Does the highly touted, allegedly advanced nuclear reactor former Microsoft found Bill Gates is financing have inherent safety issues? According to nuclear physicist Edwin Lyman (PhD, Cornell, 1992) of the Union of Concerned Scientists, the Natrium reactor being developed by Gates’s Terra Power company has a “positive void coefficient of reactivity.”
That means that when the cooling system is disrupted by voids – bubbles and other disruptions – or by a loss of coolant, the nuclear chain reaction can increase. For conventional U.S. light water reactors, the void coefficient is negative, meaning that when there are voids, the chain reaction slows down or ceases.
A positive void coefficient designed into the Soviet’s Chernobyl reactor in what is now Ukraine led to the 1986 explosion, the worst reactor accident in history. It killed 39 firefighters trying, unsuccessfully, to shut down the chain reaction and spread radiation over vast areas of western Europe. The giant plant lost its water coolant, but the graphite moderator remained and the nuclear reaction when “prompt critical,” essentially exploding.
The U.S. Nuclear Regulatory Commission has historically avoided licensing reactor designs with a positive void coefficient, including Canada’s heavy-water CANDU machines, which have a slight positive void coefficient.
In an email asking if the Gates Natrium reactor had a positive void coefficient, Lyman told The Quad Report that “based on the core size, I would expect it to have a positive void coefficient.” He added, “In a sodium-cooled fast reactor, the accidents of concern are not as much LOCAs (loss of coolant accidents) as ‘loss of flow’ events where the coolant pumps lose power or seize up.”
In a March 2021 UCS report, Lyman wrote that sodium cooled fast reactors (SFRs) “have numerous safety problems that are not issues for LWRs. Sodium coolant can burn if exposed to air or water, and an SFR can experience rapid power increases that may be hard to control. It is even possible that an SFR core could explode like a small nuclear bomb under severe accident conditions. Of particular concern is the potential for a runaway power excursion: if the fuel overheats and the sodium coolant boils, an SFR’s power will typically increase rapidly rather than decrease, resulting in a positive feedback loop that could cause core damage if not quickly controlled.”
Lyman told The Quad Report, “That’s why the notion that such a reactor is ‘passively safe’ is so absurd.”
The Natrium reactor, which Gates hopes to build on the site of a former coal-fired power plant in Wyoming, is a 345-MW unit that will melt a salt compound, which will then be stored on site. The heat will be used to make steam and generate up to 500 MWe, allowing the plant to follow load.
While not addressing the void coefficient directly in public, Terra Power’s web site obliquely addresses it: “The reactor operates at temperatures greater than 350 degrees Celsius (the equivalent of 662 degrees Fahrenheit) and far below the boiling point of sodium,” suggesting that coolant voids are not going to be a problem. Terra Power also notes that the “first-of-a-kind cost for the Natrium demonstration plant will include the reactor design and licensing, codes and methods development, fuel development and qualification, and the design, construction and operation of two supporting facilities: the Natrium Fuel Fabrication Facility and Sodium Test and Fill Facility. The sodium facility will be used to test and demonstrate the performance of first-of-a-kind equipment prior to operations in the reactor plant.”
Sodium coolant is a safety issue. There have been problems with liquid sodium cooled reactors. The most famous was the 1995 accident at Japan’s 280-MWe Monju fast breeder reactor. A vibration caused a sodium pipe to burst, spilling some three tons of sodium onto the reactor floor. The sodium reacted with air and water, producing caustic fumes and enough heat to warp the structural steel in the room.
The quasi-government agency in charge of Monju tried – and predictably failed – to cover up the accident. The effort included falsified reports, doctored videos, and a gag order to employees.
Japan restarted Monju in 2000. The machine was plagued by a series of problems, including sodium incidents. Japan gave up on Monju in 2016. The decommissioning is expected to be completed in 2047 at a cost of 375 billion yen (about $3 billion).
Another issue with the Natrium design is the lack of a conventional containment structure. Conventional pressurized water reactors have large, dry reinforced concrete domes designed to keep radioactivity from a reactor accident from spreading outside of the reactor building. General Electric boiling water reactors have less robust water-based “pressure suppression” systems designed to prevent release of radioactivity. They failed dramatically in the 2011 Fukushima catastrophe.
PWR containments are expensive and add complexity to the reactor site. Terra Power, working with GE, has proposed what it calls a “functional containment.” In a document filed with the NRC, they describe the Natrium containment as “consisting of multiple barriers internal and/or external to the reactor and its primary coolant boundary shall be provided to control the release of radioactivity to the environment.” The diverse barriers identified are “fuel cladding,” “reactor vessel & reactor head,” “concrete structure and seals,” “above grade building and structures.” The paper notes that the federal government has approved functional containment for the now decommissioned Fort St. Vrain high temperature gas reactor in Colorado [decommissioned after a less-than-stellar history—Ed.], which the Atomic Energy Commission, the NRC’s predecessor, licensed in 1972.
A key difference exists between the HTGR and the Natrium design. Gas-cooled reactors have no void coefficients, positive or negative.
–Kennedy Maize