# Why are nukes considered hard to build? **URL:** \ **Category:** Factual Questions\ **Created:** [March 3, 2009, 5:00pm UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159 "2009-03-03T17:00:51Z")\ **Posts on this page:** 20\ **Page:** 3
**Author:** ![billfish678](https://avatars.discourse-cdn.com/v4/letter/b/7bcc69/32.png) [@billfish678](https://boards.straightdope.com/u/billfish678)\ **Post date:** [March 5, 2009, 2:10pm UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/42 "2009-03-05T14:10:55Z")
> [@SenorBeef](#): > > When the US was working on refining uranium as part of the Manhattan project in Tennessee. At its peak the program took 1/6th of the entire US electrical generation capability, mostly running the cyclotrons. That must have been an expensive pain in the ass. If only they had used solar power instead 🙂 Someone asked about plutonium. You have to make it. You generally make it by bombarding uranium atoms with neutrons. Running nuclear reactors have LOTS of neutrons bouncing around in there. Some of the uranium in the reactor becomes plutonium. An analogy of why plutonium is preferable to manufacture. Imagine you have something like sand. Except every grain is virtually identical. Well, not quite. There are two major kinds of sand. One type weighs about 0.5 percent less than the other, and only makes up a few percent of the total number. Its gonna be a pain the ass the separate them and you will have to go through ALOT of sand to get the good grains. Now, imagine you have two kinds of sand particles. One is the uranium, and one is the plutonium. One sinks and one floats. Much easier to separate the two. ---
**Author:** ![Bijou\_Drains](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bijou_drains/32/2814_2.png) [@Bijou\_Drains](https://boards.straightdope.com/u/Bijou_Drains)\ **Post date:** [March 5, 2009, 2:48pm UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/43 "2009-03-05T14:48:37Z")
One of the reasons Oak Ridge was picked was that TVA had a lot of generating capacity for power. Also it was a very isolated area back then. I think back then TVA only had hydro power but that might have had some coal plants too. Ironically 20-30 years later TVA started to build a bunch of nuke plants but many of them were cancelled before completion. ---
**Author:** ![MichaelQReilly](https://avatars.discourse-cdn.com/v4/letter/m/df705f/32.png) [@MichaelQReilly](https://boards.straightdope.com/u/MichaelQReilly)\ **Post date:** [March 5, 2009, 3:53pm UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/44 "2009-03-05T15:53:16Z")
> [@Otara](#): > > As in you cant just buy plutonium on the international market, attempting to do it yourself gets some fairly heavy attention, getting the tools to build things is difficult, you have to conceal or bomb harden your facilities etc etc. > Otara I thought that by 1985 we’d be able to just walk into a drugstore and buy plutonium, no? ---
**Author:** ![Alex\_Dubinsky](https://avatars.discourse-cdn.com/v4/letter/a/dbc845/32.png) [@Alex\_Dubinsky](https://boards.straightdope.com/u/Alex_Dubinsky)\ **Post date:** [March 5, 2009, 11:15pm UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/45 "2009-03-05T23:15:59Z")
> [@billfish678](#): > > Someone asked about plutonium. You have to make it. You generally make it by bombarding uranium atoms with neutrons. Running nuclear reactors have LOTS of neutrons bouncing around in there. Some of the uranium in the reactor becomes plutonium. > > Now, imagine you have two kinds of sand particles. One is the uranium, and one is the plutonium. One sinks and one floats. Much easier to separate the two. So that’s it? You just build an ordinary nuclear reactor and it will make plutonium for you? Every reactor does this? None of the new generation reactors are any different? So if nuclear power becomes the dominant electricity (and energy) generation method in the future, the world will be overflowing with plutonium? And anyone who has the plans for the correct implosion charge can make themselves a bomb? (Not in a basement, but with reasonable equipment.) ---
**Author:** ![billfish678](https://avatars.discourse-cdn.com/v4/letter/b/7bcc69/32.png) [@billfish678](https://boards.straightdope.com/u/billfish678)\ **Post date:** [March 5, 2009, 11:24pm UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/46 "2009-03-05T23:24:31Z")
> [@MichaelQReilly](#): > > I thought that by 1985 we’d be able to just walk into a drugstore and buy plutonium, no? Only if you go back to future to the past or something like that 🙂 ---
**Author:** ![beowulff](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/beowulff/32/542_2.png) [@beowulff](https://boards.straightdope.com/u/beowulff)\ **Post date:** [March 5, 2009, 11:52pm UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/47 "2009-03-05T23:52:33Z")
> [@Alex\_Dubinsky](#): > > So that’s it? You just build an ordinary nuclear reactor and it will make plutonium for you? Every reactor does this? None of the new generation reactors are any different? > > So if nuclear power becomes the dominant electricity (and energy) generation method in the future, the world will be overflowing with plutonium? And anyone who has the plans for the correct implosion charge can make themselves a bomb? (Not in a basement, but with reasonable equipment.) Plutonium is a byproduct of all Uranium reactors, but some make more than others. The [Fast Breeder](http://en.wikipedia.org/wiki/Fast_breeder) is much more efficient in its use of fuel, but because it makes so much Plutonium, it hasn’t been widely used. The Plutonium proliferation argument is one of the more successful and hard-to-refute arguments used by the anti-nuclear energy crowd. ---
**Author:** ![billfish678](https://avatars.discourse-cdn.com/v4/letter/b/7bcc69/32.png) [@billfish678](https://boards.straightdope.com/u/billfish678)\ **Post date:** [March 6, 2009, 12:06am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/48 "2009-03-06T00:06:20Z")
> [@beowulff](#): > > The Plutonium proliferation argument is one of the more successful and hard-to-refute arguments used by the anti-nuclear energy crowd. I dont know about that. Sucessful “maybe”. Hard to refute “maybe”. but… Thats kind of like some people screeching that passenger planes are not safe, therefore passenger planes dont exist because of these people, therefore nobody is dying from plane crashes. See? We were right ! :dubious: ---
**Author:** ![Otara](https://avatars.discourse-cdn.com/v4/letter/o/a88e4f/32.png) [@Otara](https://boards.straightdope.com/u/Otara)\ **Post date:** [March 6, 2009, 2:04am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/49 "2009-03-06T02:04:49Z")
"I thought that by 1985 we’d be able to just walk into a drugstore and buy plutonium, no? " Well it might sound kind of obvious, but some countries have the material to build thier own and some dont, and its civilians uses are considered to be outweighed by the military risks, so trade is restricted. None of those are engineering obstacles as such. Iran has its own uranium supplies, which is why its major barriers are political rather than engineering, due to attempts to be seen to still comply with arms control treaties. If it ignored them and didnt have to worry about reprisals, it would only be a matter of time till it could make them. Otara ---
**Author:** ![Alex\_Dubinsky](https://avatars.discourse-cdn.com/v4/letter/a/dbc845/32.png) [@Alex\_Dubinsky](https://boards.straightdope.com/u/Alex_Dubinsky)\ **Post date:** [March 6, 2009, 2:43am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/50 "2009-03-06T02:43:02Z")
> [@Alex\_Dubinsky](#): > > So that’s it? You just build an ordinary nuclear reactor and it will make plutonium for you? Every reactor does this? None of the new generation reactors are any different? Hmm, from the Wikipedia page about Fast Breeders: “there are some designs such as the SSTAR which avoid proliferation risks by both producing low amounts of plutonium at any given time from the U-238, and by producing three different isotopes of plutonium (Pu-239, Pu-240, and Pu-242) making the plutonium used infeasible for atomic bomb use.” Of course it’s not quite “infeasibile.” You’re simply back to the enrichment problem. From HowStuffWorks re uranium centrifuges: “The creation of the centrifuges is a huge technological challenge. The centrifuges must spin very quickly – in the range of 100,000 rpm. To spin this fast, the centrifuges must have: very light, yet strong, rotors well-balanced rotors high-speed bearings, usually magnetic to reduce friction Meeting all three of these requirements has been out of reach for most countries. The recent development of inexpensive, high-precision computer-controlled machining equipment has made things somewhat easier. This is why more countries are learning to enrich uranium in recent years.” Sigh. Nukes are now firmly on my list of why the future’s gonna suck balls. Btw. No one answered the question: what’s the leap in difficulty of turning an atom bomb into a hydrogen one? ---
**Author:** ![gethsemenerose](https://avatars.discourse-cdn.com/v4/letter/g/a587f6/32.png) [@gethsemenerose](https://boards.straightdope.com/u/gethsemenerose)\ **Post date:** [March 6, 2009, 4:56am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/51 "2009-03-06T04:56:56Z")
Atom to H2 is not a big leap. IMHO the biggest thing keeping “Basement Nukes” out of play is the Radiation killing the bomb makers. The Manhatten project story already mentioned and another Incident with the army reactor in Idaho show exactly how bad that can be. Getting the Material in the correct configuration without accidently putting it into a Nuclear Power configuration (Eating up {or “Poisoning” }the Neutrons ) is a very real deterent. and a very hard step to get around. Signed [“The Gnome”](http://www.cafepress.com/gethsemenerose) ---
**Author:** ![Stranger\_On\_A\_Train](https://avatars.discourse-cdn.com/v4/letter/s/13edae/32.png) [@Stranger\_On\_A\_Train](https://boards.straightdope.com/u/Stranger_On_A_Train)\ **Post date:** [March 6, 2009, 5:04am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/52 "2009-03-06T05:04:41Z")
> [@Alex\_Dubinsky](#): > > I think the real point of computers is that you don’t need to discover and figure out how to apply all the necessary equations. They’re all embedded there in the program. You still need to use the program and that takes significant knowledge, but much less than trying to _be_ the program. The speed of computation is just icing on the cake. What is this “program” that is going to automagically design a nuclear weapon or perform a simulation for you? You seem to be espousing the same kind of naive reliance on the ability of a piece of software to substitute for interdisciplinary knowledge, practical experience, and engineering judgment that I’ve heard throughout my career. The reality is that however good the software is, be it CAD, finite element analysis, computational fluid dynamics, the quality of the results depend on setting the parameters of the model and providing the correct inputs. In other words, it’s not enough to be a proficient in driving the software (what we call a “CAD chimp” or “mesh monkey”); you need to know how to set up the problem and interpret the results, and realize when your idealized simulation may be missing some of the uncertainties of a real-world application. > [@Alex\_Dubinsky](#): > > It’s all in the theme of the modern spread of knowledge. (P.S. when people colloquially say ‘blueprints’, they mean any kind of design information. Could even be a step-by-step tutorial for n00bs.) There is no “step-by-step tutorial for n00bs” for designing a nuclear weapon any more than there is a Chilton manual that tells you how to build a spacecraft. > [@Alex\_Dubinsky](#): > > So that’s it? You just build an ordinary nuclear reactor and it will make plutonium for you? Every reactor does this? None of the new generation reactors are any different? > > So if nuclear power becomes the dominant electricity (and energy) generation method in the future, the world will be overflowing with plutonium? And anyone who has the plans for the correct implosion charge can make themselves a bomb? (Not in a basement, but with reasonable equipment.) Manufacturing and separating weapons grade plutonium ([sup]239[/sup]Pu) is actually very time consuming and difficult. While it is possible to chemically separate uranium and plutonium (which is why there was such an impetus to develop implosion weapons) the irradiation of [sup]238[/sup]U to produce [sup]239[/sup]Pu inevitably produces a fraction of [sup]240[/sup]Pu and [sup]241[/sup]Pu, both of which will “poison” a fast nuclear reaction by absorbing neutrons. In addition, [sup]240[/sup]Pu is a strong alpha emitter, and both have high rates of spontaneous emission. This can cause a phase change in the nuclear material making it unstable. These isotopes are even more difficult (in fact, technically infeasible in useable quantities with current centrifuge technology) to separate than isotopes of uranium, and so the cycle for weapons grade plutonium production involves frequent reprocessing to achieve a product that is at least 93% [sup]239[/sup]Pu (and frequently higher for more modern weapons). You can’t just use plutonium from a normal energy production fuel cycle and obtain a workable or reliable weapon. Regarding building a fusion weapon, it is in many ways much easier than building the fission Primary (though the fact that you need a fission explosion to produce the conditions for fusion still harkens back to the above discussion) and the [sup]6[/sup]Li-D fusion fuel is commercially available and easy to handle; however, the need to precisely simulate and control not just the chemical explosive lenses for the fission reaction but the radiation pressure from the Primary requires very precise modeling and manufacture. In addition, because you to get enough radiation from the Primary before it blows the fusion Secondary apart, Primaries are usually a boosted fission design, requiring tritium to provide a neutron source. (This is how “dial-a-yield” weapons work, by gauging the amount of tritium inject.) Tritium is not commercially available in the requisite quantities (or at least, not without attracting enough attention) and the production of tritium is currently highly restricted. Nuclear proliferation will continue apace, and it seems likely that sometime in this century there will be an exchange between nuclear powers, but barring some unforeseen developments it is unlikely that it will become the provence of anything below an industrial nation, at least in any practical sense. If you want to fear for the future, consider the rapidly developing ability to manufacture and modify living and life-like microorganisms at will, to the point that it may one day be feasible for a moderately experienced individual with a computer and a gene sequencer to built custom viruses and bacteria capable of infiltrating and affecting the human body. Stranger ---
**Author:** ![Otara](https://avatars.discourse-cdn.com/v4/letter/o/a88e4f/32.png) [@Otara](https://boards.straightdope.com/u/Otara)\ **Post date:** [March 6, 2009, 5:17am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/53 "2009-03-06T05:17:39Z")
My understanding is hydrogen requires extensive work and research, as you pretty much need to get fission weapons working before you can even start on trying to develop the fusion side of things, because they’re the trigger. Then you need to make your fission weapon fire in the right kind of way to get the fusion aspect to work, which is apparently quite difficult and needs extensive experimental as well as theoretical work to perfect. So unless you can get the plans, you have a lot of work to do, with test firings etc before you end up with a reliable weapon. You will umm not be popular while making said test firings. Otara ---
**Author:** ![EvilTOJ](https://avatars.discourse-cdn.com/v4/letter/e/ee59a6/32.png) [@EvilTOJ](https://boards.straightdope.com/u/EvilTOJ)\ **Post date:** [March 6, 2009, 9:13am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/54 "2009-03-06T09:13:39Z")
Especially if your bomb is made out of used pinball machine parts. ---
**Author:** ![SpectBrain](https://avatars.discourse-cdn.com/v4/letter/s/a87d85/32.png) [@SpectBrain](https://boards.straightdope.com/u/SpectBrain)\ **Post date:** [March 6, 2009, 4:59pm UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/55 "2009-03-06T16:59:16Z")
> [@SenorBeef](#): > > When the US was working on refining uranium as part of the Manhattan project in Tennessee. At its peak the program took 1/6th of the entire US electrical generation capability, mostly running the cyclotrons. nitpick…calutron, not cyclotron [http://en.wikipedia.org/wiki/Calutron](http://en.wikipedia.org/wiki/Calutron) ---
**Author:** ![Crafter\_Man](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/crafter_man/32/458_2.png) [@Crafter\_Man](https://boards.straightdope.com/u/Crafter_Man)\ **Post date:** [March 7, 2009, 4:21am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/56 "2009-03-07T04:21:40Z")
Has anyone here worked in a faculty that made parts for nuclear bombs? I have. I used to work at the [DOE Mound facility](http://www.epa.gov/region5superfund/fed_fac/npl_sites/ff_npl_mound.html). I can’t go into detail, obviously, but you must trust me when I say that at building a nuclear bomb takes a _ **lot** _ more work that you think. A nuclear bomb is the most engineered device on the planet. The specifications for each component are at least 100 pages long, and there are thousands of components that go into a nuclear bomb. ---
**Author:** ![Markxxx](https://avatars.discourse-cdn.com/v4/letter/m/5daacb/32.png) [@Markxxx](https://boards.straightdope.com/u/Markxxx)\ **Post date:** [March 7, 2009, 5:22am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/57 "2009-03-07T05:22:45Z")
I’ve read a few books on this, and basically what it comes down to is, building them is not the issue. They had a few examples of college students at reasearch universities with physics studying for their doctorate or masters, were able to start from scratch without any specific knowlege of nukes and build workable bombs with two years of their studies starting. The problem was getting material. Even in a country hell determained to get nukes. It’s estimated to take about 5 to 10 years to get enough of the proper type of uranium to fuel the nuke. It’s my understanding the Russians had some American assistance in controlling their fuel stockpiles after the Soviet Union collapse. And there was quite a lot of sales of fuel that was supposedly uranium but was worthless. Apparently it was easy to make big bucks selling bogus uranium and plutonium after the collapse of the Soviet Union. I mean who were the people that bought it gonna complain too. So that helped as the people who were selling it got money and didn’t need to touch the “real stuff” apparently ---
**Author:** ![billfish678](https://avatars.discourse-cdn.com/v4/letter/b/7bcc69/32.png) [@billfish678](https://boards.straightdope.com/u/billfish678)\ **Post date:** [March 7, 2009, 5:32am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/58 "2009-03-07T05:32:12Z")
> [@Markxxx](#): > > I’ve read a few books on this, and basically what it comes down to is, building them is not the issue. They had a few examples of college students at reasearch universities with physics studying for their doctorate or masters, were able to start from scratch without any specific knowlege of nukes and build workable bombs with two years of their studies starting. I think they meant DESIGN them. The principles are relatively simple. The math kinda nasty, but you can even get around that. I am sure there are plenty of practical gotcha’s, a few theorectical nobodys gonna tell anyones, and a shitload of better build this part JUST RIGHTs to make a working bomb a PITA to make. In theory there is no difference between theory and practice, but in practice there is 🙂 ---
**Author:** ![Alex\_Dubinsky](https://avatars.discourse-cdn.com/v4/letter/a/dbc845/32.png) [@Alex\_Dubinsky](https://boards.straightdope.com/u/Alex_Dubinsky)\ **Post date:** [March 7, 2009, 7:20am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/59 "2009-03-07T07:20:08Z")
> [@Crafter\_Man](#): > > A nuclear bomb is the most engineered device on the planet. Bullshit. > [@](#): > > The specifications for each component are at least 100 pages long, and there are thousands of components that go into a nuclear bomb. The specifications for each chip in your computer are 100’s of pages long, and it ain’t because of government asininess. The factories needed to produce a single modern day CPU (or other chip) are far more complex, more expensive, more precise than _anything_ used to make a nuclear bomb. By orders of magnitude. (The difference of course is lithography is cooperatively developed by a trillion-dollar industry, without hoarding of knowledge, clandestine facilities, or government inefficiency. But let us have a bit of perspective.) ---
**Author:** ![ivn1188](https://avatars.discourse-cdn.com/v4/letter/i/e480ec/32.png) [@ivn1188](https://boards.straightdope.com/u/ivn1188)\ **Post date:** [March 7, 2009, 9:12am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/60 "2009-03-07T09:12:47Z")
> [@Alex\_Dubinsky](#): > > Bullshit. > > The specifications for each chip in your computer are 100’s of pages long, and it ain’t because of government asininess. > > The factories needed to produce a single modern day CPU (or other chip) are far more complex, more expensive, more precise than _anything_ used to make a nuclear bomb. By orders of magnitude. > > (The difference of course is lithography is cooperatively developed by a trillion-dollar industry, without hoarding of knowledge, clandestine facilities, or government inefficiency. But let us have a bit of perspective.) If you already know the answer, why did you ask the question? I’ll need a cite for the fact that chips are more complicated than nuclear weapons. Most of the energy released by a fission bomb comes from later generations, so you have to time the weapon precisely that it manages to stay supercritical for about .5ns for a 1kt yield. This is called assembly time. I think a 30 ns difference in assembly time equates to an order of magnitude difference in power. The timing even for a “simple” gun device is tight. Getting the material is difficult, but going from having some fissile material to a nuclear weapon is a lot more difficult and engineering intensive than following some blueprints and downloading some sort of bomb-simulator program and then putting some parts together. In contrast, chips with flaws were (and may still be) sold as capable of running at a lower clock rate. A flawed chip is often just a slower chip, and the timing on a modern cpu is orders of magnitude more lax than that of a nuclear weapon. It might be easy to build a fisson bomb in principle, but in practice it requires tighter tolerances than any other device I can think of. ---
**Author:** ![Stranger\_On\_A\_Train](https://avatars.discourse-cdn.com/v4/letter/s/13edae/32.png) [@Stranger\_On\_A\_Train](https://boards.straightdope.com/u/Stranger_On_A_Train)\ **Post date:** [March 8, 2009, 1:16am UTC](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159/61 "2009-03-08T01:16:39Z")
> [@Alex\_Dubinsky](#): > > Bullshit. Given your here demonstrated ignorance of manufacturing, the design process, mechanical and electrical engineering, and nuclear physics, what makes you believe that you have any standing to authoritatively dismiss answers to _your_ question as “bullshit”? A VLSI microchip is indeed a complicated part…but it is one single component, and not build to the reliability requirements or critical tolerances of the components of a nuclear weapon. It is accepted by semiconductor manufacturers that a certain number of chips will be defective, and most of those will be caught in testing and disposed of. But many of the critical components of a nuclear device (explosive lenses, x-ray reflectors, initiators, et cetera) have tighter mechanical tolerances than anything except very high end optics and timing requirements that are measured in tens of nanoseconds. Most of these components can only be inspected, not functionally tested before use (and of course cannot be tested in a “full up” configuration) and yet are expected to operate at three sigma or better reliability after being subjected to acceleration, shock, and vibration requirements that would destroy pretty much any commercial electronic or mechanical hardware. Stranger [Previous page](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159.md?page=2) [Next page](https://boards.straightdope.com/t/why-are-nukes-considered-hard-to-build/488159.md?page=4)