Hacker News new | past | comments | ask | show | jobs | submit login
The Nuclear Option (city-journal.org)
117 points by akakievich 37 days ago | hide | past | web | favorite | 185 comments



Some of nuclear energy's challenges get a good amount of coverage: politicians unwilling to support it, citizen opposition, and cost overruns and delays at plant currently under construction. Others, less so. And if we discount politics for a moment, the day-to-day economics of new nuclear barely pencil out. Nuclear needs immense capital expenditure vs. other types of generation. Solar gets generous subsidies and doesn't need a lot of opex, wind is similar. And in the US, natural gas is cheap and plentiful, and pairs well with intermittent renewables, and load-following gas turbines are blurring the distinction between baseload generation and peakers.

Despite the wishes of many, pollution externalities are not priced in, and while that is the case and fracking continues to bring up cheap gas, the rule of natgas will continue.

But nuclear's immense capex is a problem nonetheless. While reactor designs purport to be off-the-shelf, the each plant itself is a one-off. There's no institutional memory and hardly any re-use of expertise, there's no economies of scale -- just all the hallmarks of difficult megaprojects. Sound familiar?

This is why research into 'small modular reactors' is promising. Plenty of political barriers remain, but removing some institutional barriers and easing some of the economic ones would improve nuclear energy's prospects.


Existing nuclear plants got as big as they are because of economies of scale. Turns out that a plant twice as big costs less than twice as much as two smaller ones.

But, due to nuclear expansion faltering we build too few of these plants, and thus all of them become essentially first-of-a-kind, which tends to mean a lot of (very expensive!) delays.

I also think SMR's are very promising, trying to break the cycle of ever-increasing costs of big plants through mass production (and passive safety) rather than just making them as big as possible. Think airliner factory rather than one-off megaprojects.


They also got big because of economies of scale of operation. Making a reactor smaller doesn't proportionally scale down the staff you need to operate it. This is relevant when as many as a third of reactors in the US are not even making an operating profit.

SMRs will require changes to operational regulation if they are to compete.


Indeed. Many microreactors of the past like the PM3A in Antarctica were just too expensive to operate and maintain so they got replaced with oil. SMRs are great for lower capital cost but they won't just automatically be cheap to operate. The nuclear industry has to focus heavily on getting maintenance and operation costs way down. Some of that is regulatory, but lots of it is inherent.


This is the most important point regarding nuclear if you ask me. If it is to have any chance at a rebound, it has to start from smaller reactors again. And it has to do be a design that's geared towards load-following rather than baseload, so that it complements renewable rather than compete with it.

The focus now should be on massive investments into R&D and smaller scale developments


There are lots of people considering steam bypass, where you can shift the heat from any reactors away from the turbogenerator and into some industrial process like desalination or pulling carbon in from atmo.


I have heard that France standardized it's nuclear design, and has been very successful with that. Not exactly COTS, but not like starting over each time either.


So has Korea and for a time, Japan. That is absolutely key and has led me to promote the idea of open source nuclear reactor design.


...Less radioactive material than coal ash as well. Very excited for he NRG Thorium reactor test going on in the Netherlands.

Coal Ash more radioactive: https://www.scientificamerican.com/article/coal-ash-is-more-...

NRG Thorium Reactor tests: https://www.technologyreview.com/the-download/608712/a-thori...


Sigh.

I've been arguing this for so long, and gave up about 5 years ago. I can't think of a more scalable way than fission to combat carbon, yet it won't happen for so many reasons.

* There are too many solar and wind success stories to point to. "Why can't NYC be like Scotland with wind power?!" You start to list the reasons and the response is "we can accomplish it if we put our minds to it!". It's an endless cycle of optimistic non-problem-solving.

* No politician will back nuclear. If there's an accident, they'll be run out of town. Not worth the risk.

* Along the same lines, there is zero interest in technology research/investment. This became obvious when they shut down reprocessing in the 70s and IFR was shut down in the 90s. And the reprocessing ban was by a President who happened to be a nuclear engineer! If he won't do it, who will?

* The scare-mongering in the media over accidents is impossible to overcome. There's no way to have a rational discussion when all someone has to do is point to the endless exclusion zone documentaries to win their argument.

* And then much of the blame needs to be shared by the nuclear engineers. They had decades to make nuclear safer before TMI, Chernobyl, and Fukushima (1942 -> 1970s, 80s, 2010s). And yet, here we are, with the same light water reactors that Rickover speced out in the 50s! Yes, I get that has something to do with building restrictions since TMI... so how about before that?


The overall incredible ignorance around nuclear is staggering. If you ask the average person how many people have died as a result of the Chernobyl incident, they’re usually going to estimate high by orders of magnitude. In general what people don’t understand about radiation is effectively everything about it, and worse, they don’t want to know. Meanwhile trying to explain how many million die each and every year from burning hydrocarbons rolls right off them. It’s pathetic and infuriating and I’ve also given up completely.

And then much of the blame needs to be shared by the nuclear engineers. They had decades to make nuclear safer before TMI, Chernobyl, and Fukushima (1942 -> 1970s, 80s, 2010s). And yet, here we are, with the same light water reactors that Rickover speced out in the 50s! Yes, I get that has something to do with building restrictions since TMI... so how about before that?

It has to do with fuck all investment, a dead regulatory environment, and inability to test new designs at scale. For all the other reasons you mentioned the investment isn’t there. I’d add that the waste issue is seen by most of the public as technically insurmountable, whereas in fact it’s technically simple, but politically insurmountable. At this point you’d need it to be the platform issue for a popular president, along with a truly huge education initiative and a lot of arm-twisting in the senate. It would cost a fortune, but less than the inevitable disaster were hurtling toward without nuclear.

Unfortunately even on a site like HN, optimism about renewables and new tech combines with ignorance of nuclear power; how can we expect better from the average American?


> The overall incredible ignorance around nuclear is staggering. If you ask the average person how many people have died as a result of the Chernobyl incident, they’re usually going to estimate high by orders of magnitude.

Downplaying it doesn't help either.

In France the medias and politicians repeated again and again that the radioactive cloud didn't cross the border. Not only lying for such a thing is disgusting, but it's also moronic to think people would believe it.

Decades later, now have a decent overview of the increase of thyroid cancer near the border in the years after the incident. I have a friend that had his life shattered because of it.

I'm pro nuclear, but every time there is that much on the line, those in charge stop respecting the people, so of course it's going to create strong opposition.

How do you want to establish any kind of trust in the technology if when there is a catastrophy, you make sure to betray the victims ?

Everytime there is a problem, the authorities pretend everything is under control. Not it's not. If it had been, we woudln't have had the accident in the first place.

Honesty would be to have very mediatic continuous monitoring of the situtations for the next 10 years. The president should state every year an honest status on the matter. We all know fukushima is still leaking in the ocean radioactive materials: following the impact of it should be not only transparent, but vocal, and honest.

Nobody respect kids hidding after they fucked up, saying "no I didn't do it".

Nuclear is more of a political problem that a technical one, and we, humans, suck at those.


Maybe this is an issue that could garner bipartisan support, even if for different motivations? Instead of trying to bring back coal, we should be trying to reinvigorate the nuclear power industry. Maybe even turn some of those coal miners in to Uranium miners :)


More like uranium refiners. Uranium occurs at a relatively common rate at or near the surface (forget the specifics). Yes, it’s absolutely the energy we should be using. But our first use of nuclear blew up two cities. It will forever be marred by its first impression.


Nuclear power is cost prohibitive. Additionally, it's hard to predict and account for that one-in-a-million catastrophe that creates a nuclear wasteland. The sample size is too small.


Nuclear power is cost prohibitive.

Really, because it’s widely employed around the world. If you have some evidence that it’s “cost prohibitive” though I’d love to see it. Evidence mind you, not flat declarations devoid of substance.

Additionally, it's hard to predict and account for that one-in-a-million catastrophe that creates a nuclear wasteland. The sample size is too small.

Many decades and thousands of reactors on land and naval applications with only a few notable disasters beg to differ. I’m also curious where this “wasteland” comment is coming from. The worst nuclear disaster, one singular in its extremity, has led to something far from a wasteland, although quite a few people had to be relocated.

Either way, trading in the known mass casualtiesin the millions per annum, and environmental catastrophe of burning hydrocarbons, for scaremongering “what if’s” so profoundly out of touch with reality as yours is the falsest of dilemmas. The total death toll from everything nuclear, including weapons can’t match a year of deaths and illness from burning coal. Never mind mining coal, drilling for oil and transporting it and the associated sociopolitical disaster of the Middle East, the environmental harm of fracking, and the sheer stupidity of not addressing climate change now.


A weaker statement than a flat out "nuclear power is cost prohibitive": current-gen nuclear power seems to be having serious cost problems, even in countries that previously successfully built out nuclear, and even where there is a relatively favorable political environment. Whether this is due to fundamental reasons or not is of course another question.

One example is the new-generation Électricité de France EPR design, where the first unit scheduled to be completed, at Flamanville, is way late and over budget. France has almost 100% nuclear power and has both a good safety and cost record historically, so the seeming lack of success of this project has dismayed many people: https://www.reuters.com/article/us-edf-flamanville/edfs-flam...

It's not quite as positive a political environment overall, but Westinghouse also had strong local and state-level support for its new unit in Georgia (the U.S. state), but ended up going bankrupt after cost overruns: https://www.ajc.com/business/power-seeks-best-path-forward-p...

Obviously cost overruns at two plants is a very small sample size, so it could be due to non-fundamental features, like severe mismanagement. But given that there aren't that many large-scale projects going on, the two biggest Western ones both running into major trouble has really hurt prospects for more funding/support. If even one of these had been a clear success, the outlook would be much better.


In my locality, a glut of natural gas is putting pressure on Nuclear costs. They simply can't compete because NG is so plentiful.


>The worst nuclear disaster, one singular in its extremity, has led to something far from a wasteland

I believe the phrase you are looking is "wildlife sanctuary". By far the biggest effect that nuclear disasters have on the environment is their function as a human repellent.


Nuclear power is cost prohibitive.

How about: https://www.bbc.co.uk/news/magazine-36160368?

For the cost to be prohibitive, it only really needs to be dearer than an alternative that's at least as good. That alternative is now wind: https://www.bbc.co.uk/news/business-41220948.


Wind is not a like-for-like replacement for nuclear. The amount of land wind would need to replace just one nuclear reactor is many orders of magnitude greater.

Renewables must be supported by a reliable base-load supply, which effectively means you need to choose between either nuclear or fossil fuels.


1. The amount of land or sea, which latter makes for a dearer installation but is much easier to find in densely populated countries like the UK.

2. My understanding is that, sure, you can't have a grid that runs purely on wind, but you can go a lot higher than we currently do without much trouble. The issue with massive new nuclear is that outages can take a sizeable percentage of your generation off-grid unexpectedly at a moment's notice (Hinkley will I believe represent 7% of UK grid capacity), which in a sense is a worse intermittency problem than you get with renewables (which tend to ramp up and down more smoothly and broadly in line with weather forecasts). Also, battery storage can help.



This won't be the case with modern reactor designs. Previous designs were based on an approach that is fundamentally unstable and thus requires a huge investment in safety critical systems to keep things in line.

Newer designs are fundamentally stable and fail in an inherently safe manner. Much, much less surrounding infrastructure and a greatly reduced need for system level fixes to problems since the catastrophic issues have been designed out.

It will take some time to turn the ship, but it is our best chance for producing 24 hour scalable power that is also carbon safe. Just search "modern nuclear reactor design" and start honestly reading the literature with an open mind.

A couple of random articles (plenty more available): https://www.technologyreview.com/s/512321/safer-nuclear-powe... http://science.time.com/2013/08/05/amidst-economic-and-safet...

Plenty of great information out there for anyone interested in letting go of their preconceived ideas and exploring the amazing work that is happening now.


The new designs are not going to be available for many years. They have to be proved in operation, particularly if they involve new chemistry like MSRs.

By the time they're proven, renewables will have expanded to global dominance and likely run down their experience curves to cost points that even Gen IV nuclear will find very hard to beat.

New nuclear designs are a hedge against renewables suddenly stopping getting cheaper. An insurance policy, but hopefully one that won't be needed.


There are quite a few new companies that are very interested in nuclear R&D. In the U.S. their biggest problem is the NRC.

A couple years ago I got to sit in a meeting between people from over a dozen of these companies, and a former head of the NRC. The reactor people said their biggest problem was that the NRC requires the design to be nearly complete before they'll even look at it. It takes several hundred million dollars to get to that point, then the NRC gives a simple yes or no. If no then you're done, if yes then you've still got nothing but paper. It's a very difficult environment for investors.

They said if the NRC at least went with a more phased approach it would help a lot. The NRC person was unsympathetic, said it wasn't the NRC's job to help promote or develop nuclear power, and was uninterested in climate change.

Fortunately Canada's regulators are more rational. Terrestrial Energy has spoken highly of them, has gotten through the hardest part of their process, and expects to get a molten salt reactor to market within a decade. Moltex has moved to Canada as well, and possibly others.

But it's unfortunate that the U.S. is so difficult, since with facilities like Oak Ridge it could help a lot.


This is as designed. Prior to the anti-nuclear NRC, we had the pro-nuclear AEC. Progress stopped, as intended, when the NRC took over.

https://en.wikipedia.org/wiki/United_States_Atomic_Energy_Co...


There's a long shadow in this which isn't very much discussed: nuclear weapons.

Up until the 1990s it was generally accepted that the greatest risk to humanity was not the slow problems of global warning but the near-instant destruction of an exchange of the thousands of ICBM-delivered nuclear warheads possessed by the two-and-a-half superpowers. These weapons depended on a supply of plutonium. Therefore many of the earlier reactors (eg Windscale/Sellafield) were built in a tremendous hurry to ensure enough plutonium.

Conversely, on the other side, environmentalists focused on cutting off the supply of plutonium to weapons. Recall that Greenpeace got started campaigning against nuclear bomb testing. This attracted hostility from governments, up to and including the French DGSE sinking the Rainbow Warrior and murdering the photographer who was on board. This has affected the subsequent debate because there is no trust.

The nuclear trust issue affects which countries are allowed to have nuclear reactors (see Iran, Iraq).

In the 2000s the big question was terrorism: would nuclear plants be vulnerable to attack, including crashed planes, and how could they be secured? Meanwhile a trillion dollars was spent on a largely pointless war, proving which kinds of threats are taken seriously and which aren't.


A plane crash attack is no different than any other type of accident; it can be integrated in the incident response plan (much less dangerous than tsunami). This is not the big problem.

The bigger problem is proliferation. All nuclear power plants produce plutonium from fuel stocks, which goes straight to the waste. Plutonium can be chemically separated from the nuclear waste, which is not quite easy, but much much easier than isotope separation. Thankfully, a single neutron capture transforms Pu-239 to Pu-240, which is unusable in nuclear weapons (if it wasn't the case, human civilization would have ended decades ago), and it is (or was) nearly impossible to separate these two isotopes. To obtain relative pure Pu-239, fuel assemblies or special U-238 rods need to be extracted early in the campaign, which is observable (unless a reactor is specifically designed for that). Unfortunately, modern laser isotope separation methods (spin chemistry, magnetic isotope effects) make it possible to separate Pu-239 from Pu-240. Most of the details are classified, but secrets can work only for a limited time.


I sense some cognitive dissonance here. You want that great carbon-free energy provided by nuclear, and you lament all the reasons it won't happen. Except you have no rebuttal to those reasons and almost sound convinced by their soundness yourself.

- Politicians won't back nuclear because of the risk of accident, thus giving legitimacy to the risk and the public's fears.

- There hasn't been adequate research and advancement in the nuclear industry (and other threads mention the lack of applied design research). So previous scientists and industrials have already determined it's too hard/expensive/difficult. Global warming may change that equation, but it won't make up for lost research.

- So people see the exclusion zones and shouldn't be afraid? We currently have 2 molten reactors with no long term solution found. At what point does the fear become irrational?

- Same point as above: lack of advancement, bad designs, and human mistakes, how can any of these be fixed to allay fears and be operational in time?

- Another comment mentions how storage of waste is a question of will, yet we have pools of it all over, and the underground solution doesn't seem reliable either.

All I hear is that nuclear is the perfect solution if it were perfect and people weren't irrational.

Here's a qustion to help cut through the Gordon knot: at what point is it scientifically rational to accept/shun nuclear?


The reason politicians don't back nuclear is because of the way our political system operates.

A politician might believe our nation will need more power plants within the next 20 years. But they're incentivised to do the thing that, relatively, will be best for their near-term re-election prospects.

And just think of all the things that are better for their re-election prospects than a nuclear power plant in their constituency:

* A nuclear power plant in someone else's constituency. Play chicken with keeping the lights on and maybe someone else will swerve!

* Running an existing coal or gas power plant for longer

* A solar or wind power installation. Or better yet, one in someone else's back yard.

* Demand reduction like subsidised LED lightbulbs.

* Vague technology promises, like electric car batteries stabilising the grid.

* Finding of creating some organisation to take the blame once the brownouts and bill increases start.

* Just plain doing nothing and retiring before the bill comes due.

Why risk your job today supporting a nuclear plant, when you can kick the can down the road and keep yourself employed for another 20 years, maybe more?

Politicians are subject to those incentives, and it's hardly realistic to expect them to place the nation's interests above their own. Frankly it's a wonder any infrastructure gets built at all.


How are car-to-grid schemes any more unproven than next Gen reactors? You're not helping the side of science with untested hypotheses like that.

But my main point is realpolitik, or like you say "the way our political system operates." We have a political system that has evolved to handle issues in what we think is the least bad way (worst system except for all the others). Is nuclear in its blind spot, or would untested science-ocracy lead to worse outcomes?


No, the reason nuclear is shunned is largely one of cost. Rate payers don't want higher rates. Nuclear has led to huge financial risk displaced to the rate payers, and they don't like it. And they let their politicians know they don't like it.


Well, I think there's more than one reason, corresponding to the multiple points in the posts by trimbo and 205guy.

And one of those reasons is political, independent of finance. I agree other reasons are financial.

The reason I think there are political reasons independent of the financial is because I've seen politicians oppose waste incinerators and windfarms in the face of local opposition, even though they were entirely privately financed.


Politicians in South Carolina and Georgia embraced new AP1000 reactors at VC Summer and Vogtle. It was politically popular to support the new reactors, up until it became clear that they were catastrophically mismanaged financial disasters.

For every community in the US where you can find voters too afraid of radiation to permit reactors, I can find one where voters' first thought about new nukes is enthusiasm over the increased employment and property tax base. But that's a two-edged sword. It's also the local community that is going to be stuck paying off huge costs if the projects go over budget. The recent projects in South Carolina and Georgia went hugely over budget. The SC project has been halted. If the Georgia project finishes it's still going to be one of the most expensive energy projects in US history.

So it would be hard to pitch another AP1000 project in the US today, but mostly because the first ones had every sort of problem that the builders promised to avoid. Less than a decade ago there were a dozen places in the US that had tentative plans to follow up with AP1000 reactors of their own, if the initial projects performed to spec. They've evaporated because the initial projects were years behind schedule and billions over budget. If you want to have a serious conversation about building new reactors in the US, start with the potential host communities that aren't afraid of radiation but are afraid of multi-billion-dollar project overruns. What do you think it would take to build a new reactor at VC Summer now, given the painful recent experience with the AP1000?


>Politicians won't back nuclear because of the risk of accident, thus giving legitimacy to the risk and the public's fears.

No, that’s not the case. Politicians won’t back it because of their perceived risk based on their own ignorance. Or, even if they aren’t ignorant, they are focused more on staying in office rather than helping humanity so they don’t take up any projects that the ignorant public might pitchfork about. Neither of these “give legitimacy” to incorrect risk analysis.

A politician being afraid of flying does not give legitimacy to flying fears either. Risk isn’t measured by popularity.

AOC will gain far more popularity from her completely unrealistic green new deal than she ever would proposing something that would actually help based on nuclear. Politicians propose what’s popular, not what’s moral, ethical, or scientifically sound.

>Here's a qustion to help cut through the Gordon knot: at what point is it scientifically rational to accept/shun nuclear?

When peoples’ fears are based on a real understanding/evidence and not their own ignorant imaginations.


See my point about realpolitik above. Politicians more or less reflect the people, for better or for worse. Politicians are also the leaders and deciders, if not them, who else? There is a healthy nuclear industry, lobby, and research, so it's not like nuclear is unfairly excluded from the solutions table. As others have pointed out, cost overruns have sunk many modern nuclear projects.

> real understanding/evidence

Well, where are they? I originally replied to trimbo's comment because he seemed to be pro-nuclear and scientific, but then all his arguments seemed to acknowledge that well, nuclear wasn't up to the task. So where are the rational arguments that nuclear can be a solution?


The argument is that there are no rational reasons for humans not to support nuclear. The only reason it’s “not up to the task” is because humans as a group are behaving incredibly stupidly by ignoring the numbers and murdering people with coal while concern trolling an energy that kills fewer people than your standard annual mass shootings.

So the rational arguments are everywhere. The problem is that people want to clutch onto their irrational fear and believe in eithe the fantasy of non existent climate change or the fantasy of solar/wind storage becoming feasible in 10 years or less.

In other words, as a technical solution there isn’t anything nearly as good. There are no rational arguments against that. It’s arguing with anti-vaxers at this point.

As an actual solution, well how do we deal with the anti-vaxer problem? The same could be applied there.


But there's a very rational reason to not support nuclear: it's too expensive.

I used to support nuclear, but this argument against it has become too strong.


I agree with most of your points, however: most of the accidents have more to do with flaws in regulations than mistakes from the nuclear engineers. Chernobyl, which is arguably the worst and most iconic nuclear disaster of all times was mostly caused by a severe violation of procedures.

When you take into account he amount of energy that has been produced by nuclear power since the 50s, I'd say we have a pretty good record when it comes to overall safety.


I agree with most of your points, however: most of the accidents have more to do with flaws in regulations than mistakes from the nuclear engineers. Chernobyl, which is arguably the worst and most iconic nuclear disaster of all times was mostly caused by a severe violation of procedures. When you take into account he amount of energy that has been produced by nuclear power since the 50s, I'd say we have a pretty good record when it comes to overall safety.

The biggest problem with Chernobyl was the reactor design, which had a positive void coefficient, which outside of something like CANDU reactors is considered insane today. It’s true that their experiment with simulated blackout was the proximal cause, but the fact that increased steam production led to higher power output was the big factor that made it such a huge disaster.


When I learned about Chernobyl reactor, I was surprised how bad the design decisions were. Casing the reactor core in wooden housing?


IIRC the control rods were also bizarre and had graphite bottoms, so the first part of a trip (scram) was adding reactivity.

Prompt critical is prompt.


I believe the rods were graphite, just like the moderators. Of course the problem with that was that by the time they were inserted the reactor was already out of control. All the graphite rods did was burn, which certainly didn’t help matters! Worse, after a while the blocks of graphite used as the moderator also burned, and was difficult to extinguish. What they needed (other than better reactor design) was a ready store of neutron poisons that would have flooded the reactor vessel before rod insertion. Unfortunately, by the time they were using neutron poisons the explosion and fire had already destroyed anything like containment.


Chernobyl didn't have a containment or shield building. Manually disabling the safety systems didn't help.


The control rods had a plug of graphite on their ends, I believe, which made reactivity briefly increase as they were first inserted.


Chernobyl was a "military reactor", ie. could produce plutonium. Normal reactors don't even have the same failure mode possible. So you can blame the accident on the cold war or something, not nuclear in general.

http://www.phyast.pitt.edu/~blc/book/chapter7.html


also not helped by the fact that recent plant construction wasn't superbly handled.. delay, additional cost, failure of quality control for main parts..

Sadly I think nuclear energy needs a branding motive. Something that pushes everybody to make it clear, fast, precise and shiny. To instill trust in the eyes of people (that would require 10 years to educate on the physics)


There's a new book out, reviewed recently by the New York Times, that wants to change that perception. "A Bright Future: How Some Countries Have Solved Climate Change and the Rest Can Follow" [1] is co-written by an engineer who lays out the case for nuclear power. The authors think that rapid deployment of nuclear power is the only way to avoid a climate catastrophe.

[1] https://www.nytimes.com/2019/02/05/books/review/bright-futur...


How rapid is "rapid deployment of nuclear power", given that the plants take decades to build?


1) We can (and should!) build multiple plants in parallel. The sooner we start the better.

2) Countries like South Korea manage to build nuclear plants in around 4 years or so. The secret is that they have standardized on a design, and are building it again and again. There is hope that SMR's will make it easier to produce plants on time and budget, but those are still some years off.

3) Compare how the carbon emissions from electricity production in Germany has changed since they started their "Energiewende" project where they are phasing out nuclear and are deploying a lot of solar and wind. For comparison, look at e.g. the nuclear buildout in France (mostly from the late 1970'ies to the early 1990'ies).


If that is actually true, then we should just hire the South Koreans to build some power plants for us. If that experiment actually produces safe functioning USA-located power plants within budgeted costs, then we can worry about Americans learning to build similar plants.


The South Koreans only build reactors that fast when building in South Korea.

http://www.world-nuclear.org/information-library/country-pro...

Barakah 1, built in the UAE by South Korea, was originally supposed to take 5 years, operating in 2017. It has been delayed a couple of times. Bad news from December 2018:

"Cracks found in containment building of UAE nuclear power plant built by S. Korean companies"

http://english.hani.co.kr/arti/english_edition/e_internation...

It may still start in 2019. We'll see. But we're already at nearly 7 years.

Also, "countries like South Korea" is a club consisting of South Korea and nobody else. Other countries touted for their continued embrace of nuclear power, like Russia, China, and France, don't build reactors that quickly. If you want information right from the source, use the IAEA's Power Reactor Information System and look at the time between construction start and first grid connection for commercially operating reactors.

Here's France:

https://pris.iaea.org/PRIS/CountryStatistics/CountryDetails....

Their newest commercially operating reactor Civaux 2 took 8 years between construction start and grid connection. They have another, Flamanville 3, that will be newer when/if completed, but it has been under construction for 11 years.

Russia:

https://pris.iaea.org/PRIS/CountryStatistics/CountryDetails....

Latest commercially operating reactor Leningrad 2-1 took 9 years.

China:

https://pris.iaea.org/PRIS/CountryStatistics/CountryDetails....

Newest commercial reactor Hayiang-2 took 8 years.


Very little of what you're talking about there matters.

What matters is that new nuclear reactors have come in grossly over budget. Who is going to buy a reactor when the vendor's promises about cost are demonstrably untrustworthy?

In contrast, new utility scale solar and wind installs typically come in within 10% of estimate.


"No politician will back nuclear. If there's an accident, they'll be run out of town. Not worth the risk."

I would not be so pessimistic (or optimistic, depending on one's stance on this issue).

One does not have to look far in our current political landscape or in the history books to find plenty of examples of politicians doing stupid, unpopular, or even criminal things, that one could argue they should have avoided like the plague before they even started.

When or if those things fail, if the politicians get caught, or if there's some public outcry, it rarely lasts. The news media has a very limited attention span, and when the story is no longer in the media people tend to forget about it. Not only do most people tend to have a poor memory of political events and poor knowledge of history, but protest movements also rarely last long. Many "disgraced" politicians have managed to make a comeback in politics, and be rather popular, if that's what they chose to do.

The government-corporate partnership is also a revolving door where actors frequently move to highly desirable positions from one to the other. It's quite common for politicians to leave politics and get hired in lucrative positions in the companies they regulated, assigned contracts to, or passed legislation on, and for industry execs or lobbyists to get jobs in politics, after which they go back lucrative jobs in the company they left. Companies have many other ways of rewarding politicians who are on their side.

There's every likelihood that politicians who assigned multi-billion-dollar contracts to corporations will be very well taken care of, even in the event of some sort of political firestorm over their decision. Though politicians tend to also be quite adept at covering their asses and pushing the buck to someone else. "I wasn't told". "I didn't know." "I wasn't the one making the decisions." You can't open the papers these days without seeing those words come out of some politician's mouth.

There are few real consequences except in the case of truly egregious crimes, and even then they tend to be not so bad for white-collar crimes. Not to mention that the politicians tend to have top notch legal teams to keep them out of jail -- that's when they're deemed to not be acting in an official capacity, when they effectively can't be sued at all... and good luck suing the federal government.

Of course, in the case of a truly major nuclear disaster, such lawsuits may actually be successful. But if the risk of such a disaster with modern designs really is so low, then these politicians have nothing to worry about, do they?


>And the reprocessing ban was by a President who happened to be a nuclear engineer!

Jimmy Carter was not a nuclear engineer. He began a program to become a nuclear officer but quit before the Seawolf (his anticipated assignment) even began construction. Carter was, however, assigned to assist in the cleanup of a partial meltdown incident in 1952.

[0]https://en.wikipedia.org/wiki/Jimmy_Carter#Naval_career


> I can't think of a more scalable way than fission to combat carbon

I think it's a techno-optimism blindspot to prefer the "make more" solution to these things.

We waste enormous amounts of energy to non-essential things. Not just electricity production. Consider the electric cars craze. It's obviously just going to result in more emissions unless oil extraction ramps down at the same rate.

Solutions involving adaptation to lower consumption are met with protests like "then we would have to change our way of life", and for some reason that's perceived as a good argument.


Given the construction costs, the case for keeping existing, already paid-for nuclear plants open seems a lot stronger than building new plants that won't pay for themselves for decades. Who knows what the electricity market will look like then?

Solar and wind infrastructure is now mass-produced and competition is driving costs down. It's unclear what forms of energy storage will win, but I would expect it to be competitive as well.


Except that's why we have so many power plants based 1950's era designs instead of safer, more modern plants.


Just 15 years ago the outlook was completely different, and investing in nuclear made way more sense than in unobtanium cheap renewables. Thus it can not explain why people stopped investing in nuclear 70 years ago.


“Safer, modern” plants in the US, the UK, and China are being scrapped because they’re not affordable.

Solar, wind, and storage are already cheaper than nuclear. So why build nuclear?


As far as I know scalable storage for base-load is not a solved problem. Has that changed?

Personally I find it hard to believe that we can live on renewabls until a european country goes 1 year completely on renewables (I guess Germany is probably the best candidate for that since they are going all-in).


This touches on an issue, that renewables don't need to be paired with base-load plants (which nuclear has traditionally been), but with load-following and peaker plants.

Traditional nuclear power plants are base load plants, but any development in new nuclear plants should focus on designs that are better at load following, in order to complement rather than compete with renewables.


I agree that we should be investing in R&D, but what renewables have to be paired with is the technology that we have available to us here and now. If you want load-following energy then that means fossil fuel. If you want base-load then you have either fossil fuels or nuclear.

If we want to have any chance of meeting our climate goals then it seems like nuclear, optionally paired with renewables, is our only option.


If batteries and renewables are cheaper than nuclear, how is it not a solved problem? You still need to build the new renewables and the batteries, but it’s cost effective and can be done.


Because renewables are regional. They make exceedingly good sense in places like the US midwest or most of Europe where you can build solar / wind and have substantial availability. Solar, by contrast, is much less performant if you live somewhere with less sunlight due to any number of reasons - weather, smog, latitude, etc. Same with wind - it only makes sense to build wind where you get the most wind.

And the trick with those is that its not a zero sum. Even if solar would be viable somewhere less optimal panels are still going to go where solar is cheapest first. And then you are looking at the costs of transporting that power gradually towards areas of reduced renewables efficiency and/or storing it.

To put it even more generally, if the state of Maine were willing to build a fission plant but not buy solar, wind, hydro, etc LET THEM BUILD IT. You can manufacture all the solar panels and windmills you can while also building nuclear plants. The priority has to be shutting down all oil and gas plants as soon as possible by any means available.


No one is stopping anyone from building modern nuclear reactors except debt markets and utilities. The problem is that no one wants them (compared to other generation technologies).

Timoth3y in another thread explains it extensively: https://news.ycombinator.com/item?id=19168057


That’s wrong. Debt markets are reflecting the risks of the plants being stopped by orgs like the NRC, etc. The debt markets don’t favor it precisely because so many different parties can block new builds.


I disagree with your assertion. It's simply time value of money. As an investor, you don't want to wait decades for a higher risk low return that may never materialize, or the bond defaults entirely because your nuclear plant is shuttered for safety or political reasons. Could government insure the bond? Yes! But then they're subsidizing nuclear when that money could go into renewables and storage.


You didn’t disagree, you just affirmed it. The debt markets treat it as high risk precisely because so many parties can easily stop it.

Nobody refuses to invest in nuclear because of the negligible meltdown risk.


That is a good point and I don't know enough to answer (if you have any good info sources about this I would appreciate them).

Aside from the Tesla battery in Oz (which is more about stabilization rather than primary supply) I don't know of any large-scale battery storage systems.

I get the feeling that we are ignoring the scaling part here (We Just have to scale-up). Since the battery systems would need to be big enough to cover dips in renewable output in a worst case scenario (brown-outs / black-outs are not going to be popular).

Is there any info on how many / how big / how long to manufacture these batteries and renewables would have to be to cover an example country?


You’d need 0.6% of the continental US land mass for solar panels, and 1 mile by 1 mile of batteries to power the entire US off of solar.

This doesn’t include tens of thousands of EVs sold each month, which is rapidly increasing the aggregate amount of distributed energy storage available to the grid.


That’s not right. That’s only about 5 times the size of the gigafactory and there is no way that has nearly the capacity to power the entire US for 9+ hours of continuous darkness. You need to check your math...

Unless you meant a cubic mile of batteries, in which case that’s completely unrealistic to manufacture.


No, storage is not cheaper. That’s a fantasy. If it were true they would be rapidly displacing all energy production in the US.


He was saying that renewables + batteries are cheaper than building new nuclear powerplants. This doesn't imply new renewables + batteries would be cheaper than continuing to operate existing power plants (most of which are not nuclear, and all of which have their capital cost sunk so it shouldn't be counted), which would be required for your inference of "rapidly displacing all energy production in the US".


Just about every builder and operator of large nuclear fleets of reactors have all outright said or implied by their actions that nuclear is uneconomical vs renewable power options. The massive bulge of capital and scheduling needed points to humans not being able to build a next generation of nuclear plants in time to head off the worst climate change, and renewable energy sources being both faster and cheaper to build.

France: 44 Billion would be saved by investing in rewables instead of nuclear

https://www.bloomberg.com/news/articles/2018-12-10/french-po...

Private operators in the US have said new plants are unlikely due to the economics of nuclear:

https://www.usnews.com/news/national-news/articles/2018-04-1...

The last few completed plants in the US (that didn't get canceled) were 20-30 years in the making - and typically over budget and schedule by more than 2x. With some of the cancelled projects over budget by 3-4x.

Even China, which embarked on building it's own fleet of nuclear plants, and has all sorts of regulatory advantages found that construction was slow and overbudget:

https://www.technologyreview.com/s/612564/chinas-losing-its-...


None of this has anything to do with technical problems with building nuclear reactors. The US Navy has been hiring private contractors to build them for decades with no such issues. (And Naval reactors have more stringent requirements than commercial ones, because of the widely varying loads they have to deal with.)

The so-called "cost savings" in France are due to unrealistically optimistic assumptions about renewables, not any issues with nuclear.

The issues in the US are entirely legal and political: nuclear power plant projects get tied up by interminable lawsuits based on NIMBYism and ignorance, and the US political policies on fuel reprocessing have turned what should have been an easily solved problem into a quagmire.

I suspect the "problems" in China are more due to their cavalier attitude towards coal than anything else; they don't mind firing up a lot of new coal plants and having lots of people suffer respiratory disease and death as a result, if that's cheaper than nuclear.


It depends how you defined technical issue. For example the French Areva power plant projects in Europe have been met with all sorts of issues ranging from subcontractors that are inexperienced with NPP level of quality and documentation requirements all the way to challenge of proving automation system reliability to the regulatory body. These happen in the plant vendor's native France as well, so shouldn't be cultural issues either. See https://en.wikipedia.org/wiki/EPR_(nuclear_reactor)

These are not NIMBY or political issues, but rather stem from the fact that there's been a long pause in building new NPPs in the west. All the senior people involved with earlier designs are long since retired and even the juniors that were around might have moved on. And also some of the technology choices used in old designs are not even available anymore, so there's new families of tech to prove to the regulators, digital automation being one of these to my understanding.


Also, we've seen more things go wrong.

When you design a nuclear plant, you need to do a really careful risk analysis. That's why they are so safe.

But as time progresses, and actual designs meet reality and Murphy's law kicks it, it turns out some of the assumptions are factually incorrect. More elaborate safety requirements are needed to compensate, increasing cost.

You can't build a car the way you could 50 years ago, and the same is true for nuclear plants.


Navy reactors have completely different design parameters, one of which is the use weapons grade fuel...


Why don't we just buy a few of those Navy reactors and hook them to the grid?


They wouldn't be close to competitive.


I assumed as much; that's the point.


I can get the "We cannot build them in time" part but the whole "It is not economical to build them" strikes me as a pretty crazy position to take.

Uneconomical compared to the massive disruption that climate change will bring to our civilization? Are we fighting to save our future or not?

Discarding a proven technology that will massively reduce our carbon emission for an unproven "We still have problems to solve around wide-scale baseload" seems to be ass-backwards to me and pinning more on hope than common sense.


That's not the right metric. The question is whether you can sell electricity for a lower price than the competition or will end up selling electricity at a loss, resulting in "stranded costs" for rate payers. To do that comparison, you need to predict prices in the electricity market for many decades. Predicting the future is hard, but the trend for solar and wind costs is pretty clear and it seems foolish to bet against them.

A high carbon tax would make nuclear more competitive against fossil fuels, but not solar and wind.


> That's not the right metric.

I guess that opinion comes down to whether you think Climate Change is an existential threat to our civilisation or not.

If you are of the opinion that it is then costs be damned and build everything: Renewables, Storage, Nuclear.

If we end up with more power than we need then we can shutdown excess or funnel that into powering carbon capture technologies if we ever get there.

If you are not of the opinion then current economics, as you have described, rule the day.


> I guess that opinion comes down to whether you think Climate Change is an existential threat to our civilisation or not.

Not necessarily. I don't think climate change is an existential threat to our civilization; human civilization has already dealt with problems much bigger than the average temperature going up by a couple of degrees and sea level rising a couple of feet. (Personally, I think the biggest existential threat to human civilization comes from humans with false beliefs and too much political power.)

And yet, I think nuclear energy should be a no-brainer. (I have no problem with renewables or storage either; I just don't think they can provide enough reliable base load power by themselves.) Why? Because we need to stop buying foreign oil for national security reasons. Which is not an economic reason--economically speaking, oil is still pretty cheap--but I still think it's a good one.


> already dealt with problems much bigger than the average temperature going up by a couple of degrees and sea level rising a couple of feet

With respect, I think you may be underselling the effects! The chaotic knock-on of species loss; water pressures causing human migration; a loss of half of the ice on the third pole (the Himalayas) that provides water to over a billion people; acidification of the oceans. We could go on but the point I'd like to hammer home is this isn't anything at all like the previous crises in our brief 60k year sojourn: the anthropocene is an apocalyptic event that will wipe out half of the species on earth (probably).


> this isn't anything at all like the previous crises in our brief 60k year sojourn

We do not know that this is true. Alarmists claim it, but they have no evidence to back it up. For example, we don't know what the ice in the Himalayas was like, say, 6000 years ago. We do know that as glaciers are retreating now, they are uncovering evidence of human habitation in many mountain areas, which is evidence that there have been previous warm periods where there was no ice there as well. That is evidence against the claim you are making. But you don't hear about that from alarmists.


It's always going to be a matter of economics.


Not as much as you think. When countries decide that there is an existential threat that they must fight, they can mobilize enormous resources. Look at what happened during the last World War. Imagine what we could do if we decided that Global Warming is as much of a threat as Germany was in 1939.


Climate is a matter of economics though. The difference is that the cost is in the future. That's why the poster was talking about the cost of climate and if you consider it an existential crisis. I guess it also matters if you care about current costs vs future costs.


As a sidenote. I also have trouble understanding the mindset of people who argue purely on economic basis.

It seems that telling out grandchildren "Sorry you inherited a ruined climate, but we didn't want to spend the money at the time to avoid it" is a rational decision. (And hey, maybe some people think it is; but I can't)


I honestly think they fall into one of 2 camps:

1) They don't believe in anthropocentric climate change (probably deflecting by taking this stance)

2) They don't believe the economic destructive power of climate change. Be that by disbelief, thinking it is over stated, or just not being good at thinking about the future (which a lot of humans aren't. We're terrible at it)


There are two issues here. One is, are we currently pricing carbon appropriately? And obviously not. But if we did then it would be an economic issue -- make emitting carbon more expensive than alternatives and people would naturally gravitate to alternatives.

But then we have the second issue, which is that we don't want the alternatives to be significantly more expensive than the status quo, and we may want to take some steps to ensure that.

If you tax carbon and so people switch to renewables and batteries and figure out how to make it scale efficiently, everything is fine. If not, people would then turn to nuclear because it would still be less than paying the carbon tax.

But if we wait to start building nuclear plants until after we've already established whether or not renewables can handle the entire grid, and then it turns out they can't, that could be problematic, because then you've waited too long.

It's a hedge. If we fix the nuclear regulations to remove the most egregious of the cost-wasting portions and start building nuclear plants now, and then we don't need them because renewables fully succeed, energy will cost modestly more than it would have. Or the investors in nuclear would lose money in that case. Whereas if we don't, and then we do need them, energy would cost dramatically more than it would have because of the shortfall. And as a result people would be inclined to pay the carbon tax, or repeal it, and then we get catastrophic climate change.

When "need it and not have it" is literally the end of the world, better to have it and not need it. That seems like a worthwhile hedge.


We already know that renewables can handle the entire grid. There are plenty of papers out there exploring this matter.

What we don't know yet is exactly what particular technologies we're going to end up with, especially for the last few percentage points. Probably a mix depending on location.

The reason for we don't know it is that wind and solar costs have been falling exponentially for many years now, while short-term storage is only now heating up and long-term storage not yet really started (well, except hydro).

We know tech that can be used, we know an upper bound on the cost (not prohibitively expensive), but anyone trying to look 10-20 years out in such a rapidly developing cut-throat market are basically just guessing.

10 years ago I was thinking we'd probably have to pay somewhat more for energy, perhaps up to 50%-100%. But in 2014 the Danish energy oversight body (Energistyrelsen) did a complicated analysis that said around 5% more in Denmark. And it didn't take more than a couple of years for their projected 2030 cost of offshore wind turbines to be reached by the market, more than a decade before anticipated.

So I'm beginning to think energy is probably going to be cheaper, perhaps a lot cheaper.


> We already know that renewables can handle the entire grid. There are plenty of papers out there exploring this matter.

This is all inherently speculation. If you assume the current cost trends continue, maybe everything is fine. But what happens if you scale production by a factor of fifty or more? Do raw materials prices for things like lithium or rare earth metals go up due to demand? Do average operating costs rise once the most cost effective installation sites are already occupied and the low hanging fruit has been picked in general? Are we analyzing numbers from existing installations that have selection bias in favor of factors that the remaining sites don't have, e.g. lower vs. higher population density?

Doing a long-term analysis on something with this level of complexity inherently has large error bars. Even if "everything will be alright" is a likely result, we still need to address the non-negligible probability that it isn't.


You explained and expanded on my thoughts brilliantly, cheers.


The Netherlands believes in climate change, understands what a rise in sea level means and will not build a single nuclear reactor.

Spending 15 billion on a reactor just isn't economically sound business.


3) they think, but don't explicitly say, that externals should be priced in. But AnthonyMouse expressed the real dilemma quite well.


It's not that nuclear is unproven, it's that _renewables at this point are proven_.

You can build huge (or tiny! it's super scalable) fleets of renewable energy sources pretty economically, with a ~0% chance of making surrounding areas uninhabitable for 20 years if there's a failure.

Yeah sure a truckload of uranium goes a long way but building the plants is really hard even when you look past the regulatory issues.


I think you're overstating the provenness of renewables. There's a lot of support infrastructure that hasn't been invented that is required to produce the roughly 3 terawatt average per day as well as the storage for the daily cycle.

I also think your overstating the risk involved in nuclear power. Even fukushima was pretty tame, and unlike Japan the USA has lots of relatively uninhabited real estate we can put the plants on.

Cost is one factor. Location is another. Reliability another. Risk another. Nuclear is a system of tradeoffs that is very hopeful for fighting emissions.

Of course this is all moot since even if the USA and Europe went to zero emissions, we won't make a dent in climate change. The developing world must be addressed with at least as much urgency as the West must be addressed.


> There's a lot of support infrastructure that hasn't been invented that is required to produce the roughly 3 terawatt average per day as well as the storage for the daily cycle.

On the contrary, two solutions already exist.

Batteries are expensive, but (IIRC) still cheaper than nuclear, though more expensive than fossil fuels.

And the losses of planet-scale grids are small compared to the cost-advantage that solar has. You could literally power your home at night from the sun hitting the other side of the planet. (That said, I have no idea what the Installation or maintenance or political costs are, only the efficiency).


> the losses of planet-scale grids are small compared to the cost-advantage that solar has. You could literally power your home at night from the sun hitting the other side of the planet.

You're comparing apples and widgets. The losses of planet-scale grids are way too large to make it even feasible to power your home at night from the other side of the planet: not enough power would be left by the time it got to you. The cost is irrelevant.


“””the authors measured the corona loss of a 765kV, 3 phase, and bundled transmission line to be about 1.87kW/km in fair weather. This amounts to only about a 0.083% loss over a 1000km line. In bad weather, however, the authors measured the loss to be 84.3kW/km, or about a 3.7% loss.””” - http://large.stanford.edu/courses/2010/ph240/harting1/

3.7% loss = (100-3.7)/100 multiplier per 1000km = 0.963

Half world circumstance = 20,000km -> 0.963^20 = ~0.47 multiplier

“””New US Solar Record — 2.155 Cents Per kWh””” - https://cleantechnica.com/2018/06/14/new-us-solar-record-2-1...

-> 2.155 cents per 0.47 kWh = 4.5 cents per kWh from the opposite point of the planet, assuming the worst case studied in the first link on the entire route.

It could be 80% losses (0.2 kWh receives for every 1 kWh produced) and still be effective both from ‘cost’ and ‘maximum possible power output’ perspectives.


Ffs, corona loss is only one small factor of transmission loss. https://electrical-engineering-portal.com/total-losses-in-po...

Ignoring that major oversight, transmission lines aren’t free and they require upkeep so building/maintaining them across the ocean needs to be factored into your costs.

It’s like claiming it’s cheaper to have sushi flown in from Japan everyday if you just ignore the air freight cost.


Yes that was a silly omission. However, adding resistive losses doesn’t change the conclusion because the margin is so large (it does make it much closer though). And this is ignoring that the line losses we see now are based on the most cost-effective designs for nation-scale grids, when one can straightforwardly (for example) use a higher voltage for lower resistive losses (no point reducing those losses further on nation-scale grids), or even use existing medium-temperature superconductors, which is thing but not widely used for power yet.

And I literally acknowledged that I was excluding the cost, financial and political, of building and maintaining the lines in my first post on this thread.

I ignored those costs because my argument is, and always was, that solutions already exist. That the line losses — large as they are for a worldwide grid — are not a fundamental problem. It’s not like we can’t build pylons or have yet to invent a way to join wires together after they come out of a factory.

And we already have a lot of national scale grids, how hard is to join the existing ones together? Sure, it’s a bit close when comparing 2.155 cents per kWh * 20% line efficiency (which is worse than even my updated estimate!) to fossil, but that’s also your midnight cost, when you use least.


If you have the power to just throw out costs as a concern, then we’ve had a solution for much longer. Geothermal.

>And we already have a lot of national scale grids, how hard is to join the existing ones together?

Very. We haven’t even managed to get the US on a national grid. To connect continents is just fantasy at this point. Do you realize the undertaking it would be to get 700kv line to Hawaii, Australia, New Zealand, etc?


Batteries are also one of the worst environmentally destructive things to mass produce given current best technology. So there's that.


Someone said roughly the same thing in another thread https://news.ycombinator.com/item?id=19167849

But as far as I know there is no proven large-scale storage system, which is a criticla part of renewables. The other commenter mentioned building batteries and then it is just a matter of scale which seems like an over-simplification to me but I cannot find much useful info on it.

Personally I would love a 100% renewable setup but I would not bet the future of our civilization on it when nuclear is another option.


We do have a very simple large-scale storage system that's already in use across the world: pumping water uphill. It's just not as economical as rain when you factor in the efficiency losses from generating and transferring power twice.


And pumping water uphill is a regional solution that requires specific regional geography, to me that does not class as "scalable".

I should be more specific in that to me "scalable" is a solution that can be deployed anywhere without require specific regional properties. Maybe you think this is unneeded and that a lot of region specific solutions can cover the storage requirements.

However another posted did point out that 1 mile by 1 mile of batteries is apparently enough to store US energy requirements.


Unless there is big improvements in battery tech I don't see it as a solution. Lithium batteries have their own environmental problems and die after 10 years.

Pumped hydro requires the geography, building new capacity tends to flood large areas and release huge amounts of greenhouse gases initially.

Simply lifting weights on cranes or rail tracks is looking more efficient and can be adapted anywhere. It will probably win out in many cases.


> Simply lifting weights on cranes or rail tracks is looking more efficient and can be adapted anywhere.

The energy density of that kind of scheme is a joke. Thought experiment: a fully charged Tesla, how often could it climb a hill the size of that crane if it skips recuperation on the downhill leg? That's how much lower the energy density of a crane storage would be. Pumped storage works (where the geography allows it) because water is by far the cheapest and the most easily transported ballast and geographic height differences dwarf almost every human made structure.

If you want something that scales everywhere, look no further than compressed air. It's usually ignored because of the big thermal losses, but if you have a direct application for coolant they are not that bad and even without, it serves as an almost trivial lower bound to the storage problem. We can calculate how much intermittent energy production we would need with compressed air to serve a power demand profile and everything else is just a possible improvement.


Since when was energy density ever been a problem for grid storage? Even in the most crowded cities on earth it's not a constraint.

Unless you are talking about portable applications it's all about round trip efficiency and cost, I'm not sure if you are getting confused about terms here or talking about an entirely different subject.

Compressed air efficiency isn't great. Using it for cooling lowers the efficiency even more.

Crane and rail gravity storage are pushing 80-90% round trip efficiency.

Previous HN discussion: https://news.ycombinator.com/item?id=17789456


Nuclear power plants are also regional solutions that requires specific geography....


Nuke plant siting is vastly less constrained than pumped hydro facilities.


It doesn't require specific geography. You can create entirely artificial pump systems isolated from any natural water. These are called closed loop systems. You could create an closed loop pumped hydroeletric system in the middle of the Sahara if so desired. Incidentally you could power the entire world with a solar area taking up a single digit percent of the Sahara.

These haven't been actively developed in the past because we don't have much need for massive storage and they also take something on the order of a couple of years to plan, develop, and execute. You need the demand to be there before the storage is built, but the demand won't exist until the storage is built. Fun problems. Because of this batteries are a more practical immediate solution. They can be deployed anywhere, at practically any scale, with negligible time requirements. And similarly for manufacturing. Since they aren't 'geo locked' their market flexibility is much greater.


The Sahara-global-solar-facility scale starts bumping up a lot when you factor in realities.

PV efficiency, spacing factors, panel replacement cycles, storage requirements, the fact that we're looking at total energy use and not just electricity, first-world rather than third-world per-capita use rates (presuming we're not going to freeze the entire world at its present state of energy consumption), and projected population growth.

You can still provide most or all the hypothetical demand from the Sahara, but you're well above 1% land use. I've sketched this out elsewhere previously, don't have numbers handy.

Beware optimistic estimates.


I worked out the numbers several years back and it was around 5%. Efficiency improvements since then should have improved this a fair amount. Even if you bump it up by an order of magnitude it's still rather remarkable how easily we could power the entire world on solar alone.

However, I completely agree on the real issue being one of longterm consumption. I would say this is something that's regularly ignored. The developing world starting to consume developed world electricity/capita alongside increasing world population is easily going to increase energy consumption by some orders of magnitude in the foreseeable future.

This poses unique challenges few are considering. For instance nuclear also runs into problems here with resource availability. The technology is already rather cost prohibitive and for future energy needs if it became a primary source you'd absolutely need to move to breeder reactors alongside saltwater uranium extraction which would both push the prices up significantly higher than even present. High energy demands alongside high energy prices might make the production owners/shareholders happy, but not much of anybody else.

In any case sooner or later we'll end up relying on solar simply because nothing else can compete on gross energy availability. The sun's a fusion reactor that could fit about 1.3 million Earths inside of it. That enables practically unlimited power out there just waiting to be harnessed one way or the other.


There are several solutions to supply-demand matching and buffering which either do or should work.

A key is to think of this in terms of matching supply and demand rather than simply as storage. We've adapted over the course of a century or so to a regime of dispatchable supply energy, with little use of dispatchable demand. There's also been little consideration of major behavioural, social, economic, and land-use changes which will be prompted by changes to the energy regime. Much the same way as major impacts of internal combustion engines on land-use, construction, transport, and trade were almost wholly unanticipated, most discussion today is framed in terms of "how do we sustain present behaviours and activities under a novel energy regime" (if not quite so explicitly). The short answer is: you don't.

Automobiles, rail, air transport, and powered shipping gave rise to suburban sprawl, transcontinental trade networks, same-day globe-spanning travel and light cargo, and transoceanic shipping centres, along with tremendous centralisation of activities in zones of maximum productivity (often, yes, through massive externalised costs). Little of those impacts was foreseen in the popular or academic literature of a century (or even half-century) ago.

There's much of economics that's badly broken, but a part that's useful is the notion that behaviours do change tremendously in the face of changes to real and expressed costs. The Jevons Paradox cuts both ways: increased efficiency increases total use, whilst increased costs will decrease total use of some resource or factor. (Increasing efficiency is equivalent to saying "decreasing cost".)

Addressing energy specifically:

Expect to see far more dispatchable load, effectively, "making hay whilst the sun shines". High-load, but bufferable uses such as thermal heating (water, space, thermal storage), electrically-driven refinement (aluminium smelting, electric arc furnaces), reverse osmosis desalination, and the like, can if not "store electricity", then cache useful activity whilst supplies are abundant. Smaller industrial, commercial, residential, and possibly transport loads may also see time-shifting on a similar basis.

For direct storage, pumped hydro, compressed air energy storage (CAES), grid-scale batteries (an area with frustratingly slow development, though some promise, especially with cheap-and-abundant if not highly-efficient electrolytes), are presently proven. There are a number of schemes which don't work particularly well -- flywheel storage doesn't seem useful for much besides replicating today's "spinning reserve", nor do supercapacitors look as if they'll offer much beyond grid-scale power conditioning.

Two areas which offer tremendous promise and fairly high probability of success are grid-scale thermal energy storage with regeneration and electrically-based fuel synthesis appear at least on a back-of-the-envelope basis to provide national-scale grid-level storage capacity good for weeks (thermal) to millennia (fuel synthesis). The round-trip efficiencies are not great, but the simplicity, safety, and in the case of fuel synthesis, exceedingly long-duration storage, transportability, and utility of the derived medium are huge advantages.

Thermal energy electrical storage is mostly used now in solar thermal generation plants, but could be utilised in other forms. Large insulated tanks of molten salt driving traditional steam turbine generation could offer weeks worth of grid storage for the US in a total storage capacity roughly of the magnitude of extant oil transport storage facilities in Oklahoma.

Synthetic fuel generation, first suggested for nuclear power at Brookhaven National Laboratory in the 1960s and researched for over 50 years at Brookhaven, M.I.T., and the US Naval Research Laboratory, has yet to be proven at national scale, but the basic chemistry works, it's similar to coal-to-oil processes used by South Africa and Germany since World War II, produces direct analogues to current fossil fuels (methane through bunker oil) but is carbon-neutral as the carbon itself is sourced from current biosphere reserves, principally seawater.

Otherwise: expect to see tremendous differences in how energy is used, in construction based around heating and cooling loads, lighting, transport, and other processes.


Can you give an example of an industrialized first-world nation using Wind + Solar to provide baseload power on a consistent basis? Because if you can't then they're far from proven as anything more than a partial supplement to other energy sources. The economics of Solar/Wind get much worse as you scale past ~30% of the grid because utilization factor goes down and the need for energy storage goes way up.

Right now grid-scale batteries are uneconomical and pumped hydro is only feasible in specific locations (not to mention its ecological impact) so the 'storage' backing renewable installations takes the form of big tanks of methane next to gas turbines. Even when conditions are good and it's not being burned, that methane has a tendency to leak and wipe out a fair proportion of the face-value greenhouse emission savings from swapping to renewables in the first place.


That's not correct. Denmark is currently is currently at something like 45% of electricity from wind, and going past 50% in a year or two IIRC. It's not base load, because you don't need base load - it's an obsolete model from a world of capital-intensive, constant output plants. What you need is to match the demand.

Regarding long-term storage: Exactly. Tanks or caverns filled with methane generated from a renewable source is one possible solution. Of course, you need to have a firm grip on leaks.


Renewables aren't any better at load following than current nuclear technology, and unlike nuclear energy there's no potential for the development of proposed designs that do have load-following capability. Never mind the fact that intermittent renewables themselves are the reason the 'mostly base load with some load following' model is obsolete, precisely because they can't be relied upon to generate electricity on demand.

Denmark is definitely the poster child of wind energy (and intermittent renewables in general), but it benefits enormously from utilizing the existing pumped hydro storage of Sweden and Norway, which is an option the vast majority of countries don't have. The economics would look at lot worse if they were having the build out batteries to provide that storage capacity instead of taking advantage of what's already there.

Just saying 'you need to have a firm grip on leaks' is trivializing a significant engineering problem. Existing natural gas plants have a 1-9% leak rate, and given than leaking fuel is a complete waste of money we can safely assume that doing better than that is non-trivial. Given that in the case of carbon-neutral 'synthetic' methane you have the possibility of leaks at the manufacturing plant as well as in transmission/storage, it's reasonable to expect a figure closer to the higher end of that spectrum, which makes lifetime emissions from Wind/Solar sources significantly worse than Nuclear on a per-GWh basis.


Renewables will replace the most expensive fossil plants before the cheaper ones, which will make the base load price to compete with pretty hard for nuclear to match.


Right now it looks like the opposite is true. Due to a lack of storage we need more expensive gas plants instead of cheap coal plants that can't follow demand as quickly.


Overnight capital costs are significantly lower for new gas plants than new coal plants. Real capital costs are lower yet since gas plants are faster to complete. Fixed O&M costs are also lower for gas. The shale revolution made fuel for them cheap too. There's no combination of circumstances that presently makes a new coal plant cost-competitive with a new gas plant in the continental US.

See table 1 here:

https://www.eia.gov/analysis/studies/powerplants/capitalcost...

The renewable costs are already badly outdated but AFAICT the relative prices of coal and gas plants are still about right.


> "It is not economical to build them" strikes me as a pretty crazy position to take.

Why, no matter what resources you allocate financially, if you have X dollars, you can do more and sooner with X building renewable sources than if you split it between renewables and nuclear.


Because that is based on the assumption that renewables and no nuclear is viable and going to work based on advances in technology.

I think there is enough debate in this thread to illustrate why that is not a universal position.

This comment explains it better: https://news.ycombinator.com/item?id=19169073


It's difficult to see how nuclear can fit in, even if renewables aren't directly covering everything.

Renewables will be intermittent, which means as they build out they will often crash the price of electrical power. Nuclear cannot do well in that environment, since it cannot dispatch economically. Its costs are largely fixed, and it depends on selling power at a good price averaged over the large fraction of time it needs to be operating. But renewables slash the price it can get much of the time, even if they cannot cover everything.

If nuclear is reduced to covering (say) 20% of the time cheap renewables are not available, it's a dead technology. Something else, like using hydrogen produced from renewables during their high output times, will be used instead.


Jean-Marc Jancovici, engineer in the energy field and pro nuclear, said that the math on renewable is bogus. And that if you don't want nuclear, money should be spent on energy saving measures like heat pumps and building insulation.

I do understand how tempting are renewables, low risk, low entry cost..


Ironically, and I say this as someone who is very pro-solar, nuclear power is the safest power source around. Deaths per TWh are lower just because of how many people generally die falling off rooftops and that being a common place to install them.

http://www.edouardstenger.com/2011/03/25/a-look-at-deaths-pe...


it's not flat statistics the issue, it's the high risk and high complexity. Solar roof can kill if they fail but it's quite easy to see and deal with. Nuclear plant failures are too extreme and complex. There was a research center not far from my town, and every year people are worried that their thyroid issues aren't due to past experiments polluting the underground water supply. It's never conclusive but always on people's mind.


If he's right, why are everyone + dog installing solar and wind, but hardly anyone is installing nuclear? Has there been a sudden massive decline in greed in the world?


Fad and clueless politicians. Govt gave credits for Solar and wind but apparently never did the math for other ways to save energy. Again nuclear is too risky politically...


Really? Everywhere? There's tens of billions of dollar in potential economic windfall (so the pro-nuclear idea goes) to extract their share of graft from, and politicians are ignoring it?

No. You are providing the usual pro-nuke excuse, one that has become threadbare and sad as it's been repeated over the years. At this point, it's just ridiculous.


> France: 44 Billion would be saved by investing in rewables instead of nuclear

>link

I can't read French so I can't comment on this report. I'm not saying it is wrong or your conclusions are wrong. But I would say that there's an unfortunate thing happening and it is that bad information and research is being done. In fact we don't have to go far back in HN history to find this post

> https://news.ycombinator.com/item?id=18950154

All I'm saying is that you have to be wary of these things (and a lot of people are upset about this, because that's not what science should be. It also gives cannon fodder to global warming deniers...)

But to comment on France specifically, they are basically the only European country making good climate decisions

https://www.electricitymap.org/?wind=false&solar=false&page=...

Sweden has a great advantage of a lot of natural hydro sources. But something to pay attention to in these countries is how their electricity is generated over the course of a day. You may also want to look at how different countries have fared over time (not including China because you can't make fair comparisons with it).

https://data.worldbank.org/indicator/EN.ATM.CO2E.PC?end=2014...

> The last few completed plants in the US (that didn't get canceled) were 20-30 years in the making - and typically over budget and schedule by more than 2x. With some of the cancelled projects over budget by 3-4x.

This is honestly a HUGE problem. And I think everyone pro nuclear agrees with you. There's a lot of reasons why. It isn't just, as others have mentioned, the NRC requiring essentially complete designs. But there are a lot of things that could be fixed by policy changes.

You do also have to consider the cost of human life. Safety is a great concern with energy:

https://www.forbes.com/sites/jamesconca/2012/06/10/energys-d...

But here's an interesting perspective (I don't think any sane person is arguing for 100% nuclear, or even 50%, but the perspective is interesting).

https://medium.com/@Jorisvandorp/how-much-would-a-100-nuclea...

But I think there is one argument that stands out with nuclear. It is the question of how we transition to "all renewable" or "renewable + fusion". We don't have the technology to handle the storage capacity of solar and wind for periods when there are no solar or wind. I mean we hear a lot about Germany's renewable feats, but look back at that world bank data and you'll notice that between 2009-2014 the trend is flat. So while I think we should sink a lot of R&D money into battery technology (and sequestration), you also need to have a contingency plan. While nuclear is expensive, it is clearly less expensive than what happens if we don't reduce our carbon footprint. The thing is just that we don't have time anymore. We can argue all we want, but with existing technology, the only option is nuclear. And we need to act NOW, not tomorrow. NOW. (now is even too late)


Countries like Sweden and Norway also don't want to build out any new hydro infrastructure at this point -- hydroelectric dams have a considerable ecological impact, something rarely mentioned when renewables are discussed. These countries are famous for their hydropower today, but some of these projects met with considerable protests at the time they were built (such as the Alta-Kautokeino waterway [1] in 1979-1981).

[1] https://www.sciencedirect.com/science/article/pii/0016718583...


In most developed countries hydro is more or less fully exploited already. It provides a lot of low-C dispatchable electricity, which is incredibly valuable, but there isn't much potential to increase it. So it isn't by itself a solution to how to decarbonize the world economy.


Waste disposal is a big issue too. No state wants the waste, so it keeps lying around and leaking at decommissioned sites.


What stinks about nuclear is it got too big too fast. There was a massive buildout of these 1950s-1960s reactor designs, before we had time to fully realize the full cost lifecycle of said designs, and their failure modes. I guess it didn't help at the time we were facing nuclear obliteration between two super powers :/

Still I really wish people could differentiate the risks behind a PWR, BWR, Liquid Metal cooled, the AP-2000 design, and a thorium cycle. They're all kind of lumped together unfortunately.


There's just no way nuclear is making a come back until the economic issues are resolved. Almost all nuclear projects are over budget and over time by 2x or 3x, except for plants built in/by Russia, India, China


As I mentioned in another comment, China and Russia don't build reactors all that quickly. Their newest commercially operating reactors Hayiang-2 and Leningrad 2-1 took 8 and 9 years respectively.

India is slower yet. Their newest reactor Kudankulum-2 took 14 years to build:

https://pris.iaea.org/PRIS/CountryStatistics/ReactorDetails....

Could the US build a reactor in 14 years like India does? I think so, even with my currently-low opinion of the industry's ability to plan. But I also don't see how to guarantee institutional support that lasts 14 years. Neither utilities nor politicians have the tools or incentives to support something that takes that long to gestate.


This highlights the foolishness of relying on magical government solutions. Nuclear power has plenty of promise, if the waste is magically taken care of by an as yet to be implemented solution from the government. Since this solution never arrived, the waste gets stored at each nuclear plant, where it needs to be actively cooled for some period of time.


It would need to stay at the plant for a few years in any case to be cooled. Once it's cool enough for passive dry cask storage, it's also cool enough yo be shipped.


There's nothing in principle preventing private companies from reprocessing spent nuclear fuel and storing what remains in long-term underground storage. It's just illegal for them to do the reprocessing and too politically unpopular for any local government to consent to having a storage facility built on land it manages.

It's misleading to refer to a 'magic government solution' as though the nuclear industry is expecting the government to invent a magic wand that will disappear spent fuel with a flourish of the hand.


It's not illegal to reprocess. Carter's ban was rescinded by Reagan. What it is is stupid. Reprocessing is a net economic loss.


Worked in nuclear industry in the late 90's. Most *WR designs are inherently not fail-safe, complex, require dedicated teams, are nearly impossible to insure and require a massive capex investment. This isn't necessary. Modern MSR/LFTR designs tend to get around most of these problems by being smaller, simpler and most robus, similar to what some startups, Taylor Wilson proposed and partially what ANP/ASTR/MSRE/DMSR proved. Downsides to many small reactors include more, smaller sites needing physical/SCADA security and more geographically-diffuse waste handling. Upsides include no exploding containment vessels and consumption of HEU down-blended.


There's nothing to keep us from putting lots of small reactors at the same site, if that's what we want. We'd still get the economy of scale from factory-built reactors.


We need nuclear. Not only is a reliable carbon-free source of nuclear energy, even if wind and solar continue to develop where they provide all the power for day to day use, we can use the excess energy generated to drive CO2 removal from the atmosphere. It is seeming less likely that we able to change course in time such that just stopping carbon emissions prevents disaster. We will likely also need to actively remove carbon from the atmosphere. All such processes are energy intensive, and having excess nuclear power will be helpful.


It's our best chance going forward (short of Fusion) but how do you go about calming people who are quite rightfully scared by it? I live in Japan which has a long history with nuclear but after Fukushima people take to the streets whenever a nuclear plant comes (or is announced to come) back online. If a country like Japan can still screw nuclear up so monumentally I don't hold out much hope for the rest of the world.


They are not "rightfully scared." Even if you take the highest number of death estimates -- all the way from the initial incident through the people who died in evacuation all the way those who will die of cancer in 20 years -- the death toll is staggeringly low compared to carbon emissions. It just so happens that radiation is scary and deaths happen all at once.


Don't just count deaths: Nuclear fallout has profound long-term negative effects that coal accidents do not. At Fukushima, nearly 80.000 people are still displaced, almost eight years later. Source: http://www.reconstruction.go.jp/english/


That's true, but OTOH those people are displaced due to fear of radiation (and due to large areas being trashed by the tsunami), not due to the actual effects of radiation.

And FFS, something like 15000 people were killed by the Tsunami, while 0 have died due to radiation released by the plant. Yet we continue to obsess about the nuclear accident. WTF?


Right, with coal it's not the accidents, it's just their normal operation. About 13,000 Americans die every year from coal emissions. A lot more than that die worldwide.

https://www.catf.us/resource/the-toll-from-coal/


"Fusion" and "best chance" don't belong in the same room, never mind the same sentence. "No chance" or "fat chance", on the other hand...


This is where leadership is required. Sadly a quality in short supply these days, when most politicians are worried more about not being offensive then actually doing any good.


China now has four AP1000 plants in operation. The US still has two under construction. We'll have to see how the AP1000 works out.


As far as I can tell every top level comment in this thread claiming that Nuclear is more expensive than wind/solar is simply wrong.

-- First of all, complaints about the capex of nuclear make little sense if you are trying to argue for wind/solar; as the only reason you could plausibly make a case that solar/wind enjoy a comparable marginal levelized cost as fossil fuels is assuming that the cost of capital is insanely low thus making the large upfront capital expenditures on solar/wind infrastructure look like a better tradeoff than the cheaper upfront cost of fossil fuel and higher ongoing maintenance/fuel to run a fossil fuel plant.

If you are making that assumption for wind/solar it stands to reason you can make the same assumptions for nuclear as well. Of course, in the "real world" only super rich countries can afford to make those tradeoffs in the first place and given that nuclear usually requires access to cheap capital (i.e. government loans) it stands to reason wind/solar have the same problems; and to use an example I will return to later Germany has considerable subsidies for wind/solar compared to fossil fuels as expected.

-- Secondly, no utility uses the marginal cost of producing electricity for a power source that is inherently variable. The economic costs of failing to produce enough power for a regional or national utility outweigh any hypothetical marginal advantage for wind/solar. Essentially every wind/solar power source has to have a reliable backup source that can step in to ensure the utility can meet the demand for electricity. In most cases this just means you build cheap combined cycle gas plants. If you are a rich country like germany you can afford to overbuild your power generation capacity.

Countries like germany also have the advantage of being able to purchase energy from neighboring countries to make up shortfalls in demand. The german energy program relies on the availability of french nuclear power for precisely this reason. Of course, some days you can trumpet that some gigantic % of total production is supplied by wind/solar but this is mostly irrelevant from the financial perspective of the utility.

-- Finally, energy storage for wind/solar simply doesn't make sense with current technology. To take one example, Hydropower can already be expensive; now add in the cost of producing enough renewable energy to power all the original power produced by hydro cycle AND the cost of machinery to somehow replicate the operation of gravity. Basically you are looking at 2.5 to 3x the cost of your original investment to make your variable energy source reliable. Current lithium battery technology is totally unworkable at scale as another alternative.

To top it off I think the levelized cost figures often trumpted on this site for wind/solar are off anyway. Basically, unless you get a bunch of subsidies from the government it almost never makes sense unfortunately.


> First of all, complaints about the capex of nuclear make little sense if you are trying to argue for wind/solar;

I'm having a hard time making sense of this claim. You compare the costs of wind/solar to nuclear basically by capex only, why does the cost of capital even matters? Why what relation do fossil fuels beating them on under developed countries change how one compares to the other?


His claim is absurd. The easily checked facts on the ground show new nuclear being several times more expensive than renewables in the US, on a levelized basis.

The president of Exelon, a US corporation that operates 23 nuclear powerplants, has stated that effective CO2 taxes would have to be $300/ton or higher for new nuclear powerplants to compete with natural gas in the US. And yet, wind and solar are increasingly competing with and winning against gas here.


I'm going to attack nuclear from an entirely new point of view: What if thorium or uranium have other, yet unknown uses? We could be burning potentially invaluable resources for relatively unimportant amounts of energy.


This argument could be applied to literally every depletable resource in existence. If utilized efficiently with reprocessing and breeder reactors the world's proven Thorium + Uranium reserves would last thousands of years, even powering 80-90% of world energy use. By that point we should be using either fusion energy or beamed power from panels in solar orbit, depending on how optimistic you are - and if uranium does turn out to have a miraculous alternative use we'll have a sufficiently developed space program to mine it from somewhere other than Earth.

I disagree with the notion that we'd be producing 'relatively unimportant amounts of energy' when we're talking about the only geography-independent near-zero-carbon energy source with the proven capability to provide base load power for an advanced economy, unless you think anthropogenic climate disruption is no big deal.


There are no other resources depletable in the same sense - even if you burn coal or oil, equivalent resources can be synthesized. The rest is usualy not depleted in the literal sense, only put into use and may need to be recycled to be used elsewhere. But the elements used in nuclear fuel are completely destroyed in the process and there may be no easy way to recreate them.


Uranium and Thorium could be produced by nuclear transmutation in the event we discover some miraculous property they have ten thousand years from now. This is non-trivial, but it's essentially the process used to produce plutonium for nuclear weapons and we managed to get pretty good at that.


I suppose recreating those would be much harder, because of the gap between them and the other elements.


I agree that nuclear power is safe and that reactors in good condition should not be retired before fossil generators on the same grid. Subsidies are justified to prevent premature nuclear retirements. But anyone who thinks that all new nuclear projects need is the tax subsidies currently going to renewables is willfully blind to recent history.

“Subsidies to wind and solar are backfiring, both economically and environmentally,” says Michael Shellenberger, head of the pro-nuclear advocacy group Environmental Progress. “That money should be shifted to nuclear.”

The most generous renewable subsidy offered across the USA [1] is the wind Production Tax Credit, which offers wind projects $23 in tax credits per megawatt-hour that they generate over the first 10 years of operation [2]. There is only 1 American reactor new enough that it would qualify if the same terms were offered to nuclear projects: Watts Bar II, which was completed in 2016 after a 30 year construction hiatus.

There are 4 partially completed AP1000 reactors that would also now qualify for wind-like tax incentives if they had met their planned construction schedules. Unfortunately, all 4 are horrendously behind schedule and over budget. Two have been canceled. These project failures drove the bankruptcy of Westinghouse in 2017.

New nuclear power has had its own federal production tax credit since 2005, though on terms less generous than those offered to wind:

https://www.nei.org/advocacy/build-new-reactors/nuclear-prod...

I would guess that it's less generous because in 2005 nuclear power was considered to be relatively mature, affordable technology already; wind was riskier and needed more incentives to develop to maturity. 14 years later, wind projects look like the predictable option for investors and nuclear projects are the crazy gamble. So I wouldn't be opposed to offering reactor builders the same incentives offered to builders of wind and solar farms, but I don't think that will move the needle much. The problem isn't that new reactors need more revenue per MWh. It's that they have trouble delivering any MWh. You can offer $50 per MWh and it doesn't help if the project can't deliver a single megawatt and is years delayed, billions over budget.

Why are all the AP1000s late and over budget? Why did schedules slip so badly? They were originally supposed to be built within 5 years in the US. In fact, they are taking 8 years or more. Even the AP1000s built in China ended up taking nearly 9 years. It wasn't obstructionist tactics from anti-nuclear activists or lawyers. It wasn't shifting regulatory requirements. It was the nuclear industry itself.

Here are the Post and Courier stories tagged “Westinghouse.” These are all about the failed VC Summer AP1000 project in South Carolina.

https://www.postandcourier.com/search/?k=%22westinghouse%22#...

Some damning headlines:

S.C. utilities knew of big problems 6 months into nuclear project but didn’t tell customers

Insight that would’ve alerted problems with nuclear project scrubbed from audit two years ago

Whistleblower says he was demoted after finding problems in S.C. nuclear project, report says

Confidential Westinghouse report details early faults with nuclear projects

Letter shows S.C. utilities knew Westinghouse’s reactor designs would lead to increased costs and schedule delays

The American nuclear industry is going to take a decade to recover from these self-inflicted wounds, if it ever does. Maybe (I hope) small modular reactors can avoid the runaway schedule and cost problems that plagued the AP1000 and the EPR.

I've been reading calls to prioritize building reactors over renewable projects for at least 15 years now. They even made a lot of sense back then. The pitch: if we can just get the irrationally fearful public out of the way, nuclear projects can displace fossils faster and cheaper than renewables. These American AP1000s were supposed to convince fence-sitting environmentalists and ratepayers alike that it was time for an American nuclear renaissance. Reactors would be as all-seasons reliable as coal and generate a MWh cheaper than any solar farm. Instead they've been complete fiscal disasters that to date have generated less electricity than a single rooftop solar system.

People who are still pitching the "cheaper and more predictable than renewables" line about nuclear projects should be dismissed as chat-bots stuck in a loop. They're not integrating new information and reconsidering past stances.

[1] I don't know details of all 50 states' renewable incentive policies, and some of them are structured in ways that make it difficult to assign a straightforward dollars-per-MWh value.

[2] https://www.energy.gov/savings/renewable-electricity-product...


The fact that yours is the only post here to mention South Carolina makes me really wonder how practical the typical nuclear advocate could be. $25B would buy a lot of solar panels.


My issue with trying to calm people down on the subject of nuclear is as far as I can tell they are scared of risks that don't exist. That can't be controlled with engineering solutions; so making nuclear energy safer won't do anything.

Take a recent article [0]. Now, they talk about getting "one chest X-ray every week", and apparently a chest Xray is about as radiation as 10 days of background radiation [1]. We're talking about a bit more than doubling a completely safe background rate to a figure that we have no evidence is of practically different, and that the experts claim is safe. I'm not going to look in to it, but I bet that the 'international limit' they talk about wasn't even derived scientifically, they probably just multiplied average background rate by some constant.

This is almost literally the same as taking the anti-vaccination crowd seriously. People who seem to be anti-chest-xray are being treated like credible sources of information! How can we expect sane policy in that sort of environment.

[0] https://eandt.theiet.org/content/articles/2018/03/fukushima-...

[1] https://www.radiologyinfo.org/en/info.cfm?pg=safety-xray


One thing we always keep forgetting: spent nuclear fuel, which is highly radioactive. It has to be kept safe for years and somehow I don't this is really calculated into price of nuclear electricity. But if we had a reactor that wouldn't have issue with highly radioactive waste, then it would be a different story.


FWIW, Nuclear fuel reprocessing tech has been developed for that. I know the French version. It shrinks and stabilises the waste in the form of a special form of glass (of course radioactive). The entire waste generated since 50 years fits in a hangar waiting while politicians still argue about where to bury them for all eternity.


Please remember that burning coal also leaves behind dangerous and mildly radioactive waste, that's usually not protected in any way or form - it just sits there, washed away by rain and blown away by rain. Where I'm from(Silesia region of Poland) the ash hills are permanent feature of the landscape, and a short trip with a Geiger counter shows easily that they have higher radiation level than background. It's not very dangerous of course, but it's there.

On the other hand, actual nuclear waste is protected fantastically well, and it's not that difficult to store. We always had plans for ultra-long-term storage too, no? Deep underground in salt deposits so it would be protected against earthquakes and underground rivers.


I am not against nuclear per se, only against current nuclear technology. I think that if we want to stick to nuclear, we need to heavily invest in technology, that would be more safe and would not produce long term highly radioactive nuclear waste. If those issues are solved, I am all in.


Keeping radioactive waste far away from anyone who might be harmed by it, even for thousands of years, is a solved problem. We know how to bury stuff deep underground in a geologically stable region. Or if you prefer, throw it in a subduction zone.

As with most things nuclear-related, the real problem is NIMBY. If a country as large and sparsely populated as the U.S. can't agree on a place to bury nuclear waste, there doesn't seem to be much hope for others like France and Japan.


I am not convinced that it is a solved problem. How old are Egyptian pyramids? Yet there is a lot of stuff that we don't know about them. We are confident, that after 20 000 years, people will still know about nuclear waste storage facilities? Think about it: records could be lost, our language will be different. Why are we so sure that somebody will not dig it up?


The radioactivity of spent fuel decays exponentially. After a few hundred years, most of it is gone, although there is of course the danger due to the chemical toxicity, just like any other heavy metal deposit. Obsessing over what could happen if somebody digs it up after 10000 years is not anchored in reality.


It's funny that people are so worried about stuff like Pu-239, which is stored miles underground in steel containers and will be half of what it is after 24,000 years.

Meanwhile, "According to a recent model study the total anthropogenic mercury released into the ocean is estimated to be around 80,000 to 45,000 metric tons" [1]. We have spewed so much mercury that people are advised to eat less tuna because they have too much mercury.

And mercury is not radioactive. In other words, 24,000 years later, there will be 100% of the same mercury there is now! (Granted, some of them might have been removed from the ecosystem and deposited deep underground... I mean, just like how plutonium is stored right now.)

[1] https://en.wikipedia.org/wiki/Mercury_pollution_in_the_ocean


> The radioactivity of spent fuel decays exponentially

True, but it is still question of the half-life (i.e. plutonium)

> Obsessing over what could happen if somebody digs it up after 10000 years is not anchored in reality.

It is a reasonable concern, because we have never before in our history produced such toxic waste that lives so long. I don't know about you, but I do hope that Homo sapiens will survive next 10k years.


> > The radioactivity of spent fuel decays exponentially

> True, but it is still question of the half-life (i.e. plutonium)

To expand my statement, yes, spent fuel is a mix of different isotopes, some stable, and others unstable with different half-lives. The point being, if something is super-radioactive, then by definition it has a short half-life. And conversely, if something has a long half-live, it's per definition not as radioactive has something with a short half-life.

Thus, after 10k years, what is left is not that radioactive (to be more specific, after 10k years with a once-through cycle, spent fuel will have a radio-toxicity about 20 times that of natural uranium ore. Now, I'm not suggesting you should grind it into dust and sprinkle on your cheerios in the morning. But it's not any kind of civilization-ending toxicity either.

The idea that a future civilization will have the know-how to mine stuff 1 km underground, but somehow they have no clue about chemical or radio-toxicity just seems so absurd to me.

> It is a reasonable concern, because we have never before in our history produced such toxic waste that lives so long.

Citation needed. My guess would be that we have produced an unfortunately large amount of toxic waste, but the vast vast majority of that is not radioactive, and will thus never decay. Dig it up a million years from now, and it will be as toxic as it is today. The incredible energy density of nuclear fuel means that, in the end, the amount of nuclear waste is very small.

> I don't know about you, but I do hope that Homo sapiens will survive next 10k years.

I hope so too. In my mind, however, climate change, biodiversity loss, topsoil loss etc. rate far far higher as risk factors than a comparatively small amount of nuclear waste. In fact, by not deploying nuclear due to risks associated with nuclear waste, we significantly reduce our chances of dealing effectively with climate change, a much higher risk.


The fission products are mostly gone in centuries, but the actinides persist for much longer.


Protecting our distant descendants in case we destroy our own civilization and go back to the stone age is a niche concern at best, akin to the zombie apocalypse.

Even if there's an all-out nuclear war or runaway global warming, I think it's overwhelmingly likely that a sizable population will remain to safeguard our scientific knowledge. Instead of stone age, we'll probably end up with the situation in Alita: Battle Angel. In that case, I would be more worried about rogue groups digging up dangerous stuff on purpose than about anybody 20K years in the future.

Besides, we can read Egyptian hieroglyphs pretty well, and the Egyptians weren't even trying to create a writing system to last millennia. Armed with a much better understanding of linguistics and anthropology, I'm sure we can do a better job with our warnings signs.


Would this not become more of an issue if nuclear actually was embraced on a global scale? I mean, eventually we would run out of safe storage - the question would then be whether that would be in 20, 50, 100 years (and whether some better option would exist by then).

Disclaimer - I have no idea of the numbers involved so aware this may be a ridiculously trivial issue


One of the reasons we're producing so much radioactive waste is that we're too scared to reprocess spent fuel.

There's a lot of fissionable material remaining in the "spent" fuel that can be recovered, enriched, and thrown back into a reactor. The more cycles you put it through, the less waste you'll end up with. But the same process can also be used to make bombs, so we force everyone to keep the spent fuel in a bunker or something.


The deeper reason is that we don't need to reprocess spent fuel now, and that it's cheaper to wait.


A large portion of this problem was addressed by the Integral Fast Reactor concept developed in the late 1980s.




Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: