Tritium Breeding in the Breeder
The breeder (Hyperion) is studied to breed its own tritium in a lithium blanket, with the tritium breeding ratio treated as a design lever across 1.1, 1.5, and 1.8.
- Machine
- Breeder (Hyperion), D-T spherical tokamak
- Breeding medium
- Lithium blanket
- TBR lever
- 1.1 / 1.5 / 1.8
- Fusion power
- 85.0 MW
Breeding from lithium
A deuterium-tritium fusion device consumes tritium in the plasma and produces energetic neutrons. Those 14 MeV neutrons can be captured in a surrounding lithium blanket, where reactions on lithium-6 and lithium-7 produce fresh tritium. The figure of merit is the tritium breeding ratio (TBR): tritium bred per tritium burned.
In the Kronos studies the TBR is treated as a design lever rather than a single fixed number, and is examined across values of 1.1, 1.5, and 1.8. A TBR above 1 is what makes self-sufficiency, and a surplus for external supply, physically possible. The three values bracket a range from bare self-sufficiency toward a meaningful export surplus.
Self-sufficiency versus surplus
A TBR near 1.1 aims at replacing what the plasma burns plus modest losses. Higher values (1.5, 1.8) open the possibility of a net surplus that could serve external demand — the tritium-class output on the order of ~4 kg/yr referenced in the design set. Which value is achievable in hardware is exactly the kind of question that remains open until the machine is built and measured.
- TBR > 1 is required for self-sufficiency plus any export.
- Blanket design, neutron multiplication, and coverage all drive TBR.
- The 1.1/1.5/1.8 bracket spans self-sufficiency to surplus.
Neutron economy sets the ceiling
The achievable breeding ratio is ultimately a neutron-economy problem. Each fusion neutron can breed at most a limited amount of tritium, and neutrons are lost to structure, ports, diagnostics, and parasitic absorption before they reach lithium. Neutron-multiplying materials in the blanket can recover some of that loss, but coverage is never complete because the machine needs openings. This is why a target such as 1.8 is treated as an upper study value rather than a promise: closing the gap between an idealized blanket and a real one with penetrations is exactly the engineering work the program flags as open.
Reconciling the achievable TBR with realistic blanket coverage and losses is an active engineering question in the breeder program, and is reported honestly as a design target rather than a delivered result.