Nuclear Power: Not clean, not necessary, not affordable; not timely

The Nuclear Energy Institute, the industry’s think tank, claims “existing reactors as well as new advanced nuclear technologies will provide the backbone of our clean energy transition….”  Is that right?  Or is a dying industry re-branding itself to keep making money while locking us into a dirty and dangerous energy source – even though solar and wind are much faster and cheaper?

First, take the word “clean.” Let’s leave aside the construction of the power plant; in an extractive world, there is no energy source whose construction does not involve mining resources that cause ecological harm and environmental injustices – solar, wind, nuclear, coal, fossil gas. It is critical to recover the materials used in building power plants as completely as possible for recycling and reuse independent of the source of energy. That still needs to happen, but it is possible in principle.

The essential unsustainability of fossil fuel and nuclear plants is that they need non-renewable fuels to create the heat that is turned to electricity. That need and the processes to extract and make them suitable for use (burning or fission) create waste, pollution, and health and ecological for every kilowatt-hour produced, impacting people now and in the future. Intergenerational injustice is inherent in fossil and nuclear power production.

A nuclear power reactor is just a boiler, like that in a coal-fired power plant, except the fuel is inside the oiler and has vastly more energy potential per ton. The heat generated by “splitting the atoms” of fuel is used to boil water.  The pieces so created – “fission products” – are intensely radioactive, far more than fresh uranium fuel.  (See the basics of nuclear fission here.)   Some of the uranium fuel is converted to plutonium-239 during reactor operation and some of that also fissions to produce energy.  The used fuel – aka “spent fuel” – is so radioactive that it can deliver a lethal dose in under a second when freshly removed from a reactor; it will kill in a few minutes even after cooling for ten years.

Nuclear power is dangerous – and not only because there is the potential for catastrophic accidents like Chernobyl and Fukushima. Not all the plutonium created during power production is consumed during reactor operation. Spent fuel in the United States alone contains enough plutonium to make roughly 100,000 Nagasaki-size nuclear bombs, if it is separated from the spent fuel.  Separated plutonium can also be used as nuclear fuel.  With that hope, a great deal of plutonium was separated in several countries. But it turned out to be so complex and costly that the unused commercial surplus stock is larger than the combined military plutonium stocks of all nuclear weapon states. In a very real sense to exchange fossil fuel generation for nuclear is to exchange CO2 emissions for plutonium-239, which has a half-life of over 24,000 years.

Nuclear proponents often point to the small volume of spent fuel – roughly 20 tons per reactor per year – compared to hundreds of thousands of tons of coal ash waste typically generated for the same amount of coal-generated electricity.  But you can’t make nuclear bombs with coal ash.

Then there is the inconvenient fact that nuclear power creates vast amounts of wastes – but very little of it is at the reactor site.  Each of those tons of “enriched uranium” fuel takes seven to eight tons of natural uranium; each ton of that, in turn, takes anywhere between 100 tons and 2,000 tons of uranium ore, depending on the quality of the ore.  In round numbers, there are roughly a thousand tons of uranium milling wastes at sites that process the ore for every ton of fuel used in a typical US power reactor.  And there is roughly an equal amount of uranium mine waste – overburden and ore that was of a quality too low to process economically.  

As a metric for environmental injustice, consider this fact: there are hundreds of millions of tons of uranium mining and milling wastes in the United States from nuclear power and nuclear weapons production. There are over 500 abandoned mines on Navajo Nation land alone. Environmental justice communities, and especially indigenous communities, literally experience that injustice in their bones and lungs. Mill tailings are toxic with radioactive materials, like radium and thorium, and non-radioactive materials like arsenic and vanadium – all left over when ore is refined to produce purified uranium. The poisons will last essentially forever. Not the definition of “clean.”

The injustice is global. According to the Energy Information Administration, more than 85 percent of U.S. requirements for nuclear power plant uranium between 1996 and 2018 was of foreign origin, creating mining and milling wastes abroad, including in countries that do not have nuclear power plants or nuclear weapons.

All that said, can nuclear power help with the climate crisis? I’ll focus on new reactors in this blog, because without them the industry will die within this century.  The short answer is “no.”

• First, nuclear power plants are expensive: The Wall Street firm, Lazard, estimates that the cost of power from new large-scale solar and wind plants is about $35 per megawatt-hour (roughly the average electricity use of a medium to large house in a month), compared to $170 for nuclear.
• Cost overruns: The cost of nuclear on paper is usually much less than on the ground. The only two reactors (out of dozens) announced in the “nuclear renaissance” of the 2000’s that are being built – the Vogtle 3 and 4 reactors in Georgia – were supposed to cost $14 billion. They are yet to start, and total costs are estimated at more than $30 billion.
• Nuclear plants take too long to build: Ten years is quite common.  Cancellations are common.  Since the beginning of the nuclear power to date, more reactors planned to be built in the United States have been cancelled than have been completed and operated.  The Vogtle reactors were supposed to come on line in 2016 and 2017.
• Cost overruns and delays are regressive: Georgians are paying in advance for the two new Vogtle reactors as they wait – about $10 per month per household, month-in-month-out and year-in-year-out.  It is an excellent case study of the regressive impacts of nuclear: pay now and get the electricity later (or not – as demonstrated by the multi-billion dollar tab of two reactors in South Carolina that were cancelled in 2017).

Large nuclear plants are now generally recognized to be economic lemons; the nuclear industry isn’t ordering any. It has turned to “small modular reactors” (SMRs) even though there will be a loss of economies of scale (which is why reactors are large). The theory is that an assembly-line approach and mass manufacturing will make up for that loss. As a colleague and I noted in a recent paper prepared for the Environmental Working Group, even if that vision is realized, electricity from SMRs will still be as expensive as from the current costly large ones. So SMRs are very likely to be economic failures even when supply chains and assembly lines are established. In the meantime, hundreds of small reactors will have to be custom built; they more expensive than today’s reactors (per unit of power produced).

There have already been huge delays and cost increases in the SMR business even before a single bucket concrete has been poured. Instead of being ready to go by about 2010, as the Department of Energy told Congress two decades ago, the first ones are not due to come on line until the latter part of this decade even if there are not more delays and cost overruns.  We need to have about an 80 percent decarbonized electricity system by 2030 and a fully decarbonized one by 2040, preferably earlier.  If a few SMRs are built by 2030, they will have to be proved out. Then there will have to be hundreds of orders to justify establishing mass manufacturing and building the needed supply chains.  SMRs will be too late to the game even if all works out according to current industry plans. That is unlikely.

Of the dozens of hopefuls, only one SMR, designed by NuScale, has been conditionally certified by the Nuclear Regulatory Commission. One reservation was about the “steam generator” were hot water is turned to steam to drive the turbine. This is the very component that has had to be prematurely replaced in similar present-day reactors. The one stark difference is that in today’s reactors the steam generator is outside the reactor vessel, where it can actually be replaced.  In the NuScale design, it is inside the reactor vessel and will be essentially impossible to replace. You can recall Boeing Dreamliners for a battery defect (as in fact they were).  How do you recall a radioactive reactor? How about hundreds of them? The nuclear establishment does not appear to have any plans for such a contingency, though recalls are a common characteristic of mass manufactured products.

The NuScale reactor basically a smaller version of the same basic elements of most of the power reactors in the United States in which water under high pressure is used as a coolant.  Yet, by any realistic measure, it is likely to fail on cost and, even more, on timeliness grounds to contribute to the urgent, economical decarbonization we need.

Other reactors with different concepts – and there are quite a few – are even more unlikely to make any material contribution in time. I’ll illustrate with the Natrium reactor by TerraPower, supported by Bill Gates and the government.  (Yes, the government is actually giving money to a reactor proposed by a company founded by Bill Gates.)

The basic Natrium design goes back more than seven decades; it is cooled by liquid sodium, which catches fire on contact with air and explodes on contact with water.  Tens of billions of dollars have been spent worldwide to try to commercialize this design – to no avail. That failure is part of the reason there is so much plutonium sitting around in the commercial sector; it had been the goal to use that plutonium mainly in sodium-cooled reactors.

Each design has its safety pluses, as does the sodium-cooled reactor; they all also have minuses. One minus of the sodium-cooled design is that, in certain accident conditions, it can have a runaway chain reaction, the same phenomenon that led to the catastrophic explosion at Chernobyl (which was of a different design). The Germans weren’t as convinced as TerraPower about the safety of the sodium-cooled concept.  They built a demonstration a reactor at Kalkar in the western part of the country, but never switched it on. It is now part of an amusement park.  It will, I think, generally be agreed that there are cheaper ways of building an amusement park.

Coming soon: blog posts on (1) existing nuclear reactors and (2) what to do when the wind doesn’t blow and the sun doesn’t shine.

Opening Photo Credit: Spent fuel pool at a nuclear power plant (NRC file photo)