have a small effect on leak detectability. In recent tests with much increased flow at the location of the crack, the results were similar to those of the first phase of testing. Despite these shortcomings, the margins are sufficiently large that the results from this simulation of a leaking crack provide potent support for the LBB approach. Although procedures are in place to prevent flaw initiation, should a flaw escape detection and grow, Leak-Before- Break is a next strong defence. 8
5. EXPLOITING THE MODERATOR AS A HEAT-SINK
If LBB fails in the steel piping carrying the primary heat-transport water (see previous section) the fuel may no longer be adequately cooled. This situation is an example of a loss-of-coolant accident (LOCA). Subsequently, the hot fuel heats the pressure tube sufficiently that it balloons into contact with the calandria tube. If the moderator is cool enough, the excess heat is removed by nucleate boiling on the outside surface of the calandria tube and the whole assembly is kept intact until the reactor can be shutdown. If the heat cannot be removed efficiently because of film boiling, dry-out leads to very high temperatures in the pressure and calandria tubes, which may subsequently fail. The border between the two modes of boiling has been evaluated in a simulation [48].
The apparatus to examine contact boiling is depicted in Fig. 9. A 1700 mm long section of calandria tube is placed concentrically around a 1750 mm long section of Zr-2SNb pressure tube. The inside of the pressure tube is pressurized up to 6 MPa with argon while the gap between the tubes contains CO2 at 100 kPa. Thermocouples are placed in the pressure and calandria tubes to follow temperature history. To simulate the moderator, the assembly is submerged in water in an open tank, the walls of which contain windows to allow the experiment to be videotaped. Passing current through a graphite rod concentric with the tubes simulates heat from an uncooled fuel bundle. During a test, the “moderator” temperature is established; the pressure inside the pressure tube is raised to the required value then the power is raised to the test power. The pressure tube is heated at up to 25”C/s and its temperature can be as high as 1100°C when it first touches the calandria tube. The test is stopped when either the calandria tube rewets or ruptures. The heat-transfer is satisfactory if either nucleate boiling is observed, if the temperature spike on the calandria tube is of short duration, <5s, or if any discolouration from oxidation of the calandria tube covers ~15% of the surface.
A peak moderator temperature of 70°C is used for CANDU reactors to avoid film boiling on the outside surface of the calandria tube during a postulated LOCA. Two approaches to increasing the margins on moderator temperature are:
a) to increase the temperature to the onset of film boiling, that is, increase the Critical Heat Flux (CHF), and, b) to reduce the rate at which the heat is passed from the pressure tube to the calandria tube.
For a), roughening the outside surface by peening with glass-beads has been found to enhance bubble formation [49 - 5 11. Peening with beads in the diameter range 90 to 125 pm provided an increase in CHF of over 50%. Consequently, the temperature in the “moderator” in contact boiling experiments could be increased by 3 to 5°C before the onset of film-boiling compared with experiments on untreated tubes. 9
In early experiments on b), wire mesh was placed between the pressure and calandria tubes [52]. This configuration was very effective in promoting efficient heat-transfer to the “moderator” but was considered too difficult to install. As an alternate method, small circumferential ridges, ~200 pm in height, on the inside surface of the calandria tube are being developed to reduce the area of contact between the pressure and calandria tubes [49]. To date these have been formed by either masking-and-pickling or by radially oscillating the rollers during roll-extrusion. The “moderator” in contact boiling tests can be 5 to 10°C warmer with ridged tubes than with smooth tubes before dryout is observed, showing that the heating rate between the two tubes has been reduced.
Since the two methods of enhancing heat-transfer operate on different and independent principles, they can be used additively. On sections of calandria tube where one or the other method was insufficient to provide protection during contact boiling tests with a “moderator” temperature of 85°C when used in combination they induced nucleate boiling, Table 2. The temperature history of a test with an internal pressure of 1 MPa on a section of calandria tube containing two variations of the surface modifications illustrates the point (Fig. 10). In the section of calandria tube the inside surface contained ridges 75 pm high spaced 25 mm apart and the outside surface was either smooth or roughened. The temperature of the pressure tube reached about 900°C before it touched the calandria tube. In the section where the surface of the calandria tube was smooth, (Fig. 10(a)) after contact the pressure tube cooled about 350°C and the calandria tube heated up reaching about 900°C at the time of rupture of the assembly. This section was in film boiling and was oxidized. In the test section where the outside surface was roughened (Fig. 10(b)), the pressure tube cooled about 450°C but the calandria tube temperature only increased a few degrees centigrade. The heat-transfer was by nucleate boiling and this section of tube was not oxidized. The results from a series of tests with the modified calandria tubes indicate that suppression of film boiling is such that the “moderator” could be 8 to 15°C warmer than the specified value. Currently, the geometry of the ridges - height, shape and spacing - is being optimized.
The experimental conditions for the simulation are based on calculations of the consequences of the hypothetical accident and provide a reasonable depiction of the model. Results described here are independent of the accuracy of the model or simulation because they compare current and modified components in the same test. The usefulness of the surface modifications of the calandria tubes is in potentially improving the economics of reactor construction and allowing CANDU to be operated at sites where the cooling water is warm without compromising safety margins; for example, the calandria tubes for the two CANDU reactors at Qinshan, China, include glass-bead peening of the outside surface. 10
6. SUMMARY
To safeguard and improve the performance of reactor components made from zirconium alloys, one requires a strong underpinning from knowledge of the behaviour of the materials - corrosion, mechanical and fracture properties and heat-transfer parameters. Sometimes the application of these properties to the reactor situation is complicated and experiments have to be devised to simulate the interactions between materials. The three examples of simulations described here have proved useful by either improving the performance of fuel cladding and calandria tubes made from Zircaloy or helping support the philosophy of leak-before-break as a defence against a flaw growing by delayed hydride cracking in Zr-2.5Nb pressure tubes. 11
7. FIGURE CAPTIONS.
Fig. 1. Schematic diagram of CANDU fuel channel. Fig. 2. Schematic diagram of a CANDU fuel bundle. Fig. 3. Experimental arrangement for fuel swelling simulation test. Fig. 4. Wall thickness profiles of tube specimens fractured at 300 “C in burst test and fuel swelling simulation test. Fig. 5. Effect of number of cracks in “fuel” on strain to failure in fuel swelling simulation tests at room temperature. Fig. 6. Schematic diagram of CRACLE apparatus. Fig. 7. Temperature dependence of velocity of delayed hydride cracks: comparison of leaking crack (points) with dry crack (line). Fig. 8. Change in leak rate with crack growth. Crack length at first leakage: B2-Jl l - 25 mm; P3-F13 - 8 mm. Fig. 9. Schematic diagram of contact conductance test apparatus. Fig. 10. Temperature history of pressure and calandria tubes in contact conductance test; “moderator” temperature: 85”C, internal pressure: 1 MPa, Heat-up rate: 18”C/s. Calandria tube with 75 pm high ridges 25 mm apart on inside surface (a) smooth outside surface, and (b) roughened outside surface. 12
a. ACKNOWLEDGEMENTS
I am grateful for the help and friendship I have received over the past nearly forty years trying to use zirconium alloys to the best of their ability. International cooperation and discussion has been important and I would like to acknowledge colleagues in Japan, Korea, Argentina, Russia, Lithuania, Romania, USA, UK, France, Italy, Germany, Austria, Sweden, and China. The Canadian nuclear utilities have provided the challenges and the Canadian manufacturers of nuclear components have contributed help and innovation when needed. The staff at AECL’s laboratories has provided excellent support and service. Acknowledging the contribution of my over 100 co-authors can only be done quantitatively by naming those who have shared the work five or more times: Stefan Sagat, Malcolm Griffiths, Jim Theaker, Jim Ambler, Allan Causey, Chuck Ells, Ross Gilbert, Brian Cheadle, Peter Chow, Doug Rodgers, Bob Hosbons, Randy Fong, Brock Sanderson, Nithy Nitheanandan, Rick Holt and Don Hardie (University of Newcastle), who started it all. 13
9. REFERENCES
Ul Rickover, H.G., “The Decision to use Zirconium in Nuclear Reactors,” TID- 26740, 1975. (Admiral Rickover was the first recipient of the Kroll Medal)
PI Ambartsumyan, R.S, Kiselev, A.A., Grebennikov, R.V., My&kin, V.A., Tsuprun, L.J., and Nikulina, A.F., “Mechanical Properties and Corrosion Resistance of Zirconium and Its Alloys in Water, Steam and Gases at High Temperature,” Proceedings, Second International Conference on the Peaceful Uses of Atomic Energy, United Nations, Geneva, Vol. 5, 1958, pp. 12 - 33. (Mrs. Nikulina received the Kroll Medal for 1990.) r31 Cheadle, B.A., Coleman, C.E., Rodgers, D.K., Davies, P.H., Chow, C.K., and Griffiths, M., “Examination of Core Components Removed from CANDU Reactors,” International Conference on CANDU Maintenance, Canadian Nuclear Society, Toronto, Ontario, November 1987. (Available as AECL Report AECL- 9710.)
[41 Coleman, C.E., Cheadle, B.A., Causey, A.R., Chow, C.K., Davies, P.H., McManus, M.D., Rodgers, D.K., Sagat, S., and van Drunen, G., “Evaluation of Zircaloy-2 Pressure Tubes from NPD,” Zirconium in the Nuclear Industry: Eighth International Symposium, ASTM STP 1023, L.F.P. Van Swam and C.M. Eucken, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1989, pp. 35-49.
151 Chow, C.K., Coleman, C.E., Koike, M.H., Causey, A.R., Ells, C.E., Hosbons, R.R., Sagat, S., Urbanic, V.F., and Rodgers, D.K., “Properties of an Irradiated Heat-treated Zr-2.5Nb Pressure Tube removed from the NPD reactor,” Zirconium in the Nuclear Industry: Eleventh International Symposium, ASTM STP 1295, E.R. Bradley and G.P. Sabol, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1996, pp. 469-491.
[61 Causey, A.R., Holt, R.A., Christocloulou, and E.T.C. Ho, “Irradiation-Enhanced Deformation of Zr-2.5Nb Tubes at High Neutron Fluences,” Zirconium in the Nuclear Industry: Twelfth International Symposium, ASTM STP 1354, G.P. Sabul G.D. Moan, Eds., American Society for Testing and Materials, West Conshohocken, PA, 2000, pp. 74-85.
[71 Hosbons, R.R., Davies, P.H., Griffiths, M., Sagat, S., and Coleman, C.E., “Effect of Long-Term Irradiation on the Fracture Properties of Zr-2.5Nb Pressure Tubes,” Zirconium in the Nuclear Industry: Twelfth International Symposium, ASTM STP 1354, G.P. Sabol and G.D. Moan, Eds., American Society for Testing and Materials, West Conshohocken, PA, 2000, pp. 122-138. 14
181 Cann, C.D., So, C.B., Styles, R.C., and C.E. Coleman, “Precipitation in Zr-2.5Nb enhanced by proton irradiation,” Journal of Nuclear Materials, Vol. 205, 1993, pp. 267-272.
[91 Woo, O.T., McDougall, G.M., Hutcheon, R.M., Urbanic, V.F., Griffiths, M., and Coleman, C.E., “Corrosion of Electron-Irradiated Zr-2.5Nb and Zircaloy-2,” Zirconium in the Nuclear Industry: Twelfth International Symposium, ASTM STP 1354, G.P. Sabol G.D. Moan, Eds., American Society for Testing and Materials, West Conshohocken, PA, 2000, pp. 709-734.
[lOI Motta, A.T., Lefebvre, F., and Lemaignan, C., “Amorphization of Precipitates in Zircaloy under Neutron and Charged-Particle Irradiation,” Zirconium in the Nuclear Industry: Ninth International Symposium, ASTM STP 1132, C.M. Eucken and A.M. Garde, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1991, pp. 718-739.
[ill Robertson, J.A.L., “Learning from History: a Case Study in Nuclear Fuel,” Zirconium in the Nuclear Industry: Eleventh International Symposium, ASTM STP 1295, E.R. Bradley and G.P. Sabol, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1996, pp. 3- 11.
WI Coleman, C.E., “Simulation of Interaction between Cracked UO2 Fuel and Zircaloy Cladding,” Proceedings of International Conference on Physical Metallurgy of Reactor Fuel Elements, The Metals Society, London, UK, 1974, pp. 302-307.
I.131 Gittus, J.H, “Theoretical Analysis of the Strains Produced in Nuclear Fuel Cladding tubes by the Expansion of Cracked Cylindrical Fuel Pellets,” Nuclear Engineering and Design, Vol. 18, 1972, pp.69-82.
[I41 Wood, J.C., Surette, B.A., London, I.M., and Baird, J., “ Environmental Induced Fracture of Zircaloy by Iodine and Cesium: the Effects of Strain Rate, Localizes Stresses and Temperature,” Journal of Nuclear Materials, Vol. 57, 1975, pp. 155- 179. r151 Nobrega, B.N., King, J.S., Was, G.S., and Wisner, S.B., “Improvements in the Design and Analysis of the Segmented Expanding Mandrel Test,” Journal of Nuclear Materials, Vol. 131, 1985, pp. 99-104.
1161 Foster, J.P. and Leasure, R.A., “Simulated Fuel Expansion Testing of Zircaloy Tubing,” Zirconium in the Nuclear Industry: Eighth International Symposium, ASTM STP 1023, L.F.P. Van Swam and C.M. Eucken, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1989, pp. 5 17-534. 15
[I71 Bain, A.S., Wood, J.C., and Coleman, C.E., “Fuel Designs to Eliminate Defects on Power Increases,” Proceedings of International Conference on Nuclear Fuel Performance, British Nuclear Energy Society, London, UK, 1973, pp. 56.1-56.5.
Cl81 Lau, J.H.K., Inch, W.W.R., Cox, D.S., Steed, R.G., Kohn, E., and Macici, N.N., “Canadian CANDU Fuel Development Program and Recent Fuel Operating Experience,” Sixth International Conference on CANDU Fuel, Canadian Nuclear Society, Toronto, Vol. 2, 1999, pp. 66-77.
[I91 Wood, J.C., Surette, B.A., Aitchison, I., and Clendening, W.R., “Pellet Cladding Interaction - Evaluation of Lubrication by Graphite,” Journal of Nuclear Materials, Vol. Vol. 88, 1980, pp. 81-94.
PO1 Chan, P.K. and Kaddatz, K.J., “How does CANLUB Work?’ IS” Annual Conference of the Canadian Nuclear Association, Toronto, 1994. Paper 3C.2.
1211 Coleman, C.E., Ed., “Leak-Before-Break in Water Reactor Piping and Vessels,” Elsevier Applied Science, London, UK, 1991. (Also as The International Journal of Pressure Vessels and Piping, Vol. 43, Nos. l-3, 1990, pp. l-442.)
WI Wichman, K., and Lee, S., “Development of USNRC Standard Review Plan 3.6.3 for Leak-Before-Break Applications to Nuclear Power Plants,” The International Journal of Pressure Vessels and Piping, Vol. 43, 1990, pp. 57-65. v31 Wong, H.W., Bajaj, V.K., Moan, G.D., Huterer, M., and Poidevin, C.O., “The Role of Leak-Before-Break in Assessments of Flaws Detected in CANDU Pressure Tubes,” The International Journal of Pressure Vessels and Piping, Vol. 43, 1990, pp. 23-37. v41 Flesch, B. and Cachet, B., “Leak-Before-Break in Steam Generator Tubes,” The International Journal of Pressure Vessels and Piping, Vol. 43, 1990, pp. 165-179. v51 Perryman, E.C.W., “Pickering Pressure Tube Cracking Experience,” Nuclear Energy, Vo1.17, 1978, pp. 95-105.
WI Rodchenkov, B.S., Abramov, V.Y., Klyuev, A.E. and Zolotarev, V.B., “Delayed Hydride Cracking in Zr-2.5%Nb Pressure Tubes,” Voprosy Atomnoi Nauki i Tekniki Set-, Materials Science and Novel Materials, Vol. 48, 1993, pp.17-20. v71 Moan, G.D., Coleman, C.E., Price, E.G., Rodgers, D.K., and Sagat, S., “Leak- Before-Break in the Pressure Tubes of CANDU Reactors,” The Znternational Journal of Pressure Vessels and Piping, Vol. 43, 1990, pp. 1-21. 16
WI Chow, C.K., Coleman, C-E., Hosbons, R.R., Davies, P.H., Griffiths, M., and Choubey, R., “Fracture Toughness of Irradiated Zr-2.5Nb Pressure Tubes from CANDU Reactors,” Zirconium in the Nuclear Industry: Ninth International Symposium, ASTM STP 1132, C.M. Eucken and A.M. Garde, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1991, pp. 246-275.
~291 Davies, P.H. and Shewfelt, R.S.W., “Size, Geometry, and Material Effects in Fracture Toughness Testing of Irradiated Zr-2.5Nb Pressure Tube Material,” Zirconium in the Nuclear Industry: Twelfth International Symposium, ASTM STP 1354, G.P. Sabol G.D. Moan, Eds., American Society for Testing and Materials, West Conshohocken, PA, 2000, pp. 356-376.
1301 Coleman, C.E., Cheadle, B.A., Ambler, J.F.R., Lichtenberger, P.C., and Eadie, R.L., “Minimizing Hydride Cracking in Zirconium Alloys,” Canadian Metallurgical Quarterly, Vol. 24, 1985, pp. 245-250.
[311 Wallace, A.C., Shek, G.K., and Lepik, O.E., “Effects of Hydride Morphology on Zr-2.5Nb Fracture toughness,” Zirconium in the Nuclear Industry: Eighth International Symposium, ASTM STP 1023, L.F.P. Van Swam and C.M. Eucken, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1989, pp. 66-88.
[321 Aitchison, I, and Davies, P.H., “Role of Microsegregation in Fracture of Cold- Worked Zr-2.5Nb Pressure tubes,” Journal of Nuclear Materials, Vol. 203, 1993, pp. 206-220.
1331 Theaker, J.R., Choubey, R., Moan, G.D., Aldridge, S.A., Davis, L., Graham, R.A., and Coleman, C.E., “Fabrication of Zr-2.5Nb Pressure Tubes to Minimize the Harmful Effects of Trace Elements,” Zirconium in the Nuclear Industry: Tenth International Symposium, ASTM STP 1245, A.M. Garde and E.R. Bradley, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1994, pp. 221-242.
II341 Sagat, S., Coleman, C.E., Griffiths, M., and Wilkins, B.J.S., The Effect of Fluence and Irradiation temperature on Delayed Hydride Cracking in Zr-2.5Nb,” Zirconium in the Nuclear Industry: Tenth International Symposium, ASTM STP 1245, A.M. Garde and E.R. Bradley, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1994, pp. 35-61.
[351 Simpson, L.A., and Cann, CD., “The Effect of Microstructure on Rates of Delayed Hydride Cracking in Zr-2SNb Alloy,” Journal of Nuclear Materials, Vol. 126, 1984, pp. 70-73. 17
1361 Huang, F.H., and Mills, W.J., “Delayed Hydride Cracking Behavior for Zircaloy- 2 Tubing,” Metallurgical Transactions, Vol. 22A, 1991, pp. 2049-2060. r371 Coleman, C.E., Sagat, S., and Amouzouvi, K.F., “Control of Microstructure to Increase the Tolerance of Zirconium Alloys to Hydride Cracking,” Presented at 26”’ Annual Conference of Metallurgists, Canadian Institute of Mining and Metallurgy, Winnipeg, MAN., 1987. (Available as AECL Report AECL-9524.)
[38 Simpson, L.A., and Puls, M.P., “The Effect of Stress, Temperature and Hydrogen Content on Hydride-Induced Crack Growth in Zr-2.5Nb,” Metallurgical Transactions, Vol. lOA, 1979, pp. 1093-l 105.
II39 Coleman, C.E., and Ambler, J.F.R., “Susceptibility of Zirconium Alloys to Delayed Hydrogen Cracking,” Zirconium in the Nuclear Industry: Third International Symposium, ASTM STP 633, A.L. Lowe and G.W. Parry, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1977, pp. 589-607.
[401 Ambler, J.F.R., and Coleman, C.E., “Acoustic Emission during Delayed Hydrogen Cracking in Zr-2.5 wt. Percent Nb Alloy,” Second International Congress on Hydrogen in Metals, Paris, 1977, Paper 3ClO.
1411 Ambler, J.F.R., “Effect of Direction of Approach to Temperature on the Delayed Hydride Cracking Behavior of Cold-Worked Zr-2.5 Nb,” Zirconium in the Nuclear Industry: Sixth International Symposium, ASTM STP 824, D.G. Franklin and R.B. Adamson, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1984, pp. 653-674.
~421 Shek, G.K., and Graham, D.B., “Effects of Loading and Thermal Manoeuvres on Delayed Hydride Cracking in Zr-2.5 Nb Alloys,” Zirconium in the Nuclear Industry: Eighth International Symposium, ASTM STP 1023, L.F.P. Van Swam and C.M. Eucken, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1989, pp. 89-l 10.
[431 Sagat, S., Chow, C.K., Puls, M.P., and Coleman, C.E., “Delayed Hydride Cracking in Zirconium Alloys in a Temperature Gradient,” Journal of Nuclear Materials, Vol. 279,2000, pp. 107-l 17.
E441 Cheadle, B.A., Coleman, C.E., and Ambler, J.F.R., “Prevention of Delayed Hydride Cracking in Zirconium Alloys.” Zirconium in the Nuclear Industry: Seventh International Symposium, ASTM STP 939, R.B. Adamson and L.F.P. Van Swam, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1987, pp. 224-240. 18
[451 Coleman, C.E., Cheadle, B.A., Cann, C.D., and Theaker, J.R., “Development of Pressure Tubes with Service Life greater the 30 Years,” Zirconium in the Nuclear Industry: Eleventh International Symposium, ASTM STP 1295, E.R. Bradley and G.P. Sabol, Eds., American Society for Testing and Materials, West Conshohocken, PA, 1996, pp. 884-898.
E461 Collier, R.P., Liu, J.K.S., Mayfield, M.E., and Stulen, F.B., “Study of Critical Two Phase Flow through Simulated Cracks,” Battelle Columbus Laboratory Report, BCL-EPRI-80- 1, 1980.
r471 Coleman, C.E., and Simpson, L.A., “Evaluation of a Leaking Crack in an Irradiated CANDU Pressure Tube,” Fracture Mechanics Verification by Large- Scale Testing, EGF/ESIS8, K. Kussmaul, Ed., Mechanical Engineering Publications, London, UK., 1991, pp. 189-201. (Available as AECL Report AECL- 10246.)
[481 Gillespie, G.E., Moyer, R.G., Hadaller, G.I., and Hildebrandt, J.G., “An Experimental Investigation into the Development of Pressure Tube/Calandria Tube Contact and Associated Heat Transfer under LOCA Conditions,” Proceedings of the Sixth Annual Canadian Nuclear Society Conference, Ottawa, ON, 1985, pp. 2.24-2.30.
[491 Coleman, C.E., Fong, R. W.L., Doubt, G.L., Nitheanandan, T., and Sanderson, D.B., “Improving the Calandria Tubes for CANDU Reactors,” Presented at 18”’ Annual Conference of the Canadian Nuclear Society, Toronto, ON., 1997. (Available as AECL Report AECL- 118 15.)
[501 Nitheanandan, T., Tiede, R.W., Sanderson, D.B., Fong, R.W.L., and Coleman, C.E., “Enhancing the Moderator Effectiveness as a Heat Sink during Loss-Of- Coolant Accidents in CANDU-PHW Reactors using Glass-Peened Surfaces,” Presented at the International Atomic Energy Agency Meeting on “Experimental Tests and Qualification of Analytical Methods to Address Thermohydraulic Phenomena in Advanced Water Cooled Reactors. ” 1998. (Available as AECL Report AECL- 11854.)
[511 Fong, R.W.L., McRae, G.A., Coleman, C.E., Nitheanandan, T,. and Sanderson, D.B., “Correlation Between the Critical Heat Flux and the Fractal Surface roughness of Zirconium Alloy Tubes,” Journal qf Enhanced Heat Transfer, Vo1.8,2001, pp.137-146. r521 Sanderson, D.B., Moyer, R.G., Litke, D.G., Rosinger, H.E., and Girgis, S., “Reduction of Pressure tube to Calandria Tube Contact Conductance to Enhance the Passive Safety of a CANDU-PHW Reactor,” Proceedings of the International Atomic Energy Agency Technical Committee Meeting on Advances in Heavy Water Reactors, 1993, (Available as AECL Report AECL-10892). 19
Table 1 Effect on ductility of graphite and siloxane coatings on the inside surface of fuel cladding: Fuel swelling simulation tests at 300°C
NO COATING COATING
Irradiation Elongation Strain Elongation Strain (n/m2x1024) % Concentration % Concentration Comment
GRAPHITE 0 4.8 1.7 23 1.3 2.3 2.2 3.4 4.7 7.0 2.2 Irradiated with graphite on inside surface. 8.0 1.5 . . . 9.5 . . . Graphite applied after irradiation.
SILOXANE 0 4.8 1.7 19 1.3 20
Table 2 Behaviour of modified calandria tubes during contact conductance tests: Ridge height 75 pm, “Moderator” temperature 85”C, heat up rate 218”C/s
Pressure inside pressure Spacing of ridges, Outside surface Pressure tube contact Nucleate tube, MPa mm of calandria tube temperature, “C Boiling
’ 40% of surface covered with oxide. 21
3747-E
FEEDERS END CALANDRIA PRESSURE / FITI-ING TUBE TUBE FE:DERS
GAS FUEL \ I ‘GAS ANNULUS BUNDLES SPACERS BELLOWS ANNULUS OUTLET INLET
Figure 1: Schematic diagram of CANDU fuel channel. 22
3476-J
1. ZIRCALOY STRUCTURAL END PLATE 2. ZIRCALOY END CAP 3. ZIRCALOY BEARING PADS 4. URANIUM DIOXIDE PELLETS 5. ZIRCALOY FUEL SHEATH 6. ZIRCALOY SPACERS
Figure 2: Schematic diagram of a CANDU fuel bundle. 23
c
$L /
Figure 3: Experimental arrangement for fuel swelling simulation test. 24
Figure 4: Wall thickness profiles of cladding specimens fractured at 300°C in burst test and fuel swelling simulation test. 25
STRAIN 8 TO FRACTURE1 t
Figure 5: Effect of number of cracks in “fuel” on strain to failure in fuel swelling simulation tests at room temperature. 26
3712.G
t n .TO X-l LOOP CIRCUIT
._,:_ . _ ,._; .,.* _I,xL
1 CRACLE
t %:*A i
PRESSURE VESSEL,
:..:.;. LEAD SHIELDING.
PRESSURE TUBE \ CRACK ZONE
CALANDRIA TUBE-
COOLING’ WATER
;;= TO LEAK COLLECl -ION VESSEL ..
Figure 6: Schematic diagram of CRACLE apparatus. 27
TEMPERATURE “C
IO- 7 _ 280 240 200 rso 100 1 I 1 I' 0 HEATING A COOLlUG CRACK LABORATORY DATA VU OCITY m/s
Figure 7: Temperature dependence of velocity of delayed hydride cracks: comparison of leaking crack with dry crack. 28
LEAK RATEI A A kg/h A t3 IOI - I
I- O 2 4 6 8 10 12 14 16 16 2D CRACKGROWTH FROM INITIAL LEAK LENGTH, mm
Figure 8: Change in leak rate with crack growth. Crack length at first leakage: BZ- Jll - 25 mm; P3-F13 - 8 mm. 29
InsulationVoltage -- CalandriaTube Spa? Rink I I \\Inuu __ Leadwire. I II
PressureTube
Figure 9: Schematic diagram of contact conductance test apparatus. 30
CONTACT RUPTURE a CONTACT 1000 0 1000 I 1 1 I\ RUPTURE 800 800 I
0 0 TIME TIME
Figure 10: Temperature history of pressure and calandria tubes in contact conductance test; “moderator” temperature: WC, internal pressure: 1 MPa, heat-up rate: WC/s. Calandria tube with 75pm high ridges 25 mm apart on inside surface (a) smooth outside surface, and (b) roughened outside surface. AEXL-12132
ISSN 0067-0367 To identify individual documents in the series, we have assigned an AECL-number to each. Please refer to the AECL-number when requesting additional copies of this document from: Information Centre, Information Management Branch AECL Chalk River, Ontario Canada KOJ 1JO Fax: (613) 584-8144 Tel.: (613) 584-3311 ext. 4623
Pour identifier les rapports individuels faisant par-tie de cette series, nous avons affect6 un numero AECL-a chacun d’eux. Veuillex indiquer le numero AECL-lorsque vous demandez d’autres exemplaires de ce rapport au: Service de Distribuition des Documents Offciels EACL Chalk River (Ontario) Canada KOJ 1JO Fax: (613) 584-8144 Tel.: (613) 584-3311 poste 4623
Copyright 0 Atomic Energy of Canada Limited, 2002