Technology

The Science

The Liquid Fluoride Thorium Reactor. Proven at Oak Ridge in 1965. Marchitto Inc. is building the vehicle-scale version. Here is exactly how it works — and what remains to be solved.

Thorium to Power

Thorium-232 is a mildly radioactive element more common in the earth's crust than tin. It cannot sustain a chain reaction on its own — but when it absorbs a neutron inside the reactor, it decays through Protactinium-233 and becomes Uranium-233, which is fissile and sustains the chain reaction. This breeding process happens continuously inside the operating reactor.

Energy Density3.2 million MJ per kilogram — approximately 4 million gallons of gasoline per kilogram of thorium
Annual Consumption~5 grams of thorium per year at normal vehicle load
8kg Fuel LoadTheoretical lifetime over 1,600 years — far exceeds vehicle service life
Military Design Life25 to 100 years operational without refueling
Th-232 AbundanceMore common than tin — exists on every continent in accessible deposits
Proliferation RiskVery low — U-233 contaminated with U-232, intensely radioactive, impractical for weapons

LFTR — How It Works

The LFTR dissolves thorium fuel directly into FLiBe molten salt — a mixture of lithium fluoride and beryllium fluoride. This salt serves simultaneously as the fuel carrier, the neutron moderator, and the coolant. There are no solid fuel rods to fail, no pressurized water systems to rupture, and no zirconium cladding to oxidize. The reactor operates at atmospheric pressure — there is physically nothing to explode.

The critical safety property: as reactor temperature rises, the FLiBe salt expands, reducing the density of fissile material in the core, which reduces fission rate, which reduces heat output. The reactor is physically self-regulating. It cannot run away. No operator action required. The physics prevents it.

Fuel Carrier SaltFLiBe — LiF (67 mol%) + BeF₂ (33 mol%)
Salt Melting Point459°C
Salt Boiling Point1,430°C — enormous operating margin above normal ~700°C operations
Reactor VesselHastelloy-N — nickel-molybdenum alloy, proven at MSRE 1965–1969
Power ConversionClosed-loop supercritical CO₂ Brayton cycle turbine
Thermal Efficiency45 to 50% — nearly half of thermal output becomes usable electricity
Operating PressureAtmospheric — nothing to explode
Proven PrecedentMSRE at Oak Ridge — operated 1965 to 1969, 13,172 equivalent full-power hours

TRITON — Three Layers

TRITON — Threefold Redundant Integrated Total-containment Nuclear System — is the safety architecture that makes Project THORON deployable in military environments. All three solutions must fail simultaneously for any radiological release to occur. The probability of simultaneous failure approaches zero.

Solution 1
Black Box Capsule
Five-layer spheroid capsule designed to survive any realistic combat scenario. 3,400G impact rating. 1,100°C for 60 minutes. Rated beyond the aviation black box standard. Layers: Borated polyethylene → Titanium → Aerogel → SiC ceramic → Tungsten carbide.
Solution 2
Solid Salt Shutdown
If the capsule is breached, the FLiBe salt solidifies passively. No electronics. No operator. No electrical power required. Solid salt cannot flow, cannot leak, cannot contaminate. The freeze plug drains the fuel to a subcritical catch basin below the reactor.
Solution 6
Remote Vitrification
On breach detection or remote command, a vitrification agent converts the salt to inert borosilicate glass in under 10 seconds. Glass cannot flow. Glass cannot be weaponized. The operator triggers remotely — the enemy gets nothing.
All Three Solutions Must Fail Simultaneously for Any Radiological Release — Probability Approaches Zero

The 10 Hard Problems

These are the real unsolved engineering challenges in LFTR at vehicle scale. None of them are physics problems — physics problems mean the concept cannot work. Engineering problems mean nobody has been obsessed enough to solve them yet. Project THORON's Phase 2 research agenda is a direct attack on every one.

01
Tellurium Embrittlement of Hastelloy-N
Te-132 fission product migrates into vessel grain boundaries causing intergranular cracking. Solution: modified Hastelloy-N with Ti additions (near-term), SiC-CMC vessel (long-term).
Critical
02
Tritium Production and Permeation
Li-6 reacts with neutrons to produce tritium that permeates through hot metal walls. US domestic Li-7 enrichment lost in 1963. Solution: oxide barrier coatings, double-wall heat exchangers, domestic Li-7 advocacy.
Critical
03
Reactor Miniaturization
The MSRE occupied a building. THORON needs an engine block. Smaller core requires higher fissile concentration and more aggressive thermal management. Solution: beryllium reflector, computational neutronics, dual-core architecture.
Critical
04
Pa-233 Breeding Efficiency Loss
Protactinium-233 can absorb a neutron before decaying to U-233, reducing breeding ratio below 1.0. Solution: accept sub-unity breeding (0.8–0.9) for vehicle application. 8kg fuel load provides 25–50 years regardless.
High
05
Xenon-135 Poisoning
Xe-135 (2.6 million barn cross-section — strongest neutron poison known) can suppress reactor power during military transients. Solution: continuous helium sparging — demonstrated at MSRE.
High
06
Noble Metal Fission Product Deposition
Mo, Tc, Ru, Rh, Pd, Te deposit as metallic films on vessel walls. Solution: salt redox control via UF4/UF3 ratio, porous filter elements, surface coatings.
High
07
Beryllium Toxicity
BeF₂ (33 mol% of FLiBe) causes chronic beryllium disease at microgram inhalation levels. Solution: contract salt preparation to existing licensed facilities (ORNL, Materion Corp).
High
08
Waste Heat Rejection in Vehicle Geometries
An 800kW reactor also produces ~900kW of waste heat requiring rejection in a moving vehicle. Solution: sCO2 cycle advantage, deployable radiators, phase-change material buffers.
Moderate
09
Startup Fissile Material Procurement
Pure thorium cannot start the reactor — initial U-233, HEU, or LEU required. US U-233 stockpile being disposed of. Solution: LEU startup design, early DOE engagement.
High
10
Long-Term Radiation Damage
MSRE ran only 13,172 full-power hours — no validated data for 50-year vehicle lifetimes. Solution: accelerated irradiation testing at INL, replaceable vessel architecture, SiC-CMC qualification.
High

Ten challenges. Every one real. Every one an engineering problem, not a physics problem. The distinction is everything. These are the ten reasons this technology has been sitting on a shelf for sixty years — not because the physics failed, but because the political will to fund the engineering disappeared in 1969.

The Physics is Correct

Primary sources: ORNL-4812 and ORNL-TM-0728 — available free at osti.gov. The proof is there. We are building on it.

See the Platforms Get in Touch