The trapping of radioactive tritium fuel within plasma-facing materials and co-deposited layers — a critical safety, licensing, and fuel economy challenge for deuterium-tritium fusion reactors.
Tritium retention refers to the accumulation of tritium (T or 3H) — a radioactive hydrogen isotope with a 12.3-year half-life — in the solid materials of a fusion device's vacuum vessel. When tritium ions and neutrals strike plasma-facing surfaces, a fraction becomes trapped through implantation, diffusion into the bulk, or co-deposition with eroded wall material. Because tritium is radioactive, expensive to produce, and present in limited quantities, its retention in vessel structures has implications for nuclear safety, regulatory licensing, and the economics of fuel self-sufficiency.[1]
Tritium becomes trapped through several distinct mechanisms:
Ion implantation: Energetic tritium ions penetrate a few nanometers into the surface and become lodged in the lattice. In tungsten, the solubility of hydrogen isotopes is extremely low, so implanted tritium tends to migrate to natural and radiation-induced defects — vacancies, grain boundaries, dislocations, and voids — where it is trapped with binding energies of 1–2 eV.[2]
Neutron-induced trapping: The 14.1 MeV neutrons from D-T fusion create displacement damage cascades in PFC materials, generating large numbers of vacancy clusters that serve as additional trapping sites. Over time, the trap density increases, potentially raising the total tritium inventory far beyond what surface implantation alone would produce.[2]
Co-deposition: Material eroded from one surface by sputtering can be transported through the plasma edge and redeposited elsewhere, trapping tritium in the growing co-deposited layer. In carbon-walled devices, co-deposited carbon-tritium layers have been the dominant retention mechanism, with tritium-to-carbon ratios approaching 0.4 in some cases. This was a principal reason for the transition from carbon to tungsten PFCs in ITER.[3]
The 700-gram limit corresponds to a radiological hazard threshold established through safety assessments of hypothetical loss-of-vacuum and loss-of-coolant events. For reactor-scale devices burning tens of kilograms of tritium per year, the retention fraction must be kept extremely low — well below 1% of throughput — to avoid both safety limit violations and unacceptable fuel losses.[1]
Several methods have been developed or proposed for recovering trapped tritium. Baking the vacuum vessel to 350–500 °C mobilizes weakly trapped tritium from tungsten surfaces. Oxygen glow discharge cleaning can oxidize co-deposited layers and release tritium, though it risks damaging metal surfaces. Laser-based techniques that locally heat surfaces to release tritium without full vessel baking are under investigation for reactor applications where long bake-out campaigns would reduce availability.[3]
Tritium retention is directly coupled to the tritium breeding ratio (TBR) requirements of a fusion power plant. Every gram of tritium lost to retention is a gram that must be bred in the blanket. If retention rates are high, the required TBR increases, tightening already challenging margins on blanket design. Accurate prediction of long-term tritium retention under reactor-relevant neutron fluences remains one of the key open questions in fusion materials science.[2]