Hacker News Reader

Comment by parksy | Hacker News Reader

By population that's one reactor of this scale per million people. To replace all other sources of electricity, 8 thousand such reactors would feed the current world's population. We'd need just shy of 500 more to keep up with population growth by 2030 - or at least 500 new reactors globally per decade.

(edit - assuming everyone uses the same amount of power globally as the average person in Finland which - why shouldn't they be able to - and - obviously they don't)

View on HN

> assuming everyone uses the same amount of power globally as the average person in Finland which - why shouldn't they be able to

Finland's climate is an outlier:

One-third of energy consumption in housing was electricity in 2018. [...] 47% of electricity was used to heat indoor areas and 36% to household appliances. The remainder of electricity was used to heat domestic water and saunas.

https://www.thenomadtoday.com/articulo/finland/energy-consum...

Here in Perth a good third or more of my electricity is keeping the house cool so some of that is going to average out. We don't have saunas but we do have hot water.

There's going to be an obvious error of margin either side of my napkin calculations but I think the order of magnitude is in the ballpark.

Cooling a house is more energy efficient than warming one

Isn't it the other way round?

It’s a question of temperature difference. Cooling a house from 40° to 20° uses approximately the same energy as heating it from 0° to 20°. But it gets much colder than 0° C in Finland, while it is very rare to have temperatures above 45° C anywhere in the world.

It's not just that. To make something colder using a heat pump, you must heat up something else by the equivalent amount. Actually more, because of losses.

Whereas making something warmer can be done without a heat pump, by releasing stored chemical energy, at nearly no losses.

Ductless heat pumps are more efficient at heating than alternatives, but there's a floor on how low of an ambient temperature they can handle.

Heating something with chemical energy by combustion (for example) has a large second law loss -- there's a large increase in entropy in combustion, and in transmitting the high grade heat to the low temperature of the building being heated. Because of these avoidable entropy sources, it's possible to be smarter, for example to use a gas-fired heat pump to achieve a COP of > 1.

Yeah, a simple absorbtion refrigerator (gas fired) is already way better than a gas oven. The plumbing is just more expensive than works often be worth it, and the nice low-temperature systems run on what's essentially an ammonia destiller.

How much does construction affect the efficiency?

Heat pumps are more energy efficient than resistors, so I guess it depends on the method.

Nope, it’s all about the temperature differential. Using a heat pump like an air conditioner is the same efficiency either direction. Usually when you’re talking about heat it’s going from (say) 20F to 70F, with AC it’s more like 90F to 70F. With heat you might even need to do like 0F to 65. AC max might be 115 to 75 which is still substantially less.

How about building design? I know in colder climates buildings are designed to retain heat. Are buildings in warmer climates designed with similar heat insulation in mind? I think traditional architectures are designed for convective cooling but that doesn't play well with air-conditioning. Not too mention the modern designs obsessed with glass which I think traps heat. Good and efficient for cold climates, not so much for the hot ones.

Depends what your energy source for heat is.

Right. But remember, not all forms of green energy generation are equal. Wind and solar need some dispatchable source as a backup. Hydro works, if you have it. But it's geographically limited. Gas is the typical backup source, but that emits carbon.

Nuclear, as it stands, is the only solution to decarbonizing the energy grid for much of the world. Until some breakthrough in energy storage transpires, or fusion becomes feasible, fission plants remain necessary for decarbonization.

Storage is a solved problem, and only needs to be built out. No such "breakthrough" is needed.

Storage cost is falling much faster, even, than solar and wind generation. The only question today is whether to put a euro into solar or wind now, or into storage that will be much cheaper next year. Thus far, in most places, the former is still favored. More carbon tax can help move the choice the other way.

As solar and wind costs continue down, synthesizing methane and kerosene from captured CO2 and synthetic hydrogen will shortly be cheaper than mining, refining, and transporting them, even without the carbon taxes. Synthesis of those, and of ammonia and hydrogen itself will absorb unlimited "overbuilt" peak generating capacity, and stocks of all of those serve also as storage. That their round-trip efficiency is less than, say, hydro or batteries matters little where they rarely need to be used for that.

And which solution is that? Lithium ion batteries? To put this in perspective, global lithium ion battery output [1] annually doesn't amount to 1 hour's worth of US electricity usage (500 GWh). Hydroelectric storage? You essentially need to repurpose an alpine lake as a dam. That's geographically dependent, and it's dubious it'd scale well as viable hydro sites become more and more remote. The remaining things like electricity-to-gas, or thermal storage only exist in prototypes, and might not even be feasible let alone viable.

In response to your edit:

> As solar and wind costs continue down, synthesizing methane and kerosene from captured CO2 and synthetic hydrogen will shortly be cheaper than mining, refining, and transporting them, even without the carbon taxes. Synthesis of those, and of ammonia and hydrogen itself will absorb unlimited "overbuilt" peak generating capacity, and stocks of all of those serve as storage.

Nobody has successfully built a grid-scale energy-to-gas plant, ever. It remains the stuff of prototypes. This approach is very much in the "hypothetical" phase. There are serious unsolved problems in energy-to-gas:

* It needs a source of carbon to convert H2 to methane. This could come from biofuels, but those aren't available in sufficient supply.

* Producing large quantities of hydrogen without CO2 emissions remains difficult. Almost all methane comes from steam reformation (CH4 + 2O2 -> CO2 + 2H2), which emits carbon dioxide. Electrolysis can't be done effectively.

Currently, power to gas storage is way more expensive than either lithium ion or hydroelectric storage and it's unclear whether it'll ever be cheaper than existing options.

1. https://www.spglobal.com/marketintelligence/en/news-insights...

Obviously not lithium batteries. Even suggesting it demonstrates you are not serious.

And, obviously hydrogen will not continue to be made from hydrocarbons. It will instead be feedstock for synthesizing hydrocarbons. Carbon would need to be extracted from air, or at least from exhaust, to get carbon credits.

What you're describing is a combination of electrolysis and the Sabatier process [1]. And as I explain before there are major obstacles to actually converting electricity to methane at scale. It needs a source of carbon, and extraction of carbon from the air or exhaust is not feasible. Furthermore, massive electrolysis of hydrogen is also not something we have been able to do. Almost all hydrogen is produced through carbon-emitting steam reformation [2].

In short, converting electricity to methane and then back to electricity again is not a presently available option and it's unclear whether it'll ever be viable at stale. Grid-scale energy storage remains an unsolved problem

1. https://en.wikipedia.org/wiki/Sabatier_reaction

2. https://en.wikipedia.org/wiki/Hydrogen_production

It is fortunate that you are not making decisions, then, for the people who are directing the use of billions of dollars invested in exactly the activities you insist do not work.

Are they? Who is building grid-scale power to gas storage?

The Brits are switching to store wind-generated hydrogen in undersea cavities where they now stockpile natural gas. A similar project is going on in Utah, and another in Texas. All three are utility-scale production systems, not pilot projects.

Other storage media being built out for full-scale production use include iron-air batteries, where (IIRC) $1.5B is going into factories, a liquified-air storage system in Chile ($0.5B), and synthetic ammonia in Norway. Nobody can track even a fraction of the utility-scale pumped-hydro projects under construction world-wide.

We will need hundreds times as much of these, and of others, in the end, which will all take decades to build out.

Coverage I'm reading about this project says it's not only a prototype, it's also a planned prototype. The prototype hasn't even been built yet: https://www.bbc.com/news/business-55763356

You said it'd be stored in caves, so perhaps it's a different project. It'd be good to link to this project.

The Utah project looks like it's still trying to secure funding, and hasn't broken ground. https://power.mhi.com/regions/amer/news/20210511.html#:~:tex....

Similarly in Texas. The plans for this project were unveiled only a couple weeks ago: https://www.prnewswire.com/news-releases/green-hydrogen-inte...

To have built them, you first have to plan, fund, and then actually build them, all of which takes actual time.

Nobody announces billion-dollar physical construction projects for untested tech.

Right, these alternative energy storage systems are untested technology. Contrary to your prior claims, they're not being built at grid scale currently.

You mean, they are not being operated at grid scale, currently.

Things still being built need to finish being built before they go into service.

Just yesterday, you said some unspecified "breakthroughs" would be needed. Now you point to construction not finished. You are clutching at straws. The more honest course would be to admit you were just wrong.

The construction hasn't even begun on these projects, and it's unclear whether they will actually be built. The Utah project is still trying to secure funding, We do need a scientific breakthrough to make storage possible. Grid scale hydrogen electrolysis could be that breakthrough, but we don't know if it will live up to expectations.

Energy storage is far from a solved problem. You insist that it's been solved, but none of your proposed solutions have actually been implemented.

Well-understood energy storage methods need to be built out. Building out takes time. We already knew that. We also know we don't need any breakthroughs before we can build out. We have methods that work now.

That is not to say there won't be breakthroughs, too, that make storage even cheaper to build out. But any storage already built will continue working as well as ever. Any hoped-for breakthroughs that don't pan out will not stall build-out. The worst that can happen is costs plummeting not quite as precipitously as had been expected.

We already see utility-scale hydro storage in use, and utility-scale iron-air battery factories under construction, and utility-scale liquified-air and ammonia-synthesis plants under construction, and utility-scale investment in hydrogen synthesis and storage. Each additional storage technology that begins to come online only improves the picture.

Repeating, again and again, that things that still need to be built out have not been built out yet sheds no light.

The Swedish government made a viability study on green hydrogen a years ago, which is public as all other government papers are. The cost estimate was around 100 times that of burning natural gas, or about 10 times higher than nuclear. It has potential if it can fall in prices, but so far it has not. The study also referenced German studies with similar findings.

It is so economical nonviable that there isn't even small scale experiments to get the ball running as a storage medium. There is however some bright spots for green hydrogen when hydrogen itself can be used. Green hydrogen is only a few times more expensive than using fossil fuel in order to create hydrogen, and in that situation green hydrogen has found a place. It also reduces those industries CO2 emissions which can then be turned into profits in terms of trading existing emissions rights. As fossil fuels prices goes up, the economical viability of green hydrogen in hydrogen using industries goes up, but the price compared to nuclear remain the same.

Why isn’t pumped-water energy storage more commonplace?

Because it's not much (if even that) cheaper than batteries in practice, it's very limited by geography (best sites for this are often already used for it), and there are significant environmental concerns (mountains are often protected areas, because humans have not managed to destroy them as much as other places).

I'll consider storage a solved problem when we actually have it.

In the mean time, nuclear is a nice thing to invest in.

Each dollar put into building out nuclear instead of renewables + storage brings climate disaster closer. Money is fungible. You cannot spend the same dollar on both. A dollar spent on nukes yields enormously less than the same dollar spent on renewables, and the difference is only growing.

Even nuclear needs fossil-fuel backup, as building enough reactors to cope with peak demand would be way, way too expensive.

Even a massively-nuclearized France produces 7% to 12% of its gridpower thanks to fossil-fuel plants.

French official historical data: https://www.statistiques.developpement-durable.gouv.fr/editi...

Nuclear plants can modulate power by more aggressively cooling the reactor. They don't do this because it's effectively wasting nuclear energy, but it can be done.

It can be done, but amounts to decreased utilization, and even greater relative incremental cost, because actual operating cost is the same either way. With nukes already the highest-cost choice, intermittent demand makes them even more impractical.

There are safety-related limits (power modulation proportion, duration of a pause needed after each modulation, modulations frequency...), and the very combustible status is a major parameter.

Pertinent document (French ahead!): https://new.sfen.org/rgn/expertise-nucleaire-francaise-suivi...

« un réacteur peut varier de 100 % à 20 % de puissance en une demi-heure, et remonter aussi vite après un palier d’au moins deux heures, et ce deux fois par jour »

Proposed translation: "a reactor power output can vary from 100% to 20% in 30 minutes, then after 2 hours can go back to 100% at the same speed, and can cycle this way 2 times per day".

This is quite a good performance when it comes to load-following (French engineers are very good at this), however it is insufficient in the real world (save any ridiculously expensive over-provision of nuclear reactor, most idling) and very weak compared to gas turbines performances.

1.6GW is a pretty standard install in nuclear. For scale the largest nuclear install Kori is the current largest in operation at 7.4GWe installed capacity[2]. It achieved 74+% capacity last year which is an important point missed when looking at energy _delivered_ to the grid. By contrast US plants achieve much higher capacity factors, with Diablo Canyon in California producing 2.2 GWe at a _lifetime_ operation capacity factor of 90+%[1]. Scaling up nuclear, at high reliability is not such a stretch as some other energy scaling problems.

[1]https://en.wikipedia.org/wiki/Diablo_Canyon_Power_Plant [2]https://en.wikipedia.org/wiki/Kori_Nuclear_Power_Plant

FYI; looks like your early comments attracted too many downvotes and your account was automatically shadowbanned.

1.6 GWe is huge for a single nuclear reactor. EPRs are on the high end. Most nuclear reactors are closer to 1 GWe. Nuclear plants with multiple reactors on a site certainly do often go well above 1.6.

> EPRs are on the high end.

This is an understatement. The three largest nuclear power reactors in the world (Taishan 1, Taishan 2 and Olkiluoto 3) are all EPRs.

Taishan 1 has been shut down last summer due to what seems to be a design fault and may not be restarted.

It is feared that all other current EPR suffer the same design mistake, but there are currently no plan to stop them.

China is not going to build any more EPRs. They are going with their home grown LWR design.

From I understand, China's foray in EPR was simply a move to leech the technology from France to improve their own ability to build nuclear plants.

A few people were vocal about this before it happened, and things did not go well for them. The Maureen Kearney case is simply unbelievable, this woman was working for Areva and blowing the whistle hard on this transfer of technology, after receiving threat she was found tied up in her bedroom, a letter A engraved in her abdomen, with a kitchen knife handle inserted in her genitals.

The police decided she probably did that to herself and jailed her, while the state attorney prosecuted her. And she was sentenced to prison time and a hefty fine. it took a while before the sentence was overturned and for the truth to come out that she had not faked anything.

In the end, what she was warning actually happened, France is no better and even worse in its ability to build nuclear plants with no actual plant put in service (ITER is a mess, Flamanville EPR is plagued), while China started both Taishan EPR reactors and is now able to sell its ability to build plants to other countries.

How much more nuclear wast does that imply per decade?

How many more meltdowns per decade?

And heat-polluted waste waters? People keep pretending there are no externalities.

If the biggest externality is heated water in a subarctic climate, or nuclear waste that gets stored on-site indefinitely in casks and produces no notable leaks or accidents (like has been done in the US for a long time now), the externalities are way less than literally any other form of energy production. That includes Dams, solar power, geothermal, wind, you name it.

What prevents heated waste water from being used for…heat? Surely Finland uses central heating plants that pump hot water throughout the city (or maybe that’s just a Chinese/Russian thing?). What makes waste heat water less suited to heating, higher entropy?

I think it just not hot enough after electricity production. So temperature is not high enough after it passes through turbines. Specially due to distances involved.

Combined Heat and Power in Finland is very common, much more so than in rest of Europe. 80% of the fossil fuel power plants output both electricity and heated water. Compared to (quick googling) 8% in the US and 11% in EU.

Fear of nuclear power. People don't want water heated by it in their homes. So it is not done. It's not a purely rational thing.

The plants are also located away from cities, so utilizing the waste heat for district heating is uneconomical.

No, people aren’t pretending that. Heat pollution in water is a tiny local issue compared to global warming.

It’s like complaining that a wool blanket is itchy so it might be better to catch hypothermia and die.

> And heat-polluted waste waters?

That's hardly unique to nuclear plants; in particular, coal plants typically have lower thermal efficiency.

Less radioactive waste than coal power plants, and significant less devastation compared to continuing funding wars by buying gas, oil and coal.

Energy generation is always a trade off. Right now the world is reacting to the fossil fuel funded wars created by one such trade off. We are also in the middle of causing irreversible climate change, which would cause more damage than any amount of meltdowns or nuclear waste could ever get near.

Naturally there are alternatives. If money were no objection then green hydrogen looks pretty nice, and one could always extract heat from the core of the earth as long the technology was safe enough to do so. As soon we have a technology that get proven to be cheaper, safer and more scalable than nuclear we should all switch to that. Buying natural gas from Russia is for multiple obvious reason not that.

> Less radioactive waste than coal power plants

Less radioactive EMISSION during NORMAL OPERATION than coal plants (and I think that ignores radioactivity released in uranium mining). The amount of radioactivity in the spent fuel rods of a nuclear plant is vastly higher than that liberated by a coal plant.

It is estimated that around 1/5 of people living around a coal ash lake has gotten cancer. Thankfully there are no such number for people living near nuclear plants, or around sites of nuclear waste disposal. The amount of people who has died to radiation thanks to coal vastly outnumber the amount of people who has died to nuclear waste. If we including mining, coal mining is a symbol of one of the most dangerous job a person could do, and it has harvested many more souls than uranium mining.

Coal ash is nasty stuff, but that's very likely due to chemical toxicity (things like arsenic, especially in Appalachian coal), not radioactivity.

BTW, your chance of dying from cancer in your lifetime is about 20%, so I'm not sure that the 1/5 figure you gave there means anything.

I don't think they were saying that the radioactivity is causing the cancer in those 1/5 people. I read it as two separate points: coal plants simultaneously emit more radiation than nuclear plants, and coal plants cause more cancer via the other chemicals and rare earth elements they emit.

Yes. one of those other chemicals is arsenic which has a tendency to end up in peoples drinking water.

There is a place in the US that has a rather peculiar name of Cancer Alley. It is not a nuclear testing area, nuclear waste deposit area or area for nuclear plants. It is an area know for its petrochemical plants. It illustrate quite well the difference of nuclear waste that people are scared of, and fossil fuel waste that people accept as just normal part of life.

Spent fuel rods are a much safer destination for radioactive material, much better than particulate matter released into the atmosphere.

Everything has downsides, and it has to be compared to the alternatives.

In Finland's case, the realistic alternatives are burning coal or burning Russian gas. (If the Finns dedicated a substantial chunk of their forests to this one generator, they could maybe use biomass.)

Coal kills two orders of magnitude more people per GWh than nuclear--and it does that when operating nominally, not when malfunctioning--and it produces three or four orders of magnitude more waste and more environmental harm from mining.

Russian gas has geopolitical/national security problems.

Biomass is a roundabout way of burning diesel fuel and gas, while degrading and eroding forest soils and polluting watersheds.

The number of new meltdowns per decade rounds to zero, to five significant figures.

> The number of new meltdowns per decade rounds to zero, to five significant figures.

That's a weird metric (one meltdown is quite a catastrophe) and the calculation seems suspicious too. Between Chernobile and Fukushima I don't see how this could be correct.

I do find your other points more convincing, though with some "citation needed" wrt. coal.

I can't speak to the meltdown statistic, but for coal: just the burning alone is responsible for hundreds of thousands of premature deaths annually[1].

And that's before we consider the environmental and health risks of ash ponds[2], which can (and have caused) heavy metal pollution in nearby groundwater supply. The largest industrial spill in US history happened barely a decade ago, and was an ash pond[3].

Edit: I can personally recommend "The Buffalo Creek Disaster" (ISBN 9780394723433) as a writeup by a lawyer involved in a similar coal ash accident (one that directly killed over 100 people).

[1]: https://www.pnas.org/doi/10.1073/pnas.2017936118

[2]: https://en.wikipedia.org/wiki/Ash_pond

[3]: https://en.wikipedia.org/wiki/Kingston_Fossil_Plant_coal_fly...

WHO has particulate air pollution from fossil and biofuel killing just about 8 million people per year (a Chernobyl of people (short + long term) every 7 hours). Add in that they also cause climate change and non-combustion sources like nuclear, wind, solar, hydro all look pretty darn awesome.

https://www.who.int/health-topics/air-pollution

Alternatives like coal give off substantially more radioactivity into the environment than a nuclear plant does under normal operations.

That's only if you ignore the radioactivity in uranium mine tailings.

That was already in the environment to begin with.

It was in the environment already in the same sense the uranium in coal was also in the environment already.

Maybe the problem is the uranium mine tailings are safely off in some poor country, not in the US where the coal ash would be?

They're in the US: https://www.energy.gov/em/moab-uranium-mill-tailings-remedia...

Most (99% as of 2018) uranium production is outside the US.

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

Very recently: U.S. utilities push White House not to sanction Russian uranium, see https://www.reuters.com/business/energy/exclusive-us-utiliti...

Moreover the more we obtain uranium (prospecting, mining, milling...), the more we add to the associated carbon footprint. Therefore a sustained growth of installed nuclear capacity will lead us to exploit mines at always lowering ore grades => more emissions.

Scientific studies are clear: M. Lenzen ("between 10 and 130 g CO2-e/kWhel, with an average of 65 g") and E. Warner et G. Heath ("9 to 110 g CO‐eq/kWh by 2050")...

https://www.researchgate.net/publication/222817608_Life_cycl...

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2051332

The uranium ore was deep underground, not inhaled by nearby residents as wind blew it off the tailings pile.

"How much more nuclear wast does that imply per decade?"

The popular YouTube channel The B1M has an interesting video on how Finland is tackling nuclear waste:

Finland might have solved nuclear power's biggest problem (2021): https://www.youtube.com/watch?v=kYpiK3W-g_0

Nuclear power's "biggest problem" has been solved for decades. You can put everything in giant metal canisters and sink it into bedrock. It's vastly more expensive than doing nothing, but now that politicians are going to be held accountable for their fuckups they'll allocate the money.

And even after all that it's still far less expensive than remediating coal output.

Each and every time someone claims that we can perfectly predict bedrock future for hundreds of thousands years to come I have to remind us that plate tectonics only was admitted by geophysicists during the 1960's.

Hoping to know enough and for sure about all this is... a hope.

Stating that we know what our descendants will need/do in a so distant future is even more funny.

That's why you put it in an inches thick metal case and drop it down low. What, you think they just didn't consider that?

Such a major parameter wasn't considered 60 years ago... others ignored pertinent facts probably exists. My point was about what we (for the time being) ignore, albeit it is pertinent.

They "considered" such facts using a somewhat light approach: 'Bob Loux, the executive director of the Nevada Agency for Nuclear Projects, expressed amazement that the US Department of Energy had only just carried out the "11th hour" drilling tests.

  "It certainly looks like DoE has encountered a surprise out there, and it certainly speaks to the fact they haven't done the technical work they should have done years ago," he told the paper.

  "It's going to have to cause some change of the design in the final analysis. It's going to impact the safety case."'

Source: https://www.theguardian.com/world/2007/sep/25/usa

Oops...

They also "considered" those metal containers as adequate during the 1990's, then... (what, you think "they" are omniscient?)... problems related to brines and high temperature arose...

https://www.nwtrb.gov/docs/default-source/board/mrs_duquette...

https://www.nrc.gov/docs/ML0335/ML033500420.pdf

Oops...

No remotely modern reactor has melted down.

Just as you wouldn't factor Amelia Earhart's plane into 2022 air safety prognoses, you shouldn't use Chernobyl reactors for nuclear safety.

That's a great analogy.

I love this reply. We can go even further. If we would approach air safety in such way as Nuclear, then de Havilland DH.106 Comet would be first and last commercial jet.

Nuclear fision reactors safety technology have moved further. There are challenges, but we havent even tried to solve them fully (as we were busy improving gas burning efficiency)

What are the biggest safety improvements that have become mainstream since Chernobyl?

Control rods aren't tipped with iron. Iron has a positive coefficient of reactivity.

The most obvious is that rods naturally fall into water now, which dampens the neutron exchange to a stop, which works automatically in case of power cut (The rest of improvements are about preventing water from overheating and forming H2, and preventing H2 from overpressuring the chambers or exploding, which happened in both Chernobyl and Fukushima).

The improvements were already mainstream before Chernobyl was built. It's all about making reactors tolerant to operator error, and if something does go wrong, make sure it goes wrong slowly, so operators have lots of time to react.

The RBMK reactor is quite an elegant design. Simple plug-and-play architecture for adding and removing fuel and other assemblies while the reactor was running, perfect for things like doping silicon for semiconductors and producing plutonium for weapons. But to get this capability you have to either give up affordability or safety, and they chose to give up safety.

> No remotely modern reactor has melted down

TMI and Fukushima were both "modern" when built. All reactors that melted down did because operators ignored construction, maintenance, or operating safety standards. Other plants not yet melted down show evidence of more construction standard failings: pumps installed despite failing to meet specifications, bolts of substituted, inadequate steel. Diablo Canyon is built directly on a fault line. Now we learn that new EPR plants are inherently flawed, by design.

Ignoring standards is, by the evidence, itself standard procedure for building and operating civil nuke plants. Our global society, as it is conducted, is by the evidence unable to produce and operate a safe civil nuke.

Nuclear waste takes almost no space at all. meltdowns are extremely rare.

Reply on news.ycombinator.com