I am reading https://www.politico.eu/article/eu-nuclear-energy-gas-green-climate-technology/ that this (2022) year the EU will have to decide on the labeling of nuclear energy - is it sustainable and/or green and/or at least can help the transitioning to the green, carbon-free and sustainable Europe and world?

Is there scientific/academic consensus on this question, e.g. in the form of IPCC report or something similar?

p.s. I am watching Don't look up in these days and I feel that nuclear scientists and engineers are something like astronomers in this movie - they have the important part of the solution by they are being isolated by the secular religion, political and business interests or just popular subculture.

p.p.s. I am also reading scientific papers about the nuclear fusion synthesis of heavy metals. Many green technologies require heavy (rare, precious) metals and the price of them can stop the greening of the world. That is why we should be bold and try to seek the synthesis of those metals in nuclear fusion reactors. Yes, this is quite a distant future because of technical hurdles, but still - scientifically it is possible and only the technical/engineering issues should be solved and the price of this technological process should be lowered. My guess is that nuclear engineering can one of the main component to both the green world and to transhuman world of superabundance. That is why I hope that we will not kill nuclear technologies.

For background information see this UN short report https://unece.org/climate-change/press/international-climate-objectives-will-not-be-met-if-nuclear-power-excluded

  • 2
    I am not sure that there us a "definitive" definition of sustainability, and I am sure that there is no definitive definition of "greenness", so it's pretty hard for anything to be definitively and authoritatively "green". Commented Jan 3, 2022 at 16:14
  • 2
    Fusion synthesis is mind-bogglingly expensive. Further, elements heavier than nickel consume energy when produced. Synthesizing a single gram of neodynium from iron in a perfectly efficient way, for example, would consume about 40 GJ of energy -- about as much energy as you'd get from burning a ton of oil. Apart from using neutron capture to synthesize specific isotopes, don't expect to ever see fusion synthesis outside of the laboratory.
    – Mark
    Commented Jan 3, 2022 at 21:28
  • To put this in perspective - 1 ton of oil costs 600$ (80$ per barrel). 1g Neodymium costs around 0.04$, 1g gold costs around 50$, 1g Rhodium costs around 400$. Rhodium with atomic number ~45 is far lighter than ~ 60 Nd, that means that energy consumption can be lower. So, from the energy cost it can be economically viable to produce Rh even today.
    – TomR
    Commented Jan 3, 2022 at 21:48
  • 1
    @TomR, cheaper, yes, but not dramatically so. At a rough estimate, synthesizing rhodium from iron is around 30 GJ per gram -- the savings from the lighter atoms is mostly countered by the need for more atoms to make up each gram. And remember: this is assuming perfect efficiency. Current techniques typically get efficiencies of a few percent per step at best, with a minimum of 17 steps needed to turn iron into rhodium.
    – Mark
    Commented Jan 4, 2022 at 22:13
  • 1
    In general in a nuclear vs chemical debate nuclear is more sustainable (chemical is fine in for energy storage unless we are burning it in an un-clean manor, but not as a source). This is because of the extreme energy density of nuclear which means much less fuel and waste per unit energy. Commented Jan 5, 2022 at 1:11

4 Answers 4


Sustainable? What do you count as sustainable? Do you consider for example solar power or wind power or geothermal power sustainable?

They aren't. If you extract solar power, eventually the sun has run out of fusion fuel, and you can't extract solar power anymore. Same for wind -- it's just harvesting solar power in an indirect way. Geothermal power relies on earth having enough thermal energy. The faster you extract geothermal heat energy, the faster it runs out.

However, if we bend a little bit the definition of "sustainability", and consider solutions that have enough fuel for 5 billion years "sustainable", then some forms of nuclear energy are sustainable. Nuclear energy based on U-235 isn't (as only 0.7% of natural uranium is U-235, so eventually we will run out of U-235 and have only U-238 left), but nuclear energy based on U-238 (converting it to plutonium in breeder reactors) is sustainable. Also, thorium reactors are sustainable too (they convert thorium to U-233) because essentially all of thorium is the isotope 232, and it can be converted entirely to U-233 in breeder reactors.1

Also good breeder reactors that have proper waste processing (a lot of the waste, especially of the most active waste, is good nuclear fuel) can provide a waste stream that decays faster than the waste stream of our once-through reactors. So the waste issue is mostly solved, because only 300-400 years of storage is needed for the waste (it decays in 300-400 years to the activity of natural ore) as opposed to hundreds of thousands or millions of years of the waste of current reactors.

Current nuclear reactors, that aren't passively safe, are by far the safest electricity generation method we currently have. Compared to it, for example wind power is fatal (it kills people servicing the wind turbine at great heights) and solar power is fatal (it kills people on the roof, servicing the solar array, resulting in people getting fallen from the roof). Any combustion-based electricity production method should be banned right away if we need the very good safety of current nuclear reactors.

Passively safe reactors, by the way, are safe even in the hands of incompetent people. For example, the General Atomics TRIGA nuclear reactor in Finland was operated by university students. It was in the middle of a university building. No accidents happened, because it was designed to be safe. The university was very near the city center, and there were huge amounts of people living in 5 mile radius. No problem there. We could build all nuclear reactors that way: so that they can be operated in the middle of a city. That way, their waste heat could be economically utilized in city district heating.

However, current nuclear reactors aren't the only way. In current reactors, every plant is a custom design and has to be proven safe enough separately. They also aren't passively safe. However, there is nothing preventing construction of nuclear reactors in factory assembly lines, so that only the general design needs to be proven safe once, and everyone would use essentially the same reactor. They can also be made passively safe, to prevent the accident of Fukushima for example (if we consider it important enough to prevent a future Fukushima -- the death toll of the Fukushima nuclear accident by the way is far below the death toll of the tsunami that caused it; completely different orders of magnitude here).

The big mistake in nuclear power was stopping research, development and construction of new nuclear reactors in 1980s. Because of this, we don't have passively safe reactors. We don't have reactors produced in factory assembly lines (very cheap!). We don't have breeder reactors that produce enough electricity for 5 billion years. We don't have reactors that produce waste that decays to natural radioactivity in 300-400 years. We don't have thorium reactors.

The best choice, right now, if we want to have any chance of stopping the climate crisis is to construct massive amounts of all clean electricity production methods, as much as we can. This includes hydropower. This includes hydrogen-based energy storage. This includes battery-based energy storage. This includes wind power. This includes solar power. This includes current nuclear reactors. And before someone claims that we haven't solved the waste problem of current nuclear reactors, we have: Finland has geological permanent storage for nuclear waste, good for the amount of time it stays radioactive.

However, we should not stop there: we should begin the R&D to create reactors produced in assembly lines, passively safe, capable of utilizing all natural uranium and thorium in breeder mode, capable of producing waste that needs storage for only 300-400 years.

About the EU labeling, by the way, it's bullshit. They are planning to consider natural gas "sustainable". The only way it can be sustainable if it's used only for peaking power plants (so its capacity factor is below 15%) and if the carbon dioxide produced is captured and stored back underground from where the natural gas came. The massive production of baseload electricity using natural gas is nothing but foolish.

Edit: 1 When reading this answer again long time after writing it, I have to note that the reason U-238 is sustainable is the fact that once you use all of the uranium, not just 0.7% of it (U-235), you can tolerate 142.86 times higher price of uranium. This means that you can extract uranium from seawater. So claiming that we have enough U-238 for 5 billion years is not equivalent to the claim that we have enough U-235 for 0.007*5 = 0.035 billion years or 35 million years (which we don't). The current reactors that utilize U-235 only require the uranium to be very cheap because they are very inefficient in using the uranium, meaning that they can tolerate only a very low price of uranium. This means that the vast majority of available sources of uranium are too expensive. Maybe I should have explained this better in the first version of the answer.

  • The problem with hydropower is that almost all the good sites are already in use. For example, the only free-flowing stretch of the Columbia is in Hanford Reach National Monument (one dam site on a watershed that already has 60 hydroelectric dams).
    – Mark
    Commented Jan 3, 2022 at 22:13
  • 3
    @stijn, "island known for volcanic activity"? Are you thinking about Iceland? It's about 2000 km from the "Onkalo" cave in Olkiluoto in Finland. And Finland isn't really known for its volcanoes, quite the opposite.
    – ilkkachu
    Commented Jan 4, 2022 at 10:05
  • 1
    This seems to be based in large parts on the idea that if humanity had put a lot more effort in nuclear research since the 1980s we would have save sustainable nuclear reactors now. This may or may not be the case, making predictions about technological progress is super hard but fact is we didn't. There are only reactors already built now which are nowhere near current technology and the option to start building news ones now with current technology which will come online in 10 years under very optimistic assumptions.
    – quarague
    Commented Jan 4, 2022 at 10:48
  • 1
    you seem to confuse "sustainable" with "renewable", especially in the first paragraphs
    – xeruf
    Commented Jan 5, 2022 at 10:15
  • 1
    @Mark And if the definition of "good site" includes not destroying ecosystems, natural beauty, and cultural treasures, then there were never any good sites. Commented Jan 5, 2022 at 19:52

No, and it's unlikely there ever will be -- because the risk is just too high and hard to quantify.

In the U.S. in the 1950s, the government wanted to encourage the development of a civilian nuclear power industry, so they spoke to nuclear scientists and engineers, investors, and utility professionals to understand the key obstacles. Technology? financing? fuel supply?

While there are challenges with all of these, the real kicker end up being insurability. At the time, the most liability insurance that was available for utilities was $60 million (about $600 million in today's dollars), and this was seen as inadequate. The solution to this was the Price–Anderson Nuclear Industries Indemnity Act. Per the Wikipedia article:

The main purpose of the Act is to partially compensate the nuclear industry against liability claims arising from nuclear incidents while still ensuring compensation coverage for the general public. The Act establishes a no fault insurance-type system in which the first approximately $15 billion (as of 2021) is industry-funded as described in the Act. Any claims above the $12.6 billion would be covered by a Congressional mandate to retroactively increase nuclear utility liability or would be covered by the federal government.

What this means is that all nuclear plants pay into an insurance fund. In the event of a disaster, the first $15 billion (or $12.6 for an individual claim) come from the fund -- anything beyond this comes from the taxpayers.

Is the risk really that bad?

You might argue that in the 1950s people were scared of nuclear, and such protection isn't necessary today. In fact, congress set a time limit on the act, thinking that private insurers would eventually catch up. But insurance companies, who are in the business of calculating risk, never changed their minds on the insurability of nuclear power. Since the plants were built and politicians are suckers for the sunk cost fallacy, the only feasible option was to extend the act, which has been done six times since by congress, with the coverage increasing each time.

Turns out the private interests were correct on this one, too -- of all nuclear reactors ever built, about 1.5% of them have melted down.

Haven't safety standards gotten better?

Yes, they probably are better and safer. But the problem is that the likelihood or severity of some risks just can't be predicted, and the impacts can span generations and continents:

One way to make plants better is to spend more money on them, and that's exactly what's been happening. To quote myself (emphasis added):

The most recent nuclear power plant to come on-line in the U.S. is Watts Bar 2, in eastern Tennessee. The plant came on-line in October 2016, after starting construction in January 1973 -- that's nearly 44 years. Of course, that includes a halt in construction that lasted 22 years, from 1985 until 2007. The last plant to come on-line before Watts Bar 2 was Watts Bar 1, which did come on-line 20 years sooner, in 1996. But that's still 23 years from start to finish. And what about the cost? When construction on Watts Bar 2 halted, $1.7B had already been invested. When construction started up again, it was estimated that an additional investment of $2.5B would be required to complete the plant. This was shockingly optimistic, as the total additional cost ended up being $4.7B, for a total construction cost of $6.4B, or $5.57/watt. Meanwhile, over at Vogtle (in Georgia, the only plant currently under construction), the current estimate of the final cost is $25B, or $11.19/watt. This compares extremely unfavorably to natural gas at about $0.89/W and solar at $2.43/W. Nuclear is reliable and clean -- but not cheap.

Why not focus on improving the technology further?

This is certainly an option, and most likely research and development will continue in this area -- there are certainly things to be optimistic about such as traveling wave reactors and fusion reactors.

But when it comes to governing bodies approving and subsidizing energy investments today, it is simply financially irresponsible to divert time and money from things such as batteries, wind, and solar to nuclear power. While these three technologies keep getting cheaper, better, and faster to build, nuclear keeps getting slower and more expensive, with no clear way to mitigate the risks.

  • 5
    I think approaching it from the perspective of insurance and risk management is difficult since nuclear accidents are incredibly expensive but incredibly rare. Insurance companies don't tend to deal well with these sorts of risk profiles, the total costs of major natural disasters such as large earthquakes are rarely well covered by insurance. It would somewhat be easier to understand if nuclear power stations had far more small accidents like other sources of energy! Perhaps the closest comparison would be hydroelectric dam failures, but even those are far more common. Commented Jan 3, 2022 at 0:39
  • 5
    Fukushima is hardly modern nor were the risks that caused the incident unmanageable. Tsunamis resulting in flooding was an expected risk in an earth quake region, additional backup power generators were installed at a higher elevation to reduce flood risks but the switching stations for them weren't flood protected. Other reactors on the site which had flood protection were operating normally after the tsunami. TEPCO ignored various reports of tsunami risks. Commented Jan 3, 2022 at 13:09
  • 6
    "because the risk is just too high and hard to quantify." Okay. But it's easier to quantify the risk of relying on fossil fuels for 80% of our energy, and the sure risk is even much higher than what could happen with nukes. Also, there's radiation everywhere, and it can be measured long before it's dangerous. If you mention radiation in Canada from Chernobyl, you should mention how strong it was. Commented Jan 3, 2022 at 18:02
  • 4
    @LShaver as usual with climate and energy problems, its not really possible to isolate them and consider them separately. Look at what happened in Germany after the Atomausstieg, for example, where coal powerplants are, sadly, still going really strong. Converting large SUVs from gasoline to electric won't help either. I totally agree that nuclear power has many drawbacks, BTW, it's just that every other energy source has some too. It's not just evil nukes vs good renewables and batteries. Commented Jan 3, 2022 at 18:33
  • 3
    "As a result of the Chernobyl incident, radiation traveled as far as Canada. As a result of the Fukushima incident, Japan will be releasing radioactive water into the ocean for the next 30 years." I think this is missing some important context. First, modern radiation detectors are exquisitely sensitive. So it is very easy to make sensationalist headlines even if the actual risk posed is miniscule. Second, the intentional release by Japan is tritium only, which has a short half life, very rapid dilution in oceans and does not bioaccumulate. Minimal risk.
    – uLoop
    Commented Jan 4, 2022 at 21:15

You are asking a couple related questions:

1. Is nuclear fission (that is, existing, conventional nuclear power plants, as opposed to fusion) sustainable?

No. The simple reason is that nuclear fuel, like fossil fuel, is a finite resource that will be depleted in the foreseeable future (order of magnitude hundreds of years). This is undisputed ("Meeting high case demand requirements through 2040 would consume about 28% of the total 2019 identified resource base recoverable at a cost of < USD 130/kgU", from a recent IAEA report). Surely not all resources are known, and more Uranium can be obtained at higher cost like with fracking and oil sands over in the fossil camp. Also, technologies like breeding may increase efficiency. But this does not change the fundamental resource depletion.

2. Is it green?

Debatable. Politically it is certainly not "green" because many Green parties have emerged from the anti-nuclear movement of the 1970s. Factually it is more complicated:

  • The carbon footprint is much smaller than with fossil fuels even if we consider the effort of mining and extracting and processing etc.
  • There are fewer deaths than from fossil fuels, both because mining coal is dangerous and because the air pollution kills people.
  • But the waste problem is not solved. An unsolved waste problem is obviously a showstopper for a green label.
  • The risk of Black Swan events. There is a small chance for truly catastrophic events which makes risk assessment difficult. The incalculable chance of large-scale destruction and pollution is a green showstopper, even if actual events in the past are less detrimental to humans than the distributed, permanent, low-key impact of fossil fuels, which is very well predictable.

3. Can it serve as a bridge/transition technology until everything is renewable?

Debatable. Probably not.

  • The time scale is rather long (decades to ramp up). By then we are hopefully mostly renewable anyway if we focus on that instead.
  • New projects in the West tend to run into cost problems and are many years late. Building nuclear power plants is extremely challenging compared to renewables.
  • Deployment is terrible compared to renewables. Everybody can slap solar cells on their roof. Wind-power is not really high-tech and scales well. Nuclear, by contrast, needs central planning, resource allocation, complicated approval procedures, large infrastructure measures etc. Renewables can be a grassroot movement, which eases implementation in developing regions; nuclear cannot.
  • Nuclear power comes with huge security issues because some fuels and waste products as well as the plants themselves can be used in terrorist attacks or can be used to build bombs. All nuclear facilities are high-security areas; typically the fuel and waste is meticulously tracked, facilities are subject to routine inspections etc. These safety issues exclude many developing countries from the list of countries that we would feel comfortable with if they had nuclear technology. But serving the energy needs of developing nations without fossil fuels is a core challenge for the next decades.
  • The technology is too expensive compared to renewables. While the bottom continues to fall out of renewable prices, the price of nuclear power stays constant or even rises. Nuclear power today is dependent on public subventions which will only make sense if a nuclear infrastructure is also needed for nuclear weapons production. As an example for the hidden public subsidies consider the fact that nuclear power plants are not insurable; the public bears the cost of large-scale catastrophic events like in Chernobyl or Fukushima. A German study from 2011 concluded that the electricity price would be Euros, not Cents per kWh if a realistic insurance would be priced in. That is the true cost of nuclear.
  • Nuclear fission is a hard-to-manage, capital intensive, centralized mega technology (pipe dreams of modular reactors notwithstanding). It is clearly not a technology suitable to satisfy the energy demand of the developing countries, which — again — is one of the main problems to solve in this century.
  • 1
    I believe this is the best answer so far. You may wish to incorporate some of what the IPCC says in chapter four, section on nuclear energy in the special report on warming of 1.5C.
    – LShaver
    Commented Jan 5, 2022 at 16:09
  • 1
    Chapter 7 of the fifth assessment report also addresses the question of uranium availability in section 7.4.3.
    – LShaver
    Commented Jan 5, 2022 at 16:20
  • 1
    @LShaver The chapter is interesting; I'd simply let your comment stand as a pointer. The chapter goes a little bit more into historical and political details concerning feasibility but it seems too much detail for this summary. Commented Jan 5, 2022 at 16:47
  • 1
    @LShaver Wow, thanks for the pat on the back ;-). Commented Jan 6, 2022 at 21:46

I would summarize the argument this way:

The argument for nuclear power being green is that it doesn't emit carbon dioxide. Also today's nuclear reactors are designed to run continuously with steady output. This makes them ideal for complementing solar and wind, whose power output fluctuates. So in that sense also, nuclear power is "green".

Also, nuclear power is sustainable because there is no shortage of uranium. Plus, despite what people think, the engineering of storing high-activity nuclear waste is a solved problem, the issue now being that people are nervous about it.

The argument against nuclear power being sustainable is that it has the potential to cause accidents that contaminate wide areas with radioactivity. This has happened twice: Chernobyl and Fukushima. In both cases power reactors vented part of their cores to the environment as a result of engineering failures that can be traced to bad management and, ultimately, to societal failures. Since failures of this kind can never be entirely eliminated, it can be argued that nuclear power is not sustainable because its use will inevitably cause changes to the environment.

We know a lot about nuclear fission technology since we have been using it for decades. We understand its risks and advantages. We face a difficult choice: to balance the certainty of poisoning the entire planet with CO2 if we don't take action now, against the risk of engineering failures that could poison hundreds of square miles but which we might make less likely by being very, very careful.

  • It is not true that it emits no CO², as plants with finite lives are built of concrete and fuel is mined, so you should give figures (and cite sources). To show it is geen you need to consider usage of all resources. There may currently be no shortage of uranium, but other answers say known supplies are limited.
    – PJTraill
    Commented Oct 5, 2022 at 16:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.