In debates on energy policy, all sides too often get fixated on magical, hypothetical solutions. Things that donât exist are always âbetterâ than things that do exist, because things that exist are messy, riddled with problems, and usually far from the theoretical ideal.
Much of my objection to a ârenewables onlyâ strategy for fighting climate change is for exactly this reason. As Iâll address later, renewables advocates are as guilty as anyone of selling people on hypothetical technological solutions which have not been proven at any scale and for which supply chains do not exist and would take decades to develop.
But among those of us who are in favor of the âall of the aboveâ strategy for climate change mitigation some are guilty of exactly the same thing.
So, I will not be talking about thorium reactors today. I will not be talking about spent fuel reprocessing (although there are interesting developments in that field). Iâm going to talk about good, old-fashioned pressurized water reactors: the kind that already provides the largest source of low-carbon electricity in the US, provides 75% of the electricity in France, and which is increasingly part of Chinaâs strategy for carbon reduction.
I do also want to address many of the misconceptions around nuclear energy being circulated above, but first I want to answer a more important question: why is it that some people believe nuclear energy to be a crucial tool in climate change mitigation in the first place?
Portable, dense energy with no air pollution
Fuels are useful because they allow humans to largely decouple their economic activity from fluctuations in the weather. For most of our existence, we did get most of our energy from renewable sources. But the development of energy-dense fuels was also essential to the development of industries capable of feeding and sustaining a large population â industries which need the energy portability that fuels provide.
The trouble with most fuels, however, is that we have to burn them to get energy from them. And burning of carbon-based fuels produces particulate matter and compounds such as sulfur dioxide and nitrogen oxides, all of which are known to be hazardous to human health. For instance, the WHO estimates that indoor and outdoor air pollution (largely from the burning of fuels) kills around 7 million people every year world-wide.
Nuclear power plants also use a fuelâââuraniumâââbut the fuel doesnât burn. Nuclear power works by fission, a process thatâs sometimes called âsplitting atoms.â Large atoms like uranium have a lot of potential energy in the nucleus, left over from the supernovae that created them. Certain isotopes of uranium are also prone to splitting when we add a bit of energy to them (in the form of bombarding them with neutrons). When this split happens, some of the potential energy in the original uranium nucleus is converted into heat, and in nuclear power plants, we use that heat to boil water and drive a steam turbine, which turns an electrical generator.
This means that nuclear plants are able to extract energy from fuel without actually combusting it. Nuclear plants produce no sulfur dioxide, no nitrogen oxides, and no particulate matter. The only airborne âemissionâ from the reactor itself is a bit of water vapor.
Relying on fission also means that nuclear energy produces virtually no greenhouse gas emissions.
Below is a chart I made from IPCC data on the life-cycle carbon intensity of various sources of energy:
Note that the Intergovernmental Panel on Climate Change is reporting the life-cycle carbon intensity. This means that these figures already include the mining, transportation, and construction that goes into the production of these forms of energy, not just the running of the plants themselves.
Note also that these figures are produced from the IPCCâs meta-analysis of many studies on the topic. Some people like the cherry-pick specific studies that support their case â the IPCCâs meta-analysis probably gives a more realistic picture.
With this in mind, itâs clear that per unit of electricity produced, nuclear energy produces about the same amount of carbon emissions as wind power, and less than every other form of renewable energy.
All sources of energyâyes, all sources of energyârequire resource extraction. Wind turbines, solar panels, and power plants do not fall from the sky. They are produced by labor using resources extracted form the earth. Wind and solar may have a practically limitless âfuelâ source, but the production and maintenance of the generators themselves, along with the actual service of the electricity, cost resources, just like anything else.
One of the advantages of nuclear energy is that itâs footprint is small in comparison to most other sources of energy. Hereâs another chart I made from data from the U.S. Department of Energy on materials use (table 10.4), along with uranium consumption data from the World Nuclear Association:
The materials use of nuclear energy is, perhaps contrary to popular stereotypes, one of its advantages. Per unit of electricity produced, nuclear requires far fewer material resources than other low-carbon sources primarily because uranium is far more energy-dense than wind or sunlight. Individual solar panels and wind turbines may not require much to produce, but many panels and turbines are required to get appreciable amounts of electricity. These generators must also be maintained, e.g. having blades and panels replaced periodically. And with many generators, many parts need to be replaced.
In contrast, while the initial construction of a nuclear power plant may be resource-intensive, it can provide so much electricity over a long period that this more than offsets the overhead costs.
Nuclear plants also require very little fuel to continue operating. Yes, the chart above does include consumption of uranium, but it barely registers on the chart because it is negligible in comparison to the resources that go into the production of other kinds of energy.
This is another strength of nuclear energy. Uranium is energy-dense enough where a soda-canâs worth of nuclear fuel can provide all of the electricity the average American will use in their entire life.
[Rationale: Uranium fuel has an energy density of about 80 million MJ/kg, or roughly 22,000 MWh/kg, in the units we usually use for electricity. With the volume of a soda can at 355 cubic centimeters, and the density of uranium oxide at 10.97 grams per cubic centimeter, a soda can of low-enriched uranium (~4% enrichment), at typical efficiencies in light water reactors (~30%), produces over 1000 MWh of electricity, more than the average person in the U.S. consumes over their entire life.]
Of course, for every 1 unit of low-enriched uranium, there are about 10 units of material extracted that is currently put to waste. So, letâs say 11 soda cans. Itâs not an ideal situation. But think about essentially anything else we consume throughout our lives: we definitely use more than 11 soda cansâ worth.
All forms of energyâyes, all forms of energyâproduce waste. But nuclear energy produces an extremely small amount of it that is totally manageable.
The way nuclear energy works, the quantity of spent fuel coming out of the reactor is about the same as the quantity of fuel that went in. So how much waste would a nuclear reactor produce providing all of the electricity you would use in your entire life? Well, we have the 1 soda can from the reactor itself, plus the ten soda cans of (currently) wasted material from the mining and enrichment process, plus some irradiated PPE and old parts from the reactor (the cost of which is spread out over the millions of people a nuclear plant can provide electricity for).
So, 11 soda cans of waste, plus a small amount of extra material (PPE, etc.), for all of the electricity an American will ever use.
Nuclear energy produces so little waste that we can safely contain and store all of it, in contrast to most other forms of electricity production which simply dump waste products into the air and water.
I find a good way to illustrate the small nature of the problem is just to have people look at some nuclear waste.
Hereâs a satellite photo of a nuclear plant near where I live:
In the northeast corner, you can see a pad with the spent nuclear fuel that the plant has produced. Yep, thatâs it. In 50 years of operation, providing electricity for hundreds of thousands of people, that is the waste material the plant has produced. There is some more inside the reactor building, and again, some more in the form of âlow level wasteâ like PPE. But this is a managed and contained problem.
This waste is already being dealt with more responsibly than any other source of electricity, including renewables.
For example, the International Renewable Energy Agency estimates that there will be 78 million tonnes of solar panel waste globally by 2050, hundreds of times more than the amount of nuclear waste in the world. And solar waste is not necessarily âsafer.â The disposed products from photovoltaics contain hazardous chemicals such as lead and cadmium which can wash into water supplies if not properly managed.
Likewise, nuclear waste is potentially hazardous if poorly managed, but there is not very much of it, and we know very well how to contain it safely. The material, while radioactive, is not dangerous inside the dry casks you see in the photo above. Those casks will eventually have to be replaced, but itâs not a particularly challenging problem to do so, especially with such a comparatively small quantity of material.
The bottom line is, while there are obviously better solutions out there, nuclear waste is safe where it is until we can conjure up the political will to do something better.
So, we know some reasons now why many people believe nuclear energy is a critical part of any realistic solution to climate change. It provides large quantities of low-carbon electricity with virtually no air pollution and minimal resources use, even in comparison to renewables.
However, nuclear energy gets a lot of backlash, as seen above. Now is the time to address some of the misconceptions that get thrown around.
1. Donât nuclear plants take too long to build?
One of the most common objections to nuclear energy is that it is just too slow to build, and we need to act on climate change as quickly as possible.
There is also an argument to be made that the carbon cost of nuclear should be counted as higher, because we still need to burn fossil fuels during their construction. This would be a cogent argument if one could demonstrate that renewables can categorically be built faster. If that were the case, then the âwasted timeâ building nuclear plants should be something we would need to take into account.
However, there is simply no solid evidence that renewables can be universally built out faster than nuclear. In fact, as researchers Junji Cao et al. note in Science, the fastest nuclear build-outs have been significantly faster, in terms of the number of kilowatt-hours added per person per year, than the fastest renewables build-outs we have seen.
Hereâs a chart on âAverage annual increase of carbon-free electricity per capita during decade of peak scale-upâ from the above journal article:
So why are current nuclear projects in France, the UK, and the US so slow today?
The main reasons are that these countries havenât built a reactor in many years, they are building new reactor designs that canât take advantage of expertise already gained by engineers and construction workers building previous reactors (as opposed to building tried-and-true designs), these projects are being managed largely by private business, and powerful lobbies have succeeded in ratcheting up regulations well beyond the point of being of benefit to the public, making things take even longer.
But examples during peak nuclear roll-outs illustrate that there is nothing inherent in the technology that makes it slow to build.
2. Isnât nuclear energy really expensive?
Nuclear plants have a high overhead cost, but this is meted out over the long life of nuclear plants, where they produce huge quantities of energy. In terms of renewables, the cost of constructing wind turbines and solar panels is low, but these sources tend to increase the service cost of electricity. A thorough study from researchers at MIT looked at the dynamics of real electricity markets and concluded that, under a wide range of different assumptions about the efficacy of various sources of energy, consistently the lowest-cost scenarios for deep decarbonization involve the least variable renewables (e.g. wind and solar), and the most of what they call âfirmâ low-carbon sources (e.g. nuclear).
3. Isnât radiation very dangerous and doesnât the waste last a long time
Cataloguing the various risks associated with radiation is a complex topic that would be its own long-winded post. For now, I highly recommend watching this talk from Gerry Thomas about why public fears about radiation are disproportionate to the actual realities. Suffice it to say that man-made sources of radiation are a minuscule part of the overall ionizing radiation we are exposed to constantly, and of man-made sources, nuclear power plants contribute an even smaller portion. And the ârisksâ that people talk about from nuclear plants are mostly based on speculation, rather than actual epidemiological studies of the concrete impacts, the latter of which show that the impacts are minimal.
As far as âlong-lived wastesâ go, these claims are less than half-truths. People will tell you that nuclear waste is highly radioactive and that it lasts for 200,000 years. However, the truth is that there are materials in nuclear waste that are highly radioactive, and there are materials that remain radioactive for 200,000 years, but these arenât the same elements. In general, things that are highly radioactive decay down to stable states more quickly, because the whole reason they are more radioactive is because they decay (and release their energy) faster. Meanwhile, radioactive elements that decay very slowly donât release very much energy in a given period and thus arenât especially dangerous to be exposed to.
The nasty stuff in nuclear waste decays to safe levels of radioation within a few hundred years. That may still seem like a long time, but itâs a lot less than 200,000. The longer-lived stuff in nuclear waste are leftover uranium and plutonium, which are alpha-emitters which donât post much danger outside of the body. Uranium, for example, can be handled safely without PPE other than a face-mask and gloves. You wouldnât want to eat it, but there are lots of things that are bad to eat that we manage effectively. Nuclear waste isnât a special hazard.
4. Doesnât nuclear energy production involve destructive / dangerous mining?
Nuclear energy production involves mining. So does wind, solar, and hydro. Wind, solar, and hydro involve substantially more mining than nuclear does, as addressed above. And although all mining does cause environmental damage, uranium mining is one of the least destructive forms of mining because uranium is quite common and reactors do not require very much of it at all.
And uranium mining can be done relatively safely if there are good labor laws.
Mining sucks but youâre not going to escape it with renewables.
5. Canât we power prosperous economies with 100% wind, water and, solar?
Probably not. The only examples of places which have been able to power their economies mostly on renewables are places like Costa Rica, which get almost all of their electricity from hydro power, not wind and solar. Hydro is one of a handful of technologies that has proven capable of actually replacing fossil fuels at large scales. But hydro is highly geographically dependent, and often displaces large numbers of people. Nuclear is more portable and takes up far less land.
There are no large-scale example of countries effectively combating carbon emissions with wind and solar. For example, âgreenâ Germany and âgreenâ Denmark have some of the highest carbon emissions from electricity in Europe, despite being some of the most aggressive adopters of wind and solar (see this tool from the OECD where you can compare the carbon emissions from electricity in different countries).
The main reason for this is that wind and solar are highly variable, having capacity factors in the 20-30% range (meaning they are producing on average at only 20-30% of their rated capacity). A consequence of this is that variable renewables struggle to actually replace fossil fuels, and the problem with variability actually gets harder to overcome the more variable renewables you have. In practice, wind turbines and solar panels are backed up with coal and gas, especially when nuclear is being phased out, like in Germany. Germany has seen its carbon emissions and air pollution increase substantially as a result of its nuclear phase-out.
Most proposals for a 100% wind, water, and solar economy either rely on global energy consumption dropping precipitously in the coming decades, or posit the use of hypothetical technologies that donât really exist.
On the first point, even if the US and other wealthy countries were to undergo âcontrolled degrowth,â energy use in the global South is sure to rise rapidly in the coming decades. Banking on world energy use going down is sure to lead to failed policy.
On the second point, one of the most (in)famous examples is Mark Jacobsonâs paper claiming that the United States can transition to 100% renewables by 2050. Christopher Clack and others thoroughly picked this paper apart, pointing out the numerous modeling errors and implausible assumptions in Jacobsonâs analysis. But one of the most striking things is that Jacobsonâs model requires a truly staggering amount of backup storage, all of which Jacobson says is to come from various thermal storage techniques (e.g. UTES) that do not exist at any scale outside of research projects. This is not to say that R&D is worthless, but trying to solve an imminent problem by banking on hypothetical technological breakthroughs is, to put it mildly, a poor strategy.
(As an amusing aside, when Clack, et al. publish their rebuttal to Jabobsonâs implausible model for a 100% renewables economy, Jacobsonâs responded by suing Clack for $10 million dollars for defamation (a sure sign that Jacobson is definitely intellectually honest and knows what he is talking about), before dropping the case.)
6. Arenât we going to run out of nuclear fuel?
Maybe. But many of the claims around this are based on misunderstandings. There are always âonly 40 years of resources leftâ because explorers have no incentive to find new reserves that they are not going to be able to sell for several decades. Existing reserves have historically been a very poor predictor of the resources available in the future. Second, if we are concerned about running out of uranium, then we should be even more concerned about running out of the resources that produce renewables, because we have to mine much more of those.
7. Arenât existing reactors aging and being operated beyond their expected life?
No. The fact that reactors in the US have to be re-licensed every 40 years does not mean they are only designed to last 40 years. And there is no reason you cannot replace components when they wear out.
8. Canât you use nuclear reactors to get nuclear weapons material?
Not using commercial reactors in any way thatâs practical. In spent nuclear fuel, the various isotopes of leftover uranium and plutonium are mixed together, and they could only be separated with centrifuges. If you are going to use centrifuges though, you may as well just dig up natural uranium and use that to make weapons. It would be way easier.
There is a way to get weapons-grade plutonium from a nuclear reactor, but that requires âtoasting the fuelâ, i.e. removing and replacing fuel rods roughly every 30 days. Since the normal fuel cycle for commercial reactors is 1.5 to 2 years, if you were toasting the fuel in a nuclear power plant to get weapons material, it would be obvious to everyone what you are doing, which is why nobody does it.
To cap off this long post I will leave you with a short summary.
Nuclear energy provides large quantities of energy with virtually no air pollution and overall minimal environmental impact, even in comparison with renewables. Thereâs no good reason not to use it, and itâs probably essential for any realistic solution to climate change mitigation.
