LFTRs reduce the possibility of proliferation. The weaponization of a nuclear reaction is unique in that only uranium-235 and plutonium-239 have been known to make a militarily effective bomb. However, while not on the same scale it is technically possible to create a rudimentary nuclear device using material produced in LFTRs – namely through uranium-233 and neptunium-237.
Yet doing so is considerably more difficult and less reliable than with traditional nuclear materials and traditional nuclear reactor designs. Further, uranium-233 and neptunium-237 – even if fashioned into a nuclear weapon – would likely make the device ineffective for military purposes. The reasons?
Purification difficulties and inherent dangers. It's been theorized that if an LFTR is using something called a fluorinator, neptunium-237 can be extracted via a chemical process, which has potential to undergo a fast-fissile reaction and enable a nuclear detonation.[76] But the critical mass requirement for neptunium-237 is roughly 60 kilograms, which is higher than even uranium-235 – and emits 2,000% more gamma emissions than plutonium-239.[77] That would make such a weapon far more dangerous to build and less practical to deploy even if the expertise to weaponize neptunium-237 existed.[78] Further, chemical purification to this degree requires highly expensive and purpose-built infrastructure that can’t be obtained by entities other than states with sophisticated science programs.
Moreover, part of the breeding process to transmute thorium-232 into uranium-233 involves the production[79] of an invariable amount of uranium-232 – which, while perfectly safe within a reactor, also emits high levels of both alpha[80] and gamma radiation.[81] For those inclined, ionizing radiation (the potentially harmful kind) is commonly measured in “rem” (Roentgen Equivalent Man).82/cite] In general, the more radiation one absorbs, the more harmful the effects become. At standard levels, the uranium-232 contaminant within a 5kg sphere of uranium-233 would generate up to 38 rem per hour.[83] For use in a weapon, uranium-233 has a minimum critical mass of 16.5 kilograms[84] - presenting an aggregate dose of up to 125 rem/hr. Serious radiation sickness begins with short-term exposure of 150 rem, and anything over that is potentially lethal.[85]
At sufficient mass to build a nuclear weapon, this would make the material too dangerous to handle by human beings and would require the employment of sophisticated (and expensive) remote-assembly robotics – traits not shared by other weapons-grade nuclear material.[86] Additionally, uranium-232’s gamma emissions damage sensitive electronics and increase material heat,[87] which can prevent a sophisticated nuclear device from detonating under precise and exact conditions – hard requirements for effective use as a weapon.[88]
These are important distinctions in light of concerns from nuclear agencies that cast aspersions on thorium’s proliferation resistance. One notable example of these concerns comes from a 2010-era report by the United Kingdom’s National Nuclear Laboratory[89] that states:
As with neptunium-237, “proliferation risk” is contextual – and much of that context derives basis from academic postulation as opposed to tangible capability. Just because a state could theoretically make a nuclear device from uranium-233 doesn’t mean it can make one practically – all the more so since there is negligible data or expertise to aid in the creation of such a device. And even if that effort was successful, none of that says such a device would be sufficiently powerful for use in a conflict – or even if it can be effectively deployed in the first place.
Once those factors enter the equation, the optics change significantly.
The UK National Nuclear Laboratory is, of course, correct. Both uranium-233 and neptunium-237 are fissile fuels, and both have potential to undergo nuclear detonations.[90] Further, the uranium-232 contaminant within uranium-233 probably wouldn’t stop a crude bomb from detonating crudely even if it killed its makers.
But it would stop a bomb that relies on sophisticated technology to implode in the precise detonations required for miniaturization to a warhead-scale. It also ignores that LFTRs can be designed to minimize the risk of weaponization barring major infrastructural investments that would draw the attention of international atomic energy monitors. And even if a state had enough neptunium-237, the critical mass requirements would hinder the ability to deploy a weapon of sufficient yield unless that weapon was carried by aircraft or large, long-range missile. The presence of both of these factors would remove either isotope from consideration as a primary charge for a thermonuclear device, and would further preclude both from use as a first-strike weapon.[91]
Further, even if we did grant credence to a rouge state’s ability to invest the time, effort, and risk to either purify uranium-233 or build a weapon with neptunium-237, it’s important to emphasize just how difficult this is to do – all the more so to do so quietly. If any rogue nation tried to build such infrastructure, any intelligence agency with a satellite would know exactly what was going on in short time (which is how we know Iran, North Korea, etc., have nuclear weapons programs).
Nuclear weapons design isn’t secret anymore – no 75-year old technology is. The hydrogen bomb, even, is old enough to collect social security. But reaching certain milestones towards making one are only possible with highly expensive and sophisticated systems built specifically for that purpose. They’re not the sort of thing one picks up at Walmart, and their procurement would certainly raise flags among international monitors and foreign intelligence services – especially since there’s only a few entities in the world that manufacture them. If that wasn’t enough, they remain among the most controlled machines on the planet.
If your life’s ever lacking excitement, try wiring a few million dollars to an offshore bank account for an order of krytron tubes, ultra-fast relay switches, large gas centrifuges and a hefty supply of lithium-6. At the very least you’ll see a whole lot of government property and personnel you didn’t know existed appear awfully fast. Should any state try the same, that property and personnel usually manifests in the form of an airstrike to destroy such a program in its infancy. That’s usually long-before said state has even conducted the multitude of tests needed to see if their bomb design even works, as (likely) happened in Syria in 2007.[92]
At the levels of sophistication and expertise required to covertly obtain the necessary materials and successfully make a bomb out of uranium-233 or neptunium-237, building a bomb with traditional nuclear materials sourced from the ground or ocean[93] becomes an easier prospect.
Even so, commercial LFTR designs would need to be required to intentionally contaminate the reactant with materials that would make weaponization harder from the start.[94] This wouldn’t permanently remove the risk, but it would make it much more difficult for all but the most dedicated actors. In those cases, if a state is advanced enough to make a nuclear weapon from thorium, they don’t need thorium to make one in the first place.
And regardless, it’s still poor bomb fuel. Even if they could be efficiently extracted, uranium-233 and neptunium-237 are ineffective fast-reacting fissile fuels compared to highly enriched uranium-235 and plutonium-239. There are only two known nuclear weapon tests that have ever used uranium-233 – none have ever used neptunium-237. Both of the uranium-233 devices were largely considered failures due to weaker-than-intended explosive yields, respectively at 22 kilotons (U.S. – 1955) and 0.2 kilotons (India, 1998) – relative pittances compared to modern nuclear weapons.[95] In the first device, the uranium-233 was chemically purified (which, again, is highly difficult to do) and was significantly complemented by plutonium-239 to increase yield.[96] As a consequence of those lackluster tests, there exists little research or expertise to weaponize uranium-233 or neptunium-237, nor avoid the inherent dangers of doing so.[97]
And even if there was research or expertise, what’s the endgame? To bring dynamite to a thermonuclear missile fight? In a realpolitik sense, the leverage gained by building a nuclear weapon is only as valuable as its effective usability in a conflict – or a hedge against the same. A single U.S. Navy Ohio-class submarine can launch 24 missiles – each armed with up to 12 thermonuclear warheads that each yield 475 kilotons – and the U.S. Navy has fourteen of such submarines. Nothing from thorium is ever going to produce anything that can hold a candle to that.
This is why every state with nuclear ambitions has instead invested in uranium-235 and plutonium-239, for their use is easier and safer than hijacking the thorium fuel cycle to produce weapons-grade material. While again, this does not totally alleviate concerns of proliferation through thorium, it does reduce them to a significance on par with a state making an in-house weapons program of their own volition – and that’s becoming increasingly more plausible as technology advances globally. With these considerations in mind, the clean energy benefits that thorium and LFTRs bring simply outweigh the theoretical risks of either being used by a dedicated actor for nefarious purposes.
LFTRs are simpler, smaller and less expensive than Pressurized Water Reactors. As traditional nuclear reactors have to be pressurized to 160 atmospheres just to function – pressure equal to a mile below the ocean’s surface[98] – they require redundant processes and complex systems to manage the reaction and ensure nothing goes wrong. Additionally, as Pressurized Water Reactors present the potential for catastrophic environmental damage should a reactor melt down or be destroyed through sabotage, they further require extensive security infrastructure. Combined, these factors cause such reactors to rank among the most expensive and over-engineered systems on the planet:
| Light Water Reactor fueled by uranium-235 | Liquid Fluoride Thorium Molten Salt Reactor (LFTR) |
|---|---|
| Fuel: Uranium-dioxide solid fuel rods | Fuel: Uranium-233 and thorium-232 in a solution of molten lithium-fluoride salts |
| Fuel lifetime: Two years at best. Requires reactor shutdown to replace. Core + fuel rods remain radioactively contaminated | Fuel lifetime: 30 years without replacement. Current graphite core lifetime is in excess of six years |
| Fuel input per gigawatt output: 250 tons uranium-235 | Fuel input per gigawatt output: 1 ton thorium-232. 250 times more efficient |
| Annual fuel cost for 1-GW reactor: $60 million | Annual fuel cost for 1-GW reactor: $10,000 (estimated) |
| Total unit construction cost: $7.0 billion | Total unit construction cost: $1.0 billion* (1-GW reactor) |
| Coolant: Highly pressurized water with a graphite moderator | Coolant: Self-regulating with passive gravity emergency shutdown |
| Weaponization potential: High | Weaponization potential: Low |
| Physical footprint: 300,000 square feet + large buffer zone | Physical footprint: 2,000-3,000 square feet (size of a house). No buffer zone required |
As LFTRs are spared the size, expense and security requirements of Light Water Reactors, they can be built much smaller and less expensively. They can also be built closer to population centers (as opposed to Pressurized Water Reactors that need to be geographically isolated), considerably reducing the infrastructural requirements to transmit power to electric grids.
LFTRs can be built in a modular, prefabricated capacity. Today’s nuclear reactors are designed as unique, custom systems that are each made to order – significantly increasing their total cost. Yet recent improvements in manufacturing today allow LFTRs to be built on assembly lines as iterations of product models in the form of small modular reactors.
This provides two main benefits:
First, efficiencies inherent in modern manufacturing enable us to reduce construction costs over time as more identical units are produced. This is often referred to as “the learning curve,” or “learning ratio” – the reduction in manufacturing cost every time the number of produced units doubles.[101]
In computing, Moore’s law has shown that computer processing power at a given price doubles every two years. In aerospace manufacturing, the reduction in per-unit cost has been roughly 20% every time the number of produced units has doubled.[102] As applicable to the manufacturing of Light Water Reactors, the University of Chicago estimates a learning ratio of 10% in their 2004 study The Economic Future of Nuclear Power.[103] As LFTRs can be built on assembly lines, that percentage would likely be higher, expected to be on the order of aerospace-grade manufacturing.
But even at 10%, this would mean that by the time the 1,000th LFTR was constructed it would cost around 40% of the first commercially produced unit. This means that if the estimated price tag for a LFTR stands at $200 million currently, as more units were produced that cost would fall over time – making them increasingly more affordable and economically viable. The following except is from Thorium: Energy Cheaper than Coal, written by Robert Hargreaves, PhD, Professor of Nuclear Physics at Dartmouth University:
The second main benefit of manufacturing LFTRs on an assembly line is standardization, and standardization provides modularity. This becomes important when building small modular reactors because not only are smaller, modular and standardized reactors considerably less expensive to construct, they are also easier to deploy.
If you recall, a core requirement of Scarcity Zero is widespread deployment, as many regions that suffer from the consequences of resource scarcity are geographically remote and/or feature terrain that’s hostile to the construction of something as large as a power plant. A smaller LFTR manufactured on an assembly line can be built rapidly and plugged into any grid in a relatively short time period.
So if, for example, a region needed to quintuple its electricity generation capacity in a matter of weeks, small modular LFTRs make this possible – and they make this possible effectively anywhere. This also would pay dividends toward disaster-relief efforts, peacekeeping missions, ocean trash cleanup and possibly even space exploration.