How easy is it to build a nuclear bomb?

[shadron]

Oddly enough, the implosion devices tend to be inherently "safe" from accidental detonation from, say, an airplane crash or a fire. Sure, the explosive charges go off and scatter nuclear material around, but because the detonation sequence is essentially random, the right containment needed for nuclear detonation doesn't happen.

Let me tell you a little story about that. Bomb design in the fifties changed from the spherical implosion model to a elliptical mode, much better for fitting inside of casings. It also moved to what was known as the sealed pit, where the core and the initiator were sealed against contamination permanently. There were only two detonators instead of the 16-100 used in the spherical bombs. With only two detonators required, the obvious question is, "What happens when one of the goes off, by accident or in a plane crash or what have you?" See nuclear weapon design^WP.

There were a whole series of tests to find out, for various combinations of cores and initiators. The series known as Project 56, 57, 58 and 58A, as well as various parts of Plumbbob, Hardtack II and Nougat. As it turned out, 15 of 22 test essentially failed by exploding in excess of the HE in the bomb. Fortunately that got better as time went on.

The interesting one was Plumbbob/Pascal-B. The bomb was lowered in a shaft, and a 400 lb, 4" think steel plate placed 2 meters above it to seal the shaft. There was some interest in what the plate would do, so a high speed camera was focused on the top of the shaft. The nuke was tested, and as had been true for several of these tests, it exploded with a yield if 300 tons equivalent. The camera caught one frame of the steel flying above the shaft. A computation was made; the disk was accelerated in the shaft to 6x Earth's escape velocity. Needless to say, they never found the disk; they figure it either burned up from friction, or is in solar orbit. This was August 8, 1957, two months before Sputnik. You can read about it here: http://nuclearweaponarchive.org/Usa/Tests/Brownlee.html
and http://nuclearweaponarchive.org/Usa/Tests/Plumbob.html, about 1/3rd down the page.

 

[nvidiot]

Ooooh.

A nuclear potatoe gun! Sweeeeeet.

 

[shadron]

Isn't it possible to build a simple weapon from dropping one just under critical mass onto another from a sufficient height? Far from efficient but it would that still be a nuclear explosion?

As it happens, that was one of the ways in which the "Tickling the Dragon's Tail" experiments to determine exact critical masses was determined, by dropping a near critical plug through the center of another near critical piece and watching the Geiger counter.

However, for a bomb things have to happen faster. What they did for Thin Man, the Hiroshima bomb, was to fire the critical mass down the tube of a naval 6" gun barrel using several bags of cordite. At the point when the two masses were together, the initiator was smashed, mixing a couple of materials which throw out a shower of neutrons to start the fun. The tricky thing was that they needed to bring together in excess of two times the critical mass to insure it all worked. Screwed onto the business end of the cannon barrel was a cylinder of U-235, about 90% of a critical mass. The bullet was a hollow cylinder that just fitted over the solid cylinder at the end, of about 120% critical mass, kept non-critical by it's geometry as a hollow cylinder.

As it turns out, they could only do this with U-235. Plutonium was much faster to react than U-235, and the bomb would fizzle as the bullet neared the target at about Mach 2. This was known by 1943; the only practical way to use plutonium was with super-critical implosion.

 

[shadron]

Its not that hard. The hardest part is getting the fissionable material.

Second hardest is building a good implosion sphere or lense shape. For the sphere two different explosives had to be molded precisely to fit together without seams. The lense shape is much more difficult, as it has to be created in two pieces with controlled mixing of the explosives to squeeze the elliptical core. When Lawrence Livermore Labs (then known as UCRI) got into the bomb business in 1955, to the obvious displeasure of Los Alamos, there was little cross-fertilization, and it took 2 years and about 6 fizzles before they created a working bomb. North Korea has fired at least one bomb, which was judged a fizzle by the experts.

 

[dtugg]

Actually, most now use a product called Uralloy, which is a mixture of Pu and U-235. The U-235 is apparently cheaper to produce than the plutonium. No, the Manhattan Project made enough for four bombs (Nagasaki, the demon core used in Xroads Baker, Xroads Able and another for Xroads Charlie that was cancelled), and by the Spring of 1946 was ramping up towards two cores a month.

As with deciphering codes, by far the easiest way to obtain fissionables is to steal them.

Nagasaki was a plutonium devices. So was the "demon core".

 

[shadron]

You think all mushroom clouds are alike? See this: http://en.wikipedia.org/wiki/File:Nukecloud.png

 

[shadron]

Nagasaki was a plutonium devices. So was the "demon core".

Yes, that's correct. Uralloy was started in use in about 1950.

 

[firecoins]

Well how hard is it? I have some knowledge about particle physics and related things but i have no practical knowledge about hard it would be. Let's discard the missile.

What things would you need? How easy would it be for people to discover the project? I mean you see the US and the Soviets having thousands upon thousands of nuclear weapons but then you see a country like north korea struggle to achieve just a little fizzle...

It comes in a set much like a child's science project or a magic set seen in stores. I picked mine up in Pakistan.

 

[aviolet4u]

super easy

 

[casebro]

If it was easy, LOTS of people would do it.

So far as we know, only countries have managed it. Not even mobs.

So far as we know...

 

[catsmate]

It's also to see that with your reply you provide some more complexities that I did not think of when writing my post. One being: what is the relationship between the degree of enrichment of U and the amount needed?

If I'm not mistaken during the Manhattan project the implosion route for a Pu based bomb was taken exactly to circumvent the predetonation problem. I don't know whether implosion designs are not prone to predetonation at all.

Yes the 'Thin Man' or MK 2 bomb was intended to be a Plutonium gun barrel design but was abandoned due to the 'fizzle' problem. The Plutonium produced from the reactors had too much Pu²⁴⁰ for this to be workable and a Plutonium gun-barrel weapon would need far greater velocities and hence barrel length. It is possible to build such a weapon with Pu containing less Pu²⁴⁰.
In an implosion design the time period over which the sub-critical form
(sphere) is converted into super-critical form (dense sphere) is too small for this to be an issue; the main failure route in implosion weapons is out of sequence detonation of the lenses.
FYI the actual blueprint of the 'Fat Man' design is now unclassified.

I'm confused. Do you mean that the presence of U238 wrt predetonation isn't a problem 'cause U238 doesn't fission spontaneously? I believe the problem of the presence of U238 is of a different nature: it absorbing neutrons that are thus no longer able to sustain a chain reaction.

OK I must have been unclear.
Actually U²³⁸has a far higher spontaneous fission rate than U²³⁵; however is U²³⁸ less susceptible to induced fission than U²³⁵. If the Uranium contains high levels of U²³⁸ a pre-detonation is possible during the detonation sequence.
The 'Little Boy' or MK 1 bomb design used 80% HEU to avoid this problem, and the usual weapons grade Uranium (>93% U²³⁵) means this isn't an issue. 50% HEU which could be used to create a fission bomb would be too dangerous for gun-barrel designs.
In general the gun-barrel design is only used when required by operational constraints, e.g. artillery shells. Implosion designs are safer and more efficient.

Critical masses for Uranium range from ~250kg at 40% U²³⁵ without a neutron reflector to less than 10kg at 94% with a Be/W reflector.

I understood that for maximum efficiency one wouldn't want to rely on an ambient neutron or a spontanious fissioning to start the chain reaction. The critical assembly might have less than optimal density/geometry by the time a spontaneous neutron arrives.

IIRC Klaus Fuchs spent a significant amount of time and effort during the Manhattan project designing a device that assured to that: release a bunch of neutrons at the moment of maximum density. Littlle Boy had a initiator as well, I believe.

Yes both the MK 1 and MK 3 bombs had internal neutron sources ('Urchin' and 'Abner') separate from the fissile material. Modern designs use external neutron sources for greater efficiency. Both initiators were the standard 'neutron howitzer' with Be/Pu producing neutrons when compressed together.
I'm not aware of Fuchs being involved in their design; Tuck certainly was but it's been a while since I studied nuclear engineering in depth.

I must dig of my copy of the Los Alamos Primer......

 

[geni]

It is very hard to build a nuclear bomb without outside help. Several countries have tried and failed.

Very few countries have tried. The first step in any budding nuclear program tends to be to aquire as much as possible from someone else.

Possible exception is south africa depending on you view with regards to the claims of collaboration with Israel.

 

[Ixion]

Plutonium and U-235 are alpha emitters with loooooong half-lives, which means they are not very intense. They are poisonous, similar to arsenic in that regard, and they are pyrophoric, which means they'll ignite at room temperature. Usually the core components are plated in nickel or some such to keep them from combusting (which only happens readily when they are thin shavings) or oxidizing. In that wrapper they are human handleable. Just don't inhale dust or gas emanations.

Plutonium is not very toxic. It can cause a variety of afflictions from radioactivity, but the toxicity is not that high (wiki says similar to caffiene).

 

[nescafe]

Isn't the amount needed variable with the speed of impact? That implosion concept is used to allow smaller amounts. So the trade off is, lots more precision in an implosion device, vs lots less Plutonium. In Tom Clancy's novel, they used lens grinding machines, and it still fizzled. Not many lens grinding machines in home shops.

The bomb in that novel fizzled because it was supposed to be a fusion weapon, but the terrorists killed off the guy who supplied the tritium before he could separate out the helium-3, which poisoned the reaction, and left them with a piddling little Hiroshima-style blast instead of the few hundred KT blast they were looking for.

IIRC, I read that book more than 10 years ago.

 

[GlennB]

I wonder how the device in "The Fourth Protocol" would stand up to this analysis? Great book, mind.

 

[Noztradamus]

Originally Posted by dtugg
Pretty much all nuclear weapons use plutonium 239. It is made in a nuclear reactor and somehow separated from the other material (don't ask me how they do this, I dunno). This is most certainly not something that an individual could do in his garage. You can also make a nuclear bomb using uranium 235. But it is very hard to enrich from natural uranium in the necessary quantities. The entire Manhattan project only made enough for one bomb.

Actually, most now use a product called Uralloy, which is a mixture of Pu and U-235. The U-235 is apparently cheaper to produce than the plutonium. No, the Manhattan Project made enough for four bombs (Nagasaki, the demon core used in Xroads Baker, Xroads Able and another for Xroads Charlie that was cancelled), and by the Spring of 1946 was ramping up towards two cores a month.

As with deciphering codes, by far the easiest way to obtain fissionables is to steal them.

They were plutonium cores for implosion bombs

It was actually something like 1950 before the US had enough U-235 available to put another Hiroshima type bomb in service.

They had enough U-235 before this. However 30Kg Pu and 60Kg U-235 could build 5 MkIII Implosion Bombs and One MK-1 Gunbomb

After the 1948 Sandstone tests the same amount of explosive metal could built 12 Mk4 bombs with mixed U/Pu Implosion cores.

 

[Noztradamus]

Very few countries have tried. The first step in any budding nuclear program tends to be to aquire as much as possible from someone else.

Possible exception is south africa depending on you view with regards to the claims of collaboration with Israel.

South Africa did it on their ownsome. If they had any help from anyone they would have ended up with more than the half dozen simple Hiroshima-type bombs they produced.

Everybody read the Afrikaaner-bomb wrong (I remember the assessments from the 80s - hundreds of warheads, fusion bombs, neutron bombs, 155mm nuclear howizer shells) from the size of their U-235 production facilities. Who could have imagined the reason they were so large was because the SAs went simple and crude, instead of trying to develop implosion.

 

[catsmate]

Very few countries have tried. The first step in any budding nuclear program tends to be to aquire as much as possible from someone else.

Possible exception is south africa depending on you view with regards to the claims of collaboration with Israel.

The South African weapons were Uranium gun-barrel designs; their weapons programme was based on a clever locally developed system for enrichment of natural Uranium without the need for reactors to breed Plutonium. This was an unusual approach, most countries developing on planning to develop nuclear weapons use reactor bred Plutonium rather than enriching Uranium to the necessary levels.

 

[catsmate]

The bomb in that novel fizzled because it was supposed to be a fusion weapon, but the terrorists killed off the guy who supplied the tritium before he could separate out the helium-3, which poisoned the reaction, and left them with a piddling little Hiroshima-style blast instead of the few hundred KT blast they were looking for.

IIRC, I read that book more than 10 years ago.

Technically it was a boosted fission weapon, the fusion reaction generated neutrons to make the fission more complete rather than adding significantly to the yield. The physics is quite plausible, He³ will absorb neutrons, and the decay of Tritium in weapons is a problem with such weapons, necessitating fairly frequent servicing to replace the Tritium.

 

[Gytu]

South Africa did it on their ownsome. If they had any help from anyone they would have ended up with more than the half dozen simple Hiroshima-type bombs they produced.

What I find very funny about nuclear bombs is that they are never used. Countries usually build them not as weapons but as a bargaining chips. For example, North Korea has been building a nuclear bomb for the past decade.They have threatened the United Nations by saying that if they don't receive food and supplies, they're gonna get it. Another example is Russia, which was locked in a nuclear arms race with the US til 1990 when they finally collapsed. It was the result of spending all their resources on nuclear bombs that were never going to be used.