American Nuclear Society
Home

Home / Publications / Journals / Nuclear Technology / Volume 212 / Number 9

LMFBR Core Analysis: Forward Problem Calculation and Assessment with the OpenNode Nodal Diffusion Code

Hicham Satti, Otman El Hajjaji, Tarek El Bardouni, Tarik El Ghalbzouri, Soumaya Oulad-Belayachi, Houda El Yaakoubi

Nuclear Technology / Volume 212 / Number 9 / September 2026 / Pages 2364-2386

Research Article / dx.doi.org/10.1080/00295450.2025.2517468

Received:January 15, 2025
Accepted:June 1, 2025
Published:July 27, 2026

This paper presents the implementation and verification of the developed OpenNode nodal diffusion code for the static neutron analysis of liquid-metal fast breeder reactors (LMFBRs). OpenNode, initially implemented and tested on pressurized water reactor benchmarks, is based on the nodal expansion method and features a graphical user interface with Blender integration for advanced three-dimensional (3D) geometric modeling and results visualization.

In this work, OpenNode is applied to fast spectrum reactor problems to assess its accuracy and flexibility in modeling LMFBR cores. Several two-dimensional and 3D LMFBR benchmark cases are analyzed, including configurations with four neutron energy groups, heterogeneous fuel assemblies, radial and axial blankets, and partial control rod insertion scenarios. Numerical results are compared with reference solutions from the KOMODO code, which is also based on the semi-analytical nodal methodology, to assess performance in terms of effective multiplication factor , neutron flux distributions per group, and power profiles.

This study also includes a mesh refinement analysis to investigate the impact of axial discretization on the stability and spatial accuracy of the solution. To reinforce the evaluation, relative error distribution plots are provided for flux and power comparisons, offering a quantitative insight into the behavior of the solver.

The results show strong agreement with reference data, with reactivity deviations generally below a few hundred pcm and spatial errors generally below 2%. These results confirm that OpenNode can be reliably extended to fast reactor simulations, and support its use as an accessible tool for research, code verification, and educational purposes in the field of reactor physics.