BSC’s Fusion Group has published a new article in the Fusion Science and Technology journal, entitled “Application and coupling of NEUTRO, a deterministic neutron transport solver for fusion applications in the Alya multiphysics environment”.
This work represents a significant milestone in our mission to develop comprehensive simulation tools for fusion reactor analysis. The article presents new results obtained using NEUTRO, the neutronics module of Alya, a massively parallel finite element framework created at BSC. This development is carried out within the Alya4Fusion initiative and continues previous work, which can be found in our blog (here and here).
For the first time, we have successfully demonstrated the coupling of three physics modules within Alya in the context of nuclear fusion: NEUTRO for neutron transport, TEMPER for heat transfer, and NASTIN for fluid dynamics. This integrated approach enables us to simulate the complex interplay between neutron radiation, heat deposition, and coolant circulation in fusion reactor components within a single computational framework.
The walls of a nuclear fusion reactor, which enclose the vacuum vessel where the fusion reactions occur, are exposed to various thermomechanical stresses from multiple sources. One of the most significant stresses arises from the constant bombardment by high-energy neutrons produced during plasma reactions. When these neutrons interact with the wall materials, they induce the emission of prompt gamma photons. Both the neutron and gamma radiation contribute to heat generation, which must be effectively removed by the coolant circulating through internal wall channels. Developing future fusion reactors demands a deep understanding and accurate prediction of these processes, which in turn requires highly advanced simulations to analyse the effects of neutron interactions with the reactor’s numerous components.
As described in previous publications and blog posts, Alya is a parallel finite element framework developed at BSC for simulating complex physical phenomena on large-scale supercomputers. It supports modular, independently compiled physics models. One such module, NEUTRO, handles radiation transport by solving the stationary Boltzmann equation using finite elements, discrete ordinates for direction, a multigroup energy model, and accounts for anisotropic scattering with spherical harmonics. Other modules of Alya include TEMPER, which solves the heat equation and NASTIN, which deals with incompressible fluids solving the Navier-Stokes equation.
The published work compares results obtained with NEUTRO with the well-known Kobayashi benchmark, specifically developed by OECD/NEA to compare the performance of different deterministic neutronic codes, such as NEUTRO. Our results show a good performance compared to the reference results, with a deviation within the range presented by the original group of codes. Then, our code is compared with MCNP, the reference neutronics code, which applies the Monte Carlo method. The case involves a simplified domain for a lithium-lead breeding blanket surrounded by EUROFER steel. Neutron flux, nuclear heat deposition and tritium generation are compared along the thickness of the domain with good accuracy. Total heat and tritium are also within 97% and 94% accuracy, respectively.
The final case is the main point of the publication. The domain is part of the vacuum vessel wall of one sector of ITER, the tokamak currently under construction in Cadarache, France, with international collaboration and efforts to achieve net gain in energy production and prove fusion as a viable commercial energy source. It presents the solid steel piece with water flowing within to extract and transport the heat transferred from neutrons that reach the wall, the dark green part on the left of the schematic figure above.
For this simulation, NEUTRO was coupled with the heat and incompressible fluids modules, essentially putting together the neutronics and thermohydraulics simulation into one. A neutron source was calculated to generate a certain amount of total heat. Afterwards, NEUTRO calculates the neutron flux distribution on the entire domain and the subsequent heat deposition. After that, the thermohydraulics analysis simulates the heat transfer, coolant flow and the heat removal. All the steps are carried out within one execution, effectively a tightly coupled multiphysics simulation.
Results are compared to those of other codes, with excellent performance on coolant temperatures. Maximum solid temperatures are lower in Alya than in the rest of the codes, due to some advantages presented by our code. Its efficiency and scalability in high-performance computing environments enable the use of a very fine and dense spatial mesh, which allows the use of a no-slip boundary condition on the fluid-solid contact surfaces. This prevents water accumulation in convoluted sections of the coolant channel, causing lower differences between average and maximum temperatures.
The paper has been published in open-access form in a Fusion Science and Technology special issue on fusion neutronics following the 15th ITER Neutronics Workshop, which took place in April 2024. The results detailed in the published article are extremely positive, reinforcing our commitment to further developing NEUTRO and other fusion-related physics modules as part of Alya and the whole Alya4Fusion initiative.
This work has been carried out within the framework of the EUROfusion Consortium, funded by the European Union via the Euratom Research and Training Programme (Grant Agreement No 101052200 — EUROfusion). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Commission. Neither the European Union nor the European Commission can be held responsible for them.
