Thorium Power Dreams as Mythical as the Gods

Image of Thor not to scale. Art by Bonnie Urfer, for Nukewatch.
Image of Thor not to scale. Art by Bonnie Urfer, for Nukewatch.

NOTE: Nukewatch co-director Arianne Peterson produced this investigative report on thorium-fueled power reactors. I present it here in view of recent promotion of thorium as a possible replacement fuel for our faulty and retirement-age nuclear reactor fleet.

By Arianne Peterson, Nukewatch

Named after Thor, the Norse god of thunder, thorium has been touted in recent years as the nuclear industry’s latest “magic bullet,” a fuel solution that will provide cheap, safe, clean, abundant energy while simultaneously helping to solve climate change. Its proponents, like G. Erickson in these pages, argue that thorium deposits, which are about three times as plentiful as uranium throughout the world, are so abundant that it could power the world for many thousands of years. Some even claim that a thorium fuel cycle would not only produce less radioactive waste than conventional reactor fuel but would actually provide a “safe” means of disposing current waste products from both nuclear weapons and energy production.
Perhaps these kinds of fantastic promises sound familiar to those who remember when conventional nuclear power was going to be “too cheap to meter” and was supposedly unrelated to the development of nuclear weapons. As the heavily subsidized conventional nuclear power industry struggles to overcome the stigma of its deadly waste legacy and increasingly uncompetitive production costs, it clings to thorium as one among very few theoretical lifelines. However, the actual advantages of thorium are as mythical as its namesake. In fact, thorium reactor technology is likely to be even more costly, more dangerous, and more polluting than its uranium-based counterpart.


Natural thorium occurs in a metallic form composed almost entirely of the isotope thorium-232 with a radioactive half-life of 14.05 billion years. (“Half-life” describes the amount of time that it takes for an element to release half of its radioactivity, decaying to another element or isotope.) The important fundamental difference between thorium and uranium is that natural thorium is not fissile, meaning it requires an outside source of neutron bombardment to start a reaction. Uranium-235 -- which only makes up 0.7% of worldwide uranium deposits -- is the only naturally-occurring fissile material, meaning uranium-235 is essential to start every first-generation reactor in the world. The use of thorium as a “fuel” will not change this fact.
When thorium is bombarded with neutrons from uranium-235 (or a reactor-produced fissile material), it undergoes a radioactive decay chain that yields uranium-233, which is the actual fissile fuel in a thorium-based reaction. So either the thorium fuel cycle must be started with conventional uranium fuel -- producing a long-lived radioactive waste stream similar to current reactors -- or it could be started with reactor-borne uranium-233 from a previous reaction. As a 2012 report from the Bulletin of the Atomic Scientists points out, this second method would require a prohibitively large infrastructure investment: “The only means of producing a constant source of uranium-233 on the scale necessary to drive a commercial reactor loaded with thorium at start-up would be a massive ‘separations installation’ where spent fuel previously used in the reactor would undergo a process to extract bred uranium-233.” Uranium-233 has a radioactive half-life of 160,000 years.
Though it is often portrayed as a new technology, thorium has been used to produce fissile material since the 1950s. The Atomic Energy Commission (AEC, which later became the Department of Energy or DOE) processed thorium into uranium-233 at Oak Ridge, Tenn. (1954-‘58), Savannah River, S. Carolina (1964-‘69), and Hanford, Wash. (1965-‘70), in part because of its great potential for use in H-bombs. In fact, between 1955 and 1968, several nuclear weapon tests were conducted using uranium-233. In the 1960s, the AEC extensively explored thorium for civilian nuclear energy use, because it assumed the US would have 1000 commercial nuclear power reactors by the year 2000, necessitating an alternative to uranium-only fuel use. In reality, they were off by an order of magnitude, as just over 100 reactors were operating in 2000. Currently, stable uranium prices do not “incentivize” alternative reactor fuel development.


The AEC launched several reactor projects in its effort to prove that the thorium fuel cycle was a safe, economical energy source -- including the Elk River Reactor in Minn., the Molten Salt Reactor at the Oak Ridge National Laboratory, and the Light Water Breeder Reactor at Shippingport, Penn. By 1977, however, thorium fuel cycle research was effectively abandoned after the AEC and DOE had spent billions on its research and development.
The commercial nuclear industry also invested heavily in thorium fuel cycle experiments before they were abandoned. New York’s Indian Point I was the first commercial reactor to use thorium -- in a pressurized water reactor opened in 1962. According to an Oak Ridge report, the cost of recovering uranium-233 from this reactor was a “financial disaster,” with less than 1% of the irradiated thorium successfully converted to uranium-233. The utility abandoned thorium in favor of uranium fuel. The Peach Bottom I prototype – a 40 megawatt, high-temperature, gas-cooled reactor in Pennsylvania -- also used thorium while it operated between 1967 and 1974. This reactor was closed after repeated fuel failures caused significant, expensive down time. The Fort St. Vrain reactor in Colorado -- a high-temperature, gas-cooled 330 megawatt reactor using both thorium and uranium-235 -- operated from for 10 years, until hundreds of accidents involving equipment failure, gas leaks, fuel failures, cracked piping, and human error led to its ultimate closure in 1989.
If thorium were such a miraculous reactor fuel source, the DOE and the private sector would not both have abandoned it decades ago after investing billions of dollars in its development. Without government subsidies on a grand scale, why would a market-driven nuclear industry repeat such a costly failed research experiment?


In fact, waste fuel from thorium experiments still presents a serious radioactive waste problem today -- one which wouldn’t exist if the thorium fuel cycle were a viable energy alternative. As part of its Cold War era nuclear programs, the US produced about two tons of uranium-233. From that amount of waste, which the industry calls “spent fuel,” 1.55 tons of fissile uranium-233 was separated at a cost estimated between $5.5 and $11 billion. Of this, about 428 kilograms (kg) of uranium-233 (about half a ton) is stored at the Oak Ridge National Laboratory in Tennessee, in a building inadequately secured against terrorist and other threats and old enough to be on the National Register of Historic Places. In fact, former DOE official Robert Alvarez calculates that up to 96 kg of the separated uranium-233 may actually be missing. (According to the International Atomic Energy Association, it takes only 8 kg of uranium-233 to make a nuclear bomb.)
In 2001, the DOE determined there was no use for the leftover uranium-233 except possible medical isotope production, which it abandoned in 2005. In an effort to rid itself of this particular environmental and economic liability, the DOE is waiving its own waste security criteria and disposing of most of the uranium-233 in a shallow landfill in Nevada -- a process it expected to complete in August 2014. In 2005 Congress directed the Defense Department to dispose of its remaining 3,222 tons of stockpiled thorium.
Despite this troubled history, thorium promoters still claim it would be more efficient and safer than uranium, while producing less radioactive waste and posing less of a proliferation risk. However, upon closer inspection, each of these claims becomes more and more theoretical, relying heavily on the miraculous development of technologies that remain unproven. Barring an extraordinary technical discovery, the consequences of a thorium revival would be disastrous.


Making electricity from thorium would require either: 1) research, development, and building of a whole new fleet of thorium reactors (it is likely to be at least 10 years before the technology could be built in a commercial-ready format, and another 10 years for the NRC to develop a whole new set of regulations for thorium); or 2) development of a thorium fuel that works within existing reactor technology. The only feasible way to use thorium fuel within the existing fleet of nuclear reactors is with thorium dioxide, the metallic element’s ceramic form. This kind of a “once-through” thorium cycle within existing reactors still requires uranium “enrichment” or “plutonium separation” and thorium “target rod” production for starting the reaction -- technologies which would require costly research and development. Also, a once-through fuel cycle means no extraction of uranium-233 would be possible once the fuel rods are removed. In a price-driven electricity market, reactor operators would have no economic incentive for using thorium this way. Not only would they have to invest in the development of a new fuel rod technology, they would lose the potential cost benefits of re-using the uranium-233 from the used or “spent” fuel. Additionally, the unknown factors involved in using a new reactor fuel pose significant economic risks.
Used within the existing reactor infrastructure, it is true that thorium’s properties could allow it to generate more heat than conventional fuel. However, most of the country’s aging reactors are already operating near the limit of their capacity to turn heat into electricity. A switch to thorium-based fuel would thus require hugely expensive upgrades of heat exchangers and turbines to turn that heat into electricity. Again, this kind of investment would be difficult to justify in the current electricity market. Studies also suggest that it would be exceedingly difficult for thorium -- introduced into existing reactors in thorium-uranium dioxide pellets -- to perform well in longer fuel cycles.
Thorium promoters often refer to its potential for use in a “breeder” reactor system -- that is, one that creates more useable fissile fuel products than it requires as initial fuel (in thorium’s case, recovering the uranium-233 that results from the reaction, for use to fuel the next reaction). This sounds great in theory, but the reality is that no thorium breeder that does not require “reprocessing” (see section below) has been successfully developed -- though several countries continue to try. The Oak Ridge National Laboratory tested “molten salt” reactors in the ‘60s, with efforts culminating in an early prototype of a thorium breeder reactor that used uranium and plutonium as fuel, but -- importantly – the lab was never able to deploy the “thorium blanket” that would have been used to actually “breed” the uranium-233 to be recovered from spent fuel.


In its natural state, thorium is far more radioactive than uranium, so its handling requires extra safeguards. The surface dose rate from a 55-gallon drum of thorium is about 60 milli-Roentgens per hour (60 mR/hr), or about 13 times higher than a similar drum of uranium. A worker in a thorium storage facility could expect to encounter estimated dose rates of 60-100 mR/hr, meaning they could reach the US occupational exposure limit of 5 rem-per-year in just over six days. Clearly, this has serious economic and safety implications for thorium’s use in commercial reactors.
Uranium-232 -- a byproduct of the decay of thorium-232 into uranium-233 -- is 60 million times more radioactive than uranium-238, which makes up the vast majority of uranium ore. Uranium-232’s intense gamma radiation penetrates body tissues and is difficult to shield against. With a radioactive half-life of 72 years, uranium-232 is a dangerous product of the thorium fuel cycle that needs to be containerized for 10 half-lives, or 720 years.
Andrew Nelson, a research scientist in the Materials Science and Technology Division at Los Alamos National Laboratory, wrote in Bulletin of the Atomic Scientists, “The claim that thorium fuels are ‘meltdown proof’ has no basis in reality, barring the design, development, and construction of completely new reactor types.”
Thorium-uranium dioxide fuels used in conventional reactors could indeed have favorable safety attributes compared to conventional fuels -- studies suggest they would melt down more slowly in a loss-of-coolant accident, and thorium fuels may be more resistant to cracking or fracture (which creates more surface area). In the event of a fuel cladding failure or accident, extra surface area would more rapidly release collected fission products. But, as Nelson noted, “[I]t is important to consider that the response of the fuel is, in reality, of engineering importance only if the [fuel] cladding fails. If enhanced accident tolerance is the primary goal, it makes sense to first address the cladding material itself.” Research dollars would be better spent on these types of safety measures for existing reactors, instead of on an unproven new technology.


The use of thorium-uranium dioxide pellets in current reactors would provide a small reduction in the production of transuranic elements (heavier then uranium) such as neptunium, plutonium, and americium. However, as Nelson points out, unless the federal government changes its policy in taking financial responsibility for long-term radioactive waste produced by private companies, the potential for realizing this small transuranic reduction will not provide incentive for reactor operators to go through the process of developing a thorium fuel source.

Using thorium within an “advanced fuel cycle” reactor (including a “breeder” reactor as mentioned above), still requires “reprocessing” under current technology. Reprocessing means dissolving the used, and extremely radioactive, waste fuel in acids and then chemically separating the fissile uranium-233. This reprocessing leaves liquid and highly radioactive wastes that hold dozens of fission products (pieces of split atoms, “activation products” and “transuranic elements”). At the US government’s Hanford and Savannah River sites, reprocessing for weapons plutonium has produced hundreds of millions of gallons of these ferociously radioactive wastes. Over a million gallons of this high-level liquid waste has escaped into the soil above the nearby rivers and wetlands. Thorium reprocessing poses the same problems and the same degree of hazard as this historical reprocessing.
Like uranium mining, thorium mining also produces long-term hazards. Beyond leaving behind millions of tons of radioactive mine tailings, both types of extraction are associated with hazardous non-radioactive metals in their mill tailings, not to mention fossil fuel inputs and their accompanying carbon emissions.


Its proponents often say thorium is not a proliferation risk because the gamma radiation from its uranium-232 byproduct is so deadly. However, Robert Alvarez argues that the uranium-232 concentrations in uranium-233 do not keep it from being used by a suicidal terrorist, because its deadly effects are not immediate. Based on a DOE report about uranium-233 waste currently held at Rocky Flats, Alvarez found that at a 5 to 10 parts-per-million content of uranium-232 (some of the current waste does have this low a concentration), the material would emit about 5 R/hour at a distance of one foot; therefore it would take one to two weeks of exposure at 12 hours per day before the accumulated dose would be lethal.
Again, uranium-233 can be used to make nuclear bombs. In fact, it is in many ways preferable to plutonium for use in weapons, because it does not require advanced shaping and implosion technology to be made into a bomb. Recognizing uranium-233’s proliferation risk, the DOE requires stringent safeguards for any quantity of uranium-233 over two kilograms -- to prevent “onsite assembly of an improvised nuclear device.”
Some thorium promoters also claim that the thorium cycle (with reprocessing) can incorporate uranium-238 mixed with the uranium-233 produced by the reaction, so it is not a proliferation risk. While this does dilute the uranium-233, it results in the production of more plutonium-239, which is itself a bomb-making material.
Like most new technologies proposed by the nuclear industry, the supposed benefits of thorium are, in fact, too good to be true. There is a tiny chance that, despite decades of research and billions of dollars already invested in the technology without success, further investments could yield a method for using thorium as a fuel that is better than the uranium used currently. There is also a chance that we could discover a new element on the moon that would solve all our energy problems. But with climate change, aging reactors, radioactive waste, and other energy crises looming, investing in proven wind, solar, conservation, and energy efficiency solutions avoids all the unnecessary risks of staying nuclear. Studies like the Rocky Mountain Institute’s Reinventing Fire and the Institute for Energy and Environmental Research’s Carbon-Free and Nuclear-Free have proven it is possible to mitigate climate chaos right away using readily available, safe, clean, renewable technologies. Unproven, environmentally hazardous, radiotoxic “solutions” like thorium are only promoted by corporations and their helpers that stand to profit from their implementation -- and the governments foolish or corrupted enough to have taken on the financial burden of new research and development.
– Arianne Peterson is a co-director of Nukewatch. A shorter version of this report appeared in the Summer 2014 Nukewatch Quarterly, and a longer analysis with all 29 footnotes is available at .