home
fr|
en

    Impact of nuclear data on molten-salt reactors safety

    Research team: https://www-subatech.in2p3.fr/en/research/research-team/sen/research/reactor-simulationsand-decay-heat

    This work is carried out in collaboration with the MSFR group at LPSC, Grenoble https://lpsc.in2p3.fr/index.php/en/groupes-de-physique/msfr

    PhD Director : Lydie Giot – Associate-Professor, SUBATECH Laboratory, IMT Atlantique Adress : 4, rue Alfred Kastler – La Chantrerie – BP 20722 – 44307 Nantes Cedex 3, FRANCE Tel : 00 33 +2 51 85 86 66 Email : giot@subatech.in2p3.fr

    Context: Molten Salt Nuclear Reactors (MSRs) have great potential in terms of safety and flexibility. These are reactors in which the fuel is dissolved in a molten salt (liquid), acting as a coolant. The salt circulates in the fuel circuit through a zone called the “core”, where it is made critical by geometry, producing heat. The salt releases the heat by passing through an exchanger, enabling the energy produced to be recovered, either in the form of heat (heat-generating role) or electricity (power-generating role). This type of reactor is characterized by its intrinsically stable behavior, and its versatility (choice of cycle, choice of neutron spectrum, choice of salt, etc.) and hence the versatility of its applications (power reactor ranging from small to very high power, incinerator of high-level, long-lived waste through transmutation, etc.). With these qualities in high demand in the current nuclear context, Molten Salt Reactors are attracting interest in France and abroad, with the emergence of new players such as startups. Identifying and improving the understanding of safety-related phenomena associated with MSR concepts and their modeling is a major R&D challenge, in order to fill the gap between proposed concepts and industrial deployment, with also meeting the requirements of safety standards. From this perspective, assessing and improving the predictive capabilities of modeling tools and the nuclear data they use is a major R&D focus. The CEREIS code (Python + SERPENT 2), developed as part of the MSR modeling activities at Subatech in collaboration with the MSFR team at the LPSC laboratory in Grenoble (European SAMOSAFER project 2019- 2023, national ISAC project 2022-2026), allows to calculate the source term and decay heat to be taken into account in accidental situations. In parallel, as part of the national NEEDS/SUDEC project, the COCODRILO code has been developed to use a Monte-Carlo approach to determine the impact of uncertainties in nuclear decay data on decay heat calculations.

    PhD offer : The PhD offer proposed here will contribute to the safety assessment of MSRs and the impact of nuclear data and associated uncertainties. The part of the PhD funded via the European ENDURANCE project will focus on calculations of source term, decay heat and associated uncertainties due to nuclear data (cross sections and decay data) using a Monte-Carlo approach. The part of the PhD funded via the European APRENDE project is dedicated to the sensitivity profiles of nuclear data on MSR safety parameters. The project will also investigate the impact of new fission yield measurements at different neutron energies on decay heat calculations for MSR concepts. As part of the European ENDURANCE project (EU kNowleDge hUb enAbling molteN salt reaCtor safety development and dEployment) and WP4 (Modeling and simulations to enable safety assessment and licensing) coordinated by IRSN, codes aimed at providing a BEPUtype calculation (Best Estimate Plus Uncertainty) are being developed. The BEPU approach is a methodology to gain a better understanding of the uncertainties and biases inherent in safety analyses. The CEREIS code developed with on-line composition and reactivity controls will be used to model the source term and decay heat in the different locations of the MSR concepts to be studied: the 3 GWth MSFR reference concept selected by the Generation IV Forum for the Th/U cycle, and a 3 GWth MSFR using chloride salts in the U/Pu cycle foreseen as a solution for closing the fuel cycle. An assessment of the uncertainties associated with nuclear data (cross sections, decay data) will be carried out by coupling the COCODRILO and COCONUST codes currently under development at Subatech and LPSC. The student may also take part in discussions within the CNRScoordinated WP1 aimed at identifying R&D needs related to the design, operation and safety of MSRs. As part of the APRENDE (Addressing PRiorities of Evaluated Nuclear Data in Europe) project, the sensitivity of nuclear data uncertainties (JEFF3-3, JEFF4.0) will be assessed on reactivity-related parameters (keff, beff, Doppler and density feedback coefficients) for the two 3GWth MSFR concepts (Th/U and U/Pu cycles) for static and evolving calculations. Recent measurements carried out at GANIL in inverse kinematics have given access for the first time to the distribution of neutron-induced 239Pu fission yields at different excitation energies. Some fourth-generation concepts, such as molten-salt reactors, may have a neutron flux spectrum with different energy components. These new data will help to better constrain decay heat and source term calculations, as well as the associated safety analysis.

    Developed skills:

    • Modeling, Monte-Carlo methods
    • Nuclear Physics, Reactor Physics and depletion calculations
    • PYTHON3 (analysis of the results)
    • Monte-Carlo codes : SERPENT 2, COCODRILO
    • Use of cluster computing facilities

    Required profile :

    • Student in Master 2 or in last year of engineering school (A3) with some knowledge in reactor physics, reactor modeling, reactor codes, numerical methods and/or Monte Carlo methods
    • Good proficiency in the use of computer tools (Linux, computer clusters…), and/or languages such as Python, C or C++ will be an asset .
    • Definite interest in programming requested.

    A M2 or final-year engineering school internship on the same topic is offered prior to this thesis offer. The funding of this thesis is guaranteed (already been secured).