The project advanced computational tools for the safety assessment of Generation IV fast reactors, with a particular emphasis on sodium‑cooled systems. Core neutronics were modeled in the deterministic code FENNECS and benchmarked against the Monte‑Carlo reference Serpent and experimental measurements. Mini‑core configurations of seven and nineteen fuel elements were simulated, and the resulting power distributions agreed with Serpent within a mean‑square deviation of less than 3 %. Control‑rod exchange reactivities calculated by FENNECS differed from Serpent by less than 2 % for single‑rod movements and remained below 4 % when multiple rods were displaced. Sodium‑void reactivity predictions from FENNECS matched laboratory data within 5 % of the measured values, demonstrating the fidelity of the coupled neutronics–thermal‑hydraulics treatment.

For the China Experimental Fast Reactor (CEFR), a comprehensive reference model was generated in Serpent, including detailed fuel‑element geometry and material composition. Deterministic core models were then built in FENNECS, and simulations of start‑up tests reproduced the measured core power evolution with an error margin of only 2.5 %. The Experimental Sodium Fast Reactor (ESFR) core was modeled in a similar fashion: nuclear data were generated, a reference Serpent model was constructed, and a FENNECS representation was validated against Serpent power and control‑rod worth data. Integral control‑rod worth deviations were below 1.8 % and axial power‑distribution differences were under 3 %. Coupled neutronics–thermal‑hydraulics simulations were performed with the ATHLET code, and the coupling interface between FENNECS and ATHLET was refined to ensure consistent temperature feedback and power‑dependent reactivity changes. The coupled ESFR simulations reproduced the measured fuel‑temperature distribution within 4 % and the axial power‑density peak within 2.5 %.

Thermohydraulic modeling was extended to subchannel representations of CEFR fuel elements in ATHLET, capturing local temperature gradients and flow distributions. A CFD model of a CEFR fuel element was developed in OpenFOAM, providing detailed velocity and temperature fields that were used to calibrate the subchannel ATHLET model. For the ESFR core, a full thermohydraulic model was built in ATHLET, and its predictions were validated against experimental data from the ESFR test facility, achieving agreement within 3 % for core‑average temperatures and 5 % for peak cladding temperatures. The project also contributed to the EU ESFR‑SMART initiative by performing 3D ATHLET/OpenFOAM simulations that quantified the impact of sodium‑void and fuel‑expansion effects on transient behaviour.

Collaboration involved the German Reactor Safety Society (GRS), the Technical University of Munich, and partner institutions in China and Belgium. The project was funded by the German Federal Ministry of Education and Research under the GRS‑417 and GRS‑553 programmes, spanning the period 2016 to 2020. In addition to the core and thermohydraulic work, the team participated in the IAEA Fast Reactor Knowledge Preservation initiative, ensuring that the developed models and data sets are preserved for future safety analyses. The combined effort produced a suite of validated, high‑fidelity simulation tools that enhance the reliability of safety assessments for sodium‑cooled fast reactors and support the design of next‑generation nuclear systems.