1.PRIMELE ELEMENTE COMBUSTIBILE PHWR FABRICATE IN ROMANIA.

In Romania primele elemente combustibile nucleare au fost realizate în 1978 la Institutul de Tehnologii Nucleare aflat pe platforma de la Magurele si au fost iradiate in reactorul BR2, Mol, Belgia. A fost primul mare succes înregistrat de institutul nostru. Ma bucur ca am avut privilegiu sa fiu in miezul acestei activitati si in final sa redactez la SCK/CEN din Mol, Belgia, impreuna cu colegii belgieni, raportul BLG 530 în care prezint performanta primelor elemente combustibile fabricate în România. Fiind raportul extern al institutului SCK/CEN, a fost transmis ca lucrare stiintifica la toate institutele de cercetari nucleare din lume, facandu-se cunoscut si in acest mod succesul Romaniei,  singura tara din Europa de Est care produce si testeaza, in reactor si celule fierbinti, combustibilul nuclear experimental. Raportul BLG 530 ramane pentru mine o lucrare istorica in care am prezentat o activitate de început in domeniul combustibilului nuclear.

………………………………

……………………

2. COMPORTAREA COMBUSTIBILULUI NUCLEAR PHWR CANDU IN REGIM SEVER DE CICLAJ.

La Conferita CANDU 2010 de la Niagara Falls, Canada, am prezentat o lucrare privind testele de iradiere în regim sever de ciclaj al puterii, teste efectuate în reactorul TRIGA de la ICN Pitesti in colaborare cu AECL Canada. Prin aceste teste am analizat si evaluat, în premiera internationala, performanta combustibilului PHWR CANDU în regim de urmarire a sarcinii (ciclaj de putere). Lucrarea prezentata la conferinta si publicata mai apoi în revista de specialitate Kerntechnik din Germania, reprezinta pentru mine o lucrare simbol cu care mi-am încheiat activitatea de cercetare în institut

CANDU FUEL ELEMENTS BEHAVIOUR IN THE LOAD FOLLOWING TESTS.

The authors of this contribution:
Grigore Horhoianu
PhD, Head of Nuclear Fuel Engineering Laboratory
Institute for Nuclear Research, P.O. Box 78, Pitesti, 0300, Romania

Steve Palleck
Manager of Fuel Design Branch, Sheridan Park Research-AECL
Mississauga Ontario, L5K 1B2, Canada

Two load following (LF) tests on CANDU type fuel elements were performed in TRIGA Research Reactor of INR Pitesti. In the first LF test the 78R fuel element has successfully experienced 367 power cycles, mostly between 23 and 56 kW/m average linear power. In the second LF test, the fuel element withstood 200 power cycles from 27 to 54 kW/m average linear power as well as additional ramps due to reactor trips and restarts during test period. New LF tests are planed to be performed in order to establish the limits and capabilities for CANDU fuel in LF conditions. This paper presents the results
of the LF tests performed in TRIGA Research Reactor and their relation to CANDU fuel performance in LF conditions.

Das Verhalten von CANDU Brennelementen bei Tests nach der Beladung. Zwei Tests wurden nach der Beladung (LF) mit Brennelementen vom CANDU-Typ am TRIGA Forschungsreaktor des INR Pitesti durchgeführt. Beim ersten LF Test hatte das 78R Brennelement erfolgreich 367 Leistungszyklen meist bei durchschnittlicher Leistung zwischen 23 and 56kW/m absolviert. Beim zweiten LF Test widerstand das Brennelement 200 Leistungszyklen zwischen 27 and 54 kW/m durchschnittlicher linearer Leistung, sowie zusätzlichen Anstiegen aufgrund von Reaktor Schnellabschaltungen und Neustarts während der Testperiode. Neue LF Tests sind geplant, um Grenzen und Leistungsvermögen für CANDU Brennelemente unter LF Bedingungen zu bestimmen.

1. Introduction
CANDU type reactors are usually operated at steady high power (base load). Recent years have brought to attention the problem of load following operation of CANDU power reactors. When the nuclear generated power represents a significant fraction of the grid, the load following mode of operation becomes a necessity. In this context, one of the current research objectives of our fuel behavior studies was to investigate and to reliably predict the performance of CANDU type fuel during power cycling operation conditions. In support to this project, an experimental program has been established at the Institute for Nuclear Research (INR) Pitesti [1].
The experimental work was designed both to expand our detailed knowledge of fuel element response to power changes and to provide sound data for benchmarking of existing fuel performance modeling codes.
During the first LF irradiation test performed in TRIGA Research Reactor of INR Pitesti, the CANDU type fuel element 78R has successfully experienced up to 367 power cycles, mostly between 23 and 56 kW/m, average linear power and with a maximum linear power of 64 kW/m at the bottom endregion [2]. A new LF irradiation test performed in TRIGA Research Reactor on the ME 01 fuel element has been successfully experienced up to 200 power cycles from 27 to 54 kW/m, average linear power, as well as additional ramps due to reactor trips and restarts during test period [3, 4].In order to develop strain-life fatigue curves for Zircaloy-4 cladding, out reactor fatigue tests has been performed at INR Pitesti [5].This paper presents experimental results of the CANDU fuel elements behaviour under LF conditions.

2. Irradiation facility
To determine fuel behaviour during power cycling and LF conditions, a special irradiation device (capsule C9) was designed and manufactured at INR Pitesti [6]. The C9 capsule is a pressurized, single-wall tube capsule contained in a calorimeter. A computer-controlled displacement device provides back-and-forth movement of in-pile section, ensuring fuel element power variation.
3. Experimental results
3.1. Test element 78R
The power cycling test has been performed in the capsule C9.
The specified variation of fuel element linear power was obtained by mechanical movement of the device into the TRIGA Research Reactor core. The capsule was instrumented for neutron flux and temperature measurements during the test. The power output of the test element was determined through calibration of the flux detectors as power sensors. The test fuel element (coded 78R), fabricated in INR Pitesti Fuel Technology Laboratory, was a short CANDU type fuel element, containing enriched fuel pellets. The design parameters of 78R fuel rod are summarized in Table 1 and the irradiation parameters are summarized in Table 2. The power
history comprised 367 power cycles, mostly between 50% and 100% of the specified average linear power (56 kW/m). The power cycle was 0.5 h – 4 h – 0.5 h – 4 h which is 0.5 h for reduction of power from full power to 50% power, 4 h at the lower power, 0.5 h to recover the power to the full power
and 4 h at full power. Due to axial profile of neutron flux the maximum local linear power was 64 kW/m at the bottom end region of test element.
3.1.1 Post-irradiation examination
The post irradiation investigation performed in INR Pitesti Hot Cells Laboratory included both, non-destructive examinations (visual, profilometry, axial gamma-scanning, eddycurrent testing) and destructive examinations (puncturing, fission gas volume and composition, element void volume, chemical burnup determination, metallography/ceramography and mechanical tests) [7–10].The axial profile of the gamma scans are shown in Fig. 1.Intensity dips are seen at the pellet interfaces. The intensity along the fuel stack is not uniform, indicating an increase along the fuel stack from top end of fuel element end to bottom end. The gradient reflects the axial power distribution in the 78R element during irradiation with the maximum at the bottom end region. The diameter of the element was measured by scanning along the length and by rotating the element in an in-cell profilometer. The axial cladding hoop strain profile of 78R fuel element is shown in Fig. 2. The highest cladding hoop strains were measured at the bottom end region of the 78R fuel element where the local linear power had a maximum value. The mid-pellet residual sheath strain ranges from 0.51% to 0.94% in the bottom end region and from –0.06% to 0.15% in the top end region. The pellet interface residual sheath strain ranges from 0.68% to 1.33% in the bottom end region and from 0.11% to 0.27% in the top end region. Post-irradiation measurements show that the bow of irradiated element was about 0.3 mm. The distinct ridges on sheath at pellet interface locations indicated that strong pellet-cladding mechanical interaction (PCMI) had occurred. The axial variation in cladding hoop strain was significant for element 78R and consistent with axial variation in the local linear power. The measured element gas volume was 1.66 (±0.03) ml at STP and the measured element void volume was 1.14 (±0.05) ml. The pellet macrostructure can be seen in Fig. 3a. The cracking pattern is typical of UO2 operating at about 56 kW/m (radial cracking with some circumferential cracking around a plastic core) [7 – 9]. A photograph of the etched microstructure of the pellet is shown in Figs. 3b, 3c. The microstructure is very similar to the as-fabricated microstructure, with the exception of the removal of the small, distributed pores throughout the fuel matrix. Fuel microstructure showed small grain growth at the centre of the pellet, particularly at the bottom end of the element where local linear power was at maximum value. Evidence of oxidation, white oxide, was observed on the outside of the fuel sheath. Oxide thickness varied along the sheath length from about 2 to 4 microns on the hotter bottom end, consistent with the measured axial flux gradient.
Post-irradiation appearance of the pellet-cladding gap is shown in Fig. 4a. The PI measured burnup was found to range from 82.3 to 111.1 MWd/kgU across the length of fuel element [7]. Ring cladding samples were tested on a Instron tensile machine at 3008C [10]. A method described by Uchida
[11] was used for interpretation of the results. A macrophotograph of the test ring specimen after fracture is shown in Fig. 4b. The UTS was find to be 497.2 MPa accompanied by 12.7% total elongation.
3.2. ME 01 test element
The ME01 fuel element was a short CANDU type fuel element fabricated by Chalk River Nuclear Laboratories, containing 17 enriched fuel pellets and 2 natural uranium endpellets [3, 4]. The end pellets were made with natural uranium to prevent over power in the end pellets from end flux peaking. The design parameters of ME01 fuel rod are summarized in Table 1 and the irradiation parameters are summarized in Table 2. The power history comprised 200 power cycles, mostly between 50% and 100% of the specified linear power (54 kW/m average linear power), as well as additional ramps
due to reactor trips and restarts during the test period. The power cycle was 3 h–3 h–4 h–14 h which is 3 h for reduction of power from full power to 50% power, 3 h at the lower power, 4 h to recover the power to the full power and 14 h at full power.
3.2.1 Post-irradiation examination
The post irradiation investigation performed in INR Pitesti hot cell laboratories included both non-destructive examinations (visual, profilometry, axial gamma-scanning, eddy-current testing) and destructive examination (puncturing, fission gas volume and composition, element void volume, chemical burnup determination, metallography/ceramography and mechanical tests) [9, 10, 12,13]. The axial profile of gamma scans are shown in Fig. 5. Intensity dips are seen at the pellet interfaces, and there is a clear distinction between the natural uranium end pellets and the enriched pellets. The intensity along the fuel stack is uniform, indicating a uniform distribution of fission products along the fuel stack, and thus a uniform distribution of feed material and fissile distribution in the pellets.
The gamma scans appear to be typical of UO2 that has operated in the 47–54 kW/m range.The diameter of the element was measured by scanning
along the length and by rotating the element in an in-cell profilometer. The axial cladding hoop strain profile of ME01 fuel element is shown in Fig. 6. The highest cladding hoop strains were measured at the central region of the ME01 fuel element. The maximum ridge height is 0.033 mm. The mid-pellet residual sheath strain ranges from –0.06% to 0.16%. The pellet interface residual sheath strain ranges from 0.17% to 0.46%. Post-irradiation measurements show that the bow is about 0.3 mm. The distinct ridges on sheath at pellet interface locations indicated that strong pellet-cladding mechanical interaction (PCMI) had occurred.
The measured element gas volume was 1.46 (±0.03) ml at STP and the measured element void volume was 1.2 (±0.05) ml.The pellet macrostructure can be seen in the Fig. 7a. The cracking pattern is typical of UO2 operating at about 54 kW/m (radial cracking with some circumferential cracking around a plastic core) [9, 13]. Photograph of the etched microstructure at the pellet mid-radius is shown in Figs. 7b,7c. Metal particles are still clearly visible at the grain junctions. The microstructure is very similar to the as-fabricated microstructure, with the exception of the removal of the small, distributed pores throughout the fuel matrix. The pellet centre has not experienced significant micro structural changes. The grains in the centre of the pellet are larger than those in the periphery due to the temperature gradient in the pellet.
Evidence of oxidation, white oxide, was observed on the outside of cladding surface. Oxide thickness varied along the sheath length from about 1 to 7 microns. Post-irradiation appearance of the pellet-cladding gap is shown in Fig. 8a. The PI measured burnup was found to range from 176 to 191.1 MWd/kgU across the length of fuel element [13]. Ring cladding samples were tested on a Instron tensile machine at 3008C [10]. A method described by Uchida [11] was used for interpretation of the results. A macro-photograph of the test ring specimen after fracture is shown in Fig. 8b. The UTS was find to be 431.1 MPa accompanied by 9.2% total elongation.

4 Discussions
Ideally, the LF demonstration should be done in a power reactor where the fuel orientation and the operating condition are typical for a CANDU NPP. While fuel will experience negligible strain during LF there seems to be some possibility of random stress raises occurring due to released pellet chips which fall into the fuel/sheath gap during low power operation. This effect has been noted in vertically oriented fuel irradiations tests [14]. Pellet particles could move down the fuel column in vertically oriented fuel, to perhaps lodge at pellet interfaces where the pellet/sheath gap is at its minimum because of pellet-sheath interaction. Relocation of pellet chips within the fuel/sheath gap during power reduction could result in localized clad strains at the positions of the chips on returning to full power. A reactor shutdown cycle should be more severe than a load following cycle for this mechanism. Pellet chips have been identified as sources of high sheath strain in fuel irradiated when depressurization had occurred[14]. In normal LF operation no depressurization of coolant will occur but some pellet/sheath gap will be present at major axis of ovality of fuel during low power operation. For this reason we consider that tests in the vertical oriented capsule C9 of the TRIGA research reactor are a more severe tests of
fuel performance during load following than would be the case in a CANDU power reactor. Therefore, successful completion LF tests in the TRIGA reactor, with 78R and ME01 fuel elements, would adequately demonstrate the absence of life limiting mechanisms for CANDU fuel from load following conditions. The experimental results obtained in these tests suggest that the
CANDU fuel would survive the more frequent load following modes of operation in a NPP. If needed, the performance margins can be increased by a variety of operations such as reduction of sheath stress by improved shapes of pellets and modified clearances. We have very little information on the performance of CANLUB coatings in a stress cycling regime. It is possible that coatings may prove less effective with stress cycling than
for steady high power under stress, in which case stress corrosion or corrosion assisted fatigue of the cladding could result. As limit, a power cycle with a hold time of twelve hours like in ME01 fuel element tests could be considered to be similar to a multiple power ramps test. However, while the use of CANLUB coatings has effectively prevent SCC in ramps, i. e., in a single cycle with high stress, it is not proven that CANLUB coatings will effective for long term cycling stresses. Sheath coating has been made for both 78R and ME1 fuel elements using a CANLUB graphite layer. Metallographic examinations of 78R and ME1 fuel elements indicates that
most of the graphite layer is missing probably due to cyclic sheath deformation and vertically oriented irradiation in C9 capsule
5. Conclusions
Severe LF tests performed in vertical oriented C9 capsule of TRIGA research reactor shows no evidence of fuel failure from 367 power cycles at 78R fuel element and from 251 power cycles at ME01 fuel elements.
Profilometry measurements and distinct ridges observed on sheath at pellets interfaces show that both elements experienced high tensile strains in pellet interface regions. The axial variation in cladding hoop strain was significant for element 78R and consistent with axial variation in the local linear power. Power cycles do not appear to enhance fission gas release over that from fuel burned up to the same level without cycling.
Pellet and sheath microstructure yields results typical for CANDU6 fuel with similar burnup without power cycling. Metallographic examination indicates that most of the graphite layer is missing probably due to cycling sheath deformation and vertically oriented irradiation in C9 capsule.
The above results suggest that CANDU fuel continue to show good performance not only in base-load operation mode but also in a severe load-following operation mode .

Acknowledgements
Many individuals contributed to these tests. In particular, the authors would like to acknowledge the contributions of the persons of the Fuel Technology Laboratories for their effort in manufacturing the experimental fuel elements, the staff of TRIGA research reactor, who conducted the irradiation tests and the Hot Cell Laboratory staff for the PI examination of the irradiated fuel elements.

References
1. Horhoianu, G.: Nuclear Fuel R&D Program of INR Pitesti for the
Period 2006–2010, Internal Report No. 7212, I.N.R. Pitesti, Romania, 2005
2. Horhoianu, G. et al.: Experimental Aspects of Load Cycling Capability of CANDU Type Fuel, Annual Meeting on Nuclear Technology 1996, Manheim, Germany, 21–23 May, 1996
3. Palleck, S. et al.: Load Following Testing by AECL Collaboration
with the Institute for Nuclear Research Romania, Seventh International Conference on CANDU Fuel, Kingston, Canada, September
23–27, 2001
4. Horhoianu, G. et al.: Load Following Tests on CANDU Type Fuel
Elements in TRIGA Research Reactor of INR Pitesti, 11th International Conference on CANDU Fuel, Niagara Falls, Canada, 17–
20 October, 2010
5. Roth, M. et al.: Low Cycle Fatigue Tests on Zircaloy-4 Fuel Cladding, Internal Report No. 8094/2008, INR Pitesti, Romania, 2008
6. Dumitru, M. et al.: Power Cycling Test in C9 Capsule on 78R Fuel
Element, Internal Report No. 4210/1993, I.N.R. Pitesti, Romania,
1993
7. Parvan, M. et al.: Post Irradiation Examination of 78R Fuel Element, Internal Report No.4351/1994, I.N.R. Pitesti, Romania, 1994
8. Uta, O. et al.: Post Irradiation Examination of 78R Fuel Element
(Destructive Tests), Internal Report No. 6038/2001, I.N.R. Pitesti,
Romania, 2001
9. Uta, O. et al.: Post Irradiation Examination of 78R and ME01 Fuel
Elements (Destructive Tests), Internal Report No. 8889/2010,
I.N.R. Pitesti, Romania, 2010
10. Ionescu, S. et al.: Ring Tests on Zy-4 Fuel Cladding of 78R and
ME01 Fuel Elements, Internal Report No. 8834/2010, I.N.R. Pitesti,
Romania, 2010
11. Uchida, M.; Ichikawa, M.: J. of Nucl. Sc. and Tech. 13 (1976) 585–
590
12. Parvan, M. et al.: Post Irradiation Examination of ME01 Fuel Element, Internal Report No.7425/2006, I.N.R. Pitesti, Romania, 2006
13. Uta, O. et al.: Post Irradiation Examination of ME01Fuel Element
(Destructive Tests), Internal Report No.6038/2001, I.N.R. Pitesti,
Romania, 2001
14. Carter, T. J.; Ferenbach, P. J.: The Load Following Capability of
CANDU Fuel. Report AECL 7096, Chalk River Nuclear Lab., 1983
15. Horhoianu, G.: Nuclear Fuel R&D Program of INR Pitesti for the
Period 2011–2015, Internal Report No. 8779/2010, I.N.R. Pitesti,
Romania, 2010

Fig. 1. Axial gamma scan profile of the 78R test element

Fig. 2. Post-irradiation cladding deformation profile of the 78R test element

(a) (b) (c)
Fig. 3. Post irradiation appearance of 78R fuel element: (a) cross section view, (b) crystal grains in the peripheral part of pellet, (c) crystal grains in the central part of the pellet.

Fig. 5. Axial gamma scan profile of the ME01 test element

Fig. 6. Post-irradiation cladding deformation profile of the ME01 test
element

(a) (b) (c )
Fig. 7. Post irradiation appearance of ME01 fuel element: (a) cross section view, (b) crystal grains in the peripheral part of pellet, (c) crystal grains in the central part of the pellet .

Fig. 8. Post irradiation appearance of 78R fuel element: (a) cross section view at the pellet-cladding gap (b) ring test samp.

……………………………….

………………..