Computing Library › Applications
Applications

Computing for Neutronics and Shielding Design

Simulating how 14 MeV neutrons travel, breed tritium, deposit heat, and damage materials so blankets and shields can be sized.

Why neutrons dominate the engineering

D-T fusion releases 80 percent of its energy as 14.1 MeV neutrons. These uncharged particles pass through fields and thin walls, breed tritium in the blanket, heat structures, activate materials, and damage crystal lattices. Almost every wall, magnet, and shield decision follows from where the neutrons go.

How it is computed

Neutron transport is solved either by Monte Carlo methods, which track individual particle histories through detailed geometry, or by deterministic methods that solve the transport equation on a mesh. Monte Carlo captures complex geometry faithfully but needs many histories to beat statistical noise; deterministic methods are fast but discretize angle and energy.

The magnet-protection constraint

Superconducting magnets tolerate only a limited neutron and gamma dose before performance degrades. Shielding must knock the flux down by many orders of magnitude in a compact space, which is especially tight in a spherical tokamak like Hyperion where the central column is slender. Neutronics sizing of that shield is a first-order design driver, not a detail.

Verification

Neutron cross-section data carry uncertainty, so results are checked against benchmark experiments and sensitivity studies identify which nuclear data most affect the breeding estimate. The distinction between idealized local breeding and net breeding in the ported, gapped real geometry is preserved throughout.

All Hyperion neutronics is design-stage simulation; no hardware net-gain is claimed before first-of-a-kind first tritium around 2030.