[cletus]

There are a number of problems with nuclear power.

1. The world's uranium supply won't actually go that far [1].

2. A lot of places aren't suitable for nuclear reactors (eg they're seismically active).

3. We have to trust either corporations or governments to run such plants. Corporations will tend to maximize short term profits at the cost of long term safety. Governments will tend to do the same for budgetary reasons.

4. We'd create a whole bunch of radioactive slag that we honestly have no good way of dealing with.

So far it seems the best power source we have is hydroelectric. Of course it's only possible in some places. This can devastate certain species (eg salmon) but in terms of cost, risks, environmental impact and power output, hydro is hard to beat if it's an option.

Solar has been on a stellar (pardon the pun) rise for some years simply because the cost of cells has decreased by way more than I ever would've predicted.

Widespread electric vehicles are still hindered by the relative expense and scalability issue of batteries, notably how much lithium we have available as well as the environmental impact of mining the necessary materials. It does seem like we're one big battery breakthrough away from completely changing this landscape however.

Wind has a place but I think will remain a niche energy source.

I increasingly have the view the the economic production of electricity from nuclear fusion is a pipe dream. The temperatures are too high, magnetic containment seems too problematic and, worse, the neutron emissions are a big problem (yes, yes, I know about He3).

This does seem like it's a problem we're going to have to solve this century.

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

[acidburnNSA]

Thanks for posting. I'm in the nuclear industry and would like to try to address some of these issues if you don't mind.

1. Uranium in seawater is replenished by erosion through streams at a rate faster than we could ever burn it. It is effectively totally sustainable and even renewable on a 4+ billion year time frame, even without breeder reactors. But really, breeder reactors will become a thing when uranium gets expensive enough to warrant them. The reactors that were attempted so far just weren't needed because it turned out we had way more uranium that we originally thought.

2. With passive cooling systems, even seismically active regions can be powered by nuclear. Also, there are exciting possibilities for offshore nuclear, floating 10km out to sea where tsunami wavelengths are huge. Cooling is guaranteed and huge shipyards can do the construction. I agree that it's clear from Fukushima that designs requiring active cooling are no good in such places.

3. There are teams of nuclear engineers and other reactor designers in San Francisco right now at the American Nuclear Society meeting whose goal is to reduce the price of nuclear so that corporations can make a profit from them without compromising safety. I believe it can be done.

4. The Finns are about done with their deep geologic repository [1] and it's looking great. We definitely know how to dispose of nuclear waste. It's just a matter of political will and outreach. The tradeoff you have to keep in mind is that you can get all your primary energy for literally your entire life while personally producing 2 soda cans of nuclear waste. Compare that to fossil fuel (2 million times more waste) or to the vast energy harvesting resources for wind and solar (concrete bases, fiberglass, steel, silicon, glass, semiconductor processing, etc.) and you see that nuclear is extremely low footprint and that's its magic.

I sympathize with your concerns and am working to mitigate them. Biggest hurdle in my mind is standardizing designs. I dream of coordinating an open source reactor design effort some day.

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

[jacquesm]

> Uranium in seawater is replenished by erosion through streams at a rate faster than we could ever burn it.

Isn't there a substantial energy cost associated with collecting that Uranium?

The same goes for gold, there is a huge amount of gold in seawater and yet, nobody has been able to extract it in a way that left money on the table after the process had run.

[acidburnNSA]

Great question, and one that I don't have a quantitative reply to on hand. The thing with uranium is that it is so unimaginably energy dense that with the success we've seen so far in seawater extraction technology, it's hard to imagine the energy required for extraction being anything but minuscule in comparison. Obviously there's a limit of concentration where this becomes not true but my understanding is that we are not near that limit.

Here's a summary of some recent work on seawater uranium extraction: https://www.forbes.com/sites/jamesconca/2016/07/01/uranium-s...

And here's a special edition of Industrial and Chemical Engineering dedicated to the topic: http://pubs.acs.org/toc/iecred/55/15#UraniuminSeawater

[jacquesm]

Thank you for the links and the answer.

[mseebach]

This article suggests a total cost of $1000-$1500 which is ~20x the cost of mined uranium (and 40x less the cost of gold). It does not mention energy as a significant driver of this cost.

http://scienceandglobalsecurity.org/archive/2013/06/the_cost...

[wuschel]

Hi, interesting answers.

1) Could you tell me more about replenishment of Uranium in seawater, or point me towards your source?

Wouldn't it mean that we would reach a maximum saturation of Uranium in the ocean after some time (with precipitation of excess Uranium in the form of salts)?

Have you seen / access to estimates of Uranium electrode deposition cost from water? Are they cost competitive in any way in comparison to other energy systems that will be employed in the future?

At the end, you would need to use electrical energy converted from another source to gain Uranium. For me, that scenario is a bit 'fishy'.

[acidburnNSA]

Thanks, and sure. You're right to be skeptical. The renewability of uranium isn't a widely accepted idea and unfortunately is hard to definitively prove experimentally due to the hilariously long time scales. So we're left with less rigorous claims of this to treat rationally more than scientifically. Also, haha: 'fishy'!

Here's a good source: http://www.mdpi.com/2071-1050/4/11/3088/htm And it points to an older one: http://tmtfree.hd.free.fr/albums/files/TMTisFree/Documents/E...

From the first:

"One additional aspect of nuclear sustainability—noted long-since by Bernard Cohen—is that a significant fraction of the nuclear fission energy resource is in fact completely “renewable” in the same sense as wind and solar energy [32]. Wind and rain constantly erode the Earth’s crust, which contains an average uranium concentration of 3 parts per million. Rivers then carry this dissolved uranium into the oceans, at a rate of approximately 10,000 MT per year [33]. In a breeder reactor energy system, this is a sufficient rate to supply the world’s entire electricity demand at the present time more than five times over—or is roughly one quarter of what’s needed to supply a continual 100 TW to a hypothetical global civilization of 10 billion persons which is energy supply-replete by any contemporary measure.

As the crust is being eroded by rivers, it is constantly replaced by new layers of rock being pushed upward by plate tectonic processes. The supply of uranium in the Earth’s crust is effectively inexhaustible, on the order of 40 trillion metric tonnes, a factor of 10,000 more than is present in the oceans. At present erosion rates, this source of uranium would last on the order of 4 billion years, similar to the timespan over which the Sun will become a red giant.

Therefore, this assured source of “continually mined-by-Nature and oceanically presented” uranium will last as long as life on Earth does—even if burned at rates sufficient to supply a large fraction of a fully-developed human civilization—and represents an astronomical amount of nuclear energy, one that is in fact truly renewable and inexhaustible by any human measures."

EDIT: As for increasing concentrations of uranium in the sea, it's already in equilibrium so the incoming stuff that leaches in through rivers is leaching back into the rocks, giving us the equilibrium concentration we observe today. The point here is that if we start extracting it that concentration is not expected to drop because we can't offset the balance very much.

[wuschel]

Hi,

thank you very much for your detailed elaboration on the subject, and the references. I was not aware of the bigger, geologic perspective:

   ~ 3 mg Uranium in 1 m^3 ocean water. 

Makes sense, when one thinks about the relative ratio of chemical elements on earth. I guess, as always, it is the demonstration of chemical enrichment procedures that is the key here, and there is currently no economic incentive to do such a thing. Although, I have not looked into current fuel production procedures. How expensive is 1g (sufficiently) pure Uranium fuel?

The sentence that our energy requirements could be covered with breeder reactor systems for 50K years made my smile, though. It will be probably the lack of higher "level fuels" such as proteins and carbohydrates that give us headaches, first.

Please note that I am not at all hypercritical regarding nuclear energy. There is a lot of industrial and green propaganda out there, and it is sad to see that we scientist are loosing the battle in media. Too often, rigorous scientific explanations in not sexy enough for us apes.

[seb1204]

I also had my fair share with nuclear power plant design and power plant design in general in the past. From that time I have kept a rather moderate/positive view of the technology. However 'we' seem to be implementing it so badly that I am concerned. The total (from end to end) are unknown and even based on what is known not fully accounted for.

While you point out that there are technical solution to the challenges I strongly believe Countries/Corporations/Governments are not willing or capable of operating these plants to a safe standard and earn money doing so.

The whole issue of decommissioning and deconstruction is only in its infancy and already a huge money drain on the backs of the people and not the Corporates who earned money for 30 years running the plant. I think we should speak of deconstruction and not of demolition. Demolition is mainly smashing buildings to pebbles with a big steel ball, for nuclear power plants that is not the case. See [1] or [2] - sorry both in German

Re 1. Until the technology has actually been built and scaled up we won't know what technical details and challenged await us. I don't see that as a reason to not do it but rather a call for caution on over-optimistic timeline.

Re 2. Indeed many aspects of the reactors have been improved on paper/in theories. See for example the European Pressurized Reactor [3]. The first 3 plants to that nature are riddled with technical detail problems and budget overruns, see [4]. When I was at Unit learning about EPR most lecturers where approaching retirement or already past retirement. The nuclear power industry - at least for the German part - has had the issue of very few skilled successors. This might have a big role in the building of the actual new types of plants. 35-45 years ago when the current old plants where built there was a large pool of experts - engineers, construction, QA/QC - they are all retired now.

Re 3. That is good, and there surely are technical solutions but I am very sure they are no easy find. Just think about the discussion of material selection for certain vessels in light of the current material embitterment news from Belgium Power plants [5]. I'm not saying that these issues are insurmountable but that the process to get to an agreed (experts and governments) new solution is slow and painful with no ETA.

Re 4. In Germany at least the search for a repository was driven by politics into the wrong direction. This should have been driven buy science and technology people. By now the topic is so far down the drain so that changing course is political suicide. Another point regarding the waste is the discussion of marking the nuclear wasted depositories for >1000 or rather 10000 years. Interesting YC discussion here [6]

[1] http://www.focus.de/finanzen/eon-spaltet-akws-ab-muessen-jet... [2] https://www.handelsblatt.com/unternehmen/industrie/milliarde... [3] https://en.wikipedia.org/wiki/EPR_(nuclear_reactor) [4] https://en.wikipedia.org/wiki/Olkiluoto_Nuclear_Power_Plant [5] https://en.wikipedia.org/wiki/Nuclear_energy_in_Belgium#Hydr... [6] https://news.ycombinator.com/item?id=14530978

[jjoonathan]

> in terms of cost, risks, environmental impact and power output, hydro is hard to beat if it's an option.

The death tolls tell a different story.

You aren't alone in this belief, but I've never really understood it. Banqiao dam bursts and kills 170,000 people: hydro is fine, we should do more. Chernobyl melts down and kills 30: nuclear power is inherently unsafe!

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

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

[drvd]

If a dam burst it kill a lot of people in short time. And 36 hours later rebuilding the infrastructure and towns begins.

If a reactor bursts it might not kill lots of people but it will devastate a large area for a very long time. We still do not know internal details about the Fukushima reactors because not even robots can go there.

A dam burst is a nightmare but it is a nightmare which is easy to clean up afterwards. A reactor catastrophe is a nightmare nobody knows how to clean up at all.

[mseebach]

Not quite. The majority of deaths from the Banqiao Dam burst was from subsequent epidemics and famine. The 11 million displaced also weren't exactly spending three nights in a hotel, then going home.

The construction of the Three Gorges (2012) dam flooded 13 cities, 140 towns and 1350 villages, as well as 1,300 archaeological sites, and caused relocation of 1.24m citizens -- this is permanent and was done entirely on purpose, planned years in advance. Fukushima (2011) has a 20km exclusion zone that is now gradually being reopened a few years later, and 100,000 persons are still displaced.

One of these are acceptable collateral damage in the battle against climate change and the other is so bad that it constitutes conclusive evidence of the fundamental futility of the very technology itself. But you need to drink a lot of koolaid to see which is which on face value.

[fusiongyro]

Water is less scary than uranium. I agree with you, but it seems to be futile to convince people because there's just more movies about the harm of radiation than deaths by flooding. They keep rebuilding New Orleans and living in coastal Florida too.

[mseebach]

There's a chicken-and-egg situation here -- people are never going to get comfortable with nuclear while politicians are leading the FUD-charge.

[x0x0]

Yet we keep being told, "Oh no, it's safe this time. Really." and then Fukushima happens. Not to mention cleanup costs that started at $50B are now estimated to surpass $0.2T. I would be surprised if they don't double a couple more times before all is said and done, and that's fractions of a trillion dollars.

[mkesper]

You're not taking into account all the people who didn't die immediately but long before their time because of cancer.

“I know three women my age (between 30–40) who have experienced thyroid cancer. When one of them was surprised to get the diagnosis, her doctor told her they see women our age from the Soviet Union very frequently with the same. Not a coincidence.” — Z. K.

[mseebach]

Even the heavily agenda-driven Greenpeace report on Chernobyl deaths gives an upper limit of 200,000 (and if we're counting non-lethal casualties, the benchmark is 11 million displaced, so let's not go there) - so the absolute insane, totally reckless and avoidable nuclear disaster was in the worst case as bad as a hydro disaster that most people haven't even heard of because everybody had just decided that it doesn't matter. I think GPs point stands.

[boomboomsubban]

>You're not taking into account all the people who didn't die immediately but long before their time because of cancer.

Neither are you, you're speculating that they must exist in high numbers. That quote doesn't show anything, that could be a coincidence and frequently is a vague term for a very busy individual.

There has been a noticeable increase in thyroid cancer incidents in the area, but treated thyroid cancer is the second least fatal cancer. About 93% of people are alive thirty years later.

The highest peer reviewed studies estimate 27,000 deaths, still far below the Banqiao dam failure.

[ckastner]

The apparently most extensive study lead by the WHO even puts this at around 4.000 deaths [1].

[1] http://www.who.int/mediacentre/news/releases/2005/pr38/en/

[boomboomsubban]

That's a writeup on the report from the Chernobyl Forum, a group created by the UN's International Atomic Energy Agency. The WHO was a member of the forum, and it won a Nobel Peace Prize, but there are some large issues with their projection methods.

That said, mine wasn't peer reviewed, and used the same method as the Forum report with some small tweaks. I didn't check my sources well enough, but I also can't find anything credible listing it far above that number.

[Shivetya]

and how many will die from starvation this year? Here's a hint, more than all those who died from nuclear accidents and dam breaks. You could total them all up and you might get to this year's starvation numbers.

The point is, there is no good outcome in a pissing match over what has the most negative impact when we routinely ignore any number of causes of death that we can be fixing but don't.

The threat to hydro is the reclamation of environment which tends to go unchallenged in many parts because of the feel good lobby.

[boomboomsubban]

>The point is, there is no good outcome in a pissing match over what has the most negative impact when we routinely ignore any number of causes of death that we can be fixing but don't.

That isn't the point, any time you mention nuclear power you get a swarm of people saying it's so unsafe and we need to move towards renewables. It's safer than the largest renewable source currently, and the UN considers some nuclear power a renewable source. The US doesn't, largely due to fossil fuel lobbying.

[slackingoff2017]

I think it should be kept in mind that the design of Soviet reactors was incredibly unsafe compared to their western counterparts. Fukushima had four meltdowns with only a few casualties.

[kodfodrasz]

This is not true in this form. There were unsafe soviet designs, but also there were unsafe western designs. Soviet designs were not all categorically unsafe. The design used at Chernobyl was a problematic design, but still many layers of human error had to be involved to create the accident.

Military reactor designs are generally less safe. Civilian designs were usually OK in the USSR.

[slackingoff2017]

Soviet era designs generally didn't use containment buildings. The RMBK (Chernobyl) in addition used graphite moderator with a positive void coefficient.

In combination this made such reactors intrinsically unsafe compared to their western counterparts, especially the positive void coefficient which can't be found in any western reactors as far as I'm aware.

Not all Soviet designs were as unsafe as RMBK but most were less safe than the average western reactor.

The meltdowns at Fukushima would have been as bad as Chernobyl if the design was similar. And the Fukushima reactors used an old western design dating from roughly the Chernobyl era. The cause of the meltdowns(loss of power) was similar.

The major difference was that Fukushima had containment buildings, no graphite in the core to burn, and a reactor designed to become less critical as the water boils (void coefficient less than 1.0).

[rb1]

It's not just the design(s), it is the implementation too.

At Chernobyl the powers at be (accountants?) got involved and decided to use flammable bitumen coverings on the roof of reactor 3, one would assume to save money. Unsurprisingly, the roof of reactor 3 caught fire.

[fusiongyro]

This is a good example of one of the failings at Chernobyl, but it took many of them coinciding to make the disaster as bad as it was.

They were required to use another material to build the roof, but they were also required to be finished with construction by a certain time. Soviet central planning led to a shortage of the proper roofing material. I'm sure the decision to use bitumen was either made in ignorance or with the assumption that it wouldn't matter for other reasons. Lots of mistakes were made with the assumption that the rest of the system was safe so it wouldn't matter.

[makomk]

If we do that, we also need to bear in mind that dams can be pretty safe if designed and operated correctly too.

[sampo]

> Fukushima had four meltdowns with only a few casualties.

The tsunami killed two workers, but the nuclear incident as such didn't kill a single person.

[dragonwriter]

> The tsunami killed two workers, but the nuclear incident as such didn't kill a single person.

The evacuation made necessary to avoid direct casualties from the nuclear incident killed 34.

[boomboomsubban]

I wouldn't say they were direct. That number is based on a comparison with a 1916 tsunami and earthquake death toll, and with around one of those thirty-four being under sixty that seems hard to calculate exactly.

[eni]

May be because we are used to dangers of water, and floods for millions of years.

On the other hand, dangers of nuclear is a very recent thing. Common people's primary recollection point of nuclear is probably nuclear bomb.

[ralfd]

A hundredseventythousand?! :-o I have never heard of that dam accident.

[slackingoff2017]

Hydro is great, so great that around 90% of dammable rivers are already used for power generation. Further hydro generation is not really possible so it will never fulfill more than a fraction of energy needs.

Wind is an excellent source for some countries. Much of the US is great for wind generation and unlike hydro the resource is largely untapped. Wind is unpredictable but averages out over large areas. It will never be a base source but has much potential remaining.

Solar is held back by cost alone. All predictions point to solar providing the majority of our electricity in the future.

Nuclear fuel is essentially unlimited. Uranium is in seawater and if prices rise high enough we can extract from that and basically never run out. On top of that thorium is also usable as fuel if we ever get around to building anything that uses it.

In some countries the best power source is nuclear. If you don't have a lot of wind, water, or sun, your options become limited to fossil fuel or nuclear.

[wbl]

Notice how reprocessing or fast breeders solve problems 1 and 4. As for 3 the US has never had an release of radioactivity to the environment from a nuclear reactor aboard a ship or submarine in several thousand years of operation time.

[sitharus]

That's not correct, there have been several releases of radioactivity from US military vessels. This is all prior to the late 1980s, since the mid 1970s many controls and monitors were added to prevent releases.

Personally I'd attribute this to lack of awareness rather than lack of budget. The Soviet program looks quite different.

December 12, 1971, The USS Dace spilled 1,900 litres of radioactive coolant water in to the Thames River, Connecticut.

Undisclosed location, radioactive resin being dumped at sea from the USS Guardfish blows back on to the vessel.

May 22, 1978, USS Puffer releases another 1,900l of radioactive water near Puget Sound, Washington, United States

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

Apparently 'millions of gallons' of primary loop coolant was released in to the environment prior to 1973. This even references some sources if you want to dig deeper http://www.environment-hawaii.org/?p=3836

[wbl]

Thanks for correcting! It seems that these are low-level intended releases, not core damage. Bad for a power plant, but not horrifically bad.

[ckastner]

> So far it seems the best power source we have is hydroelectric. Of course it's only possible in some places. This can devastate certain species (eg salmon) but in terms of cost, risks, environmental impact and power output, hydro is hard to beat if it's an option.

Note that according to one recent study, reservoirs also contribute significantly to greenhouse gas emissions. Previous HN discussion [1].

It's just one study and its quantified result should probably be debated, but it's general argument that reservoirs contribute to greenhouse gas emissions sounds well-reasoned to me.

[1] https://news.ycombinator.com/item?id=12884030

[reitzensteinm]

How can you list scarcity of uranium as an issue for nuclear, then advocate for hydro without mentioning the same?

There are only so many suitable locations for hydro, and the further down the list we go the worse the incremental environmental impact.

[jacquesm]

Tidal is hydro and has barely been tapped.

[fennecfoxen]

Tidal's problem is that it's low-density energy and has a low capital-to-extracted-power ratio. And that capital is sitting in salt water.

[reitzensteinm]

My intent was not to advocate for nuclear, just to push back at a one sided argument.

You can make a compelling case for anything if you leave out inconvenient details.

[kodfodrasz]

Hydro is not an option at many places. Also dams also have negative environmental impact, which is often simply ignored.

Fusion will work eventually, but currently it is in the constant "we'll be there in 30 years" since the 1970s. Yet we are not getting further, but closer to it. Just thnik about it: now you have technology in your pocket that was used in putting humans on the Moon. More and more durable and special materials becoming cheaper and affordable. We are getting close to handle the heat, EM and possibly the neutron flux. It is doable in the near future (<100 years), thus worth pursuing. What we must avoid (this also applies to the energy storage tech for renewables) is the wishful thinking: "we can shut down nuclear powerplants, as fusion/battery tech will be solved in ten years: extrapolation from 2 conviniently chosen datapoints..."

Breeder reactors and Thorium based fuel cycle will mitigate the problem of running out of U235. Radioactive residual reprocessing will eventually be solved. Maybe peak-uranium will make it economically desireable. Btw there is current work on reprocessing spent radioactive fuel in reactors, at least in Russia and China.

I firmly beleive that while solar and wind will stay with us, they cannot be the fundamentals of our energy mix, and nuclear is here to stay. Simply many politicians are not ready to admit it yet.

[fulafel]

Your #1 is wrong in at least 3 ways. (Seawater, Thorium, breeders).