Reactors need to be built on time and at lower cost to remain competitive.
At the Vogtle power plant near Augusta, Georgia, the first new large nuclear reactors to start construction in the US for more than 30 years are taking shape. Units 3 and 4 are scheduled to start up in November 2021 and November 2022, respectively, and are intended to keep the lights on in Georgia and Florida, with no carbon emissions, into the 2080s and possibly beyond.
The project has been so fraught with difficulties, delays and cost overruns, however, that it seems likely to be another 30 years at least before anyone tries building another such plant in the US again.
Nuclear power appeals as being a source of reliable electricity without causing greenhouse gas emissions. But new reactors are so expensive that in many countries they are unable to compete with cheap gas and coal or renewable energy sources. If new nuclear plants are to play any significant role in curbing future emissions in developed economies, their costs are going to have to come down a long way.
That is the argument underlying the recent upsurge in interest in new nuclear technologies, including small modular reactors (SMRs). When Fatih Birol, executive-director of the International Energy Agency, gave evidence to the US Senate Committee on Energy and Natural Resources in February, he suggested there were two priorities facing the US nuclear industry. In the short term, it needs to find ways to keep open plants that are running well but faced economic challenges, he said. In the longer term, developing new reactor technologies “will be of crucial importance to have the US leadership continuing in the nuclear domain”.
In the broadest terms, new nuclear technologies divide into two varieties: first, there are those that use water for temperature regulation and enriched uranium fuel, like the standard reactors in use today; second, there are advanced reactors that can have a wider range of coolants including molten sodium or salt and use a wider range of fuels including depleted uranium.
A recent report on breakthrough technologies from the Energy Futures Initiative, a think-tank, and IHS Markit, a research company, suggests that the light water SMRs could start coming into service in 2020-35 while the advanced reactors might be in operation from 2025-30.
Backers of both technologies advocate building new reactors in factories rather than entirely on location to improve productivity and reduce costs.
“The reactors they are building at Vogtle are enormous construction projects, with many thousands of workers on site,” says Ernest Moniz, who was energy secretary in President Barack Obama’s administration and founded the EFI think-tank. “If you have to build a workforce of many thousands, that is not a very easy labour pool to assemble.”
The company that has made the most progress with an alternative approach is Oregon-based NuScale. It has designed a nuclear power module, a 75-foot steel cylinder that can produce 60 megawatts of electricity, and power plants that put one or more of those modules together, sunk mostly below ground level in pools of water.
It is the only design for a small modular reactors that is in the review process with the US Nuclear Regulatory Commission. NuScale has a project under development in Idaho, intended to start up by the end of 2026 and to be in full operation the following year.
The construction budget is $3bn for a plant that will deliver 684MW, working out at about $4.2m per megawatt of capacity. That might look high compared with a gas or wind plant, but is only about a third of the cost of Vogtle units 3 and 4.
NuScale is aiming for a levelised cost of electricity — a commonly used metric for the economics of power generation — of $65 per megawatt-hour, which is within reach of that of a modern combined-cycle gas-fired power plant.
Tom Mundy, Nuscale’s chief commercial officer, argues that its reactor design is “a near-term deployable technology”, unlike some of the more radical approaches.
Other companies, however, argue that the fundamental concerns about nuclear safety and cost can only be addressed with yet more innovative technologies.
TerraPower, a Washington company co-founded by Bill Gates, has designs for two reactors using technologies sometimes described as “advanced nuclear” or “Gen IV”.
Based on “fast reactor” technologies pioneered in the US from the 1950s to the 1970s, they run much hotter than a light water reactor, using molten salt or sodium as a coolant, and do not need to be kept under high pressure.
“That has great benefits in cost and safety,” says TerraPower’s chief executive Chris Levesque. “It’s a game-changer.” Operating at atmospheric pressure avoids some of the demands placed on the equipment in light water reactors. TerraPower is looking for “several billion dollars” of government support to build its first reactor, Mr Levesque says, but he argues that the investment would be worthwhile because the end result will be a lower-cost nuclear power.
“We simply can’t achieve our climate goals, and we won’t have an affordable source of electricity and heat for the future, if we don’t invest in nuclear energy,” he says. “If we are trying to compete with natural gas, we have got to move away from light water technology and get to Gen IV.”