Generation III nuclear reactors have not shown much ability to overcome the weaknesses of conventional Gen-II light-water reactor technology, offering at best evolutionary approaches. Is there room for a more revolutionary approach? Many parties are exploring new technologies, but it’s impossible to tell which, if any, will succeed.
Last August, Andy Revkin, The New York Times’s “Dot Earth” blogger, waxed enthusiastic about a new fusion reactor design from a team of students and researchers at the Massachusetts Institute of Technology (MIT). Revkin reported (based mostly on the MIT press release) that the team had come up with a plan for a “demonstration-scale fusion energy power plant that could actually produce a fusion energy machine that is affordable, robust, compact.”
MIT claimed that the new design takes the long-disappointing “tokomak” donut-shaped fusion reactor and shrinks it with new materials. This would allow the use of a much higher magnetic flux to contain the superhot plasma needed to fuse hydrogen atoms, which would produce heat far exceeding any conventional fission reactors, yet with a much smaller footprint. The hype suggested a 10-year time frame.
Despite the credulous coverage of MIT’s dream machine, there are important skeptics, many of whom note that fusion is a technology whose horizon has receded despite years of research and billions of dollars of government investment. Robert Hirsch, who ran the fusion program for the Atomic Energy Commission (AEC, the predecessor of the Nuclear Regulatory Commission) and the Energy Research and Development Administration in the 1970s, told POWER, “Higher magnetic field tokamaks have been around since the early 1970s, but high magnetic fields contain high stored energy, which can be released when [superconducting] magnets quench, which the regulators will be very sensitive to.” Magnet quenching—abrupt termination of the magnetic field—can result in destructive forces inside the machine and considerable damage.
The MIT reactor design, which MIT is calling the ARC (Figure 1), hits all the proper notes to attract attention: small, modular, and efficient. But it’s just one of a number of “new” (actually mostly old but previously discarded) reactor models various engineers and entrepreneurs are advancing as the solutions to the well-known woes of conventional, large-scale light-water reactors (LWRs).
|1. Pocket powerhouse? This fusion reactor design from the Massachusetts Institute of Technology (MIT) promises larger power through a smaller footprint. Courtesy: MIT|
A Nuclear Gen-X
Call them “Gen-Next” reactors, as they do away with the conventional numerical nomenclature of Gen-I (small, early plants such as Indian Point I in New York, now long closed), Gen-II (most of the large plants ordered in the 1970s and operating today), and Gen-III (today’s designs, such as the Westinghouse AP1000 and AREVA’s EPR, under construction but not yet operating). Gen-IV, the industry’s label for advances over Gen-III designs, implies more of the same, while Gen-Next implies radically different approaches, with much promise and plenty of risk (see sidebar).
Earlier conventional reactor designs are being phased out. The Gen-Is are gone. Many Gen-IIs are nearing retirement. But Gen-IIIs have not met stated goals for plants that are cheaper and easier to build, feature much greater standardization, and offer modular construction advantages over the prior generation.
Investment portfolio manager Henry Hewitt wrote in Greentech Media recently that Gen-III reactors “have been a disappointment.” None are currently operational, and many of the plants under construction have seen delays and budget overruns—some of them huge, as with the EPR. The latest World Nuclear Industry Status Report (a publication that is skeptical of nuclear power) attributes these delays, including those at the four Westinghouse units under construction in the U.S., to “design issues, shortage of skilled labor, quality control issues, supply chain issues, poor planning either by the utility and/or equipment suppliers, and shortage of finance.”
Those looking to nuclear power as a long-term component of a plan to limit carbon dioxide emissions have for more than a decade been examining and touting new generations of nuclear concepts that escape the limits of the LWR (see “Nuclear Industry Pursues New Fuel Designs and Technologies” in the March 2015 issue). These new technologies include designs that rely on thermal (slow) neutrons, fast neutron breeder reactors, various cooling approaches, higher-temperature machines that are more efficient, and the ability to burn spent nuclear fuel from those conventional LWRs, which look to be around for a very long time.
Salt of the Earth
The most recent leader of the Gen-Next hit parade is the molten salt reactor. Nuclear scientists Leslie Dewan and Mark Massie are designing what they call the “Waste-Annihilating Molten Salt Reactor” (Figure 3). Starting with designs originated at Oak Ridge National Laboratory in the 1950s and 1960s, Dewan and Massie are developing plans for a liquid-fueled reactor that overcomes some of the problems with the 7.5-MW Oak Ridge plant that operated from 1966 until 1969, when the money ran out. Their design also addresses important current problems with LWR plants.
|3. Salty prospect. Experimental reactors using molten salt as a coolant and fuel carrier have been around since the 1960s, but new designs hope to overcome past problems. Courtesy: Transatomic Power|
Dewan and Massie have formed a company, Transatomic Power, to commercialize their concepts. They have raised $6 million in venture capital so far. Unlike the old Oak Ridge reactor—championed by nuclear power founding father Alvin Weinberg (1915–2006)—their design does not use fast neutrons in order to breed plutonium. It is a thermal device, with the fuel in suspension in the coolant, a key feature of Weinberg’s Oak Ridge machine. According to their website (transatomicpower.com), the fuel can be either “fresh” (unenriched) uranium or spent fuel, unlike the 33%-enriched uranium in the Oak Ridge prototype. The developers claim that the “main technical change” they have made from the Oak Ridge days “is to change the moderator and fuel salt used in previous molten salt reactors to a zirconium hydride moderator, with a LiF [lithium fluoride]-based fuel salt.”
This machine, according to its developers, can run on spent fuel, can burn up as much as 96% of the energy in the fuel, and should provide exceptional safety. It avoids the intense radiation damage from fast neutrons. If the plant loses all of its electric power, a phenomenon known as “station blackout” (such as what occurred at Fukushima), the fuel drains into a tank and freezes solid. There can be no meltdown.
But there are tough challenges to overcome, particularly handling the highly corrosive molten salt, which carries the fuel, serves as the moderator, and cools the reactor. Licensing will be a problem in the U.S. because of the novelty of the design. The reactors could require on-site chemical plants to deal with the coolant and fuel mixture. Nor are the economics of the technology at all clear.
In 2006, multi-billionaire Microsoft founder Bill Gates and former Microsoft chief strategist Nathan Myhrvold concluded that raising living standards globally—including providing access to electric power to all—requires the private sector to step into the action. They founded Intellectual Ventures, located in Bellevue, Wash., to come up with ideas for reducing global poverty. Two years later, they spun off a new company, TerraPower, to focus on a new approach to an old, largely unsuccessful, nuclear technology: fast breeder reactors, which generate electricity while producing more plutonium fuel than they consume from natural uranium and fast (unmoderated) neutrons.
TerraPower’s wrinkle on breeders was called the “traveling wave reactor” (TWR) first proposed in the 1950s, according to the Alvin Weinberg Foundation. It was a sodium-cooled fast breeder design to burn the plutonium it breeds internally from conventional spent reactor fuel, without the need for plutonium reprocessing. That’s a big technical, economic, and environmental advantage. Nuclear reprocessing is fraught with problems, including diversion of plutonium from civilian reactor fuel into atomic weapons.
An article in MIT’s Technology Review reported, “As it runs, the core in a traveling-wave reactor gradually converts nonfissile material into the fuel it needs.” Another description, from ZDNet, likened it to a candle, burning from one end to the other.
The technology got a lot of hype. New York Times reporter Matthew Wald (who wrote the Technology Review article and now works for the Nuclear Energy Institute), wrote that TerraPower’s idea could answer the quest “for a new kind of nuclear reactor that would be fueled by today’s nuclear waste, supply all the electricity in the United States for the next 800 years and, possibly, cut the risk of nuclear weapons proliferation around the world.”
It was an intriguing concept. But TerraPower soon abandoned the traveling wave in the face of an engineering conundrum. The reactor’s liquid sodium cooling system had to follow the wave, a very tricky task. Instead, TerraPower changed the design so the uranium-plutonium conversion does not move. “It’s just the practical considerations associated with making the most of every neutron, and the engineers’ love of keeping the cooling system in one place and not chasing the wave,” TerraPower’s CEO John Gilleland told Weinberg Foundation blogger Mark Halper. TerraPower continues to call its machine the traveling wave reactor, although now it’s more of a standing wave. The company says it hopes to achieve startup of a 600-MW prototype TWR in the mid-2020s.
TerraPower’s move away from its original technology recognized the problems with liquid sodium coolant. While it has excellent heat transfer properties, it also has a tendency to leak and come into contact with air and water. When that happens, it can spontaneously ignite in a wild reaction. That’s a real problem, because sodium fires are very difficult to fight. They create caustic fumes, release explosive hydrogen, and can’t be quenched with water or CO2. Breeders in operation in the world so far have had serious sodium coolant problems.
TerraPower has broadened its interests in Gen-Next nuclear, including looking at molten salt reactors.
On Sept. 22, the company announced a memorandum of understanding with China National Nuclear Corp. (CNNC). A press release sent to POWER said that “The two companies plan to work together to complete the traveling wave reactor (TWR) design and commercialize TWR technology. Cooperation between TerraPower and CNNC will speed technology development, promote clean energy growth, and enable global economic growth in both countries” (Figure 4).
|4. Traveling wave reactor design traveling the world. In September, Bellevue, Wash.–based TerraPower announced it had signed a memorandum of understanding with China National Nuclear Corp. to complete the design and commercialize the new form of nuclear technology. Courtesy: TerraPower|
Breeder reactors remain the nuclear Holy Grail for many plant developers, despite the mixed record of earlier generations, because they offer an endless supply of fuel generated by transmuting uranium into plutonium. Among various projects, the most advanced appears to be Astrid (Advanced Sodium Technological Reactor for Industrial Demonstration), a sodium-cooled fast breeder project led by the French government’s energy research agency CEA, working with AREVA, Electricité de France, and Toshiba.
Japan and France, countries with few indigenous energy resources, have long seen breeder reactors and plutonium reprocessing as the path to greater energy independence. But the breeder programs in both countries have had problems. France’s Superphenix reactor closed in 1998 after serious engineering problems with the liquid sodium coolant. Japan’s Monju reactor also closed after similar problems, including a nasty coolant fire and a utility and government cover-up.
Last May, the governments of France and Japan announced they would move forward with Astrid, aimed at developing a demonstration breeder burning spent nuclear fuel at CEA’s Marcoule site near Avignon on the Rhone River. In 2011, France put up $900 million to fund Astrid through 2017. Japan Prime Minister Shinzo Abe and French President Francois Holland signed a joint agreement to “intensify their civilian nuclear research,” while both countries were restructuring their nuclear programs. Japan is still recovering from the Fukushima disaster, which led to the shutdown of all of the country’s nukes, now slowly coming back into service. France, meanwhile, is aiming to reduce its dependence on nuclear, from 75% of its electricity production down to 50%.
According to the World Nuclear Association, “Astrid is envisaged as a 600 MWe prototype of a commercial series of 1500 MWe [sodium fast reactors] which is likely to be deployed from about 2050 to utilise the half million tonnes of [depleted uranium] that France will have by then and also burn the plutonium in used MOX fuel. Astrid will have high fuel burnup, including minor actinides in the fuel elements, and while the MOX fuel will be broadly similar to that in PWRs [pressurized water reactors], it will have 25-35% plutonium. It will use an intermediate sodium coolant loop, and the tertiary coolant is nitrogen with Brayton cycle.” CEA says a final decision on whether to build the Astrid prototype will come in 2019. As is the case with all advanced nuclear technologies, the economics of commercial plants are entirely unknown.
Hot, Hot, Hot
One weakness of LWRs is the low quality of their steam, which reduces the efficiency of the plant. Typical steam outlet temperatures for PWRs are under 400C, which compares unfavorably to modern ultrasupercritical coal-fired plants that operate with steam temperatures above 600C.
Very-high-temperature nuclear reactors with outlet temperatures around 1,000C are possible and have actually been demonstrated, although not commercially in the U.S. The AEC and the Philadelphia Electric Co. teamed up on a high-temperature gas reactor (HTGR), a 40-MW graphite-moderated, helium-cooled design originated by General Atomics in the 1950s. The unit located at the company’s Peach Bottom site operated well from 1967 to 1974, with a lifetime capacity factor of 75%.
As a result, Public Service Co. of Colorado, with AEC financial backing, ordered a 300-MW HTGR based on a scale-up of Peach Bottom—the Fort St. Vrain plant. But that plant operated poorly from the day it opened in 1979 until it shut down just a decade later. Both Peach Bottom and Fort St. Vrain used large prism-shaped fuel blocks of enriched uranium surrounded by a graphite moderator.
Since then, much time, effort, and money have gone into attempts to revive HTGRs, mostly focused on a General Atomics design that uses billiard ball–sized “pebbles” consisting of many 9-mm uranium fuel spheres embedded in graphite (the moderator), surrounded by a ceramic coating. These are known as “pebble bed reactors.”
According to noted nuclear engineer Andrew Kadak, an HTGR is fundamentally different from LWRs. The differences include higher thermal efficiencies, an inert and noncorrosive helium coolant, lower water requirements, the use of gas turbine technology, and a less-complicated design, because there is no emergency core cooling system like those required in LWRs.
Philadelphia Electric, before it was acquired by Exelon Corp. in the early 2000s, spent some $20 million for a 12% share in a joint venture with South Africa’s state-owned Eskom utility to develop a commercial 110-MW pebble bed HTGR, aimed both at South African and U.S. markets. After Chicago-based Exelon took over, spending on the project stopped in 2002. Eskom continued work but dropped it in 2010, citing “run-away costs and technical problems.” According to Kadak, among the disadvantages of HTGRs are the poor operating history, “little helium turbine experience,” and “licensing hurdles due to different designs.”
Yet industry interest in the HTGR concept remains. A consortium of New York utilities, the New York State Energy Research and Development Agency, National Grid, and South Africa’s Pebble Bed Modular Reactor Co. in 2013 pitched the design into the U.S. Department of Energy’s (DOE’s) second-round competition for financial support for small, modular reactors (SMRs). That bid was unsuccessful, losing out to NuScale’s more conventional LWR-based SMR technology. NuScale is now negotiating with the DOE over terms and conditions for a cooperative funding agreement.
Nuclear reactor designers have produced a wide variety of fascinating concepts for alternatives to the light-water technology that is the world’s go-to choice for atomic energy. But these exotic designs, no matter how elegant, exist mostly on paper and may not be practical or economic. For now, LWRs rule the real world. ■
—Kennedy Maize is a frequent contributor to POWER.
[Note: Phrasings of some passages in the original version of this article suggested that Transatomic Power Corp. was affiliated with MIT, which it is not, although its founders are MIT graduates. Minor edits to remove that confusion have been made 11/10/15.]