Hacker News new | past | comments | ask | show | jobs | submit login
The Visible Atomic Bomb (unmakingthebomb.com)
98 points by callum85 on Nov 13, 2014 | hide | past | favorite | 41 comments



Coincidentally, I just came across a link someone posted on Reddit to an annotated copy of the so-called "Farm Hall Transcripts."

I had heard of the ALSOS project before. As WWII was winding down in Europe, the Allies attempted to interview as many German nuclear physicists as possible in an effort to learn how far Germany had progressed in their own bomb-making program. But I didn't realize they (we) had essentially captured these scientists and isolated them from contact with the rest of the world, and bugged the building where they were held.

The resulting transcript was declassified in the early 1990s, and it is easily the most fascinating document I've seen all year: http://books.google.com/books?id=pzNjntMMq-oC&printsec=front... You basically get to eavesdrop on Heisenberg, Hahn, and other household names in the physics community as they try to figure out why they're being held, wonder what's going to happen to them, react incredulously to the news of the Hiroshima bombing, blame themselves for it, try to puzzle out how it was done, and try to predict how geopolitics will play out over the coming decades.

Do not click that link if you need to do anything else for the next six hours or so.


There were enormous debates within the nuclear physics community for decades regarding "the German bomb", which many wanted to exist because the possibility of it existing was the justification for the Ally's bomb (bombing Japan was something of an afterthought, albeit one that likely saved tens of thousands of lives on both sides, not that anyone on the Allied side cared about Japanese lives at that point in the war.)

There is a quite good play by English playwright and novelist Michael Frane called "Copenhagen" about Heisenberg's visit to Copenhagen in 1941 to see Bohr, that to my mind convincingly argues for the non-existence of any German bomb program, or at least Heisenberg's complete innocence of it.

SPOILER ALERT: don't read the rest of this comment if you want to see the play and be surprised by the revelation!

================== ================== ==================

I'll bury this in other words so hopefully it'll be difficult for anyone to see if they don't want to, but the basic argument is drawn from the Farm Hall transcripts and the reaction that the German physicists had to the bombings in Japan, which was astonishment, followed fairly quickly by a calculation by Heisenberg that the mass of the bomb might be as little as a thousand kilograms or so. He quickly refined this calculation to something more reasonable, like twenty kilograms, but if he had had any knowledge of or interest in building a bomb he would have done the calculation long before, because a first-order, back-of-the-envelope approximation to the chain reaction based on known properties of uranium is exactly what any physicist would do the very first day they set out to build a bomb. If Heisenberg had been part of a programme to build a bomb he would not have already done that and he wouldn't have needed to do it again. This argument is absolutely compelling to any working physicist, because that is exactly what any of us would do under those circumstances, so there is no doubt that if there was a German bomb program that Heinsenberg was not part of it.


This play was adapted to Portuguese and was played in Lisbon around 2002. I was working at a company that did interactive installations and I was tasked to write software that illustrated the characters physics discussions and projected these on stage under control of the actors via cameras and machine vision. It was one of the most fun projects I worked in.


This is really cool, is there anything similar for hydrogen bombs? I've always been curious how they are different from fission bombs (although obviously a lot of their true workings are still classified).


Probably not. This appears to be the Mark 3 "Fat Man" fission bomb developed by the U.S. during WWII and represents one of the few actual nuclear weapons whose designs are known in reasonable detail (due to the age of the design and some impressive historical detective work[1]).

The detailed layout of components in a real hydrogen bomb is still officially secret and a matter of much speculation. Journalist and activist Howard Morland pieced together a rough outline from unclassified sources and the government sued to block publication of his article.[2]

That suit had the practical effect of confirming at least some of Morland's deductions, and much additional information has come to light since then including at least one alleged cutaway of a real weapon[3].

[1] e.g. http://www.newyorker.com/magazine/2008/12/15/atomic-john

[2]http://fas.org/sgp/eprint/cardozo.html

[3] http://rafaelminuesa.files.wordpress.com/2010/05/morland_ima...


IANANP (I an not a nuclear physicist) but I have read a great deal, and your best bet is to read http://nuclearweaponarchive.org/ which is chock full of interesting information, including a comprehensive list of every nuclear detonation, ever. Go read there about how Hydrogen bombs work.

TLDR for that site; there are two major factors in play with an H-bomb. The first is the addition of a fair amount of deuterium (Hydrogen with an extra neutron) around the explosive core. This acts as a large source of potential energy, and after the initial fission reaction is kicked off, the deuterium ignites and fussion occurs, similar to that which happens on the sun: 2 Hydrogens become 1 Helium, and 2 neutrons go flying off.

The Fission reaction has to be managed differently, however, as there needs to be a secondary reflector system in use. This secondary core reflector system captures the X-rays after the initial fission begins. These x-rays are reflected back into the core, causing the implosion to have a second stage. This "second stage" ignites the third stage fuel, and boom. The documentation will call it only 2 stages, though that x-ray reflection device is an essential part of kickstarting fusion.

The first H-Bomb detonation was essentially a A-bomb surrounded by liquid deuterium in a massive cooling system. The cooling system was actually the size of a building. http://nuclearweaponarchive.org/Usa/Tests/Ivy.html

Of interesting side note is the second H-bomb Castle Bravo: http://nuclearweaponarchive.org/Usa/Tests/Castle.html ) included a Lithium-deuteride solid fuel around the detonation device. The detonation was accidentally the largest ever conducted by the United States. It irradiated a huge swath of the Pacific ocean, and it's follow-up test had similar results.

Devices which used Lithium-6 or enriched Lithium instead of Deuterium would also be considered H-bombs, despite their not having Hydrogen. That's why we call them thermonuclear.

This was because of the added Lithium in the solid fuel. You're basically tripling the amount of neutrons available to the fusion reaction, and thus, the detonation was completely out of hand and exceeded calculations by an exponential factor.


Your description also leaves out the strictly optional, but common 3rd stage of a thermonuclear device: the thick Uranium blanket surrounding the entire assembly. Even "depleted" U-238 will undergo fast neutron fission, which the deuterium / tritium fusion reaction produces in extreme abundance. In most TN weapons, most of the explosive yield results from fission of the Uranium blanket, not the fusion reaction. In a "neutron bomb", or "enhanced radiation warhead", neutrons are the desired product rather than explosive yield, so these typically omit the U casing.

Richard Rhodes' "Dark Sun: The Making of the Hydrogen Bomb" provides an excellent description and paper bibliography.


Hydrogen bombs are basically this plus extra stuff nearby that undergoes fusion when exposed to the fission explosion. Wikipedia has pretty good illustrations and explanations, although not quite to the same level: https://en.wikipedia.org/wiki/Thermonuclear_weapon


I'll second the recommendation for http://nuclearweaponarchive.org/ , especially the "nuclear weapons faq" section, it's a great site though some of the material can be heavy going.

Thermonuclear bombs use a multi-stage design. A nuclear bomb (such as a fission bomb) can be thought of a bit like a light-bulb, only an extremely bright one that is so hot it shines mainly in the x-ray spectrum. A multi-stage thermonuclear bomb takes the light from a fission bomb and traps it inside of a "hohlraum" which is a fancy german word for a box or chamber that happens to be made out of very heavy metals (often depleted Uranium) that is reasonably opaque to the soft x-rays emitted by the nuclear explosion.

Of course, no mere metal container is going to contain the energy of a nuclear explosion for long, the inside surface layers of the hohlraum will absorb some fraction of the x-rays and be vaporized (this is called ablation). This ablation of surface material is so energetic it creates a thrust, like a rocket exhaust coming off the surface of the material it pushes the material away, blowing up the hohlraum like a balloon. The same forces play out on the other occupant of the hohlraum, a capsule containing fusion fuel (often lithium deuteride paste) with a shell made out of similar "high-z" x-ray opaque material. The ablation of the surface of the fusion fuel capsule creates incredible forces, powered as it is by the energy of a nuclear explosion. This causes the fusion fuel capsule to implode, creating conditions of extraordinarily high matter density, pressure, and temperature (due to adiabatic heating). At the center of the fusion fuel capsule there is a fission bomb core, a chunk of Plutonium or Uranium. Because of the speed and strength of the "radiation implosion" (though strictly speaking it's a radiation powered ablation implosion) the amount of fissile material can be small compared to a chemically powered implosion design (keeping in mind that critical mass goes down as density goes up). This "spark plug" then kicks off a super critical fission chain reaction which starts releasing energy that heats up the fusion fuel and kicks off a fusion chain reaction. The fusion reaction then releases a metric crap-ton of very high energy neutrons which facilitate fission reactions including fissioning of depleted Uranium (U-238) in the bomb components (such as the hohlraum, the tamper around the fission primary, the fusion fuel capsule casing, etc.)

In principle, much is the same. The main difference is that the explosive force to implode the main bomb (or the "secondary") comes from a nuclear explosion.


H-bomb works like this: You need to initiate fusion. For this you actually don't need Uranium or Plutonium, you just want fusion to occur in Hydrogen. However fusion only occurs when temperatures are extremely high and kept that way for small slice of time. This can be achieved by - guess what - firing fission bombs (i.e. regular atomic bomb)! However you can't start fusion by firing just one atomic bomb because fusion material will just fly off. You want fusion material to experience extreme pressures and temps. So - again guess what - we fire two or more nuclear bombs precisely at the same time with fusion material in the center. Another way is to use reflector so that energy from fission can be concentrated on fusion material. That's pretty much it. So each H-bomb may carry one or more regular A-bombs.

A side story, which is the most amazing part, is that this whole process was worked out as mathematical theory and huge amount of time and money spent on the faith that math would work out. The math was so complicated that physicists needed computers to do that. That was back in late 40s and computers used valves and almost never worked. The story is that they finally gave up on computer and physicists would sat through many many days and nights calculating everything by hand and double checking. And all that finally worked exactly what math had predicated.


Read the Richard Rhodes books.


This diagram only shows one type of weapon design, and in particular, the type used in the Fat Man bomb. The Little Boy bomb had a different design:

http://en.wikipedia.org/wiki/Gun-type_fission_weapon

[EDIT: Originally had FM and LB reversed. Thanks to droopybuns for pointing this out.]


You're inverted. Little boy is the gun bomb.

Another interesting anecdote is that Little Boy's design was a fall back. The "Thin Man" was plutonium based and discovered to be unlikely to yield an efficient explosion, so they changed to Uranium and made the gun bigger.


You're right. I fixed it. Thanks!


so the explosives cause the jawbreaker like layers of radioactive materials to compress creating nuclear fission that unleashes ton of energy?

So what stops so many other countries from creating such seemingly simple device? Is it the trouble in acquiring the materials?


While this looks simple, many of the details of the construction of the first nuclear devices are still classified. So part of the reason is that simple explanations of the design do not necessarily reflect a simple construction.

On the other hand, you get a taste of a story about how John Phillips got his Undergraduate senior thesis classified for being to informative on how to resolve some important technical details around the actual explosive design for a pit based weapon.

http://www.universitypressclub.com/archive/2013/02/the-a-bom...

Plutonium is difficult to handle and accumulate in significant quantities. Here is a fun video that gives you an appreciation for the danger and rarity of the element: http://www.youtube.com/watch?v=XLufmakbiU0

*Edit: I linked the wrong youtube video. While the above is fun, this one gives you a better idea about the manufacturing difficulties of working w/ Plutonium: https://www.youtube.com/watch?v=89UNPdNtOoE

The bottom left corner of the periodic table is apparently filled with wildly interesting elements that do super bad things to your body. They get hard to come by- and in some cases they only get created in the laboratory. Maybe that's for the best.


It's not quite as easy as it looks. The implosion has to be perfectly symmetrical, or else the core will pop out the side instead of being crushed. Think trying to spear a cherry tomato with your fork. That in turn means that your chemical explosives have to be carefully arranged and they have to be detonated at the precise correct times. In fact, one theory of how the Permissive Action Links (basically, the stuff that makes it so a bomb needs a code to be detonated instead of any random person just pushing a button) work is by encrypting the timing sequence, so that without the code you don't have the information needed to detonate the bomb, rather than relying on some physical lockout. Working out the timing and building the device accordingly are both tough.

Additional difficulty comes in when you want to either increase the bomb's power or decrease its size and weight, both of which are important for making them practical to use against an enemy. There's a lot of work between Fat Man, which weighed 10,000 pounds had a yield of 21kt, and a modern weapon like the W88 which weighs 800 pounds and has a yield of 475kt.

But still, once the basic research was completed and available, producing the materials becomes the primary difficulty in building a bomb.


Making enough plutonium is pretty difficult from what I understand. In WW2 some of the biggest buildings ever built were necessary to house all the equipment to make just enough material for a few bombs (see http://en.wikipedia.org/wiki/K-25). Nowadays you need a nuclear reactor and fancy equipment like uranium centrifuges. See the Stuxnet worm and how it targeted Iran's uranium centrifuges in an attempt to disable their ability to make bomb material.


Here's a TEDx talk from 2009 that goes into some of the difficulties: https://www.youtube.com/watch?v=AeodWP0B-pw

The TLDR is that you have to do a whole bunch of things that give away that you're doing it, take a lot of time and money, make you an easy target to be disrupted in your task, and failing all of that, can kill you pretty easily along every step of the way.


Plutionium is one problem, but AFAIK not the greatest for this type of device.

Greater problem is precise machining and fuse technology necessary, which makes it an enormous and expensive project even today.

The rifle design is just the reverse. There technology is a lot simpler, but rich enough Uranium 235 is a lot harder to come by.


another complexity is that the plutonium core is not just any weapon-grade plutonium: it has to be casted in unstable delta phase (which has unusually low density; the implosion's task is to switch it into alpha-phase which is much dense: this process makes core critical and explosion happens). Not to mention toxicity of the material itself.


See this article from Der Spiegel [1]. Many methods are outlined - including getting unwitting cooperation (wink wink, it's just for energy / research, we promise [2]) from developed nations - along with the caveat that until the bomb is ready it's best to deny, deny, deny and avoid all the headache.

The article discusses Israel, South Africa, Brazil, Libya, and of course Iran.

That said, being part of the nuclear club means a huge responsibility - assuming you just want influence and don't want to kill somebody, now you've got to make sure nobody messes with your new toy or tries to take it.

[1] http://www.spiegel.de/international/world/the-bomb-for-begin...

[2] "Um ... we just want to learn more about submarine propulsion, nothing else. Trust us."


Not much at all, and an interesting consequence of this is that manufacturers of high-precision CNC machines have to be very selective about which countries they're willing to sell to, and many of these machines carry GPS trackers and other sensors which will alert the manufacturer and prevent the machine from functioning if somebody tries to move it.[1]

[1] https://www.techdirt.com/articles/20140109/03060325817/lates...


From reading about Stuxnet, I believe they also have something involving IP address pings and feedback.


Nothing.

Remember: this was done the first time in less than four years, from 9 October 1941 to 16 July 1945, using technology from the stone age.

No solid-state electronics beyond crude diodes. No electronic computation. Approximate estimates on most of the cross-sections. Hopeful guesses on many of the processes (including multiply redundant approaches to some things, including the basic design, because people weren't sure what would work or what would be possible to build.)

Today, a team of reasonably intelligent engineering students with a budget of a few billion dollars could build a bomb in a year, simply because they would know vastly more than the original team.

Again: it took three years and $2 billion ($30 billion in today's dollars) to build the very first bomb from scratch knowing almost nothing about the process and having to custom build and design and invent almost every component.

To suggest that any particularly rare expertise today would be required or that the cost would be more than a tenth of the original development is implausible.

Remember this the next time you hear about Iran's nuclear program, which has been going on for twenty years, supposedly in mad pursuit of nuclear weapons. It's a ridiculous claim: if the Iranians wanted a Bomb very badly they would have one by now. If they don't have one by now the only reasonable conclusion is that they don't want one very badly.

The extreme disapprobation they get from the rest of the world on this issue may well contribute to their tepid enthusiasm, but no amount of Stuxnet and the like is going to put a dent in a strongly committed program. The Iranian's commitment just isn't that strong.

Gas centrifuge technology in particular has been a game-changer in the lowering the nuclear bar over the past couple of decades. People were warning about it in the '80's, and their warnings seem to have had some prescience. It makes practical enrichment of natural uranium to weapons grade, bypassing the need for a reactor program to breed plutonium, and all the nasty, difficult chemistry involved in plutonium extraction.

And unlike plutonium weapons, uranium weapons are ridiculously easy to detonate. Trinity was a plutonium test. Little Boy was one of the very few uranium weapons ever detonated, and it was live tested because the scientists were as certain as anything it would actually explode.

There are uraninite (pitchblende) ores that are getting on for 10% uranium, or 0.07% 235U, so a tonne of ore would yield 7 kg of 235U. It'll only get worse as seawater extraction gets better (it's already capable of producing macroscopic amounts of uranium from the relatively uranium-rich Japan Current, although post-Fukushima the primary research, which was Japanese, has likely been curtailed.)

Processing a couple of tonnes of ore is sub-industrial. It isn't that hard to hide. On the other hand, if countries want to go the plutonium route, plutonium producing reactors aren't hard to hide either: the Russians were running them underground--a few hundred metres below the surface--in the 80's and 90's.


All correct, and last but not least: You know for sure it can be done. That alone is a huge motivator.


I've always read that the primary difficulty in building a nuclear weapon isn't necessarily the difficulty in producing the bomb or detonator, but in producing the fissile material in large quantities. (U235 is something like 0.7% of bulk uranium we find in nature, and separating it from U238 is prohibitively difficult. You need 80% for "weapons-grade" according to wikipedia.)


It really makes me wonder if a non-government entity is currently a nuclear power.


Following your reasoning, it makes sense why North Korea was able to produce a small amount of it in a relatively short time. They were well funded thanks to the left-leaning South Korean administrations from 1998-2008 who practiced Sunshine policy. The last bit is that they certainly want it bad enough but have failed to produce a nuclear weapon with significant power (nothing in the range of WW2 nuclear weapons). So if a country that wants it so bad like North Korea have been able to produce a controlled nuclear explosion, why can't they produce anything significant enough to be recognized as a nuclear state? The only thing that would stop them seems to be either expertise or lack of key material or perhaps lack of monies.


Producing some kind of nuclear explosion is mostly only about procuring the raw materials. Getting to significant yield is the actual hard part as it involves precision machining, measurements and timing.

Machining heavy metals is complex problem in itself even without requirement of extreme precision. It is hard to procure machining equipment with sufficient precision and tooling capable of machining hard metals without raising lots of red flags.

Almost anything that is capable of switching of significant powers with small or repeatable latency is essentially non-exportable with the reasoning that it is not useful for much else than triggering nuclear weapons (one could think of lots of other applications, but the devices are too bulky and expensive for most of them).


This is off-topic, but the agreement between North Korea and the US was made in 1994 (well before there was any "left-leaning" president in South Korea), in which the US agreed to pay for two Light Water Reactors in exchange for the North Korea discontinuing its nuclear program[1].

As expected, South Korea ended up funding most of the cost, while both the US and NK violated the agreement in numerous ways until the deal completely broke down in 2003. The reactors never saw the light of the day. Bummer.

There's a story (not sure how much of it is true) that, shortly before the deal, the Clinton administration went within hours of air strike on North Korean nuclear facilities, while Seoul was completely left in the dark. The realization that South Korea was essentially reduced to a spectator (and a deep pocket) in NK-related matters doubtlessly affected the two following administrations' policy toward North Korea, trying to have at least some of our own leverage. ("The US will take care of us" is not a strategy.)

But we get to blame our "leftist" presidents for their nukes. Niiice...

[1]: http://en.wikipedia.org/wiki/Agreed_Framework

> Terms of the pact and consequent agreements included ... In exchange two light water reactors would be constructed in North Korea by 2003 at a cost of $4 billion, primarily supplied by South Korea.


In the case of North Korea it is likely the raw material they lack. They are a small and insular country, and may not have any significant indigenous reserves of uranium ore.

This is a very good thing, and unfortunately one that sea-water extraction technology--which is otherwise a very good thing because it makes natural uranium a renewable resource--may eventually have an effect on.


Yes, in fact the website in the OP is for a book which advocates for a disarmament approach based on eliminating weapons-grade fissile material stockpiles:

http://unmakingthebomb.com/sample/


Main thing that stops any country is availability of refined fission material. First you need to have access to Uranium mines which US and Russia pretty much has covered all over the world. Second, you need to dig tons of ore and go through very tedious and complicated refinement process that is very expensive and requires very large scale operation. To avoid these, countries like Iran would just use the material generated from reactor fuel that can be obtained more easily.


Yep, also enriching uranium to anything near weapons grade is a cumbersome process.


What stops so many other countries from building nuclear weapons?

That's an interesting question, perhaps the best answer is simply affluence, the perception of geopolitical safety, and the perception of unwanted negative repercussions of obtaining nuclear weapons. Any developed country with a significant industrial base can build nuclear weapons. Trivially so for countries with significant local nuclear power industrial infrastructure such as fuel reprocessing or Uranium enrichment. Brazil, Mexico, Canada, Japan, South Korea, Germany, all these countries, and more, could build nuclear weapons quite rapidly and in great quantity just by deciding to do so.

For everyone else, well, one hopes that it is sufficiently difficult and costly that it's just not worth it for most countries, and difficult to hide as well, though mostly that's a lie we tell ourselves to make us feel safe. The hardest part is acquiring the fissile materials, which turns out to be slightly tricky. Plutonium requires production in reactors, but more so it requires a rather inefficient process to produce weapons grade Plutonium in reactors. Reactors are hard to hide, and if a reactor is under IAEA observation it's non-trivial to use the reactor to produce Plutonium. I say non-trivial but it is certainly far from impossible, as it's been done, but it makes it difficult to produce in large quantity while under the IAEA's eyes.

Uranium enrichment is a bit of a different story, and not a particularly heartening one either. It does require very sophisticated equipment and significant industrial resources to enrich Uranium, but it turns out that this equipment can be hidden fairly well and it actually takes much less of an investment to create enrichment facilities for making weapons versus for making reactor fuel (the enrichment level needs to be much higher, but the quantity needs to be vastly smaller). For example, to process natural Uranium into 5% enrichment takes about 75% of the work as to process it into 90% enrichment (weapons grade), and you only need perhaps 16kg or so of highly enriched Uranium to make a bomb. Getting back to observability, these facilities can be easily hidden and operate without notice since they blend in well with other industrial facilities externally. The world learned of Saddam's vast Calutron facilities only due to the first Gulf War. And the world learned of North Korea's centrifuge cascades only when that regime announced their existence.

In the last 50 years half a dozen countries have acquired nuclear weapons without sanction from existing nuclear powers and in most cases without the world being certain of the existence of those nuclear programs until they were announced (such as through weapons tests). That gives some perspective on the true difficulty of preventing nuclear weapons proliferation.

As a side note, if the entire worldwide civilian Uranium enrichment capacity was diverted from making reactor fuel to making nuclear bombs it could produce around 400 nuclear bomb cores per week.


tres cool


This is pretty cool.


Here's the funny thing: If you had 6kg of U-235 and through it forcefully on the ground it would most likely explode. This almost happened at Las Alamos. All the machinery in nuclear fission bomb essentially just to protect unintentional explosion and make sure explosion is symmetrical and as vigorous as possible when triggered. The triggering mechanism is essentially firing usual explosives around the sphere of fission material.


> If you had 6kg of U-235 and through it forcefully on the ground it would most likely explode.

First, you mean "threw", not "through." And second, this is absolutely not true. To achieve an explosion you have to induce super-criticality with very precise timing. Otherwise the uranium will melt and become sub-critical before it explodes.

> This almost happened at Las Alamos

No, it didn't. You are almost certainly thinking of these incidents:

http://en.wikipedia.org/wiki/Demon_core

but those were completely different. It was a plutonium core, not uranium, and it went critical not because it was thrown forcefully to the floor, but because it was inadvertently enclosed in a neutron reflector that caused it to become critical.


>To achieve an explosion you have to induce super-criticality with very precise timing. Otherwise the uranium will melt and become sub-critical before it explodes.

True, but the complexity of the Fat Man plutonium bomb (shown in the link here) belies the simplicity of Little Boy uranium bomb.




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

Search: