Small reactors don't add up as a viable energy source
cosmosmagazine.com
cosmosmagazine.com
The author is also clearly anti-nuclear - that’s not to say they’re not right, but they’re motivated.
It's not just the scale, it's also how few there are in planning in the west. You work out the kink (both building and operational) by building more of the things, and that's also how your builders skill up. At the height of its buildup, France had half a dozen nuclear reactors being built concurrently.
Because SMRs are smaller and you need more for the same total output, more would be built, and thus there are more opportunities to work out the kinks in manufacturing and setting up.
An other theoretical advantage of SMRs is they could be built in factories and craned in, and when the fuel is expended they get craned out, a new reactor is dropped in, and the old reactor is moved to a refueling facility (which can be nearby a reprocessing plant), the site doesn't need to be offline for refueling, and it gives higher opportunity for automation.
Why being anti-nuclear is even a legitimate stand is a good indicator that people still don’t take climate change (and general over consumption) seriously. It’s like being anti-chemicals because some of them can be toxic to ingest.
As for "nuclear being effectively dead"; the story you yourself posted in this thread says this:
quote: "By contrast, Russia, which now dominates the international market for new reactors, has 53 under construction, planned, or proposed within its own borders and another 50 in 19 countries. China is constructing, planning, or proposing to build 220 new domestic reactors, and 20 of its models are being built or are under consideration in 12 other countries."
And thats not even considering the agreement on tripling nuclear capacity before 2050, adopted by 20 nations at the COP28.
Can we stop with the ultra partisan claims, please?
I will add that with natural gas still facing no CO2 tax in the US the impressive thing is how much renewable energy is still being installed. The environment is oppressive for anything but gas; it's absolutely lethal for nuclear, to the extent nuclear can count it as a victory if existing plants can remain operating.
I reject the argument that being against nuclear energy is "partisan". I suggest instead that new construction nuclear is just bad when examined objectively. Nuclear partisans (and there the epithet is warranted) do not exert quality control on their "arguments".
I'm specifically referring to your extreme rhetoric in this thread. Despite your continued claims, nuclear is experiencing a renaissance these years, and I see that it's purposeless to argue the point further with you, so I'll let you observe the developments yourself in the coming 10 years as they will speak for themselves. End.
Could you please indicate sources?
BTW: you count a lot of "proposed" reactors. Proposing is easy, building is harder.
1) Have we built enough reactors consistently enough over time to say for sure the experience effects don’t build? I know we effectively stopped in the 80s, but I’m curious if there’s evidence from elsewhere to say nuclear’s indeed different here.
2. Hypothetically, if SMRs lead to a 10-100x increase in volume building, would we see experience effects? At what volume would you expect to see genuine experience effects, or again is nuclear fundamentally different for some reason?
I’ll say my biases are: I’m not surprised we’ve seen limited learning over time for various reasons, I think the size of nuclear projects almost precludes repeat learning, and I do expect if we lower the project size and increase the volume we’d see efficiencies of scale, but I could be convinced that either nuclear is sufficiently different or that even with the SMRs we’re not going to see enough scale to actually recognize efficiencies or learning.
Experience effects occur when an individual or a cohesive group gains experience. But with NPPs taking a long time to build, any individual doesn't go through many iterations in their career. Experience also decays, so if the learning rate is low enough decay should cancel it out.
In contrast, the facilities for building renewable technologies iterate much faster, both on production of individual modules, and in design updates. Installation and planning also have similar high rates of activity.
SMRs might provide a solution, but if they're so fast we have to ask why none of the SMR companies have built any yet.
> The climate crisis is urgent. The world has neither the financial resources nor the luxury of time to expand nuclear power. As physicist and energy analyst Amory Lovins argued: “… to protect the climate, we must save the most carbon at the least cost and in the least time.”
> Expanding nuclear energy only makes the climate problem worse.
> The money invested in nuclear energy would save far more carbon dioxide if it were instead invested in renewables.
> And the reduction in emissions from investing in renewables would be far quicker.
These statements may wind up being correct, but they’re speculative, and they’re anti-nuclear.
To the point of the Wikipedia page, a couple selected publications - although in fairness I haven’t read the publications and am just judging from the titles:
> Nuclear power: Economic, safety, health, and environmental issues of near-term technologies, Annual Review of Environment and Resources 34, 2009, 127-152
> Beyond our imagination: Fukushima and the problem of assessing risk, Bulletin of the Atomic Scientists, 2011, 19 April
> Nuclear Power in India: Failed Past, Dubious Future, 2007, Available at www. npec-web. org/Frameset. asp
Again, all of these could be entirely factual and well defended, but the author is clearly not a disinterested observer.
I’m more pro-nuclear than the author, and mostly I’m skeptical about good scalable solutions to the storage problem. I’d prefer renewables + storage to be the answer, but I think we made a mistake when we halted nuclear development and I don’t think we’re doing ourselves any favors by not pursuing the technology.
It is, however, something to be aware of when reading an article they wrote: they are not a fully disinterested observer, they have a pre-existing belief about nuclear, and their arguments for this article agree with those beliefs. That’s a thing to be aware of when one evaluates the set of facts they’ve chosen to present.
Settle down, champ.
> Why is disinterest an ideal to you?
It isn't, because I don't think it's possible. Because of that, I think it's relevant to know what an author's preexisting opinions and beliefs were when writing the article. If, for instance, the author had been strongly pro-nuclear, then their conclusion that SMRs are a dead end would carry more weight, as they would have had to overcome their own biases. Articles that confirm an author's beliefs, on the other hand, are much easier to write.
I have no idea what you are talking about. Are you mistaking me for someone else?
> To me it seems that the article describes a career doing research and applying it to public policy.
That's not incompatible with what I wrote earlier...
That's not what affordable means. People are building smaller reactors because it's more likely that a project will be completed, not because of unit costs.
The article ignores this, and in fact complains about the cost of the first six reactors proposed by NuScale (which, fwiw, is just a smaller LWR, not one of the more innovative designs).
Before anyone talks about the negative learning curve of large reactors in the US, bear in mind that we mostly build those as one-offs, so no learning curve exists. We don't build many, regulations change pretty frequently, and sometimes the NRC requires design changes after construction starts.
And coal is not exactly fertile ground for innovation, given that we'd like to stop burning coal entirely.
Matching the price of coal, with clean dispatchable power, is not such a bad outcome. There's a bit of wiggle room for extra complexity since coal pays 2 cents/kWh for fuel, and has to handle large amounts of incoming fuel and outgoing ash.
For a reliable zero-carbon grid of only wind/solar/battery, we need about 2X overproduction and 4 days of battery storage[1]. Maybe that still ends up the cheapest option, but it's not so obvious that a backup plan isn't worth considering.
[1] https://caseyhandmer.wordpress.com/2023/07/12/grid-storage-b...
Switch off (curtailment) is almost instantaneous for wind and PV.
And otherwise, batteries, hydropower and gas turbines are dispatchable on the second/minute scale.
How dispatchable nuclear is depends on the specific design, which is important given our context of small reactors with new designs.
According to the DOE, "Certain designs, like DOE-supported NuScale Power, LLC, can vary their energy output over days, hours and even minutes."[1]
According to NuScale, "The NuScale Power Module is capable of a ramp rate of 40% per hour in reactor power change, which aligns with specifications set by the Electric Power Research Institute (EPRI). For even quicker responses to electricity demand, the NuScale SMR can rapidly lower its electric power output up to 10% per minute and return to full output at the same rate utilizing turbine bypass. This is significantly faster than conventional nuclear power."[2]
More advanced designs may be even better.
[1] https://www.energy.gov/ne/articles/department-energy-report-...
[2] https://www.nuscalepower.com/-/media/nuscale/pdf/publication...
Solar+battery gets you there today, is getting cheaper every day, and there’s almost zero project risk. If the install is small enough, you could likely beat the realistic costs for SMRs with solar + battery + diesel generators if you need a guarantee of 100% uptime.
Debatable. In Ontario nuclear costs 10¢/kWh while wind costs 15¢ and solar 50¢ (Table 2):
* https://www.oeb.ca/sites/default/files/rpp-price-report-2022...
And when wind goes to zero at night, then (natural/methane) gas generators are often spun up (generating carbon emissions).
Meanwhile the refurbishments of Ontario's nuclear plants are on-budget, and often ahead of schedule:
* https://canada.constructconnect.com/dcn/news/infrastructure/...
If I go to https://model.energy/ and ask it to solve for a cost optimized renewable + storage system for Ontario, I get a cost out of 55 Euro/MWh for providing synthetic baseload (2030 cost assumptions). That solution doesn't include using hydro to deal with variance in supply and demand; Canada has large amounts of hydro.
Well, the price is irrelevant when solar supply (and wind supply) goes to zero:
* https://www.ieso.ca/power-data § Supply
> […] Canada has large amounts of hydro.
Some parts of Canada have large amounts of hydro. Others do not.
I'd rather use hydro to reduce/eliminate gas plants:
* https://www.ieso.ca/power-data § Supply
Looking at the variability of renewables, I do not wish for the grid in Ontario (where I live) to be more dependent on them.
It's not globally representative though.
[1] Yes, really: the populated places of Canada are comparable to northern Italy in terms of latitude. The Gulf Stream really does an unbelievable job hiding that fact.
Global problems need global solutions and most people are not in nations near the poles with decades old nuclear plants they can coast on.
They need to build new energy plants, and live in areas where solar plus battery beat nuclear on multiple dimensions.
That's why every prediction has solar accelerating past nuclear deployment and heading for multiples of nuclear output.
A bit less actually. But you cannot dismiss a technology that works for 10% of people. And we still weight more than this in terms of CO2.
> Global problems need global solutions
No, global problems need custom solutions suited to all the different situations. Thinking there exist one technology to rule them all that will solve the problem everywhere is not helpful to anyone.
> That's why every prediction has solar accelerating past nuclear deployment and heading for multiples of nuclear output.
This isn't a race! Stop thinking of technologies as if it was sport teams.
> > Solar+battery gets you there today, is getting cheaper every day, and there’s almost zero project risk.
> Debatable. In Ontario nuclear costs 10¢/kWh while wind costs 15¢ and solar 50¢ (Table 2):
Please stop acting as if we don't exist.
SMR are a huge opportunity for 300+M people to dramatically reduce C02 emissions from electricity generation, for which there is no credible alternative.
Solar + battery cannot be sufficient for most Europe except Mediterranean countries, it's not a matter of cost, there's just not enough sun in Winter and you need months of electricity worth of storage which isn't happening in our lifetime.
In general, talking about energy price ($/kWh) only make sense when you have fossil fuel as a near majority of your mix (because you have practically unlimited power as long as you spend money on fuel), but cease to make sense without it, because nobody cares about energy (Wh), what you (and the grid) need is power (W). With either nuclear or renewable, energy is practically free, but power is what costs money, and as we move towards a decarbonized mix, we'll need to change how the economics work to adapt to the underlying changes (including how we price electricity to consumers and businesses), because when you don't align the economics with the how the supply works the system collapses (like it did in Texas as few winters ago).
Germany has not tried to roll out hydrogen yet. CO2 charges are not yet at the point where natural gas must stop being used for long period leveling.
Also, a big part of Germany's large expenditure was in 2009-2012 when solar (in particular) was far more expensive. Funny how you didn't mention that, isn't it.
And for good reason…
> CO2 charges are not yet at the point where natural gas must stop being used for long period leveling.
During that period, Germany has emitted more CO2 than hundreds of millions of peoole, and we're way past the moment where we should have stopped using coal and gas really.
> Also, a big part of Germany's large expenditure was in 2009-2012 when solar (in particular) was far more expensive. Funny how you didn't mention that, isn't it.
And why hasn't Germany completed the transition now that “solar is dirt cheap” for years now, then? Solar makes no sense in Europe, period. Wind, hydro and nuclear, yes, but every solar panel installed in non-mediteranean European country has been a tragic waste of taxpayer's money (giving the panels to Greece or Arab/African countries would have been a much better investment, by an order of magnitude)
For Germany to be saved from this fate by nuclear, nuclear has to be as cheap as solar is in the best locations in the world, not just better than solar in Germany.
Germany (and Europe) should probably be putting more money into CO2 sequestration, so they can keep burning fossil fuels.
Are you extrapolating rooftop solar rates in California to a large, industrial solar installation in Canada, or where do these figures come from?
Batteries would best be used for diurnal leveling; don't overestimate demand by assuming they are used for (say) seasonal leveling.
The batteries used by electrified motor vehicles would exceed those needed for the grid.
Elsewhere it needs to compete on price against other viable options.
AFAIK, that's typical consumption for data centers being built today.
Note also that the cost figures NuScale gave for that was the cost after 40-50 reactors had been built (and with federal subsidy), not FOAK costs.
No one has ever built a nuclear plant to sell into a competitive power market. One can trace the downfall of nuclear in the US to the time when markets were opened to competition, with PURPA.
One can trace the downfall of nuclear in the US to the end of the New Deal coalition and the divestment of public infrastructure, which meant short-term planning only. It has nothing to do with "markets". Markets in energy don't exist now and did not exist then.
> The Public Utility Regulatory Policies Act of 1978 (PURPA) triggered a restructuring of the previously monolithic utility sector, stipulating in particular that electricity produced by independent power producers must be purchased by utilities at "avoided cost." The new power from independent producers, combined with lack of demand for electricity, further eroded utilities' need for new nuclear plants. In large part owing to the provisions of PURPA, nonutility generation rose steadily from 71 billion kilowatt-hours per year in 1979 to almost 400 billion kilowatt-hours per year by 1995 -- this new, nonutility generation was the equivalent of adding more than 50 typical 1,000-megawatt nuclear plants (Energy Information Administration, 1996). As Peter Bradford (2011), a former member of the Nuclear Regulatory Commission, argued in the Wall Street Journal:
> "Nuclear-plant construction in this country came to a halt because a law passed in 1978 [PURPA] created competitive markets for power. These markets required investors rather than utility customers to assume the risk of cost overruns, plant cancellations, and poor operation. Today, private investors still shun the risks of building new reactors in all nations that employ power markets."
https://pubs.aip.org/physicstoday/article/71/12/26/904707/US...
Some of those fears are founded, but some are not. But overall, the regulation required is what drives the cost up.
One way to subsidize nuclear is to artificially reduce the interest rate on its financing, but market rates for nuclear financing would make a 60 year lifespan almost irrelevant. The NPV of the out year revenue would be very low.
Put another way: for nuclear to pay out over 60 years, it also has to compete with the cheaper energy sources that will be discovered and improved over those generations. This obsolescence risk cannot be ignored, and gets reflected in interest rates charged. Now that wind/solar are becoming dominant, their inherent rapid evolution has pulled in the time horizons for all other energy sources. I sometimes think this has kept natural gas going longer than it otherwise might have, since the uncertainty adds incentive for sources with more operating cost and lower investment cost, as these have less obsolescence risk.
Expecting AP1000 to show NOAK improvements is optimistic.
I’m not presenting environmentalists as powerful. I’m presenting them as innumerate feelers. Unfortunately that’s contagious.
What we do know is that France can no longer do what they once maybe did. Their recent attempts to build nuclear plants have been disastrous. They've also given up on fast reactors, which is a clear tell they do not expect the world to go nuclear anytime soon (if the world did, it would quickly need breeders).
The "environmentalists did it" argument is not a good one. It falls apart when examined closely. If they were so powerful as to suppress nuclear worldwide, why can't they (for example) stop oil pipelines? Or coal combustion? The argument has all the signs of something cooked up to save a tenuous position, not because the evidence actually supports it.
For example, Greenpeace has recently successfully lobbied the Phillipne government to ban golden rice despite its obvious advantages:
https://phys.org/news/2024-04-philippine-court-blocks-gmo-go...
So please don't present the environmental lobby as being selectively powerless.
That’s what got Fukushima btw - when they shut the reactor down and then the backup generators got destroyed, they lost their ability to pump water to cool the reactor (which requires significant electrical power), which proceeded to start to melt down the core, and causing massive hydrogen buildup, eventually blowing up the reactor building.
Some new designs allow more effective emergency passive cooling, but the issue remains - nuclear plants are great for baseline power, but they aren’t good for sub-day, hourly, or finer grained peaks. Both economically and technically. Think ‘fully loaded container ship’ or ‘multi-mile long train’.
Pumped storage, battery, or fossil fueled turbines are great for those faster reactions - and often can provide useful sub-second grid stability too. Think ‘speed boat’ or ‘passenger car’.
From what I've read, "green" hydrogen made from electrolysis with energy from non-carbon-emitting sources extremely energy intensive, as is compressing it for transport. Transport is also expensive. To my knowledge, there aren't any pipelines in Western Canada that could transport it, so I guess it's trucks and train cars. But maybe there's potential there because hydrogen wouldn't need to be the backbone of the grid, it would just need to pick up the (substantial?) slack when renewables weren't producing.
Green hydrogen certainly is more expensive than hydrogen derived from fossil fuels, but if we are imagining a 100% RE grid, that hydrogen isn't available. It does make sense to burn natural gas directly instead of burning green hydrogen as long as the CO2 charge isn't high enough to rule out use of natural gas. This is why we're not yet seeing a large green hydrogen role out in Europe.
The interesting question is whether fossil hydrogen with CO2 sequestration is acceptable. It may be preferable to green hydrogen, depending on things like methane leakage.
Another possible alternative would be artificial geothermal, with very high temperature rock at fairly shallow depth. Obviously that's not transportable.
There’s a really interesting nomenclature thing in Canada with respect to electricity. Every province (except Alberta) has a government-owned power company. If you want to tell whether or not a given province has good water resources for generating electricity you just need to look at the name of their power company: BC Hydro, Manitoba Hydro, Ontario Hydro, Quebec Hydro… all good places for hydroelectricity. SaskPower? Not so much.
We have recently bought into the SMR idea and one of the questions I used to have was “why the hell didn’t we build nuclear 20 years ago here”. The problem, I understand now, is overall grid sizing. You don’t want to have a single power plant that provides more than about 10% of your grid capacity. In SK, our total generation capacity is around 3500MW; a conventional 1+GW reactor would have dramatically exceeded the 10% cutoff. Smaller 300MW reactors provide a much better fit for our grid.
* https://en.wikipedia.org/wiki/Point_Lepreau_Nuclear_Generati...
CANDUs are pretty flexible in that there's a lot of maintenance (including refuelling) that can be done while it is running, but there's still some stuff that needs to be done when the system is powered down, which means taking down a large source of power for the grid.
If there were 2-3x300MW reactors, when there could be rotating maintenance without much impact to the grid.
There's a certain amount of concrete and such that needs to always be built, and if you go with a "cheaper" SMR, then the fixed cost becomes a large portion of the total project budget.
So unless there's a specific local need for ≤300MW, it might be better to go with a 600/900(+) MW design if you can tie into a large grid where all of those 'extra' MWs can be soaked up.
Flyvbjerg mentions SMRs as an example for modularity in his book, How Big Things Get Done, and predicts that they will be much less prone to overruns because experience can be accumulated along a series of reactors, whereas traditional reactors are one-off, bespoke projects which directly implies that they will be built with a lack of experience. Even if a nuclear power plant gets built that is "like" an existing one, it is never the same.
TBF they don't have to be, obviously they don't number in the hundreds and site-specific concerns matter (especially as they have a large surface, they matter for SMRs too but the footprint means they matter a lot less). However you have to commit, hard.
During its buildup, France built 54 reactors of just two classes (34 C-class, with 3 variants, and 20 P-class, with 2 variants) in 22 years: https://fr.wikipedia.org/wiki/Liste_des_réacteurs_nucléaires...
What small ones can do is afford either government/public/private energy sources in localized areas. Infrastructure was built upon technology stacked on top of previous; dirt to stone, stone to asphalt, and on; etc.
The same is inherent with nuclear. It is easy to tie in to the existing grid, but the grids are extremely out of date for the growth of populations in general.
A large mix of SMR's could absolutely fuel energy needs in both the short and long term as technology continues to improve. The cost is a metric of current economics/interest. That's the problem right now - perspective states it's unaffordable because we've pivoted it that way.
Large reactors are highly bespoke, and therefore increasingly expensive, and the whole idea of small reactors is that they can be produced identically from an assembly line, and so you get the same kind of learning-curve price reductions as for solar panels or whatever.
I don't think that's the case, yet, though, so the argument might be compared to criticisms of solar in the 1980s and anyway this article seems to be a simple attack on nuclear in favor of 'renewables', which have already experience a learning curve.
There are arguments that for a small modular reactor the civil works that happen onsite (e.g. foundations) could make up a larger proportion of the costs than for a large modular reactor!
- Health care provided by the VA
- Education, both before the job (officer school or trades training) and for post-military life (GI Bill)
- Decent pension after retiring
Etc
Also, it is estimated that it would take about ~7 million acres to power the US entirely with solar. The US currently uses about 40 million (!!!!) acres for corn for ethanol.
Land is just not a problem, at all.
https://elements.visualcapitalist.com/how-much-land-power-us....
https://frontiergroup.org/resources/ethanols-outsized-place-....
https://ec.europa.eu/eurostat/statistics-explained/index.php...
At this price, the cost of land is quite small compared to the cost of a PV field on that land.
From the outside, this SMR situation looks a bit like monolith vs micro-services where there is a great deal of non-technical reasons why SMRs are a route being taken and it appears it's mostly political and organizational ie. approve and build the damn thing.
It was finished on April 29 and has been online since that time.