Safety assessment of nuclear fuel: thermodynamic investigation of the (Ba-Mo-O) and (Cs-Te-U-O) fuel-fission product systems

Transcription

1 Safety assessment of nuclear fuel: thermodynamic investigation of the (Ba-Mo-O) and (Cs-Te-U-O) fuel-fission product systems Duration: 6-9 months Light water reactors (LWRs) are currently the reference in the nuclear industry, but do not provide a fully satisfying response to the social-political concerns. To replace these reactors at the end of their operating licenses, Generation IV reactors are currently being developed. In Europe, the research has been focused mainly on two designs: the sodium-cooled fast reactor (SFR) and the lead (and lead-bismuth)-cooled fast reactor (LFR). The safety of a nuclear reactor is directly linked to the performance of the fuel, which is at the heart of the energy production process. UO 2 is the most common fuel used in LWRx, whereas the (U,Pu)O 2-x mixed oxide fuel (MOX) is currently the reference for the SFR and LFR. The nuclear fuel is subjected to extreme conditions of temperatures, gradients and radiation damage. In addition, the fission reactions lead to the formation of fission products (FPs), making the nuclear fuel an extremely complex multi-component system. Several questions are still open on the thermodynamic properties of the fuel-fission product systems in LWRs. For instance, microprobe analysis of irradiated LWR fuels have revealed the formation of the so-called grey phase with the general formula (Ba 1-xySr x Cs y )(Zr,U,Pu,Mo)O 3, whose thermodynamic properties are not well known. Moreover, little is known on the MOX behaviour in FNRs, for which several differences can be expected in comparison to LWRs. For instance, irradiation tests in prototype fast neutron reactors (for example Phenix, in France) have highlighted the formation of a fission product layer between the fuel pellets and the cladding, called Joint-Oxide-Gaine or JOG, never observed in LWRs. This is due to the higher temperatures and gradients of the fast neutron reactors: fission products such as Cs, Te, I, Mo migrate from the hot inner of the pellets toward its colder periphery and there accumulate, forming the JOG. This leads to the formation of new phases, with different thermal properties (as melting temperature, thermal conductivity etc.) than the (U,Pu)O 2 fuel. Therefore, the investigation of the JOG thermodynamic properties- which are currently mostly unknown- is fundamental for the safety assessment of the fuel for fast neutron reactors. In this research project, we will look at the structural and thermodynamic properties of three different systems: Ba-Mo-O, Cs-Te-O and the quaternary Cs-Te-U-O system. The Ba-Mo-O and Cs-Te-O systems will be investigated by combining experimental studies and thermodynamic modelling. The experimental work will start with the solid state synthesis of the ternary compounds expected in the reactor conditions and their characterization using X-ray diffraction, neutron diffraction and, possibly, X-ray Absorption Spectroscopy at the European Synchrotron Radiation Facility (Grenoble, France). Phase transitions will be studied using High temperature X-ray diffraction and Differential Scanning Calorimetry. Enthalpy of formation measurements will be performed by solution calorimetry. In parallel, a thermodynamic model of the Cs-Te-O system will be developed using the CALPHAD method (with the Thermocalc software). For the Ba-Mo-O system, an uncertain thermodynamic model (based on few experimental data) is available and thermodynamic computations will be performed and compared to the experimental data obtained in this work. Finally, the study of the quaternary Cs-Te-U-O systems will be approached with experiments aiming to determine the solubility of tellurium in Cs uranates.

2 Disciplines covered: Materials Science, Nuclear materials, Physical Chemistry, Thermodynamics Contact details: Dr. Enrica Epifano, Dr. Anna L. Smith Tel: Reactor Physics and Nuclear Materials (RPNM) Radiation, Science & Technology Department Faculty of Applied Sciences Delft University of Technology Mekelweg 15, 2629 JB Delft - The Netherlands

3 Study of the oxide fuel-cladding chemical interaction in a fast breeder reactor Duration: 6-9 months The chemical interaction between (U,Pu)O 2 mixed oxide fuel, fission products and cladding materials is one of the major factor limiting the integrity and lifetime of a fuel pin in a fast breeder reactor. Type 316 stainless steel is currently the reference for fast reactors such as the Sodium-cooled Fast Reactor (SFR). During irradiation, the highly corrosive fission products cesium (Cs), tellurium (Te), and iodine (I) are generated in the fuel pins, which are particularly volatile. These elements are gases at the fuel temperature and migrate radially and axially to the colder temperature regions (towards the cladding) where they accumulate in the gap between fuel and cladding. The generation of Cs, Te and I is the main cause of attack of the stainless steel cladding. Understanding the chemistry of the interaction between Cs, Te and I and the cladding depending on conditions of temperature and oxygen potential as well as the corrosion mechanisms is essential for the safety assessment of the reactor during operation and accidental conditions. The effects of Cs and I are fairly well-understood but there are little data on the role of tellurium in the chemical attack. In this research project, we will look at the structural and thermodynamic properties of the ternary phases formed in the M-Te-O (M=stainless steel components Fe,Cr,Ni,Mo,Mn) and Cs-M -O systems (M =Te,Cr,Mo,Ni,Mn). The work will involve solid state synthesis and characterization using X-ray and neutron diffraction, Mössbauer spectroscopy, the study of the compounds thermal expansion using high temperature X-ray diffraction, as well as calorimetric measurements (Differential Scanning Calorimetry and Thermal-relaxation calorimetry). In addition, the experimental results will be coupled with thermodynamic calculations using the Factsage and/or Thermocalc softwares. The work will be done at the Delft University of Technology (TU Delft, The Netherlands) partly in the Department of Radiation Science & Technology ( and partly in the Department of Materials Science & Engineering ( Disciplines covered: Materials Science, Physical Chemistry, Thermodynamics Contact details: (1) Dr. Anna L. Smith Tel: a.l.smith@tudelft.nl Reactor Physics and Nuclear Materials (RPNM) Radiation, Science & Technology Department Faculty of Applied Sciences Delft University of Technology Mekelweg 15, 2629 JB Delft - The Netherlands (2) Prof. Jilt Sietsma Tel: j.sietsma@tudelft.nl

4 Materails Science & Engineering Department Delft University of Technology Mekelweg 2, 2628 CD Delft - The Netherlands

5 Physico-chemical properties of molten salt fuel for MSRs Duration: 6-9 months The molten salt reactor (MSR) was selected as one of the promising designs by the International Generation IV Forum for the next generation of nuclear reactors. Running on a liquid molten salt fuel as opposed to the current generation of nuclear reactors, the Molten Salt Reactor technology provides a safe and truly innovative concept. Moreover, it can be coupled to a thorium fuel cycle, which produces much less long-lived radioactive waste and allows a more sustainable energy production, as thorium is three times more abundant on Earth than uranium. However, before the Molten Salt Reactor technology can be realised, a thorough safety assessment of all components of the reactor must be carried out. In particular, understanding the chemistry of the liquid fuel and acquiring a thorough knowledge of its physical and chemical properties such as density, viscosity, heat capacity, thermal conductivity and vapour pressure, is essential. The proposed research will look at the physico-chemical properties of the molten salt fuel and will involve both experimental studies and theoretical modelling. The experimental investigations will include solid state synthesis, characterisation using techniques such as X-ray and neutron diffraction, measurement of thermodynamic properties using calorimetric techniques, as well as thermodynamic calculations using the Factsage software. The work will be carried out at the radiological laboratories of the Reactor Institute Delft. Disciplines covered: Materials Science, Physical Chemistry, Thermodynamics Contact details: Dr. Anna L. Smith Tel: a.l.smith@tudelft.nl Reactor Physics and Nuclear Materials (RPNM) Radiation, Science & Technology Department Faculty of Applied Sciences Delft University of Technology Mekelweg 15, 2629 JB Delft - The Netherlands

6 Study of radionuclides transport behaviour in clay (Chemistry) (BEP/MEP) Denis Bykov The project is a part of the Dutch research programme on geological disposal of radioactive waste called OPERA (Onderzoeks Programma Eindberging Radioactief Afval). Clay and salt formations are considered as potential host rocks in OPERA. In particular, the feasibility of Boom clay is being investigated both in Belgium and in the Netherlands (Boom is a municipality located in Flanders, one of the three regions of Belgium, and in the Flemish province of Antwerp). Boom Clay has excellent characteristics as main barrier against the release of radionuclides: a low hydraulic conductivity, a reducing redox potential, a slightly alkaline ph, a high cation exchange capacity and a high plasticity. Due to these properties, migration in clay is extremely slow. Therefore, in the current project we use electric field as an efficient mean to accelerate migration (the process is called electromigration). Our recent research showed that transport properties of different radionuclides in Boom clay are influenced by the copresence of other cations. This phenomenon becomes important in view of the complexity of radioactive waste compositions. Ceramic materials for radioactive waste disposal (Chemistry) (BEP) Denis Bykov Ceramic materials based on the structures of natural minerals draw attention as an alternative to glass for radioactive waste disposal. Ceramics generally surpass glass with respect to chemical, radiation and thermal stability. A variety of natural minerals that can potentially accommodate nuclear waste components are mainly based on oxides, phosphates, silicates and the examples of mineral types are represented by monazite, langbeinite, kosnarite, whitlockite, pollucite and other structure types. The elements that potentially can be immobilized in ceramics as isolated waste fractions are represented by the fission products (for example, Cs, Sr, I, lanthanides) and the transuranium elements. Thorough investigation of new materials is needed in order to understand their suitability for fixation of radiotoxic components. It should be noted that in most of the cases for the research purposes radioisotopes can be replaced by their stable analogues. For the characterization of materials structure and properties we apply a number of analytical techniques such as X-ray diffraction, infrared spectroscopy, thermal analysis, Mössbauer spectroscopy, scanning electronic microscopy and others. Our multidisciplinary research is at the intersection of materials science, solid state chemistry, radiochemistry and crystallography. Reconstruction of pore structure in concrete (Physics/Chemistry) (BEP/MEP) Denis Bykov Concrete and clay play important role in the deep geological disposal of nuclear waste. In the Netherlands, the Boom clay formations have been identified as possible final destination of the reprocessed radioactive waste. Concrete has several functions in the underground shafts, such

7 as encapsulation of waste canisters, backfilling and lining of the galleries. The research activities nowadays are aimed at building a strong safety case in order to ensure safe final storage of radioactive waste. One of the research questions addresses concrete-clay interface. During concrete degradation, highly alkaline waters from the concrete will diffuse through the clay barrier. Pore waters of concrete have high ph (12.5 to 13.4) compared to the ph of pore waters in a clay (7 to 8). This may produce structural transformations in the clayey materials, cause dissolution/precipitation of minerals in the clay/concrete interface and, consequently, lead to changes in the clay properties. This research project is a part of the studies of interface between concrete and clay and will be focused on the new ways to characterize the porosity in concrete. This will be done with a strong X-ray source and CT-scanning devise available at RID. Moreover, a collaboration with the University of Groningen will be established to investigate size distribution of nanosized pores by small-angle X-ray scattering (SAXS), which is a Dutch SAXS facility. The goal of the project is to optimize a 3D-modelling of the pore structure. The concrete samples to be investigated will be supplied by the Dutch nuclear waste management organization (COVRA). Dissolution and analysis of molten salt simfuel in support to irradiation experiments (Chemistry) (BEP) Elisa Capelli Molten Salt Reactors, which are a promising type of future fission reactors, utilize a liquid molten salt mixture as nuclear fuel. The typical composition of the MSR fuel, which depends on the specific reactor design, is a mixture of alkali/alkali-earth fluoride (e.g. LiF, BeF2,..) and actinide fluorides (e.g. ThF4, UF4,..). Although the salt composition is accurately controlled to ensure reactor safety, new elements are inevitably formed during reactor operation as consequence of the fission reaction. With this respect, irradiation experiments are essential as they can provide invaluable information on the behaviour of salt under irradiation and production of fission products. In order to support irradiation experiments and accurately determine the fission product concentrations in the salt after irradiation, a method for quantitative analysis of irradiated salt is required. A suitable method for quantitative analysis is also vital for issues related to the processing and storage of radioactive waste. At present, such a method is not available. In this project, the dissolution of simfuel salt will be studied and a procedure for analysis, for example by means of ICP-OES, will be developed. The potential of alternative analytical techniques will also be explored.

8 Determination of noble metals particle size in molten fluoride salts (Chemistry) (BEP) Elisa Capelli Molten Salt Reactors, which are a promising type of future fission reactors, utilize a liquid molten salt mixture as nuclear fuel. Since fluoride salts are one of the best option as candidate salts, their physico-chemical properties are under investigation since many years. However, little information is available on the behaviour of salt containing fission products. The present work focus on a class of fission products, the so called noble metals of which Mo, Nb, Te and Ru are the most important contributors. This elements are expected to be reduced to the metallic state under the reducing conditions which would be normally mantained in the reactor. For fission product management purposes, an important parameter to be determined is the size of these metallic precipitates. During the project, different ways of dissolving noble metals in fluoride mixture will be explored, such as direct dissolution or reduction of noble metal fluorides. The particles size will be determined, for example by means of SEM/EDX, and correlated with the most relevant experimental parameters. Fission product extraction in MSR (BEP/MEP) Elisa Capelli The Molten Salt Reactor is an innovative type of nuclear reactor that meets the Generation IV goals in terms of safety, reliability, proliferation resistance and economics. The very unique feature of this reactor is the liquid nature of the fuel, which is composed by a mixture of molten fluoride salts. One of the advantages of having a liquid fuel is the possibility of controlling and extracting the fission products while operating the reactor. Helium bubbling is one of the reprocessing steps and is used to separate the gaseous fission products from the fuel in order to keep good neutronic properties. Furthermore, this process has the potential to remove via flotation the insoluble particles, such as noble metals, that might otherwise deposit on the metallic surfaces. In the current project, the efficiency of particles removal via flotation will be assessed. The aim is to use the experimental setup available at TU Delft to study the underlying mechanisms of the process and understand the dependence of the extraction efficiency with the process parameters (e.g. bubble size, particle properties, fluid properties). Finally, simulation work could also be performed to support the experimental results and optimize the process design. Fluorine-containing waste forms for the needs of future energy technologies (Chemistry) (BEP/MEP) Denis Bykov The Molten salt reactor (MSR) has been identified as one of the six Generation IV reactor technologies to serve future energy needs of the human society. The concept uses a molten salt

9 as primary coolant in which the fuel is dissolved. A mixture of molten fluoride salts is typically proposed, for example LiF-BeF2-ThF4-UF4 as fuel for thermal reactor concepts and LiF-ThF4 for fast reactor concepts. But also other combinations of solvents are investigated. A possibly good neutron economy makes the MSR attractive for the neutron-poor thorium thermal fuel cycle, based on abundant fertile isotope 232Th. In the MSR reactor design an on-site clean-up of fuel is foreseen, i.e. purification of the salt by removing fission products and actinides produced by fission. Nonetheless, volumes of spent contaminated salts are expected to be accumulated at the end of operation. In view of the existing approach for the geological disposal of nuclear waste, a durable chemical form for the waste containing the fluoride salts must be developed. The proposed research project is focused on the fabrication and characterization of properties of glass and glass-ceramic materials, based on fluorophosphates glasses and natural minerals like sodalite, apatite etc. An optimal synthesis route for the novel materials needs to be developed, while their properties must provide long-term chemical, thermal and radiation stability. Investigation of alkali and alkaline-earth fission products in oxide fuel: safety assessment of fast breeder reactors (Chemistry (BEP/MEP) Anna Smith Alkali and alkaline-earth metals such as cesium (Cs), rubidium (Rb), potassium (K), lithium (Li), barium (Ba), strontium (Sr), calcium (Ca) and magnesium (Mg) are generated during irradiation in fast breeder reactors, which makes the chemistry of the nuclear fuel particularly complex. (U,Pu)O2 mixed oxide fuel is currently the reference for fast reactors such as the Generation IV Sodium cooled Fast reactor (SFR). With a view to improve the sustainability of the fuel cycle, the incorporation of minor actinides elements (Np,Am,Cm) to the fuel is also envisaged, corresponding to the composition (U,Np,Pu,Am,Cm)O2. Some of the fission products generated during burn-up are soluble into the fuel matrix, while others form oxide precipitates depending on conditions of temperature and oxygen potential in the reactor. Ternary alkali and alkaline earth actinide oxide phases can be formed with lower density and thermal conductivity than the mixed oxide fuel. This leads to fuel swelling and temperature increase, which affects the fuel behaviour. Understanding the chemistry of the interaction between fuel and fission products, in particular the conditions required for the formation of the ternary phases, their thermal stability and thermodynamic properties, is essential for the safety assessment of the reactor during operation and accidental conditions. In this research project, we will look at the structural and thermodynamic properties of the ternary phases formed with uranium in the (Cs,Rb,K,Li,Ba,Sr,Ca,Mg)-U-O systems. The work will involve solid state synthesis and characterization using X-ray diffraction and neutron diffraction, the study of the compounds thermal expansion using high temperature X-ray diffraction, as well as calorimetric measurements (Differential Scanning Calorimetry and Thermal-relaxation calorimetry) and possibly Knudsen effusion mass spectrometric measurements. In addition, the experimental results will be coupled with thermodynamic calculations using the Factsage and/or Thermocalc softwares.

10 Study of the oxide fuel-cladding chemical interaction in a fast breeder reactor (Chemistry) (BEP/MEP) Anna Smith The chemical interaction between (U,Pu)O2 mixed oxide fuel, fission products and cladding materials is one of the major factor limiting the integrity and lifetime of a fuel pin in a fast breeder reactor. Type 316 stainless steel is currently the reference for fast reactors such as the Sodium-cooled Fast Reactor (SFR). During irradiation, the highly corrosive fission products cesium (Cs), tellurium (Te), and iodine (I) are generated in the fuel pins, which are particularly volatile. These elements are gases at the fuel temperature and migrate radially and axially to the colder temperature regions (towards the cladding) where they accumulate in the gap between fuel and cladding. The generation of Cs, Te and I is the main cause of attack of the stainless steel cladding. Understanding the chemistry of the interaction between Cs, Te and I and the cladding depending on conditions of temperature and oxygen potential as well as the corrosion mechanisms is essential for the safety assessment of the reactor during operation and accidental conditions. The effects of Cs and I are fairly well-understood but there are little data on the role of tellurium in the chemical attack. In this research project, we will look at the structural and thermodynamic properties of the ternary phases formed in the M-Te-O (M=stainless steel components Fe,Cr,Ni,Mo,Mn) and Cs-M -O systems (M =Te,Cr,Mo,Ni,Mn). The work will involve solid state synthesis and characterization using X-ray and neutron diffraction, Mössbauer spectroscopy, the study of the compounds thermal expansion using high temperature X-ray diffraction, as well as calorimetric measurements (Differential Scanning Calorimetry and Thermal-relaxation calorimetry). In addition, the experimental results will be coupled with thermodynamic calculations using the Factsage and/or Thermocalc softwares. The work will be done at the Delft University of Technology (TU Delft, The Netherlands) partly in the Department of Radiation Science & Technology ( and partly in the Department of Materials Science & Engineering ( Reactor Physics and Nuclear Materials (RPNM) research group: brief descriptioin The research section Reactor Physics and Nuclear Materials (RPNM) of the Department Radiation Science & Technology (RST) is engaged in research activities in the field of nuclear reactor technology, including the solid state chemistry and thermodynamics of nuclear materials, as well as the analysis and design of innovative nuclear reactor systems to improve the safety, economy and sustainability of nuclear energy. The research focuses on Generation-IV systems and beyond, among which the LFR, SFR, MSR, Accelerator Driven Systems, VHTR, GCFR, SCWR. Our research areas encompass the safety assessment of nuclear fuels (fuel performance and chemical compatibility with coolant), the management of fission products (chemical state and transport behaviour), and the development and analysis of waste materials. Our activities are focused on the coupling between experimental investigations (using X-ray and Neutron Diffraction, Calorimetry, X-ray Absorption Spectroscopy) and thermodynamic modelling using the CALPHAD method.

11 Infrastructure / technical equipment The NERA research group has a wide range of experimental laboratories and facilities, including radiological laboratories at the Reactor Institute Delft ( for the experimental work with uranium and thorium. Moreover, our close collaboration with the Reactor Institute Delft (2MW experimental neutron research reactor) allows us access to the reactor and its irradiation facilities, in particular the neutron diffraction PEARL beamline

12 Assessment of uncertainty in CALPHAD predictions for Molten Salt Reactor fuel (BEP) Understanding and predicting nuclear fuel behaviour in innovative type of reactors, such as the Molten Salt Reactor, is a key step for their future development and deployment. The CALPHAD (CALculation of PHAse Diagrams) method is a proven and indispensable tool in this endeavour and it is currently used to generate customized thermodynamic databases. Based on the developed models, thermochemical equilibrium calculations are performed allowing the prediction of the thermodynamic properties and the evaluation of the safety margins for the potential fuel mixtures. For reactor designer and regulatory bodies, knowledge of the uncertainty associated with the results from CALPHAD is crucial. Currently however, the reliability of these predictions is only inferred by comparing the calculated results to experimental phase equilibria data, resulting in a region of confidence which is limited to known composition and temperature regimes. Rigorous, comprehensive evaluation of uncertainties is lacking. At the same time there is substantial knowledge about uncertainty propagation and evaluation methods applicable to generic computational problems. The goal of this project is to make the first step towards the accurate and computationally affordable prediction of uncertainties in CALPHAD calculations by using and adapting advanced uncertainty quantification methods. Info: Dr. Elisa Capelli, Dr. Zoltán Perkó, E.Capelli@tudelft.nl, Z.Perko@tudelft.nl