Community Research and Development Information Service - CORDIS

Final Report Summary - FAIRFUELS (FAbrication, Irradiation and Reprocessing of FUELS and targets for transmutation)

Executive Summary:
As part of the EURATOM 7th Framework Programme (FP7), the 4 year project FAIRFUELS (Fabrication, Irradiation and Reprocessing of FUELS and targets for transmutation) was aimed to a more efficient use of fissile material in nuclear reactors by implementing the transmutation of long-living radioactive waste from actinides in GEN-IV fast neutron spectrum reactors or sub-critical Accelerator Driven Systems. Such a transmutation scheme provides a route to a reduced volume, lifetime and hazard of high-level radioactive waste by recycling the most long-lived components. In this way, the nuclear fuel cycle can be closed in a sustainable manner. FAIRFUELS has contributed to the realization of waste lifetime reduction by investigating the material behavior under irradiation of the innovative fuels carrying the actinide waste.
In FAIRFUELS, The focus of the investigation was on the transmutation of americium. The rationale behind this choice is that the in-core behaviour of plutonium, which is the first lifetime-determining component in the waste, is already reasonably well known. Following the destruction of plutonium, americium determines the lifetime of the final waste, but it introduces new issues, in fabrication, behavior under irradiation and reprocessing. Compared to known uranium and uranium-plutonium oxide fuels, americium-bearing fuels have significantly higher (helium) gas production, which a priori could result in increased fuel swelling as well as increased pressurization of fuel pins. Based on current knowledge, the third candidate for transmutation, neptunium, does not present additional issues beyond those of americium.
The innovative fuels studied in FAIRFUELS represent the current agreement on three main routes to a closed nuclear fuel cycle:
• Admixture of low amounts of actinide waste in fast reactor (mixed oxide) driver fuel, with a single (homogeneous) recycling route for the driver fuel through a process of dissolution – separation – conversion and fuel fabrication. In this route, the actinide waste content is kept low to reduce possible negative impact on core neutronic behavior.
• Incorporation of the minor actinides in UO2-based targets located in the blankets of fast-spectrum reactors, with separate (heterogeneous) recycling of the driver fuel and the MA-bearing targets.
• MA destruction in dedicated Accelerator Driven Systems. Here, the minor actinides are carried by ‘inert’ instead of uranium-based matrices in order to optimize the rate of destruction.
The material response of of candidate materials have been investigated both experimentally, through neutron irradiation in Material Test Reactors and post-irradiation examinations in hotcell facilities, and theoretically, through simulations using a fuel performance code and of the underlying physics of gas migration. Two irradiation experiments were carried out successfully within FAIRFUELS in the HFR Petten. In MARIOS, an UO2 matrix containing around 15% of americium intended as a blanket material was irradiated at well-defined temperatures of 1000 and 1200 oC, while in SPHERE, fast reactor MOX driver fuel incorporating 3% of americium was irradiated in pellet and sphere-pac forms. Post-irradiation examinations on these fuels will be carried out in the FP7 project PELGRIMM.
In addition, post-irradiation examinations were performed on several inert matrix fuels, based on magnesium oxide, zirconium oxide and nitride, and molybdenum metal matrices. Results highlight the pros and cons of the different fuel forms. Zirconia provides a stable irradiation matrix which is however limited by low thermal conductivity, which induces fracture as well as restructuring at lower power levels compared to (reference) UO2. Zirconium nitride in contrast shows excellent in-core behavior up to high power levels, except for relatively strong swelling, necessitating larger fuel-cladding gaps. Another matrix showing excellent irradiation behavior is magnesia, with little swelling due to confinement of irradiation damage to the incorporated actinide oxide particles, and no fuel restructuring due to high conductivity. Compared to magnesia, molybdenum fuel has slightly higher swelling rates especially at higher powers, which was observed to be related to irradiation damage and gas bubble formation in the matrix surrounding the particles. This temperature-dependent swelling may result in increased probability of PCI failure during power transients, although other factors play a role too.
The experimental data and supporting models generated in FAIRFUELS have provided, together with parallel efforts in France and around the world, a significant basis for the evaluation and selection of transmutation fuels and targets, allowing for more focused future efforts to qualify the most promising candidate(s). The next step in the qualification of transmutation fuels will be the experimental assessment of behavior in off-normal and accident conditions, notably through power ramp testing at selected burn-ups as well as RIA and LOCA tests.

Project Context and Objectives:
2. Project context and objectives

Nuclear energy constitutes currently a major source of energy in Europe, providing highly competitive, CO2-free electricity. However, exploiting the benefits of nuclear power for the future, necessitates the demonstration of sustainability of the nuclear fuel cycle, specifically with regards to natural resources to (nuclear) waste, and to safe operation of the fuel.
As part of the EURATOM 7th Framework Programme (FP7), the 4 year project FAIRFUELS (FAbrication, Irradiation and Reprocessing of FUELS and targets for transmutation) was aimed to a more efficient use of fissile material in nuclear reactors by implementing the transmutation of long-living radioactive waste from actinides in GEN-IV fast neutron spectrum reactors or sub-critical Accelerator Driven Systems. Such a transmutation scheme provides a route to a reduced volume, lifetime and hazard of high-level radioactive waste by recycling the most long-lived components. In this way, the nuclear fuel cycle can be closed in a sustainable manner.
In order to implement such a sustainable, closed nuclear fuel cycle, a system of partitioning and transmutation (P&T) would be required. P&T requires a coherent approach, where all necessary steps should be compatible and integrated with one another, thus building a closed cycle. This chain is schematically represented in Figure 1. FAIRFUELS has contributed to the realization of waste lifetime reduction by investigating the material behavior under irradiation of the innovative fuels carrying the actinide waste.
In Europe, and generally speaking world-wide, one can distinguish two main approaches (or a combination thereof) towards closing the nuclear fuel cycle by means of Partitioning and Transmutation:

• Transmutation in future Gen IV fast reactor systems;
• Transmutation in dedicated sub-critical reactors, such as Accelerator Driven Systems (ADS).

The latter system can be seen as dedicated “nuclear waste burners”, whereas Gen IV reactors constitute a new generation of nuclear power reactors, directed towards a continued generation of nuclear power in a sustainable way, i.e. with a closed fuel cycle, minimisation of high-level nuclear waste, and optimal use of natural resources.

Figure 1. Schematic picture of the chain of Partitioning and Transmutation.

In accordance with the road-mapping exercise undertaken in FP6 PATEROS and the SNE-TP vision report, and in accordance with the FP7-Euratom Call Fission 2008-1-2-2, FAIRFUELS is devoted to
• Achieving advances in fuels and targets preparation at the laboratory scale along with intrinsic property determination,
• Understanding their long term transmutation behaviour,
• Addressing in detail, both experimentally and by modelling, transmutation properties of previously irradiated fuels and targets.

In fuel and target preparation and in the assessment of transmutation performance, some scientific and technological challenges are outstanding, and numerous studies are to be undertaken to assess the transmutation of minor actinides in oxide, nitride or composite (CerCer or CerMet) inert matrix fuels and targets. In Europe, MgO and (depleted) Mo metal are the preferred inert matrix materials under consideration, while ZrN inert matrices are considered as a back-up solution.

Since these irradiated fuels and targets should be compatible with subsequent dissolution (head-end steps) and separation processes, which in return should provide suitable pre-cursors for generation of fuels and targets for transmutation (conform Figure 1), FAIRFUELS addressed these challenges in a coherent way in close cooperation with other European projects in the field of Partitioning and Transmutation. Particularly, close links have been maintained with ACSEPT and ASGARD for their dissolution and partitioning programme, and with F-BRIDGE for fuel understanding, studies of fundamental properties and prognosis.

The innovative fuels studied in FAIRFUELS represent the current agreement on three main routes to a closed nuclear fuel cycle, based on implementation in Gen-IV fast reactor or ADS:

• Admixture of low amounts of actinide waste in fast reactor (mixed oxide) driver fuel, with a single (homogeneous) recycling route for the driver fuel through a process of dissolution – separation – conversion and fuel production. In this route, the actinide waste content in the fuel is kept low to reduce possible negative impact on core neutronic behavior, and the significant past experience with (U,Pu)O2 mixed oxide driver fuel can be built upon.
• Incorporation of the minor actinides in UO2-based targets located in the blankets of fast-spectrum reactors, with separate (heterogeneous) recycling chains for the driver fuel and the actinide waste bearing targets. A higher actinide waste content is required to realize transmutation effectively, and therefore the potential impact of the waste on fuel material performance is larger. On the other hand, because this fuel is located around the reactor core, conditions are milder.
• MA destruction in dedicated Accelerator Driven Systems (ADS). In an ADS dedicated to transmutation (as opposed to low-cost power production), the destruction rate of the actinide waste may be applying an ‘inert’ instead of a uranium-based matrix. An ideal inert matrix is transparent to neutrons (improving neutron economy), has high thermal conductivity (allowing for higher powers and therefore transmutation rate), and does not contain uranium or plutonium, which would reduce the net transmutation rate through the breeding of new waste.
Independent of this choice, the fuel form has widely converged on pellets, either as solid solution in oxide form or as composites between actinide oxide particles and a continuous ceramic (MgO) or metallic (molybdenum) matrix. However, sphere-pac columns may be a viable alternative for the pellet form. Sphere-pac fuel consists of a packed bed of spheres in 2 or 3 size fractions to occupy volume fractions above 80%. This particle fuel form has distinct advantages, especially as the interstitial space allows for helium and gaseous fission product release and limiting fuel swelling, thereby decreasing stresses, and mechanical interaction with the cladding, which for pellet fuel is substantial e.g. 18% in the Am-transmutation experiment EFTTRA-T4. The Sphere-Pac concept will reduce swelling and as a consequence the fuel-cladding mechanical interaction will also be reduced, which is especially advantageous for high burn-ups. The concept also could ease manufacturing, with substantial reductions in the number of processing steps, a wet (dust-free) production process and the potential for remotely handled process.

The focus of FAIRFUELS has furthermore been on the transmutation of americium. The rationale behind this choice is that the in-core behavior of plutonium, which is the first lifetime-determining component in the waste, is already reasonably well known. Following the destruction of plutonium, americium determines the lifetime of the final waste, but it introduces new issues, in processing, behavior under irradiation and reprocessing. Compared to known uranium and uranium-plutonium oxide fuels, americium-bearing fuels have significantly higher (helium) gas production, which a priori could result in increased fuel swelling as well as increased pressurization of fuel pins. Based on current knowledge, the third candidate for transmutation, neptunium, does not present additional issues beyond those of americium.
Within this context, FP7 FAIRFUELS has taken up the study of oxide, nitride and composite fuels and targets in a 4-year research programme based on the following main pillars:
• Preparation of fuels and their irradiation
• PIE of previously irradiated fuels
• Education and Training
Dedicated fuels were prepared with a high content in minor actinides using several different fuel methods. In addition, a comprehensive irradiation programme was undertaken to address the transmutation performance of these minor actinide bearing fuels and targets. These R&D activities were supported by high level fuel safety performance modelling as well as fundamental studies related to gas migration and release.
FAIRFUELS has also taken the opportunity of the availability of fuel and target samples that have been previously irradiated within other European projects, such as FP5-CONFIRM and FP6-EUROTRANS, to perform Post-Irradiation-Examinations and their modelling at an appropriate level in order to help assessing the transmutation performance and understanding key properties of nuclear fuels, such as helium gas migration, fuel swelling, evolution of thermal conductivity etc. The obtained results from post irradiation examinations provided essential information and experimental data on the behaviour of innovative fuels.
The FAIRFUELS consortium consists of ten European research institutes, universities and industry. The project started in 2009 and is coordinated by NRG. It has finished end of July 2015, after two contract extensions, related to fuel transport issues and breakdown of facilities. The project is closely related to FP7 PELGRIMM, which compares the performance of granulate (sphere-pac) transmutation fuel with that of pellet fuel.

Project Results:
3. Main S&T results/foregrounds

The structure of FAIRFUELS divides the Management and the R&D work into two Science and Technology Domains and one separate Domain for Management and Education & Training, as shown schematically in Figure 2 below.

Each of the two S&T domains deals with a specific aspect of the partitioning and transmutation chain and can therefore be mapped on the sustainability chain of Figure 1. Domain 2 (DM2) Fabrication & Transmutation Performance addresses the front end of the P&T cycle, and DM3 is devoted to experimental Post Irradiation Examinations (PIE) and Modeling in support of the PIE, giving prognosis of the dedicated fuels behaviour. Each Domain is subdivided into Work Packages, which address specific, more focused research tasks.

The proposed RTD activities of FAIRFUELS are complementary with major other FP7 programmes in the field of P&T, to allow for an effective integral approach towards safe development in FP7.
F-BRIDGE (through multiscale modelling and complementary laboratory experiments) and PELGRIMM (by comparing the irradiation behaviour of different fuel forms) focus on fuel preparation and irradiation behaviour, while ACSEPT and ASGARD study dissolution, separation and conversion.

Figure 2. FAIRFUELS organizational chart.

3.1 Fabrication and Performance (DM 2)

The deployment of partitioning and transmutation strategies be they for Gen IV or ADS, requires the preparation of excellent samples to investigate performance of the fuel under irradiation. Domain 2 “Fabrication & Performance” is therefore directed to preparation and transmutation performance. The R&D activities comprise of the synthesis and performance assessment of Am-containing fuels.
The domain addresses important issues of importance for Gen IV and ADS reactor systems, and especially focuses on the management of minor actinides. Within this approach, the Domain answers the high level goals of Gen IV, whereby increased proliferation resistance is achieved through the use of group separation leading to a homogeneously fuelled reactor core, while in addition addresses industrial and deployment issues, within a more practical approach based on the heterogeneous recycling of minor actinides in dedicated assemblies, possibly located in the reactor blanket. The latter concept is fully aligned with the deployment of Gen IV fast reactors or ADS.
Domain 2 focuses on the testing of fuel preparation methods, which is technically demanding even at laboratory scale technology through the technologically demanding fabrication of actinide-containing inert matrix targets. Specifically, the behaviour of Americium under irradiation needs further assessment, considering the potential impact of the process technique as well as whether the microstructure of the fuels/targets is able to accommodate a large production of helium. This is the subject of the MARIOS irradiation (WP2.1), of which the post-irradiation examinations will be performed in FP7 PELGRIMM.
The scientific information on actinide-target behaviour gained from MARIOS will determine in part which partitioning and preparation strategy is most promising to pursue (i.e. homogeneous reprocessing, taking out all actinides, or heterogeneous reprocessing). In this assessment, an evaluation of preparation methods is needed as well. As minor actinides are difficult to handle due to their high radioactivity, processes need to be kept as simple as possible. Both the choice of homogeneous vs. heterogeneous recycling, and the safety performance assessment of pellet and/or Sphere-Pac-type fuels, will be specifically addressed in the SPHERE fabrication and irradiation work package (WP2.2).
Domain 2 is divided in two work packages.

WP 2.1: High MA content fabrication and transmutation
The aim of the MARIOS program is to investigate both the synthesis of americium-bearing targets based on a uranium oxide matrix carrier, and the behavior under irradiation. In these targets, large amounts of helium are produced, which causes significant damage to the material under irradiation. It is the first time that americium (241Am) is included in a (natural) uranium oxide matrix – Am0.15U0.85O1.94 – to conduct an experiment in order to study the behavior in terms of helium production and swelling. MARIOS is a disk irradiation in which small samples of the material are irradiated at well-defined temperatures (1000/1200 oC) and allowed to swell freely. The temperature of the pellets is well defined by using thin pellets (1.5 mm) embedded in a molybdenum block acting as a heat bath. Besides two temperatures, also the safety performance of two different microstructures were investigated: a regular porosity structure, and a tailored microstructure resulting in increased open porosity.

Figure 3: The two microstructures of U0.85Am0.15O2-x fabricated by CEA for the MARIOS irradiation.

After determination of optimum sintering parameters, pellets of 4.5 mm diameter and 1.5 mm height were fabricated by CEA in hot cells and then cut into discs in an inert glove box environment. Pellet microstructures are shown in Figure 3. The preparation of the MA-bearing blanket materials remains difficult due to the particular thermodynamic properties of AmO2-x oxides, which exhibit a very high oxygen potential in comparison to other actinide oxides such as UO2. To study and optimize the sintering process, sintering tests were performed in Ar – 4%H2 atmosphere with a variable oxygen partial pressure.

Transport of the pellets to NRG Petten was achieved on 20th April 2010 using a leak proof double container conceived on the basis of the dose rate limitation and the risk of sample re-oxidation. At NRG, four pins with six pellets each, were assembled (as in Figure 4) and welded in a Helium atmosphere. The four pins were assembled into a sample holder, X-rayed and tested for leak tightness at JRC-IE, Petten, after which it was commissioned for irradiation in the High Flux Reactor (HFR) in Petten.

The MARIOS irradiation was then conducted in the High Flux Reactor in Petten, the Netherlands, and has run for 11 cycles from March 2011 to May 2012, with regular 3-day stops in between roughly 28-day reactor cycles, longer stops after cycles 1, 4 and 10 for scheduled reactor maintenance and reload, and several short shutdowns - reactor SCRAMs or APDs.

In conclusion, Minor Actinide Bearing Blanket material has been successfully produced following a powder processing route. Both regular and high open porosity microstructures were realized. The small pellets were irradiated in the HFR Petten, where the MARIOS irradiation was finished

Figure 4: The design of the MARIOS irradiation capsules was aimed at reaching high temperatures that are stable against power changes in the fuel pellets (pellets indicated in blue in the top image). Heat is produced by the TZM trays, and a central thermocouple is used to monitor temperatures online. successfully after 11 reactor cycles and 304 full power days. MARIOS will provide information on temperature-dependent helium release and swelling in Am0.15U0.85O2-x, a material that is representative of the blanket material foreseen for transmutation of minor actinides in sodium-cooled fast reactors. The power histories, fuel temperatures and burn-up will be reported together with results of post-irradiation examinations within the FP7 PELGRIMM project. A low-temperature (600 oC and 800 oC) counterpart to MARIOS, DIAMINO, is being prepared outside FAIRFUELS by CEA for irradiation in OSIRIS (outside FAIRFUELS). Together, these experiments will give an overview of temperature-dependent swelling and gas release in minor actinide bearing blanket material.

WP 2.2: Sphere-Pac fabrication and safety performance

There are two main objectives of this work package:
• To assess the safety performance of homogeneous (i.e. Pu-containing) fuels for transmutation
• To assess the synthesis methods and potential for improved irradiation performance of Sphere-Pac fuels compared to conventional pellet fuels
To reach these objectives, a Sphere-Pac / pellet-type irradiation experiment “SPHERE” has been performed where Am containing fuel was prepared by JRC-ITU and subsequently irradiated in the High Flux Reactor in Petten. In order to assess both the influence of Am on mixed oxide driver fuel and the expected improved irradiation performance of Sphere-Pac fuels compared to conventional pellet fuel, a dedicated irradiation experiment has been conducted in which sphere-pac and pellet fuels were directly compared. This is a first of a kind irradiation test, as MA bearing Sphere-Pac fuel has not been irradiated before.

Sphere-pac fuel has the advantage of an easier, dust-free preparation process. Sphere-pac technology (leading to production of beads), would also bring a significant simplification of the steps thanks to the elimination of some process steps as milling, pressing and grinding. Especially when dealing with highly radioactive minor actinides, dust-free processes are essential to reduce the risk of contamination. Moreover, sphere-pac fuel performance under irradiation could also provide significant advantages, thanks to a better accommodation of solid swelling (compared to pellets) through the re-arrangement of the free inter-particle areas, and in better management of the He generated during irradiation.

Two short fast reactor MOX pins have been irradiated in a sodium-filled sample holder, each containing almost the same amount of 241Am, present in one pin in the form of 10 pellets with a total length of 58 mm for the lower pin, and in the other as a stack of sphere-pac fuel of 48 mm length with spheres of two different dimension (0.8 and 0.05 mm) for the upper pin. The sphere-pac column was produced from MOX (U0.8Pu0.2O1.98) sol-gel beads in two size fractions, in which around 3% americium-241 was infiltrated before sintering, a method developed at JRC-ITU that allows to perform a significant part of fuel synthesis without the need to handle americium. Fuel pellets were produced from the sol-gel beads (without americium) obtained by the sol-gel external gelation route. This process produces very homogeneous fuels. Microstructures of the fuels are given in Figure 5.

Figure 5: Ceramographs showing microstructures of the pellets (top) and large sol-gel beads (bottom).

Irradiation of the fuel pins was conducted by NRG and JRC-IET in the High Flux Reactor in Petten, The Netherlands, and has run for 11 cycles between August 2013 and April 2015, for a total of 295 Full Power Days (FPDs). Besides a long outage of the reactor between September 2013 and April 2014, irradiation has been uninterrupted and showed no remarkable deviations from designed operational performance.

The targeted burn-up of both pins is ~5%FIMA, starting irradiation at an initial linear power of ~300 W/cm. The design of the irradiation rig, choice of irradiation position and duration of irradiation was chosen in order to accommodate these irradiation targets. Thermochemical calculations indicate that the measured temperature of 450 oC in the Mo-shroud correlates to a cladding temperature of ~520 oC during irradiation, and a maximum fuel temperature of around 2000 oC near the beginning of irradiation.
Before and after the first irradiation cycle neutron radiograms have been made, which are shown in Figure 6. Whereas the pellets only appear somwhat fractured, the sphere-pac fuel has strongly restructured and produced a central hole, indicating significantly higher central temperatures at the same linear power rating. This would have produced significantly higher release of fission gases.
In conclusion, fuel pellets and sphere-pac beds containing 3-4% of americium have been successfully synthesised by a combination of sol-gel and infiltration techniques developed at JRC-ITU, and were irradiated in the HFR Petten for a total of 295 full power days. Post-irradiation examinations and burn-up calculations will be carried out and reported within the FP7 PELGRIMM project. The burn-up calculations will also confirm the achieved power rating and accumulated burn-up.

Figure 6: Neutron radiographs of the pins with the pellet (top) and the sphere-pac (bottom) fuel column after the first irradiation cycle.

3.2 Experimental Post Irradiation Examinations and Modelling (DM 3)

Fuels to be used in Accelerator Driven Systems dedicated to Minor Actinide transmutation can be described as highly innovative in comparison with driver fuels of critical Fast neutron Reactors. Indeed, ADS fuels are not fertile, so as to improve the transmutation performance, and they contain high volumetric concentrations of minor actinides and plutonium compounds. This unusual fuel composition results in possible degraded performance under irradiation, implying the need for a fuel test irradiation program for characterization, optimization and selection.
Domain 3 of FAIRFUELS contains Post Irradiation Examination (PIE) investigations of fuels that were under irradiation in FP5 / FP6 projects, notably the CONFIRM, FUTURIX FTA and HELIOS safety irradiation test, carried out to provide scientific information about the irradiation performance of oxide, nitride and inert matrix fuels and targets. These examinations provide essential information on the safety behaviour and performance of MA-bearing fuels and targets under irradiation. The results from PIE will ultimately set the guidelines for regulating fully implementing partitioning and transmutation.
In support of the PIE results, modelling activities were also carried out. Development of Fuel performance codes, such as MACROS, and helium/fission gas behaviour, swelling and thermal conductivity modelling form an essential task to support and improve understanding of PIE results and predict fuel performance in power reactors. Vice-versa the PIE activities in FAIRFUELS provide a validation approach for existing fuel and target performance codes. In addition, modelling of the underlying physics of helium migration in the different inert matrices, crucial in assessing effects of minor actinide incorporation, was carried out. These activities help to extend the application of the current knowledge towards novel systems that have not been experimentally studied.

The results of the PIE programme of DM3 provide vital scientific information with respect to the behaviour of oxide and nitride inert matrices as well as ceramic-ceramic and ceramic-metallic composite fuels. In addition, the results highlight the most suitable form for (minor) actinide recycling.

Domain 3 is divided into three Work Packages.

WP 3.1: PIE of HFR irradiation programmes

Work package 3.1 is devoted to the Post Irradiation Examinations of the irradiation programmes, performed in the field of transmutation at the High Flux Reactor. Specifically these are the CONFIRM and HELIOS irradiations.

CONFIRM is a nitride irradiation of two Phenix-type (Pu,Zr)N pins, irradiated in the HFR Petten to a plutonium burn-up of 9.7% (88 MWd/kgHM) in 170 full power days. On the CONFIRM pin, both non-destructive (gamma spectrometry, profilometry, visual and X-ray inspection) and destructive PIE (gas puncturing, ceramography, SEM/WDS analyses) were performed at NRG.

Both pins remained fully intact. Post-irradiation examination performed at NRG and PSI showed that the overall swelling rate of the fuel was 0.92 vol-% / %FIHMA and that gap closure had occurred. Fission gas release was 5-6%, while helium release was larger than 50%. No fuel restructuring was observed, and only mild pellet cracking (Figure 7). No corrosion was visible on the inside of the

Figure 7. Optical images of the CONF-30-U sample after irradiation.

cladding, and no fuel-clad interaction zone had developed. These findings suggest that uranium-free nitrides allow for high linear power and therefore high waste destruction rates without being affected by the heavy restructuring and fission gas release of oxide fuels, but also that the larger swelling of nitrides compared to oxides may limit the overall burn-up unless the fuel-clad gap is wider to take this into account.

Almost all investigated fission products were homogeneously distributed in the matrix with respect to the local burn-up, indicating good solubility. This was also true for Xe, as no gas pores are found as in oxide fuels. The only microstructural changes of note are some grain subdivision (a sign of onset of the high-burnup structure formation also occurring for oxide fuels below 1000 oC) and the appearance of a Pu-rich phase indicating some phase separation between Pu and Zr in the nitride.

In the HELIOS irradiation, performed at the HFR, targets were specifically investigated with respect to helium gas behaviour and swelling, as well as the related microstructural changes in the fuels. Destructive PIE was carried out by JRC-ITU and NRG. Note that the original description of work assumed that the PIE on HELIOS pin 1 would be performed by CEA. However, since it proved impossible to transport the HELIOS pin 1, it was decided that NRG would undertake this activity instead.
The HELIOS test matrix consists of 5 Phenix-type pins, based on a MgO (pin1), zirconia (pin 2 and 3) and molybdenum matrix (pin 4 and 5), see Figure 8. The MgO matrix for pin 1 contains a network of open porosity in order to allow studying its impact on gas release during irradiation. Similar to the Mo matrix (pin 4 and 5), MgO forms a 2-phase composite with the actinide oxides it carries. The situation is different for the zirconia matrix of pins 2 and 3, which forms a solid solution with the actinide oxides. Plutonium was added to pins 3 and 5 to increase temperature and be able to assess its effect on helium and fission gas release. Due to the high Pu content the central temperatures and burn-ups are significantly higher in these two pins. Estimated central temperatures are shown in Table 14. Burn-up is 12,97 % FIHMA for pin 3 and 19.97 %FIHMA for pin 5. All other pins showed significantly lower burn-up of 5.11, 3.51 and 4.64 %FIHMA for pins 1, 2 and 4 respectively.
All pins were intact after irradiation as concluded from the visual inspection. No serious damage was found except for a slightly corroded area of all pins at the location of the fuel stacks. However, clear effects of both temperature (from Pu inclusion) and matrix type could be observed. The zirconia shows a fission gas release <1% at moderate irradiation temperatures, increasing to ~30% when Pu is included.

Figure 8: Post-irradiation ceramographs of HELIOS pins 1-5, from top to bottom.
At similar fuel central temperatures, the molybdenum matrix releases ~20% and ~80% of the fission gases with and without the presence of plutonium, respectively. More comparable are the helium release rates, which for both fuel types are roughly 20% at moderate central temperatures and >50% when Pu is included.
An observed issue with the molybdenum cermet fuel was significant swelling at higher operating temperatures compared to low swelling at lower temperatures. This would translate to higher chances of PCI failure during power transients. From the optical and scanning electron microscopy results it was concluded that in Pin 4 the gas is retained in about 1-10 μm size bubbles mostly inside but also around the actinide oxide particles. Much stronger bubble formation due to higher temperature had occurred in the Pu-containing pin 5, together with nearly full fission gas release. In addition, clad oxidation could be seen where oxide particles were located close to the pellet surface.
Similar effects of strongly temperature-dependent swelling were not observed for the zirconia. However, the relatively low thermal conductivity of the zirconia fuel produced more significant fuel fragmentation. The microstructure remained similar to that of the low-temperature pin 2, but in the higher power pin 3 the onset of central hole formation was observed, even though the maximum linear heat rate of 200 during irradiation, almost a factor of two less than that applied for uranium plutonium MOX fuel. Structural changes appear in the vicinity of the central region and in the mid radius range as well, where close to the central region there is evidence for bubble precipitation along grain boundaries.
The MgO composite fuel combined high fission gas and helium release with very low swelling. Magnesia (MgO) was chosen because of its high resistance to irradiation damage and amorphisation. Moreover, tailored porosity was included in the MgO matrix, to provide channels for helium release and accommodate the swelling. A relatively high temperature is chosen for this Pu-free fuel on the basis of the following two criteria:
• The possibility of annealing of irradiation damage in the target.
• The release of helium outside the americium compound particles.
From the results of post-irradiation optical and electron microscopy it was concluded that the fuel particles were well fixed within the matrix. Neither sintering, cracking nor significant swelling of the pellets were observed. The gap between the pellet and the cladding was preserved, and the cladding was in good shape with no signs of strong chemical interaction. In a few instances presence of a new different phases at the fuel particles was observed. One type of the “new phase” looked like metal droplets, the other type had a very porous structure. The WDS mappings showed that in both cases the suspected new phase had the fuel particle composition.

A tentative conclusion from the CONFIRM and HELIOS irradiations is that the MgO and ZrN matrices present the most promising candidates for actinide transmutation at higher destruction rates, combining practically unrestructured fuel with high helium and moderate fission gas release. MgO unlike ZrN additionally features low swelling, but for MgO irradiations at higher linear powers are needed for further qualification. The strongly temperature dependent swelling behavior of molybdenum based matrices, taking place due to local bubble growth in and around the oxide particles, suggests an increased probability for PCI-induced fuel failure during power ramps, although this also depends on the creep behavior of the molybdenum at those temperatures. Zirconia based inert matrix fuels tend to restructure already at lower powers, which could limit their potential to reach high destruction rates safely. It should be noted that ramp testing at burn-up would be needed to further qualify the investigated fuels.

WP 3.2: Futurix FTA Post Irradiation Examinations

This Work Package addresses the Post-Irradiation Examinations (PIE) of Accelerator Driven Systems (ADS) fuels tested in representative irradiation conditions in the fast reactor Phénix. The FUTURIX FTA irradiation was loaded in the fast reactor Phénix on May 2007 to March 2009 for an irradiation duration of 238 EFPD. This test aimed at comparing the behaviour of several innovative fuel types under similar irradiation conditions in the PHENIX reactor. Part of the programme (dealing with molybdenum-based CERMET and MgO-based CERCER fuels) was performed under the umbrella of the European project FP-6 EUROTRANS. Post-irradiation examinations on CERCER and CERMET fuels have been implemented within the FP-7 FAIRFUELS European project. A transport of the FUTURIX-FTA CERMET capsules from CEA to ITU has been performed in this framework, to allow the study of the Mo-based fuel by JRC-ITU.

The purpose of the FUTURIX-FTA irradiation experiment was to increase the knowledge and understanding of the irradiation behaviour of low fertile and uranium-free fuels containing significant quantities of minor actinides for transmutation in fast reactor systems. This experiment provides essential data concerning behaviour under irradiation and allows qualification and validation of models developed to predict fuel performance.

During irradiation, pins 5 and 6 carrying molybdenum cermet fuels reached burn-ups of 18 and 13% FIHMA, while pins 7 and 8 carrying the MgO-based cercer fuels reached burn-ups of 8.8 and 5.6% FIHMA, all at linear heat rates of around 100 W/cm.

The post-irradiation examinations of the cermet pins 5 and 6 revealed that no significant pin length or diameter change had occurred, indicating that the fuel-clad gap was not closed. However, significant fuel swelling was observed: around 15 vol-% for pin 5 against 9.5 vol-% for pin 6, in both cases around 0.8 % per % burn-up. A fractional gas release of circa 30 % was measured for both pins, while little migration was observed for the volatile fission product cesium.

Pins 7 and 8 containing MgO-based fuel revealed significantly lower fission gas release, below 1%, combined with a significantly lower swelling rate. Average swelling rate was 3.2 % and 3.1% as obtained from geometrical density measurements for pin 7 and 8, respectively, which is equal to 0.35 and 0.56 % per % burn-up. As in the case of HELIOS, porosity remained confined to the actinide oxide phase, leaving the MgO matrix almost undamaged (Figure 9). This explains the difference in swelling with the molybdenum-based matrix, which is affected relatively strongly by gas bubble formation.

WP 3.3: Modeling in support of PIE

A significant parallel effort to the post-irradiation examinations has been made to model material behavior to correlate experimental information from the PIE campaigns in WP3.1 and WP3.2 to physical mechanisms, in order to improve prediction power for these innovative fuel types.

In particular, helium is produced in large amounts in fuels and targets containing minor actinides, and the release rate of helium is essential to the performance of the fuel in terms of swelling, mechanical integrity and change of gap conductance. The relation between diffusion of a.o. helium (at the atomic level) in the matrix and coalescence at pores and bubbles and on the other hand the effects the effects of this diffusion behaviour on the macroscopic scale swelling and release have been investigated.

Figure 9: Ceramography of FUTURIX-FTA pin 7. (top) optical macrograph, (bottom left) a more detailed image of microstructure, (bottom right) detail of a Pu0.5Am0.5O1.88 particle inside the MgO matrix.

Imperial College London (ICL) contributed to this effort by progressing in the development of multiscale modelling of properties and gas migration in actinide oxides. Multiscale modelling encompasses simulation techniques that operate from the atomic to models that describe entire fuel sub-assemblies. An essential aspect of the multiscale concept is the ability to successfully propagate information between different time and length scales.

In recent years, a substantial effort has been devoted to advancing the understanding of nuclear fuel behaviour under irradiation through modelling on atomistic scale. A special phenomenon occurring in minor actinide bearing fuels and targets is the very high production of helium gas resulting from alpha-decay of Cm-242.

Thermo-mechanical simulation codes model gas migration using effective diffusion coefficients. These may either be derived from experiments or from multi-scale modelling. In the case of UO2, grain boundary diffusion coefficients of Xenon recently were derived from molecular dynamics simulations, employing empirical potentials. In a similar manner, diffusion coefficients for helium in minor actinide bearing oxides can be obtained. The prerequisite for such simulations is a set of empirical potentials for the actinide oxides themselves. For UO2, several potentials have been published in the literature, albeit their ability to reproduce properties like thermal expansion and elastic constants as function of temperature is of varying quality. Hence, Imperial College focused their work in FAIRFUELS on creating a set of potentials for actinide oxides that could provide a better description of these compounds, prior to introducing helium into their framework.

MD simulations were then performed with the LAMMPS code in the NPT ensemble. The resulting temperature dependence of lattice parameter and bulk modulus is in reasonable agreement with experimental data, where such are available. The potentials obtained by Imperial should now be tested for defect formation and diffusion properties, in order to pave the way for implementation of potentials describing the interaction with noble gases.

KTH from Stockholm has studied how the stability of voids in molybdenum (Mo) and magnesia (MgO) are influenced by helium formation using a range of computational techniques, including rate theory, density functional theory and molecular dynamics.

When Cm-242 decays, the emitted alpha-particles may exit the fuel particles and enter the surrounding matrix. Once there, they interact with voids which form due to neutron irradiation. Thus, the ”conventional” swelling of neutron irradiated metals and ceramics is affected by the presence of helium. It was studied how the stability of voids in molybdenum and magnesia are influenced by helium using a range of computational techniques, including rate theory, density functional theory and molecular dynamics.

The rResults may be interpreted in the context of the BODEX and HELIOS irradiations. For the BODEX irradiation, carried out within the EUROTRANS project, boron (10B) doped molybdenum and magnesia pellets were neutron irradiated at temperatures roughly equal to 1073 and 1473 K, as 10B is a strong helium producer through the (n,α) reaction, mimicking the effect of inclusion of americium. At the lower temperature, the measured swelling of molybdenum was limited to less than 2%, whereas at 1473 K, the volume increase exceeded 8%. These results are similar to the observations made on HELIOS, confirming that the behavior is dominated by the helium effect. The calculations by KTH indicate also that the incubation threshold for swelling at the lower temperature might not have been reached.
A detailed calculation of the radial burn-up profile of the MARIOS irradiation was carried out by CIEMAT as part of the modelling work package in FAIRFUELS. The modelling was performed using EVOLCODE, which couples MCNPX with ORIGEN or the ACAB fuel depletion code. Since the MARIOS irradiation was carried out in a thermal spectrum, the transmutation rate of americium is higher in the periphery of the samples, where also the major fraction of the fissions occurring in the samples would take place. One therefore expects that helium production rates are higher in the periphery. These calculations help interpret the results of the MARIOS post-irradiation examinations performed in the P7 PELGRIMM project.

Model fuel safety performance code analysis was performed by SCK on the five fuel pins from the HELIOS irradiation as well as one MgO-based pin from the ECRIX-H irradiation, which was carried out under a locally moderated neutron flux in the Phénix sodium-cooled fast reactor (SFR) for 318 Effective Full Power Days (EFPD). Burn-up of the micro-dispersed fissile fuel particles in the ECRIX-H pin was half an order of magnitude larger compared to HELIOS-1 pin, i.e. ~ 25 % against 5.1 % FIMA (pellet average).

The full-scale fuel safety performance code MACROS developed at SCK-CEN was used to simulate in-reactor properties of fuel targets and behaviour of encapsulated fuel pins. This code is verified and already used in a number of European projects including OMICO, FUTURE, EUROTRANS. Within the scope of the project, first a review of the correlations and recommendations developed for the main thermophysical properties of the oxides of interest and selected inert matrix materials was carried out, and a set of empirical or semi-empirical correlations for estimation of these properties (thermal expansion, density, heat capacity, thermal conductivity) was proposed. Detailed material data for the inert matrices (MgO, YSZ and Mo) have been collected. Effects of porosity and, when available, of neutron irradiation (f.i. degradation of thermal conductivity for MgO) were included.

Following this review, all five HELIOS fuel pins and the ECRIX-H pin have been analyzed and modeled. The scope of in-reactor behavior modeling includes calculations and evaluations of material properties and their evaluation during in-reactor period; temperature calculations, calculations of the observed axial and radial anisotropic swelling of HELIOS-2 and HELIOS-3 fuel pins; and full analysis of fission gas release for all five rods and comparison with available data.
Calculated and measured centre-line temperatures agree reasonably well with experimental data within a 50 0C band. Analysis of fission gas release showed good agreement with measurements except for the HELIOS Pin-4, where the MACROS code models under-predicted integral release of Xe and over-predicted release of helium. Code calculations also confirmed the observed high temperatures in the HELIOS Pin-5, which could be explained by an early start of fission gas release and contamination of the fuel-to-clad gap interior. In addition, swelling and gas release results from these calculations could provide more insight in underlying effects, as may be clear for instance from the breakdown of gas release shown in Figure 10. The model analysis of the HELIOS and ECRIX-H pins behavior in thermal and fast flux irradiation conditions also shows the significant role of helium in terms of swelling.

Figure 10: HELIOS Pin-1 global balance of fission Xe (top) and He (bottom) retention and release.

Finally, comparisons between the different concepts of inert matrix fuels using the fuel performance code showed that MgO-based targets and YSZ-based target all in all demonstrated better in-reactor behaviour and better structural stability than that of molybdenum-based composite targets.

Potential Impact:
4. Impacts and dissemination/exploitation activities
4.1 Related events
The consortium was represented at the SNETP general assembly during a poster session on 20 September 2010.
The scientific results of FAIRFUELS were presented during scientific conferences, such as Nuclear Materials 2010 (Karlsruhe, October 2010) and Global 2011 (Makuhari, Japan, December 2011) and 2015 (Paris, September 20-24).
The Fairfuels workshop was organised on 9 and 10 February 2011 the FAIRFUELS Workshop was held in Stockholm and 7 oral presentations were related to Fairfuels topics.
Contribution to the conference: Actinide and Fission Product Partitioning and Transmutation - 12th Information Exchange Meeting (2012) p. 210-222 was assured by the presentation "Preliminary results of the MARIOS experiment on minor actinide bearing blanket concept" prepared by E. D'Agata; R. Hania; S. Bejaoui; C. Sciolla; T. Wyatt; M. Hannink; N. Herlet; A. Jankowiak; F. Klaassen; J-M. Bonnerot;
The ASGARD-FAIRFUELS-CINCH winterschool was held in Petten on 29-30 January 2013.
The Final Fairfuels workshop took place in Alkmaar on 19-20 May 2015 and hosted 10 topic specific presentations. Representatives of the project End user group were invited.

4.2 Related publications
P.R. Hania, J. Wallenius, F. Ingold, B. Wernli, J.A.N.C. Verbruggen, S.C. van der Marck, S. Knol, F.C. Klaassen, Irradiation of Uranium Free Nitride Fuel at High Linear Power in the Petten High Flux Reactor, Poster at Nuclear Materials 2010, Karlsruhe, 4-7 October 2010

D. Prieur, A. Jankowiak, T. Delahaye, N. Herlet, P. Dehaudt, P. Blanchart, Fabrication and Characterisation of U0.85Am0.15O2-x discs for MARIOS irradiation program, J. Nucl. Mat. 414 (2011), 503-507

O. Runevall and N. Sandberg. Self-diffusion in MgO, a density functional study, J. Phys.: Condens. Matter, 23 (2011), 345402.

Syriac Bejaoui, Elio d’Agata, Ralph Hania, Thierry Lambert, Stephane Bendotti, Cédric Neyroud, Nathalie Herlet, Jean-Marc Bonnerot, Americium-bearing Blanket Separate-effect Experiments: MARIOS and DIAMINO Irradiations, Proceedings of GLOBAL 2011, Paper no. 503621, Makuhari, December 11-16, 2011

O. Runevall and N. Sandberg, Helium induced void and bubble formation in MgO,
Computational Materials Science 60 (2012) 53.

E. d’Agata, P.R. Hania, S. Bejaoui, C. Sciolla, T. Wyatt, M.H.C. Hannink, N. Herlet, A. Jankowiak, F.C. Klaassen, J.-M. Bonnerot, MARIOS: Irradiation of UO2 containing 15% americium at well-defined temperature, Nucl. Eng. & Design 242 (2012), 413-419

O. Runevall, Helium Filled Bubbles in Solids – Nucleation, Growth and Swelling, PhD thesis, KTH (2012).

E. D'Agata; R. Hania; J. McGinley; J. Somers; C. Sciolla; P. Baas; S. Kamer; R. Okel; I. Bobeldijk; F. Delage; S. Bejaoui, "SPHERE: Irradiation of sphere-pac fuel of UPuO2 containing 4% Americium", Nuclear Engineering and Design (2014) vol. 275 p. 300-311.

J. Wallenius, K. Bakker, Ch. Ekberg, A. Geist, R. Hania, E. Slooten & E. de Visser-Tynova: Minor Actinide Bearing Fuels, Second edition (June 2015), LeadCold Books, Sweden, ISBN 978-91-980272-1-2

G. Cécilia, F. Delage, J. Lamontagne,L. Loubet, Post-Irradiation Examination results for MgO-CERCER Am-bearing fuels irradiated in PHENIX within the frame of the FUTURIX-FTA experiment, Proceedings of GLOBAL 2015, Paper 5059

P.R. Hania, F.C. Klaassen, B. Wernli, M. Streit, R. Restani, F. Ingold, A.V. Fedorov, J. Wallenius, Irradiation and post-irradiation examination of uranium-free nitride fuel, accepted in Journal of Nuclear Materials, August 2015

Website address:

EC project officer: Coordinator:
Georges van Goethem Ralph Hania
European Commission Nuclear Research & consultancy Group (NRG)
Directorate-General for Research Westerduinweg 3, P.O. Box 25
Directorate Energy (Euratom) 1755 ZG PETTEN, the Netherlands
Unit J.2 – Fission Tel. +31 224 564131
CDMA 1/47 Fax: +31 224 568608
B-1049 Brussels Email:
Tel: +32 2 295 14 24

List of Websites:

Related information

Reported by



Nuclear Fission
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top