Hacker News new | past | comments | ask | show | jobs | submit login
First contact made with melted nuclear fuel at Fukushima plant (asahi.com)
371 points by howard941 5 months ago | hide | past | web | favorite | 208 comments

While I realize it was likely impractical to meltdown reactors just to develop the technology to clean them up, I find the process here pretty interesting. A lot of good engineering has gone into the effort, and the engineering challenges (and risks) of post runaway reactors has gotten a lot more data.

The Chernobyl folks just entombed everything, which is a one approach, even though they can "walk up to"[1] the remains of the nuclear fuel which they call the "elephant foot."

I also find it interesting that the "China Syndrome"[2] speculation is pretty much completely debunked. Molten/melted fuel diffuses into the material it is melting into until it loses criticality and then freezes where ever it is. All modern reactors have a borosilicate sand pile underneath the reactor. The boron acts as a neutron absorber (poison) which stops the chain reaction once the fuel has been glassified by the silicate.

[1] It is too radioactive for that to be literally true, but it is on a floor with direct access rather than being underwater and only accessible through cracks as the Fukishima fuel is.

[2] Speculation that a blob of melted core material would just keep melting into the ground until it "got to china" (which in fact would be interrupted by getting to the earths core).

[3] EDIT: Corrected for the effects of Boron on the reaction acting as a neutron poison (absorbs neutrons) rather than a moderator (which just bounces them around)

> the "China Syndrome" speculation is pretty much completely debunked

You would be dissapointed. Cou-cou! surprise! The ocean strikes back!.

Unexpected source of Fukushima-derived radiocesium to the coastal ocean of Japan. 2017. Sanial, Buessler, Charette, and Nagao. PNAS. DOI:10.1073/pnas.1708659114

"Sanial, Buessler, Charette and Nagao, scientists from Massachusetts and Japan found a new source of Caesium-137 to the environment after the Fukushima accident. From 2013 – 2016 they visited 8 different beaches within 100 km of the accident site, took sand and water samples and measured its radioactivity.

They found that at the bottom of their sand samples, up to 2 m deep into the beach, the 137Cs concentrations were incredibly high, up to 480,000 Bq per square meter in their deepest samples. The soil in the restricted zone of the Fukushima site, up to 100 km away, measured ~100,000 Bq per square meter.

This means even tens of kilometers away, the radiation can actually be higher than near the disaster. (Because, huh?, 137Cs sticks to the sand grains).


That is a great paper[1], thanks for the reference I've added it to my nuclear effects library.

It doesn't involve the China Syndrome speculation though, what it does find is much much more interesting.

The paper reports that beach sand is capable of 99% adsorption of Cs137. And then goes on to discuss how it gets de-adsorbed (up to 11%) when Cs137 free sea water is mixed with it. What this says to me is that you can use really cheap beach sand to filter the discharge water from a nuclear reactor and collect 99% of the Cs137 in the water which will stick to the sand. The sand in your filter can be replaced and the cesium rich sand can be sequestered for the Cs137 to decay over time.

This is an excellent finding that suggests reactors have a reservoir of sand for pumping cooling water through especially during reactor compromise. Knowing this a future even would release significantly less Cs into the ocean, collecting it instead in the sand filter.

[1] The experiments showed that beach sands have an adequate ion exchange capacity for this level of 137Cs, with an average adsorption fraction of 99% regardless of the salinity, sediment grain size, or mineralogy (Table S3). ---https://www.pnas.org/content/early/2017/09/26/1708659114

Decontamination factor of 100 is rather terrible compared to DF of over 8 million of actual filter materials, which are being used at Fukushima: https://www.fortum.com/sites/g/files/rkxjap146/files/documen...

See also: https://www.fortum.com/products-and-services/power-plant-ser...

Adding another factor of 100 with something as cheap as sand still seems worthwhile.

There are reasons, why not to do that:

1) DF doesn't add up like that at low concentrations. Then it comes down to selectivity of the material. Sodium and cesium have similar chemical characteristics, but CsTreat is very efficient in selecting Cs over Na.

2) Material you are using as filter becomes radioactive waste. You want to keep the volume at minimum.

3) With CsTreat you already reach radioactive Cs levels below measurable limit. There is no reason to try to remove more Cs from that water :)

Honest question. Is is being CsTreated since eight years, why we do have radioactive Caesium still accumulating in the coast?

> 2) Material you are using as filter becomes radioactive waste. You want to keep the volume at minimum.

This is very reasonable, but storing sand for 50 years somewhere is technically feasible and maybe even cheaper than storing water. You can put the sand in concrete and obtain a solid block. Then put the package in some kind of coating and bury it in a safe vault. Water instead has the bad habit to leak, run away, and enter in the life chain when bad weather, monsoon or accidents happen. A huge concrete block is also less vulnerable to terrorism than a water tank. Is not so easy to steal a jar of it, for example. Even if you could make a hole in the coating, the product will not just flow out.

>Is is being CsTreated since eight years, why we do have radioactive Caesium still accumulating in the coast?

As far as I know, this is because about 10 PBq of Cs-137 was released during the accident and the released material is transferred by ocean currents. CsTreat is being used for purification of the water, which is pumped into and out the containment to keep core remnants cool. Nowadays the releases are very limited if any.

>This is very reasonable, but storing sand for 50 years somewhere is technically feasible...

You're right. Also CsTreat is solid, rock/sand-like material. It can be stored as mixed into concrete and disposed as low/intermediate-level radioactive waste.

Now you probably want to ask: Why there is huge amounts of radioactive water stored at the plant site? Because of tritium, which is chemically equal to hydrogen and is very expensive to separate from water. Under normal operating licenses of NPPs, you could release such water into environment, but at Fukushima standards are nowadays peculiar.

> regardless of the salinity, sediment grain size, or mineralogy

Yes, I though also about it, but I will add something to the idea of a sand filter. This is not our average dead sand, is marine sand, cleaned and oxigenated constantly by the tides. If we discard an effect of grain size, and sand composition, the really interesting hot-stuff is IMHO probably in what remains, the bacterial biofilm coating the sand.

What volume of sand would you need to make a difference? And at what rate does it collect Cs137 to reach 99%?

Would it be feasible to stack filters? (E.g., 0.01 * 0.01 * 0.01...)

What are you using to organize your nuclear effects library?

For comparison:

According to Wikipedia, granite contains 1-20 ppm of uranium [1]. One gram of natural uranium is about 25,280 Bq. Hence, one ton of granite is somewhere between 25,280 - 505,600 Bq.

In other words, if a beach is made of eroded granite, its radioactivity will be between 25,000-500,000 Bq/t: assuming the sand is more than a feet deep, I think it's safe to assume > 25,000 Bq/m^2.

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

> Some granites contain around 10 to 20 parts per million (ppm) of uranium. By contrast, more mafic rocks, such as tonalite, gabbro and diorite, have 1 to 5 ppm uranium, and limestones and sedimentary rocks usually have equally low amounts.

[2] http://www.wise-uranium.org/rup.html

Radioactivity of rocks is caused by only a few elements so is useful to consider the minerals in which those elements are hosted. i.e. the radioactivity is not evenly distributed among the minerals that comprise granite. The elements that contribute most to the radioactivity of granite are U, Th and K. The K is mostly contained in alkali feldspar and mica. The U and Th are contained in accessory minerals, such as monazite and zircon. Monazite and zircon are dense, robust minerals, which means they persist after weathering and go on to accumulate in detrital sediments, and we call the sediments in which they are concentrated heavy mineral sands. Alkali feldspar is much less robust and rarely survives in mature sediments (i.e. quartz-rich sands), instead all the K ends up in clays.

When heavy mineral sands are processed to extract the Ti and Zr ( from rutile, illmenite, zircon), the residual concentrate is rich in monazite. This material comprises the bulk of our easily accessible Th reserves. However, you can't just leave this monazite-rich material lying around heaps, as it creates a windborne radioactive dust hazard, so it gets mixed back in with the other light material. This is a bit of a shame. All that energy and effort is expended to extract the heavy minerals, but as there is no immediate market for the monazite, and it is a liability to keep it in the extracted state, all that work hard work is undone to mitigate the dust hazard and it gets mixed back with the quartz etc.

So it is perhaps somewhat ironic that we mine beaches to get the minerals that are the source of the bulk of the radioactivity of granites.

> In other words, if a beach is made of eroded granite, its radioactivity will be between 25,000-500,000 Bq/t: assuming the sand is more than a feet deep, I think it's safe to assume > 25,000 Bq/m^2.

I don't think this computation is valid, because you have to consider what and how much radiation makes it out of the solid. So converting from activity per mass to activity per surface area cannot be done, I think, by simply assuming a given depth.

Yeah this is just a version of the 'chest x-ray' fallacy/lie/propaganda.

You just simply cannot compare radiation hazzards based on gross disintegration's per second. Anyone doing that is either lying because they know better or ignorant.

> You just simply cannot compare radiation hazzards based on gross disintegration's per second. Anyone doing that is either lying because they know better or ignorant.

You know, the really nice thing about this part of your statement is that there's no way to tell which side you're standing...

This whole thread started with Bq/m^2 (i.e., "gross disintegrations per second" per square meter), after all. Of course you can accept some numbers and reject others as invalid/irrelevant, but when your whole argument looks like "I reject this number because nuclear is dangerous", it starts to look awfully like circular reasoning, or propaganda, if you prefer.

> You know, the really nice thing about this part of your statement is that there's no way to tell which side you're standing...

Two points that inform my position. There is no civilian nuclear industry. Military and civilian are two sides to the same coin. Due to it's black budget military origin, that industry lies about risks out of habit. None of the public debate deals with the complex risks, dubious economics, and long tail waste issues.

I actually know almost nothing and I can see the public debate is just a bunch of deeply ignorant people spouting off. So my position, given the industry lies, the risks, and the existence of alternatives, the sooner we stop mucking around with radio-isotopes and sequester the stuff we've made the better.

I live near Aberdeen, in Scotland, a city built largely of granite, and somewhat famous for having (relatively) high levels of background radiation as a result. Supposedly, background radiation levels are higher than they are in Fukushima!

But sand is not normally made of eroded granite, or it does not contains only granite. A lot of sand is made of shells or coral. Most beaches should have much lower levels of radioactivity.

Behold the prolonger of discussion and destroyer of easy decideability : It deepends

What I mean is that we can't just take the most radioactive beaches in the world and assume that this is the normal level in beaches so "is not so bad". In the real life, beaches aren't made (only) of granite dust. Most material is silicon, carbon and calcium.

And we should remember that unless x-chest rays, thousands of people lay practically naked in the same beach and favourite spots for entire weeks, ten hours a day. Being exposed to the UV rays of the sun also, that is another important source of cancer. In this cases, even relatively low dosis could be of concern.

People at the beach eats a lot of local sea foods also.

> This means even tens of kilometers away, the radiation can actually be higher than near the disaster. (Because, huh?, 137Cs sticks to the sand grains).

Radiation is a funky and tricky topic and frequently results in confusing cases like this. Let me try to explain (trying to keep things simple, so there's more nuance than what I'm writing).

Bq is a measurement of activity. That means the number of decays per second. That actually doesn't tell you the strength of that decay, the particle type, or the effective dosage one would receive. BUT it is still an incredibly important number. It will give you an indication of the half-life of the material and how much there is. A high Bq suggests low half life while a low Bq means a long half-life. OR a high Bq can tell you that there's A LOT of a material. This should make logical sense because Bq is telling you how fast it is decaying, and the more of a material there is the longer it takes to decay. Frequently if a material is known people will refer to its Bq as a way to tell someone else how much of the material there is. (yes, this can get confusing and there are two things it can be used to tell you. You have to pay careful attention. Context matters. In our case of the link you provided it tells you how much of the material there is.)

You'll also see Sieverts[0] listed commonly. This is a SI unit that tells you the effective dosage (REM is the equivalent). That means it considers the weighting that happens because of different particle types as well as where it impacts the body (you're in more danger if you swallow radioactive material than if you touch it with your hand). Which is important for maps like this one [1]. This is easier to understand.

On the other hand, there is grays (rads) is pretty similar to Sv but does not include particle weighting. In fact Sv is derived from Grays. They call this absorbed dosage.

So knowing this you should be able to conclude that the most dangerous material is something that has a high activity and has a high particle weighting.

For some comparisons, 137Cs has a half life of 30 years and decays with a beta- of 0.5Mev and a gamma at 0.7MeV. 235U on the other hand has a half life of 730Myr and decays an alpha particle at 4.8MeV.

[0] https://en.wikipedia.org/wiki/Sievert (There's a break down and link to the other units discussed here at the bottom).

[1] https://jciv.iidj.net/map/

> the 137Cs concentrations were incredibly high, up to 480,000 Bq per square meter in their deepest samples.

But is that high? I don't know. Let's make some uninformed back of the envelope calculations. The energy of the Cs137 radiation is 662keV, let's assume a normal person with 100lb and a surface of 2m^2 (only half get the radiation). In a day this person will get https://www.wolframalpha.com/input/?i=480000Bq%2Fm%5E2+*+662... 48uSv per day. [If I didn't make a huge mistake.]

It's like 5 times the normal background radiation, a little more than the radiation you get during a flight in a plane, less than the maximum permitted radiation dose for a worked in a nuclear plant (when there is not an emergency). https://xkcd.com/radiation/

I was expecting a much lower value, but this is still on the side of the safe level (if I didn't make a huge mistake).

You are missing there an important factor: Life and dynamics of marine ecosystems. One of the problems is that 137Cs is being accumulated by sea currents, therefore even if we have a safe level in eight years, we could have, lets say, the "double" of the safe level in 16 years, rising to "four" times in the next 30 years (of expected half life of 137Cs). Minus the Cs released again to the water, but...

The other potentially big problem is that Caesium is bioaccumulative. Clams and other animals could take it from the water again.

Good points. Bioaccumulation is a real problem. For example from https://en.wikipedia.org/wiki/Caesium-137#Radioactive_caesiu...

> In Scandinavia, some reindeer and sheep exceeded the Norwegian legal limit (3000 Bq/kg) 26 years after Chernobyl.

About the slow accumulation in the sand: If we make some simplifications and assume that without decay the flow rate is constant and the concentration increase linearly and in 32 years the concentration will be 5x the current amount. But when we add the decay the actual concentration will be only 2.5x the current amount.

(With a simple model like this using 30 years for the half life, I get the peak in the year 2043 with a concentration 2.4x of the current value.)

> 100,000 Bq per square meter

What is the Sievert dose one receives from being permanently exposed to such radiation (ideally expressed as bananas eaten per time)? To the lay person such numbers are useless without some comparison.

Boron actually mostly acts as a neutron poison, i.e. it absorbs neutrons. That way further (stimulated) decay of other nucleii that were hit by a neutron is eliminated and the chain reaction stops, reducing the production of thermal energy.

A moderator on the other hand inelastically scatters neutrons, converting some of the kinetic energy of the neutrons to heat and leaving a slower neutron. Moderators are used in reactors to bring neutrons down from their initial velocity to velocities where they are more likely to stimulate fission in another nucleus.

> A moderator on the other hand inelastically scatters neutrons,

Most moderators elastically scatter neutrons. What other kind of collision could there be with a hydrogen nucleus?

When discussing scattering of particles, "elastic" usually refers to electrostatic slingshots where almost the entire kinetic energy of the particle is conserved. Inelastic, however, refers to a direct collision where there is a significant energy loss of the particle. Obviously the former doesn't happen with neutrons, but I believe it's still a typical convention in particle physics.

No, elastic means an encounter where kinetic energy is conserved, rather than being converted to some other form of energy.

THe neutron/nucleus collisions in a moderator are largely elastic. The total kinetic energy of the neutron and the nucleus is conserved, although kinetic energy is transfered from the neutron to the nucleus.

An inelastic collision would involve (for example) nuclear excitation of the target nucleus, or ejection of particles from the target nucleus. The neutrons from fission are usually too low in energy for this to be very likely, though.

You are of course right. The scattering conserved total energy, but not the energy of the neutron.

Correction noted! Thanks!

@ Melting down fuel to determine its properties, the US just restarted a facility that does exactly that! The TREAT reactor purposely melts fuel to determine its failure points and to study new cladding materials.


Very cool with a link to a video https://www.youtube.com/watch?v=h0o4P_F4s9s where you can watch a fuel rod actually melt.

Glad to hear they restarted it last year for additional research into these topics.

Holy shit, wikipedia says it handles transients of up to 19 GW. They don't say for how long, but the transient control rod actuators take a quarter second to cover their full range, so I'm guessing that means 19 GW for up to a quarter second. That's about 4.5 gigajoules. As it happens, that's roughly the explosive energy of 1 ton of TNT. From a reactor core the size of a small closet. Makes me wonder, in fascinated terror, what power output we're seeing in that video.

Wikipedia says it can simulate six different reactor loop types. This is a seriously impressive facility.

Obviously it couldn’t melt through to China but AFAIK the way Czernobyl played out was that they had to somehow stop it from melting down to a flooded basement due to the fear of a long lasting series of hydrogen explosions that could have been truly disastrous for the region, if not Europe. Thus I think that the true worst case has so far never happened: A reactor without any of the modern safeguards (such as boronic sand pits) melting down to groundwater. Though, I guess subcritical core melts would not lead to hydrogen explosions, rather the creation of a lot of steam - it would be interesting to know how likely such a worst case would be and whether any of the world‘s reactors are configured to make this possible. Incredibly, 10 Chernobyl-type RBMK reactors are still in service.

The sad part is that the whole anti-nuclear politics often prevent building new and way safer reactors, yet they also don't manage to force shutdown all the dangers ones.

Building new reactors costs a massive amount of money, more than pro-nuclear advocates realize or are willing to admit.

Shutting down reactors doesn't save much money; they've already been built. They also weren't built when renewable energy was developed into the very cost-effective solution it is today.

The worst thing about building new reactors using existing technology (which would really no better than the existing ones) is the waste of money compared to alternatives, AND you get all the risk on top of that. Delicious government-subsidized risk.

>Building new reactors costs a massive amount of money, more than pro-nuclear advocates realize or are willing to admit.

Those costs are entirely political. Fear mongers have made it extremely difficult to get a nuclear project to completion (see the whole NRC approval process). Solar panels would be super expensive as well if each one had to go through a giant 30 year planning process that NIMBYs could easily disrupt.

>They also weren't built when renewable energy was developed into the very cost-effective solution it is today.

Renewable isn’t cost effective. That’s a lie told to placate people. In reality, renewable currently has no feasible way to replace enough fossil fuels to matter for global warming. Energy storage solutions to bring it into candidacy for base load are so far off the mark that they aren’t even deployed.

Until we have a storage solution that doesn’t 10x+ the cost of renewables, which there is no promise of, we are signing up wholeheartedly for fossil fuels as our electricity backbone (and that doesn’t even address transportation energy where the picture is even more bleak).

Reminder for anyone thinking renewables are so cost effective, here is the energy source breakdown in the US: https://www.eia.gov/energyexplained/images/charts/consumptio...

> Those costs are entirely political.

This is simply not true. There has been tremendous ongoing incompetence and mismanagement in the reactor construction process in recent years. You may blame regulation -- that's the goto excuse for pronuclear people to avoid coming to unpleasant conclusions -- but the problem is that reactors are complicated things, and reactors with safety systems are even more complicated things (and reactors without regulation are complicated things that will almost certainly fail in unpleasant ways.) Building complicated things is hard, especially if all the institutional knowledge of how to do it, and the supply chains for the parts, have decayed away.

The huge advantage solar/wind have is they are simple and scalable. Installations consist of thousands, or even millions, of replicated parts. Experience builds up rapidly as the same damned thing is built and installed over and over. Nuclear doesn't have that.

> Experience builds up rapidly as the same damned thing is built and installed over and over. Nuclear doesn't have that.

Do you think this is the reason the US Navy doesn’t seem to have major problems getting nuclear power plants built and installed in submarines and aircraft carriers? It’s interesting how civilian power companies have had a bunch of high profile failures in this area but as far as I can tell new warships still seem to be able to get reactors fine.

What's the power output of those I imagine much smaller reactors, compared to ones meant to power entire regions and millions of consumers? What did they cost to build? What risks are there, compared to regular reactors?


The A1B being installed on the newest carriers is a 700 MW unit being installed in a pair, so 1.4 GW total. That number seems comparable to a lot of commercial power generation, e.g. Vermont Yankee is a 620 MW plant.

Yes. There is a book by Serhii Plonkhy called ‘Chrrnobyl’. The lasting impression I have is that every single stage of management, design and construction was broken. One chapter alone makes the whole great book worthwhile - describing the Russian presentation at the European conference in Switzerland after the disaster.

Well it's certainly not feasible in the US with the current government in power.

That energy breakdown pic doesn't say anything about the cost effectiveness of any of the sources, it's just a breakdown of current use. Nice try, pro-nuclear activist.

There is nothing sad about it.

It's about time. Especially in countries that do not posses any reasonable ways of discarding the produced waste as well as countries that have other capabilities to generate power.

This tech IS dangerous. There WAS fallout and you can still find 9000Bq wild animals in the middle of Europe. Not even mentioning the cancer rates all over East Europe here.

So please spare us this overhyped new rush for nuclear.

> new and way safer reactors

actually the problem was never the reactor itself. chernobyl never failed because the reactor was totally unsafe. It wasn't the newest, but humans could've prevented anything. fukushima wasn't the unsafest reactor as well. however it got hit by a wave it couldn't stand and after that humans missmanaged everything to safely shut it off.

do you see a common denominator? the problem is that money is always involved even when you can build these things for cheap, however money and humans will make reactors unsafe, not the design of the reactor. at some point somebody wants to save money and most often this money saving practice will then be extremly disastorous.

I'm not saying that nuclear is bad per se, of course the waste is bad, but is it worse than coal? I'm not sure. however as long as humans control or have control over these things they will always be unsafe in a certain way.

That's somewhat my point as well. There's no perfect world, there will be human errors and there will be natural disasters, but the point with newer reactor designs, is that they have design decisions in them that account for such issues, we have learned from the past, so that even a meltdown can be contained relatively well.

well depends, even the safest system can have a bug. especially the more complex a system gets. I mean look at all the software bugs. I wouldn't count on human technology on catastrophic failures and sadly most humans make errors in situations that we can't control.

Part of that problem is that water is actually a decent moderator and neutron reflector, so it helps a lot in keeping a chain reaction with lots of energy output going. Some of the energy also goes into splitting water into hydrogen and oxygen. The hydrogen can then collect in some corner of the building until it reaches an concentration where it explodes rather easily. The explosion can then damage the building, ruining containment and scatter radioactive material around. Moral of the story? Keep your box of boronic sand nice and dry.

> Moral of the story? Keep your box of boronic sand nice and dry.

See also:

"For every three units of energy produced by the reactor core of a U.S. nuclear power plants, two units are discharged to the environment as waste heat. Nuclear plants are built on the shores of lakes, rivers, and oceans because these bodies provide the large quantities of cooling water needed to handle the waste heat discharge."[0]

[0] https://www.ucsusa.org/nuclear_power/nuclear_power_technolog...

Pretty much every energy source has a lot of waste heat produced.

Solar panels get hot, gas turbines have hot exhausts, coal plants have cooling towers etc.

They all approximate a heat engine, and they must theoretically give off heat to the cold reservoir

> They all approximate a heat engine

For each of those energy sources mentioned, what would happen if the generating station were to be completely inundated with water?

Are there lessons we can take away from further studies on the ancient reactor at Oklo? (see https://www.scientificamerican.com/article/ancient-nuclear-r... for ex)

I believe the 'chernobyl nearly caused a series of hydrogen explosions' hypothesis is not true. I will find the links I have on this subject, later.

Contaminating the groundwater is also a risk.

I also find it interesting that the "China Syndrome"[2] speculation is pretty much completely debunked.

That term was never meant to be taken literally, in the movie where it originates it is presented as a fanciful exaggeration meant to emphasize the seriousness of the risk. The actual plot concerns the risk of groundwater contamination following a nuclear accident, not a planet-piercing tunnel.

> That term was never meant to be taken literally

As a kid growing up in the 80's, I remember people, even school teachers, did take that term literally!

It was one of those things I kind of forgot about for years and took for granted as being "truth", until I actually read about nuclear power!

But is there a material that humans could build that would sink into Earth like that, for a few kilometers say, either due to melting its way down or gravity or density?

I really don't know. I've never thought of it as anything other than a colorful metaphor and don't know enough physics or chemistry to speculate.

> [1] It is too radioactive for that to be literally true

They walked up to it anyway


I think a few years ago, when all their robots kept dying, I suggested just using a big stick, possibly with mirrors, to look inside.

Funny to see that after years they finally reach the fuel with a big stick :P

> realize it was likely impractical to meltdown reactors just to develop the technology to clean them up

That's a great insight.

Maybe if some technologist approaches you and says "this is a great tech but we can't possibly recreate the worst case scenario, even in an expensive simulation" then that tells you that this is a straight "no".

Fukushima is evidence that the worst case cannot be sufficiently modelled and mitigated, at least in a commercial market.

Not saying "don't build nuclear", just that pro-nuclear lobbies should have been able to predict this situation (with a small chance of occurrence) and model it. If not, don't build. If you can predict it, but govt doesn't care about the risks, publicly document it, and that's OK.

OT, but at this point the nuclear debate looks irrelevant - direction of travel is that renewables win.

[ further OT ...moment of wild speculation: the idea that energy will be scarce will, within the 21st century, become as crazy as the idea that food could be scarce in rich nations right now. The race to renewables will look irrelevant as solar costs hit rock bottom, and no-one will remember why coal/nuclear plants were ever a thing. Within our lifetimes. I genuinely think that scenario is more likely than not. ]

I don't entirely agree with you, based mainly on when these reactors were originally designed. They started building in '67, but the actual construction designs were based on even earlier plans and tests that were conceived of in the 50s. So the original designs of the plant were born only around 15-20 years after we even invented nuclear technology. I find it similar to discussing automobile safety based on a Model-T.

Sure they have upgraded some things over the years, but there are only so much you can do when your modifications are done after it was already built or even in the middle of being built. I just don't see how we can really compare nuclear technology of the 60s, with nuclear technology of the 2019.

The Fukushima situation was predicted, several times, and TEPCO were warned about the dangers of having basement emergency generators at least 6 times. Nothing happened, because fixing that would cost money. The sea wall was artificially lowered from 30 metres to 10 metres, because keeping it at 30 during construction would have cost more money.

At the end of the day it's money.

And people. Neither of which new plant designs will address.

The Chernobyl reactors were built after Fukushima. It was a really, really old plant, that had already gone 10 years past it's intended lifespan, but they kept it open despite that because they didn't have a viable replacement source of energy.

> melted core material would just keep melting into the ground until it "got to china"

aka "China Syndrome", just like the 1979 film:


Now that convincingly debunks the China syndrome, China is obviously not the antipode to Fukushima

Could crushed borosilicate glass be used as concrete aggregate when constructing containment buildings?

Edit: apparently yes: https://www.ncbi.nlm.nih.gov/pubmed/28189090

About the engineering challenges : I love to know the camera, sensors, electronic that let them take that picture in such high radioactive environment.

In a pressurized water reactor the water is the moderator (bounces and slows the neutrons) and dissolved boron is the poison used to control the reaction.

I agree with your assessment that it's impractical to meltdown reactors just to learn how to clean them up, but apparently the US government didn't: https://en.wikipedia.org/wiki/BORAX_experiments

> [2] Speculation that a blob of melted core material would just keep melting into the ground until it "got to china"

Are people seriously considered that's a possibility? Even if the melted core went through the earth unstopped, a little bit of physic would conclude that Earth's gravity would pull the melted core toward the Earth's Core, and passing that, it would just be pulled back in the opposite direction, oscillating back and forth until stopping at the Core.

I suspect I am not the only person who saw "First contact made" at the top of HN and for one brief, shining moment lived in a far more interesting Universe.

This has happened to me here and there on various aggregators over the years, and every time I get the goosebumps, and ride it out without fully reading the headline all the way, letting myself imagine, think "yeah, this is probably how I'd find out about it!"

To me it would feel like discovering magic, or the existence of latent psychic abilities in humans. A reality shattering event.

It’s a poor choice of words for any headline that isn’t contact with ET.

Yeah, my heart raced for a second before I read the full headline. Happy I'm not the only one. One day, it will be for real. Hope that I'm still alive to witness that.

I don't want to die not knowing.

I hope an alien friend visits me at my deathbed.

People should check out Safecast, which is a citizen-led open data network for monitoring and publishing radiation readings, including an open API:


They're using low-cost open hardware, including Arduino:


Here's their latest heatmap (well, radiation map) for Japan, centered on the Fukushima area:


They also got volunteers to take radiation readings in and around the Santa Susanna Mountains near Los Angeles, to verify that radioactive waste from an old reactor meltdown there was not kicked up during the recent wildfires. (It was not, luckily.)


Do you know how they're communicating the expected health effects of dose rates like these? For instance the yellow is 4 uSv/hr, roughly 10x a normal dose rate in the USA.

I don't know that, I'm sorry, but you can e-mail them or Tweet them and find out.

There is better pictures and explanation at TEPCO site: https://www7.tepco.co.jp/wp-content/uploads/handouts_190213_...

Interesting, seems like it's been tough to build robots to withstand the radioactive materials in the water.


Has there been a proper study done on how dangerous it is to live in areas around the plant, or how safe it is to eat produce grown from Fukushima and all the fish caught in the area? There's been so much rumor and fake news around this topic that I don't know what to think.

Absolutely yes. The United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) does reports every year or so on the status. Here's the 2017 one [1].

[1] http://www.unscear.org/docs/publications/2017/UNSCEAR_WP_201...

From the executive summary (for others):

> ...its assessment of the exposures and effects due to the nuclear accident after the 2011 great east Japan earthquake and tsunami... It had concluded in that report that, in general, doses were low and that therefore associated risks were also expected to be low. A discernible increase in cancer incidence in the adult population of Fukushima Prefecture that could be attributed to radiation exposure from the accident was not expected. Nevertheless, the report noted a possibility that an increased risk of thyroid cancer among those children most exposed to radiation could be theoretically inferred, although the occurrence of a large number of radiation-induced thyroid cancers in Fukushima Prefecture—such as occurred after the Chernobyl accident—could be discounted because absorbed doses to the thyroid after the accident at Fukushima were substantially lower. It had also concluded that no discernible changes in birth defects and hereditary diseases were expected and that any increased incidence of cancer among workers due to their exposure was expected to be indiscernible because of the difficulty of confirming a small increase against the normal statistical fluctuations in cancer incidence. The effects on terrestrial and marine ecosystems were expected to have been transient and localized.

I’m a bit frustrated by some of these arguments. In particular there is all this “expected” part about low level of radiation that _should_ lead to low effects.

That all fine, until we get the numbers of actual cases, which are higher than expected.

But then we have something looking like backpedaling, a lot of “yeah it’s higher, but it’s not what you think. Because it should be low”, wich I’d take as a partly circular argument.

An extract from this study: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5770131/#!po=41...

> According to the studies reported until now, it is difficult at this point to attribute thyroid cancers in Fukushima to the direct effects of radiation exposure. The reasons are as follows. Exposure doses in Fukushima are significantly lower than those in Chernobyl (65).

There are other arguments details further down these lines, but the lower exposure shouldn’t be used to argue the effects of radiation when the whole point of the studies are to mesure these effects. It taints the whole conclusion in a veil of “we don’t want to disown or past predictions”

That paper has some wording that is hard to understand, so I think this might be what's causing you trouble. The quote you make from the paper is not the only reason. They give 5 reasons.

- The exposure rates are significantly lower that Chernobyl (earlier in the paper they explain that in previous studies there is a correlation between exposure and cancer rates -- because the thyroid problem rates here are high while exposure rates were low, it is possible that the high rates are caused by something else)

- There is no apparent difference in rates corresponding to proximity to the disaster. Similar to the above, our models would predict that as you get closer to the disaster thyroid problem rates would increase. Because this is not happening, it seems likely that the high rate is due to something else.

- The high rate of thyroid problems was detected mostly in older, not younger children. If the thyroid problems were caused by exposure to radiation, our models predict that younger children would be more susceptible. Because we have the opposite, it is likely that the cause is something else (this, to me, seems one of the more convincing arguments). However, they note that it may just be a case that by chance the younger children may not have developed problems yet and so further study is required.

- Thyroid problem rates in other prefectures are also high (and, in fact higher than in Fukushima). Again earlier in the paper, they discuss this in much more detail. It was interesting to me that even as far south as Yamanashi, there were very high rates.

- The gene mutation and rearrangement rates are different than that of Chernobyl.

From those pieces of evidence, they conclude that the high rates of thyroid problems are likely due to some other cause (many possibilities of which they describe in the paper).

Like I said, I found the wording in this paper to be difficult to understand. It appears to have been written by non-native English speakers (mostly Japanese), so that's probably why. If you read it again, hopefully it will make more sense.

My main beef is how in parralel to the pure data and experiments, there’s all this commentary going back to other events (Chernobyl) or expected results (“ours models would predict") to seed the ideas that some of these data is just non significant.

To draw a bad analogy, if we had a plane crash and had to study how 5% survived, we wouldn’t be saying ‘previous crashes had everyone dead, we should not pay too much attention to this set of survivors’ or ‘our models predict bulky males to survive, young kids surviving here is just an oddity that has nothing to do with our study’.

Also if there’s actual data not matching the model, shouldn’t we think hard and long about the the model ? And if we do, the argument wouldn’t be that data doesn’t matches the model, but that we have additional data explaining the root cause of the oddities.

I am not saying we must to tie everything back to a single source, and for instance the fact that having more screening than before actualy affects the rate of detection and changes the statistics is a very valid point among many.

But brushing away swaths of oddities because it’s not expected feels lazy. Basically I’m wondering if we would find causes aside from direct radiations that could cause some of the issues. For instance the whole country’s rate of cancer rising shouldn’t be an argument to dismiss Fukushima having something to do with it, hell the gov shipped contaminated soil around the country, there were multiple food mislabeling issues etc. We should pin down an actual cause of the rise of cancers around the country before saying it’s unrelated.

Well, to be fair, the paper you picked did no experimentation. It's a review paper. Not only that, it's a review paper whose express purpose was to put forth the idea that the high thyroid rates in Fukushima prefecture are likely not due to the disaster (they say as much in the conclusion). It is very possible that it suffers from confirmation bias (I don't know enough about the topic to comment one way or another). If you look on any topic, you will find review papers like this.

If I understand your comment, though, I think you still suffer from misunderstanding the paper. The paper states that the data is consistent with our models. It's not consistent with the idea that the thyroid problems are caused by the disaster. As the paper states, we do not know what the thyroid problem rate was before the disaster because we did not screen people in the same way or at the same rate. The equipment being used now is much better at detecting thyroid problems. Other places using this equipment (for example South Korea) are also finding high rates of thyroid problems. When doing screening in other prefectures, they are also finding high rates of thyroid problems. Discovery of thyroid problems are on the rise all over the world. (All this according to the paper -- I haven't the foggiest clue if it is true).

In other words, the paper is explicitly saying that while the rates are high, it is likely that the high rates are due to a variety of other causes. They go to a considerable amount of effort to document in detail what these are. You seem to be fixed on the idea that the high rates are especially unusual or outside the boundaries of what we would expect if there were not a disaster. This entire paper was written to dispute that point of view. It is not the case that they are simply ignoring it. I think the biggest thing to understand is that they claim that there are places in the world where the rates are higher, which didn't have a nuclear disaster. Since the data does not match that of areas that did have nuclear disasters, but does match that which did not have nuclear disaster, they conclude that the thyroid problems are likely to have been caused by something other than the disaster.

If you dispute that, it's entirely up to you. Like I said, I have no idea if it is true or not -- I'm just telling you what they wrote. Generally, I don't like review papers like this because of the problem with confirmation bias. People start with the conclusion they want to have and then find evidence to support it. You can always find that kind of evidence. It's not even a well written paper... but I didn't choose it ;-)

I think the best way to learn more about whether or not these researchers know what they are talking about is to read the citations that they make. They make a lot of claims about detection rates of thyroid problems, the rate at which these turn in to cancers, etc. If they are wrong that the current rates can be explained by non-disaster causes (or if they are wrong that data does not support a disaster cause), then you should be able to find the problem in those citations. Remember, it's a review paper! They are only gathering data from other papers and putting it together.

I suppose if I'm particularly uncharitable, if the authors of the paper are lazy, then I'd ask you to be at least reach their bar. If the data is rife with swaths of oddities, then write a review paper that points it out. Provide citations that show current data does not match our models. I bet you don't even need to do that. I bet there is at least 1 group of researchers in the entire world who has at least tried to write such a paper. Find it and we can have a much, much better conversation.

Thanks. I think page 20 has a lot of information I'd been looking for.

Yes. And I recommend having https://xkcd.com/radiation/ handy as you read through those numbers. The units are consistent.

Forgive the macabre question, but if a human (or other living creatue) were to venture in to the plant and collect this fuel by hand, how long would they survive?

I'm in no way suggesting that this is an option, of course -- just curious.

You should watch this documentary on Chernobyl:

>According to Vyacheslav Grishin of the Chernobyl Union, the main organization of liquidators, "25,000 of the Russian liquidators are dead and 70,000 disabled, about the same in Ukraine, and 10,000 dead in Belarus and 25,000 disabled", which makes a total of 60,000 dead (10% of the 600,000 liquidators)


They had guys running in and throwing lead bricks - and some of the scientists died by giving their life to get in and take video of the meltdown.


These guys knew they were doomed.

But they had freaking reportedly 700,000 men attack that problem.

I've met a dying Chernobyl first responder on a commuter train in Belarus, back in early 1990s.

These people never made it to WHO statistic as their deaths protracted over many years can not be directly attributed to the accident. As pro nuclear crowd would be happy to tell you, only couple dozen firemen died on the first day and that was it, making nuclear meltdowns almost benign incidents.

It is a pretty hard-hearted thing to say, but if the WHO doesn't think there is enough evidence to count his death in the Chernobyl statistics, what evidence does he have they are wrong? I mean, it isn't unprecedented for crippling damage to not surface for a long time (eg, asbestos/mesothelioma), but on the other hand cancer can strike at any age [0] and people struggle to attribute things to chance if they have a potential reason for it.

The WHO is using a statistical model to account for the fact that individual deaths might be cancer or might be something else. There is pretty concrete evidence that in terms of deaths nuclear meltdowns are benign, and the major concerns are land contamination and how big the exclusion zones need to be when things go wrong. Some of the better science-based counterarguments to nuclear I've heard go to contamination in the food chain.

[0] https://www.cancerresearchuk.org/health-professional/cancer-...

Most of the studies among Chernobyl liquidators were based on monitoring the cohorts between 1994 to 2004. Very little was done form 1986 to 1991 at all.

The "never made it to WHO statistics" as the difference in cancer incidence rates and other problems after Chernobyl are so small that they are hard to distinguish from the background noise. A lot of work has gone into trying to document the actual effects, but the reality is that while there were initial large spikes in e.g. thyroid cancer among children, the cancers that spiked are ones with very high survival rates. The WHO estimates were initially very high, but keep being adjusted down because every year that goes is another year without a spike in deaths above expected levels.

The breakup of the Soviet Union caused much larger scale effects on life expectancy.

The effects of the Chernobyl radiation are apparently common, that when I left my luggage in the Belorusskaya train station in Moscow I encountered a very large plaque describing exactly what categories of people were allowed to get in front of the queue when retrieving your stored luggage -- with "People suffering from radiation as consequence of Chernobyl" on the list.

If you served in the military for at least 6 months 1941-1945, you were higher on the priority list, which included around 17 different very detailed categories, all translated to English too.

The accurate number of deaths appears to be very controversial.

Most certainly - I'd expect its way higher.

They were referred to bio-robots.

There are a few documentaries about them. A lot of elderly volunteered to be a bio-robot.

There was one doc that showed them throwing lead bricks into the pit - and they were only capable of being exposed up to like ~2 minutes maximum of something.

10% of an adult population dying over the course of 20 years doesn't seem inherently crazy? The WHO says no more than 4k died due to the accident.

It sounds far too good to be true. According to the actuarial tables at [0], 10% is the expected 20-year death rate for 40-year old men. This is for Americans in 2014, I'd expect the outcomes to be worse for Soviet citizens in 1986.


In 1986, the life expectancy for Soviet men (at birth) was 65 years:


This might interest you and give you a benchmark for the lethality of approaching it: https://rarehistoricalphotos.com/the-elephant-foot-of-the-ch...

> After just 30 seconds of exposure, dizziness and fatigue will find you a week later. Two minutes of exposure and the body cells will soon begin to hemorrhage; four minutes: vomiting, diarrhea, and fever. At 300 seconds you have two days to live.

> All of the fire fighters and people who worked in building the sarcophagus died around a year or so after the event.

I've read (long time ago) the US did tests on monkeys, exposing them to large neutron doses. There is also a number of human cases of exposure to large neutron doses, starting with Louis Slotin


I'd recommend Svetlana's Alexievich "Chernobyl Prayer" although is more intimate (as stories told are very personal) prose than historical records: https://thephraser.com/2017/09/07/chernobyl-prayer-a-chronic...

This was heavily documented after Chernobyl. If you're not affraid of heavy and even gruesome recollection from people and relatives there at the time you can find documentaries on youtube

this is good that they are there. It means that the radioactive material can be sequestered. It means that the radioactive material can be inventoried, and escapee content determined.

Not there yet. They got in far enough to touch it with a grabber. Didn't get a sample yet. Removing it for disposal is years away.

Are all the rainbow colored pixels in the picture caused by radiation hitting the camera sensor?

All pixels in the picture are caused by radiation hitting the camera sensor

I assume the commenter was referring to ionizing radiation.

In a CCD, even photons of visible light are ionizing radiation. The light causes electrons to pop out of a semiconductor and get caught in a capacitor, where they are stored long enough to be read out as a voltage, and turned into a number.

I don't think so. I rather think it's water droplets acting as little prisms. Radiation would hit the camera sensor in a random pattern and the effect would decay over time. A bit like how you determine pi with the monte calro method by throwing points at a circle and then checking whether they are inside or not.

I found a video of the robot on YouTube [1] and it seems that the rainbow patterns don't follow the "monte carlo throwing" pattern, but their location depends on the scene.

I don't know how radiation would affect a CMOS sensor, but I would guess it to be similar to what strong light is doing, e.g. when you point your camera into the sun. What then happens is an effect called "blooming": CMOS sensors work by photodiodes converting light to electric potential. If the radiation that reaches one of the photodiodes in the sensor plane is strong enough, the potential would jump to the other sensors next to it, causing a large bright field in the digital image.

Edit: Oh and it seems that the material itself is reflecting light in a rainbow pattern, but that's not the same as "ionizing radiation hitting the sensor".

Edit2: it could just be oil. Oil on water causes rainbow like artifacts.

[1]: https://www.youtube.com/watch?v=cyKEVD4oC4s


My own little pet theory is that super volcanoes are made by nature's corium, slowly desolving from bedrock and migrating in channels to earth's core.

Are there not any suits that protect against the radiation to let them go in and do some things by hand? If there are but it's still not a good idea -- why?

Gamma radiation goes through everything. The only thing that stops it is mass and distance. You can only wear a certain amount of lead before you collapse.

The amount of radiation is just too high.

Gamma radiation not so much simply makes it through everything, but it does indeed take a lot of shielding to attenuate to save level if there is enough radiation to begin with.

The fact that it interacts with stuff is actually what makes it dangerous. If it went through every thing it would not interact with the worker either and would not pose a thread to them.

Btw. there is a type of radiation that is practically like that: neutrinos. Impossible to shield against, but also completely harmless if you are not standing right next to a super nova.

>Impossible to shield against, but also completely harmless if you are not standing right next to a super nova

There was actually a fun what-if based on that: https://what-if.xkcd.com/73/

Thanks for the answer, just curious would the lead sheet have holes thorough it if you put it up to that radiation?

I'm picturing a sheet of paper with bullet holes through it.

Or is there a better analogy of what is going on?

A lead apron, when exposed to a strong enough source of gamma rays, will be unable to sufficiently attenuate the radioactive flux. It will absorb gamma photons, and then emit a bunch of of them right into you. Radiation shielding has to take into account the total flux, and then it has to be built in layers to a given thickness and overall density. It doesn’t do you any good if the incident radiation is all absorbed, only for 99.9% of the original flux to be re-emitted by the shield!

You can think of it as roughly analogous to the excitation of atoms leading to emission of a photon in the visible range. It doesn’t help to wear a pair of goggles that will absorb all of the light you’re hoping to block, only to have your goggles glow brightly as a result. You’d rightly conclude that you need much thicker goggles, and the same is true here.

No holes needed, and a lot of the original gamma radiation is being absorbed and subsequently emitted, which defeats the point. There are also complications of a substance being exposed to some forms of radiation (primarily neutron) becoming radioactive themselves. It wouldn’t be visible, but it wouldn’t be good either.

A note that shielding, in addition to considering total flux, has to consider the type of radiation. Gamma (very high energy photon) is very penetrating, beta (high energy electron) is less so, and alpha (a helium nucleus) hardly penetrates at all. You also have to consider, in the case of a molten core, other contaminants and radionuclides actually getting on you or in you. Alpha emitters are pretty harmless outside of you, where your skin stops them cold. If you eat or breath them in however, they’re deveststing. So your concern with alpha emitters tends to be covering yourself and wearing a respirator, while that won’t even help a little with gamma.

It would not have holes in it. The main reason why the gamma radiation makes it through is because it never interacted with any of the lead. And even if a (unlucky) gamma photon interacts inside the lead sheet it would not do much damage there. The only thing that really do damage to the material is neutron radiation and even there the damage is on the microscopic level (atoms getting knocked out of place) than anything macroscopic.

From the point of view of the radiation, there's a lot of empty space between lead atoms and a lot of it will just slip through and not interact with the lead at all. A fraction will slam into lead atoms, a fraction will slam into your atoms, a fraction will keep going.

More like a sheet of paper held up to the light. Some light gets through, and if you have thicker paper it will allow less light through, but it doesn't noticeably damage the paper.

Depends how long you leave the paper exposed to light. if you have a sheet of paper half exposed and half shielded for a year of sunlight the effect is marked.

The equivalent in lead would probably manifest in some way, perhaps a visible patina on the surface? or, as a differential in decay products. you could call that 'damage' if you wanted to.


We've already asked you many times to please follow the guidelines and post substantively, so we've banned the account. This isn't a sarcastic flamewar site.

Can someone add (2016) as the publish date?

Edit: I don't know why but the first time I clicked the link, it went to a 2016 published piece. Please ignore!

Why? It was published yesterday, and mentions previous events having occurred in 2017...

The article says February 14th 2019.

What strikes me is just how inept the Japanese are in handling Fukushima compared to the USSR during the much worse Chernobyl disaster. The USSR responded quickly and it took just 8 months to entomb the reactor. Over 600k people worked to handle Chernobyl. Fukushima, on the other hand, is still at risk of another disaster, 10 years later. The Japanese government and power company under-handled the situation and lied for years about it's severity. For example that 300 tons of polluted water was going into the pacific every day for 5 years.

What the Soviet Union accomplished at the cleanup of Chernobyl likely wouldn't have been possible without a wanton disregard for the value of human life only possible in a totalitarian state, the details of which are discussed in other comments.

I don't think the lack of will or ability on the part of Japan to doom thousands of people to death by radiation in the course of a speedy cleanup can be described as ineptitude.

The reaction to Chernobyl was excessive. What we've learned since is that while a lot of rapid response was needed at the immediate site, at least a portion of the rapid evacuation likely did more harm than good. Large parts of the exclusion zone remained livable (and the Chernobyl plant itself remained operational for many years afterwards - the danger was low enough that several of the reactors were re-started).

We should be thankful the Japanese did not overreact as badly, but in Fukushima too there is reason to believe that more people than needed were evacuated, and that this response may have caused more harm than the accident itself.

This is an interesting perspective. It's probably unknown if it's too much or too little without some serious scientific investigation. What do you base your opinion on?

For starters there's extensive data on effects of staying in large areas of the outer exclusion zone, because a lot of people (a small proportion of the population, though) stayed or went back. The government eventually gave up on keeping people out.

Secondly, there is the experience of continuing to operate the other reactors. It took 3 years before work on reactor 5 and 6 (which were unfinished) was halted. Operations at reactor 2 continued until a fire in 1991. Reactor 1 remained in operation until 1996, and reactor 3 was finally decommissioned in 2000. While workers were subject to stringent controls, and on/off schedules intended to reduce their exposure, people were nevertheless working in some of the most radioactive areas of the exclusion zone for many years, with very limited indication of lasting problems.

Thirdly, there is the economic impact on the Soviet Union in general, and Belarus overall, both unavoidable costs of the cleanup, but also very much avoidable costs of large-scale reduction in economic activity in a large area. There is a lot of data on the subsequent effects of the economic upheaval that followed the breakdown of the Soviet Union that shows the effects of economic upheaval on e.g. life expectancy. Applying the same to the upheaval caused by evacuation of areas that were quickly known to be within safe radiation levels implies substantial harm.

In general any evacuation causes substantial harm, because of e.g. increased mortality when moving hospital patients or other vulnerable groups; increased traffic accidents etc. Just doing more stuff quickly causes harm.

> The USSR responded quickly and it took just 8 months to entomb the reactor.

At a gross cost to human life[1], and in a way that leaves the problem there - work is _still_ to contain and clean up the reactor. There are good reasons for not doing the same here.

The sarcophagus was never intended to be a permanent solution, and was unsuccessful in completely containing the accident: Sr-90 and Pu-239 were leaching into the surrounding water table, and some have suggested that the rainwater and remaining nuclear fuel posed the risk of restarting the nuclear chain reaction[2].

[1] death figures are controversial, and vary wildly, but 600,000 people were involved in the cleanup, some of whom will have been put at risk.

[2] https://outline.com/fTk4GX

I'm curious, what would have been a better approach the USSR could have taken. Also, is there a better approach that Japan could have taken?

It took 10 years to even get a robot that could get close to the fuel... and people still ask if nuclear might not be as bad as coal!

The truth is that accidents like this can never be cleaned up or salvaged. It's time to move away from nuclear, as other countries have already figured out.

Fortunately for us in the US, the last nuclear plant under construction today will likely be cancelled, as the last two were. There are no plans to build another one here. That doesn't relieve of us of the chance of a homegrown Fukushima on American soil, but it shows that the future isn't nuclear for environmental and cost reasons -- of which "decommissioning" costs billions, even if there's no meltdown.

If you look objectively at the numbers, you'll see that nuclear is one of the safest forms of energy production we know. Even considering conservative estimates from Fukushima and Chernobyl, nuclear power has net saved over 2 million lives simply by displacing air-pollution related deaths that would have occurred had the nukes not been built.

Coal kills on the order of hundreds of thousands of people per year. Fukushima killed zero. Chernobyl directly killed 60, and caused up to 4000 early cancer deaths over the next several decades on top of a population that would get millions of cancer deaths anyway, and that's with a bizarro linear no-threshold model almost unheard of in other latent estimates.

Nuclear is also very nearly carbon free. So if you start factoring in health effects of global warming, nuclear's even better.

Also we should keep in mind that the major incidents involed reactors designed in the 60s/early 70s, when nuclear energy was still very new. It's like wanting to ban planes because of the safety records of planes from the 1930s. It is because plane designs iterated, learning from our mistakes over time, that it became an extremely safe technology.

When were the majority of reactors operating today designed? Serious question. I actually want to know.

1960s and 1970s mostly. Chernobyl was designed earlier. Fukushima was a super early design. The real flaw there was they put the backup generator fuel supply in the basement in a flood zone. The reactor was designed for the US midwest, with tornados in mind as the biggest threat. Moving to Japan, they should have put at least some of the backup fuel above ground level and it would have been fine.

The flaws in an earlier design can't be a justification for absence of upgrade. Since 1960 we had build thousands of structures higher than 150m. Skyscrapers are not exactly a novely. We have the tools and the knowledge.

They had 51 years, 51! to raise a damned wall 20 m. Had all the new information, improved models, latest computers and warnings... and they didn't do anything. Neither Tesco, nor the Japanese government were able to see it coming. Their lack of vision is really infuriating.

I don't know how nuclear can be considered remotely "carbon free" if you take into account the usually-externalised emissions incurred in mining, refining and processing the uranium ores into fuels.

Then, too, I've seen no figures on the carbon costs of nuclear waste-management.

Australian mining engineer here. Based on some pretty basic evidence [0] I'd assume uranium mining is several orders of magnitude less damaging than other energy sources, and probably has a decisive "carbon free" advantage over wind and solar once you factor those externalities in.

Australia is a top-3 uranium producer [1], we have 3 mines. That rounds to 0 in Australian mining terms. Olympic Dam is a monster in relative terms, and the production numbers are ... by tonnes Olympic Dam isn't even a uranium mine, it is a copper mine. And Olympic Dam [2] is a world class uranium mine afaik.

[0] https://en.wikipedia.org/wiki/Energy_density

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

[2] https://www.bhp.com/-/media/bhp/documents/investors/reports/...

Those are all considered in the lifecycle emissions calculations done by folks like the IPCC, who released these data [1]. It considers mining, construction, transportation, waste, etc. of all energy sources so they can be compared for energy policy purposes.

E=MC2 is an amazing thing.

[1] https://partofthething.com/thoughts/wp-content/uploads/ipcc-...

Source: Schlomer S., et.al., 2014: Annex III: Technology-specific cost and performance parameters. In: Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the 5th Assessment Report of the IPCC

Uranium mining is a tiny operation due to it's absurd energy density. I wouldn't be surprised if the mining cost of rare Earth's used to maintain broken solar units surpasses it it's so insignificant.

Besides most countries with a nuclear weapons program end up with a lot of reactor-grade uranium lying about anyway, and those countries also happen to be the biggest polluters in energy

> I wouldn't be surprised if the mining cost of rare Earth's used to maintain broken solar units surpasses it it's so insignificant.

Yeah, there's plenty of very unpleasant chemistry involved in making PV panels and I strongly doubt that the wastes involved in the associated mining/manufacturing/disposal processes are treated anywhere nearly as carefully as nuclear waste (not to speak of the fact that due to sheer energy density difference, PV panels will produce far more waste than nuclear fuel/reactors).

No, it's actually a terribly polluting and inefficient mining process. Only a tiny fraction of what is dug up is usable. It requires many (even more toxic) steps in order to turned into fuel. There's a good documentary about uranium tailings and how they pollute the local water permanently.

Besides, there isn't enough uranium in the world to provide baseload nuclear power that would replace coal.

We shouldn’t be burning coal either.

What we have right now are three options : fossil fuels, renewable sources like wind/solar/hydro, or nuclear.

Fossil fuels kill millions a year related to pollution, and does make whole areas unlivable. About 16% of all fracking oil in the US gets spilled in nature every year[1], on top of crap like Flint. Coal is responsible for 80% of energy-generated related air pollution, worldwide[2]. They are not a good solution.

Natural sources are simply extremely limited in terms of access. Few countries can realistically really rely on these as a main source of electricity.

The, there's nuclear. We have fission, which is, while relatively safe (5 major impactful accidents since the rise of nuclear power, 3 of those being the three Fukishima plants), pretty damaging _when_ shit happens. However, are the damages realistically worse than fossil fuels? I'm not exactly qualified to figure this out myself but just looking at the numbers really doesn't make fission look that bad. Finally, there's nuclear fusion, which is a lot safer, while we've been making decent progress in the recent years, doesn't look like it's gonna run our cities any time soon.

So no, we shouldn't be burning coal, obviously, but if nothing's perfect, the logical option is the next best thing.

[1] https://www.bbc.com/news/science-environment-39032748 [2] https://www.ucsusa.org/clean-energy/decrease-coal-use

> on top of crap like Flint

Flint's water contamination has absolutely nothing to do with energy production of any kind. It was caused by the city still having lead water pipes.


> About 16% of all fracking oil in the US gets spilled in nature every year[1]

> [1] https://www.bbc.com/news/science-environment-39032748

That isn't at all what that article says. It says 16% of wells suffer a spill each year. The industry would never allow 16% of its product to simply go to waste, particularly considering how difficult and expensive those operations are.

I'm not american, seems like I mixed up Flint with some other american city that made the news up here for having water catching fire and shit like that.

And indeed, I misread, it's not 16% total spillage but 16% of them spilling at all. This one doesn't really change much to my point that having one in six fracking points spilling out is extremely damaging.

Of course we shouldn't be burning coal, but we are. In 2017, 17.8% of US energy production was from coal. 28.0% from petroleum, and 31.8% from gas.

Until these three numbers become 0.0, 0.0, and 0.0, building more nuclear plants to displace them is an immense progress.

But we are, so it's better to go with the cleaner option.

You are basically saying: Since you can't be perfectly clean, I'd rather you be as dirty as possible.

Okay, how should we have powered the industrial revolution?

> and people still ask if nuclear might not be as bad as coal!

It's not even close. Nuclear is so much better than coal it's unbelievable.

Just to start with coal releases orders of magnitude more radiation than nuclear. https://www.scientificamerican.com/article/coal-ash-is-more-...

Then the number of deaths from coal, take a look at the numbers:

Coal: 100,000 deaths per PWh.

Nuclear: 0.1 deaths per PWh.

That means coal is 1 MILLION times worse than nuclear. I mean seriously, what else would it take to convince someone?

Let's see, what's worse? Taking 10 years to do something, or killing 1 million times as many people?

Lest anyone forget, the Centralia (coal) mine fire has been burning since at least 1962 and no one knows how to stop it. The Fukushima Daiichi incident really doesn't seem that bad to me, comparatively speaking.


Granted that coal fires can occur due to other causes than mining (such as simply exposing a coal seam to air during construction). That said, mining greatly increases the number of incidents that occur.

The choice is not between coal and nuclear.

Nuclear kills less people than Coal mostly from asthma, but coal even causes more cancer because on a normal day it emits far more radiation in the form of small particles of radioactive dust present in coal.

The real danger of Nuclear is it can make land unusable by people for centuries (this actually appears to be good from a biodiversity and ecological heath stand point because people do more damage to that than some radiation). But by contributing to climate change coal has the potential to put far more land under water.

If we had gone for widespread adoption of nuclear power on the scale of Japan climate change would not be a problem like it is today.

Centuries? Have you been to the Chernobyl exclusion zone recently? [1] That was the worst accident imaginable. Fukushima is much less bad. Natural background radiation in places like Ramsar, Iran are way higher than what you'd experience at Fukushima even right after it happened. And people have been living healthily there for hundreds of years. Low-dose effects of radiation seem to not be all that bad.

[1] https://news.nationalgeographic.com/2016/04/060418-chernobyl...

I'm going to correct myself and say that Ramsar rates are what you'd experience around Fukushima a year out, not right after it happened. Some of those workers got way higher doses than anything natural.

> The truth is that accidents like this can never be cleaned up or salvaged.

While global warming is easily solved.

Coal or nuclear are both bad but differently. Which one is less bad depends on your risk model

The choice is not between coal and nuclear.

Yes, actually it pretty much is.

If climate change truly is incompatible with global civilization, I would rather have some defunct coal plants laying around than spent nuclear fuel. Nuclear maybe would have helped us avoid the worst of climate change if we had adopted it at a broader scale sooner, but at this point I think we have to be planning for post-civilization scenarios.

You might want to read "The Moral Case for Fossil Fuels" for a slightly less dire perspective.

Thanks, I'll check it out. I've been watching the shitshow for a few decades now though, and to be honest the only surprising thing is that it hasn't already fallen apart. In my mind it's inevitable, and not just due to fossil fuel usage. We won't change course until it's demonstrably catastrophic and then it will be "too late" for a majority of humans. The only question is, will it happen within 20 years or 50 or 100? Malthus was right in principle, only wrong in timeline.

> Nuclear maybe would have helped us avoid the worst of climate change

Not even that. People seem to forget that most utilized energy is eventually dissipated in heat in the atmosphere!

Any unlimited and extremely cheap energy source would only encourage more unsustainable consumption.

Not to mention that many industrial products that lead to pollution (e.g. plastic) are correlated to the cost of electricity. Give the world free energy and people will want a new smartphone every month.

(let's sit back and count the downvotes now...)

The bad parts of nuclear power is extremely concentrated and so has great visibility. I have asthma and I used to live near a coal plant. My lifespan is shortened considerably - not to mention the huge decrease in quality of life. And I'm just one the many millions who are affected by increased pollution and particulate matter in the air. However because the effects of coal is diffused from your perspective you have no idea the damage it causes to people like us. On aggregate coal kills more people (earlier than what their lifespan would otherwise be and that has to count for something).

Considering the failure mode of nuclear posions maybe 10 sq miles, whereas the operating mode of coal posions the entire fucking atmosphere, I'd say coal is still worse

> ... and people still ask if nuclear might not be as bad as coal!

That's playing on feelings rather than making a rational argument. Regardless of whether I am for or against widely deploying nuclear energy, this is not the way to talk about it and create informed opinions. The media are already doing their best to play on feelings on this topic, we don't need to replicate that here.

The logical argument is that we still do not have a good way to deal with the radioactive waste and that a lot can still go wrong in the next million years.

And another logical argument could be that this causes us to delay cutting emissions enough to prevent massive ecosystem changes resulting in humanity's death in the next few hundred years.

It's just not that simple. We need to look at both sides, weigh the options, and find a solution which probably involves a mix where we compromise on having X m³ of earth dedicated for a long time to nuclear waste to save us in the short term. I'm not informed enough to give exact figures and make proper arguments, but reiterating known one-liners won't get us anywhere.

The thing is that there is a rather pro-nuclear crowd on HN here that does not understand enough of the physics and the systems engineering of the large scale electrical grid that goes into making this trade-off in a rational way.

Are you suggesting that the anti-nuclear crowd does? What I've seen is that most anti-nuclear people are mostly driven by emotion in a one-sided way (such as not accounting for the environmental and real human cost of oil, gas, coal).

What parts are we missing?

Regulatory capture and "understandings" between supervising authorities and commercial operators that boarder on corruption, unsolved waste problem, very slow ramp-ups and ramp-downs which are annoying for the grid and a lot of things I don't understand well as an astrophysicist either. But both friends I have in the nuclear energy field have become very anti nuclear power and would rather see it shut down tomorrow instead of next year.

> there is a rather pro-nuclear crowd on HN here that does not understand enough of the physics and the systems engineering of the large scale electrical grid

Regulatory capture and what's essentially corruption is not "physics and systems engineering", and they should be considered independently of nuclear power's inherent technical merit. And regarding ramp-ups and ramp-downs, I have never heard of anyone suggesting nuclear power for anything other than base load, so I don't see how this would be an issue?

I'd add to that the astronomical costs of decommissioning (always passed on to ratepayers) and the fact that nuclear is basically uneconomic. It requires enormous subsidies and always goes over budget. Perhaps most disturbing is the aging fleet of nuclear plants we already have in the country, almost all of them way past their designed lifetimes. This is a recipe for accidents.

I notice that no one brings up that all recent plant construction in the US has been cancelled except one -- which could get killed any day. And no new nuclear plants are planned or proposed for the US.

> this causes us to delay cutting emissions

Nuclear VS "delay cutting emissions" is a false dichotomy.

> we still do not have a good way to deal with the radioactive waste and that a lot can still go wrong in the next million years.

With adequate transuranic/actinide burning and breeding/reprocessing capacity the time constants for waste disposal won't be anywhere that long. Discounting nuclear energy because of stupid once-through fuel cycles seems a bit unfair.

>The truth is that accidents like this can never be cleaned up or salvaged. It's time to move away from nuclear, as other countries have already figured out.

France per-capita CO2 emission is 1/3 the US rate and 1/2 the German rate. This is due to their heavy focus on nuclear power for electricity production. The data shows we should be moving to nuclear rather than away from it.

As far as "never" being able to clean up an accident, how could you take such an absolutist position? Do you really believe that with a reasonable research effort we won't understand how to reprocess nuclear waste and/or accident debris within 50 years? We couldn't even fly until a little more than 100 years ago. Our understanding of the physical world is growing more rapidly than any of us can really appreciate.

To say we can "never" clean something up is to completely disregard the exponential rate of increase of both scientific knowledge and technological capability. Given progress in the last 100 years, it is reasonable to conclude that nuclear waste is a 50 to 100 year problem at worst.

Geez, did we toss out cars or aircraft after a few accidents? Did we stop trying to go into space after some accidents? Did we stop trying to cure cancer after some deaths? No, we learned from the mistakes that were made that led to the accidents and fixed them. The development of nuclear power should follow the same iterative process all other technologies rely upon. Obviously we work as hard as possible to risk reduce the process, but iteration is how science turns into safe and practical technologies.


> Do you really believe that with a reasonable research effort we won't understand how to reprocess nuclear waste and/or accident debris within 50 years? We couldn't even fly until a little more than 100 years ago.

This probably requires a new kind of physics to be discovered and eventually applied. It's absolutely not clear how this should work.

Fluid Dynamics is Newtonian physics, Bernoulli equation already exists since the 17th century.

Frequently people worrying about environmental impact aren't as concerned about financial impact. I don't care if it cost eleventy-gazillion-bitcoins to decommission a nuclear plant if the alternative is continued global warming and an inhospitable planet when the bill comes due.

>>The truth is that accidents like this can never be cleaned up or salvaged.

I mean that depends on how pedantic you want to get with a "never" and whether you think "sequestered" is good enough or only "salvaged" is acceptable.

Current nuclear isn't great and is incapable of solving climate issues by itself. It still seems like a more viable option than continued reliance on fossil fuels.

There is nothing even close to be as cheap / clean and give as much power than nuclear right now. Not every country have wind / sun / rivers to match what produces nuclear energy.

Yes, and move away to what ?

LED lights? Highly isolated apartment buildings? Revised urban planning to accommodate a post-commuting work-life? Other energy saving technology? In EU we have a comfortable living off consumption levels that in the US would qualify as church mices' (still not enough to prevent collapse though: https://web.archive.org/web/20140913032823/http://graduatein...)

> LED lights?

you need to be addressing the larger chunks of these pie charts:


Switching to LED lights (and reducing usage) is one example of low-hanging fruit that is easily achievable without affecting lifestyle much. The fat needs to be trimmed in all areas.

Yeah, did I mention "Highly isolated apartment buildings"? (I did ;)

In fact, there are apartments here in Amsterdam that stay a cool 20C in winter with little or no heating at all, just the warmth from the water boiler flue running through the heat exchanger of the air ventilator.

Nationally sponsored infrastructure plans to replace existing buildings with energy class A at scale would put a big dent on energy consumption.

Solar, wind, storage. Fusion at a distance.

EDIT: "20 years from now" is still faster than any nuclear generator getting built.

Nuclear power plants take at most 5 years to build from the time we break ground until the reactor's hooked up to the grid. Look at how fast the Chinese are building them, or how fast e.g. the French or the US were at building them in the 60s and 70s.

That it takes so long now in the west is because we're not building enough of them, thus losing the advantage of experienced workforces and inability to leverage mass-manufacturing of standardized designs.

It's also slow because of onerous safety regulation seeking to make nuclear too safe, when we consider the result of a slowdown in commissioning of new reactors, thus increasing the reliance on coal killing way more people than if we were to hypothetically mass manufacture relatively unsafe reactor designs from the 60s today.

>Look at how fast the Chinese are building them, or how fast e.g. the French or the US were at building them in the 60s and 70s.

"The country [China] has the capacity to build 10 to 12 nuclear reactors a year. But though reactors begun several years ago are still coming online, the industry has not broken ground on a new plant in China since late 2016, according to a recent World Nuclear Industry Status Report.

Officially China still sees nuclear power as a must-have. But unofficially, the technology is on a death watch. Experts, including some with links to the government, see China’s nuclear sector succumbing to the same problems affecting the West: the technology is too expensive, and the public doesn’t want it."

https://www.technologyreview.com/s/612564/chinas-losing-its-... (China’s losing its taste for nuclear power. That’s bad news.)

> That it takes so long now in the west is because we're not building enough of them, thus losing the advantage of experienced workforces and inability to leverage mass-manufacturing of standardized designs.

Can you provide evidence that the cost of nuclear would be comparable to solar and wind if built at scale?

> It's also slow because of onerous safety regulation seeking to make nuclear too safe, when we consider the result of a slowdown in commissioning of new reactors, thus increasing the reliance on coal killing way more people than if we were to hypothetically mass manufacture relatively unsafe reactor designs from the 60s today.

I'm unsure you can sell nuclear to the public or politicians as "it's expensive because it's too safe". It's unlikely we're going to relax safety regulations when cheaper alternatives already exist (solar, wind, natural gas, storage). We're not increasing our reliance on coal, we're temporarily increasing our reliance on natural gas (which is driving coal and nuclear out of business).


"but pointed out that it taking anywhere near 20 years to build a nuclear power plant is clearly off by an order of 4x at the very least."

I’d prefer nuclear reactors not be built in the developed world to Chinese standards.

I made no claims about overall cost, but pointed out that it taking anywhere near 20 years to build a nuclear power plant is clearly off by an order of 4x at the very least.

So a mixture of unreliable, expensive, and "20 years from now".

Don't forget the chemical byproducts of producing solar and storage.

All options are trade offs, and I don't understand why "perfect" keeps preventing us from just getting "better".

With many recent breakthroughs, fusion is now seen as very viable and just needs funding to finish it off. The last of the hard problems are solved and they just need to pull it all together and "adjust for rounding errors", so to speak.

ITER won't become fully operational until 2035 and that is just for conducting experiments. We are still decades from commercial fusion power plants.

We haven't figured out how clean up or salvage atmospheric CO2 either.

Registration is open for Startup School 2019. Classes start July 22nd.

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