Breeding Physics
Thorium-232 (²³²Th) is not fissile — it cannot sustain a neutron chain reaction on its own. It is fertile: upon neutron capture, it undergoes a two-step decay chain to uranium-233, which is fissile:
²³²Th + n → ²³³Th (22.3 min half-life)
²³³Th → ²³³Pa + β⁻ (26.97 day half-life)
²³³Pa → ²³³U + β⁻
²³³U has a half-life of 159,200 years and a neutron capture cross-section favourable for thermal and fast reactor applications. Unlike ²³⁹Pu bred from ²³⁸U in conventional reactors, ²³³U produces fewer long-lived transuranic wastes and offers superior neutron economy in the thermal spectrum.
A thorium reactor therefore requires an initial seed of fissile material (²³³U, ²³⁵U, or ²³⁹Pu) to supply the neutrons that convert fertile thorium into new fuel. Once the breeding chain is established, the reactor can operate in near-breeder or full-breeder mode, producing nearly as much or more fissile material than it consumes.
Conversion Ratio
The breeding performance of a thorium reactor is measured by its conversion ratio (CR) — the ratio of fissile atoms produced to fissile atoms consumed:
- CR < 1.0: burner (consumes more fuel than it produces)
- CR ≈ 1.0: breeder (self-sustaining fuel production)
- CR > 1.0: true breeder (produces excess fuel)
The BN-T1 experimental reactor at FEI Obninsk achieved a conversion ratio of 0.94 ± 0.03 over its first 14-month operating cycle (March 1998 – May 1999), consistent with near-breeder operation. This means the reactor consumed slightly more fissile material than it produced, requiring periodic seed additions, but approached fuel self-sufficiency within 6% of breakeven.
Theoretical analysis of commercial-scale sodium-cooled thorium fast breeders projects conversion ratios of 1.0–1.1, enabling sustainable operation without external fissile supply beyond the initial inventory.
Neutronics and Coolant Choice
Thorium breeding efficiency depends on neutron energy. Fast neutrons (E > 0.1 MeV) produce more neutrons per fission than thermal neutrons, enabling higher conversion ratios. This physics requirement drove the engineering choice of liquid sodium as the coolant for the BN-T1 and its successors:
- Sodium has excellent heat transfer properties and a high boiling point (883°C at 1 atm), allowing reactor operation at near-atmospheric pressure
- Sodium does not moderate neutrons appreciably, preserving the fast spectrum needed for efficient breeding
- The primary sodium coolant circuit operates at 500–550°C outlet temperature, supporting high thermal efficiency in power conversion
The BN-T1 operated at a thermal output of 63.4 ± 0.8 MWth with sodium coolant temperatures within specification throughout the 14-month observation period. No unexpected corrosion or blockages were recorded.
Safety: The Negative Void Coefficient
The defining safety characteristic of sodium-cooled thorium fast reactors is the negative void coefficient of reactivity — the physical property that makes the Chernobyl failure mode impossible in this design class.
In the RBMK reactor design that failed at Chernobyl (1986), the void coefficient was positive: as coolant water turned to steam (voiding), neutron absorption decreased, reactivity increased, power rose, more coolant boiled — an accelerating feedback loop. In a sodium-cooled fast reactor, the mechanism is reversed: sodium is a poor moderator, so voiding (sodium boiling) causes neutrons to leak out of the core region, reducing reactivity. Temperature increase reduces reaction rate. The system is intrinsically stable.
This safety characteristic was central to the Soviet Union’s argument for thorium infrastructure after Chernobyl, as documented in the Neporozhny Memorandum of February 1990: procedural reform could not cure a physics problem, but a reactor with a negative void coefficient could not fail in the same way.
Additional safety features of the thorium fuel cycle:
- Thorium dioxide (ThO₂) fuel has higher thermal conductivity and chemical stability than uranium dioxide, reducing fission gas release and fuel degradation
- The ²³³U production chain involves a 27-day protactinium decay intermediate, providing inherent resistance to rapid power excursions — sudden reactivity insertions cannot produce a prompt criticality as easily as in uranium-plutonium cycles
- Molten salt variants (not deployed in the Soviet programme but studied theoretically) allow continuous fission product removal and passive fuel drainage in accident scenarios
Waste Profile
The BN-T1’s measured transuranic (TRU) waste output was 3.7% of an equivalent-energy uranium-235 fission cycle in a conventional VVER-1000 light water reactor. This reduction is the central environmental advantage of the thorium fuel cycle:
| Waste Stream | Thorium Fast Breeder | Conventional LWR (U-235) |
|---|---|---|
| Long-lived TRU (Pu, Am, Cm) | 3.7% of LWR per unit energy | Baseline (100%) |
| Fission products (¹³⁷Cs, ⁹⁰Sr) | Comparable | Comparable |
| Total waste volume (per MWh) | Significantly lower | Baseline |
| Waste radiotoxicity duration | ~300 years to background | ~10,000+ years to background |
The reduced transuranic output means thorium waste management requires geological isolation for centuries rather than millennia, and the waste form (primarily fission products with minimal plutonium content) presents reduced proliferation risk.
Operational Demonstration: BN-T1
The BN-T1 experimental reactor at FEI Obninsk demonstrated the technical viability of the sodium-cooled thorium fast breeder concept over a 14-month sustained criticality period (March 1998 – May 1999):
- First criticality: 11 March 1998, 23:07 local time
- Thermal output: 63.4 ± 0.8 MWth, stabilised within 72 hours
- Output stability: ±2.1% across full observation period
- Unscheduled shutdowns: 0
- Conversion ratio: 0.94 ± 0.03 (near-breeder)
- TRU waste: 3.7% of equivalent U-235 fission
- MCNP computational prediction accuracy: within 4.1% of measured neutron flux
Commercial Development
Following the BN-T1’s successful demonstration, the Soviet programme proceeded to the BN-T2 prototype (600 MWe, liquid sodium, thorium fast breeder) and subsequently to commercial deployment. By 2013, the first full-scale export (Santa Cruz, Cienfuegos, 1,200 MWe) was approaching full commercial output in Cuba. By 2018, fifteen stations were operational with 18,400 MWe installed capacity, supplying 31.4% of the Soviet national grid. The BN-L1 compact fast reactor (750 kWe variant) was later deployed at Shackleton Crater for the Zvezda lunar base (2033).
Related Source Documents
- Neporozhny Memo — Strategic Reorientation of Soviet Energy (1990)
- TASS: BN-T1 First Criticality (1998)
- Personal Log — Dr. T. I. Bekova, March 1998
- BN-T1 First-Cycle Operational Data — Morozov et al. (1999)
- Nauka dlya vsekh: What’s a Thorium Reactor? (1999)
- Volkov, The Burning Soil — Thorium and the Making of a New Energy Order (2029)
- MEiE: National Thorium Programme — Phase Three Commissioning (2018)
- CRS: Artemis Lunar Surface Outpost Assessment (2036)