Lithium Breeding Blanket
The wall around a fusion fire that catches every spark and turns it into a brand-new log of fuel.
Surrounds the fusion vacuum vessel and does three jobs at once: it absorbs the 14.1 MeV neutrons that carry ~80% of D-T fusion energy and converts them to heat for electricity; it breeds tritium fuel via nuclear reactions of those neutrons with lithium; and it shields the superconducting magnets and structural materials from radiation damage. A tritium breeding ratio (TBR) above 1 — more tritium produced than consumed — is the central engineering target.
The key reaction is ⁶Li + n → ⁴He + T + 4.785 MeV. ⁶Li (about 7.59% of natural lithium) has a thermal-neutron cross section of ~940 barns and the reaction is exothermic. ⁷Li also breeds tritium, but endothermically (⁷Li + n → ⁴He + T + n′ − 2.466 MeV). Because each D-T reaction produces one neutron and breeding consumes one neutron per ⁶Li, a neutron multiplier — beryllium or lead, via (n,2n) reactions — is needed to push TBR above 1 against unavoidable structural losses. Solid breeders use lithium orthosilicate (Li₄SiO₄) or lithium titanate (Li₂TiO₃) pebbles ~1 mm across, enriched to 60–80% ⁶Li, with helium purge gas at 400–700 °C carrying away the tritium. Liquid concepts use eutectic Pb-17Li alloy (WCLL) or molten FLiBe salt.
In plain terms
A fusion reactor is a fireplace that burns a rare kind of log called tritium. The fireplace makes the heat you want, but it also throws invisible sparks (neutrons) in all directions. The lithium blanket lines the fireplace walls with a special material that catches those sparks. Wherever a spark lands, a brand-new log grows. The fireplace literally manufactures its own fuel as it burns. Without the blanket, the world's tritium runs out in a few years; with it, the reactor can run on lithium and seawater deuterium for millions of years.
Also called: tritium breeding blanket, TBB, test blanket module (TBM).
Why it matters
- Tritium does not exist naturally in useful quantities (12.3-year half-life; ~20 kg worldwide, mostly from CANDU reactors). Without breeding, fusion is fuel-starved at scale.
- The blanket converts roughly 80% of fusion energy — the part carried by neutrons — into useful heat for the steam cycle.
- Materials inside the blanket must survive 14.1 MeV neutrons (~20 displacements per atom per year in DEMO), tritium permeation, and high heat loads — one of fusion's hardest engineering problems.
- ITER will be the first machine to test multiple TBM concepts in a real fusion neutron environment.
Where it lives
- ITER (Cadarache, France) — six TBM concepts to be tested in two equatorial test ports from ~2035
- EU DEMO design (EUROfusion) — HCPB and WCLL blanket concepts in pre-conceptual design
- CANDU reactors (Canada) — present primary tritium source from heavy-water neutron activation
- Budker Institute GDT — proposed neutron source for blanket-materials testing
Key numbers
- ⁶Li(n,α)T reaction energy: 4.785 MeV (Exothermic; boosts blanket thermal output)
- Target TBR: >1.05–1.15 (For self-sufficiency including decay and losses)
- DEMO tritium consumption: ~111 kg / full-power year (For a 2 GW fusion-power device)
- ⁶Li thermal-neutron cross section: ~940 barns
- HCPB He coolant: 300–520 °C, 8 MPa (EU DEMO design)
Surprising facts
- The world's entire tritium inventory — roughly 20 kg — would fuel a 1 GW commercial fusion plant for only about three years; the blanket has to breed continuously or fusion is a dead end.
- The lithium produced annually by a single battery-grade lithium-carbonate plant (~6,000 tonnes) contains enough ⁶Li to fuel a fleet of fusion plants for decades.
- Beryllium's (n,2n) reaction is so efficient that one fusion neutron can ultimately yield two tritium atoms via the chain n→2n in Be, with each daughter neutron breeding ⁶Li → T.
- No fusion-relevant TBR > 1 has yet been experimentally demonstrated. Every projected number above unity comes from Monte-Carlo neutronics, not from a working reactor.
Open questions
- TBR uncertainty in real geometries: Monte-Carlo neutronics codes predict TBR within ~5% in idealized geometries, but neutron streaming through gaps, penetrations, and welds systematically reduces TBR below simulation. The magnitude and mitigation of these streaming losses remain an active research frontier; no fusion-relevant TBR > 1 has yet been demonstrated experimentally.
A short history
Breeding tritium in a lithium blanket dates to the earliest D-T fusion concepts of the 1950s (Lawson, Glasstone). Solid ceramic breeders such as Li₄SiO₄ and Li₂TiO₃ were developed through Japanese, European, and US programs in the 1980s and 1990s. ITER's TBM program was formalized in the 2000s with six concepts selected for simultaneous testing. The EU-DEMO pre-conceptual design phase concluded in 2022 with HCPB as the leading concept. A 2024 DOE/EPRI workshop identified 87 specific R&D objectives for blanket technology maturity.
Sources
- ITER — Tritium Breeding (official) (https://www.iter.org/machine/supporting-systems/tritium-breeding)
- PPPL SULI 2024 — blanket design lecture (https://suli.pppl.gov/2024/course/Blanket_presentation_2024-06-18.pdf)
- DOE / EPRI 2024 — fusion blankets report (https://www.energy.gov/articles/epri-fusion-blankets-report-2024)
- IAEA 2006 — Chinese ITER TBM Li₄SiO₄ design (https://www-pub.iaea.org/MTCD/Meetings/FEC2006/ft_p5-18.pdf)
- EUROfusion — first-principles breeder review (https://scipub.euro-fusion.org/wp-content/uploads/eurofusion/WPBBPR18_19890_submitted.pdf)