Interesting segmenton Rachel Maddow tonight on the latest conservative brainstorm for cutting wasteful government spending, nuclear security, and a new documentary about a retired truck driver who put together a surprisingly plausible design for a Hiroshima type nuclear weapon.
It got me to thinking, there are a lot of people who think they know how to make a nuclear weapon. Odds are they’re deluded. While the basics physics are widely known, the details of building a working nuke are way more complex than most people realize. I know just enough to know I couldn’t do it, and I’d be willing to bet I know way more than most self-professed home nuke builders. Just to give a taste of what is involved, let’s break the hypothetical process up into three stages, design, prototype, and weapon.
The design relies on creating or obtaining weapons grade stuff — which is an article in itself. Fissionable substances are isotopes of uranium or plutonium in which a tiny percentage of the atoms spontaneously release neutrons traveling at the proper speed, what physicists call the correct neutron temperature. Too fast or too slow and the neutrons chip off the wrong kind of particles when they hit other atoms of the same substance. Just right and the neutrons knock lose more neutrons going at the right speed, which knock lose even more neutrons, etc. If enough of that substance is located in one small spot, free traveling neutrons will hit other atoms of the same stuff and the number of speedy little neutrons will grow until the reaction runs away with itself. In principle then our design is simple: you have half the critical mass in one hand and half in the other, you bring them together and, kaboom, you’re part of a mushroom cloud.
That brings us to the prototype, because the ‘hand method’ wouldn’t actually work. The two subcritical masses would be very hot, way too hot to hold (Not to mention radioactive as hell). But even if you were working in a shielded glove box it wouldn’t work. When those subcritical masses approach one another they get hot, fast. They go from red-hot to melting to gas to plasma in a micro second, in other words they get so so hot they’ll turn into blazing hot expanding gas and blow apart with the force of a stick of dynamite before the reaction can go super critical. The prototype will have to ‘weld’ the two masses together into one strong, solid chunk and this has to happen in an instant. Early prototypes did this by the collapsing sphere or gun method, illustrated right. It’s a lot harder to do than it looks, the traditional explosives must all detonate precisely, the pieces must fit together perfectly, they have to come together at the exact speed to be hot enough to mash together into a single piece but not so hot they turn into runny liquid or gas, and even then it doesn’t always work. This is a prototype device that may go off, but there’s also a chance it will create what weapons designers call a fizzle yield, producing a tiny fraction of the potential explosive power of a real nuke and creating a big dirty-bomb mess.
Which brings us to the final phase: weapon. To be a weapon the prototype has to be both reliable and secure. That means a lot of things, no doubt some of them classified. But two unclassified items, purely for illustrative purposes, are a neutron trigger and a tamper. A neutron trigger is a device which floods the interior with neutrons from another source just as the subcritical masses are uniting. A tamper is a heavy metal, like lead, that fills in around the forming critical mass and helps keep it in a small volume for a few moments longer by sheer inertia when it tries to expand violently as it approaches its boiling point. The advantage of additions like a trigger and tamper are twofold: they make the weapon much more reliable and they provide a separate, independent system which has to be armed for the weapon to work, thus giving an extra level of security. This is a weapon that detonates when you want it to and, perhaps more importantly for you Dr. Strangelove fans, does not go off when you don’t want it to.
Needless to say going from design to prototype to weapon takes a lot of research and money. And no, you won’t find reliable data on that technology on the internet despite what some people think they know — if my tax dollars are doing any good, what you should find is a bunch of misinformation! — with the possible exception of the occasional dumbass Bush WH officials who released some hard data conveniently written in Arabic while trying to backstop the invasion of Iraq. In short it takes a relatively wealthy country with a dedicated and educated professional workforce to do all that. Terrorists working in a cave aren’t likely to pull it off.
But stealing the components, or better yet an assembled bomb, is a different story. That’s what nuclear security agencies try to prevent. And that’s what these nutbags in Congress are toying around with defunding. The word idiot hardly describes them.
cyberax says
Uhm, you have several misconceptions.
First, it’s really really easy to build a gun-type bomb. Calculations required for this are certainly within reach of math/physics undergrad student.
Second, enriched U-235 is hot (around 80000 Bq/g, compared for 12000 Bq/g for U-238), but it’s not that terribly hot. Yes, if you work with highly enriched U-235 in bomb-type quantities for a length of time you’d receive a great deal of irradiation. Possibly lethal in the long run. But it won’t be “dead within minutes after you come close to it”. And it’s not like some terrorist organizations lack in suicide bombers.
Third, you have misconceptions about gun-type bomb. In such a bomb U-235 must NOT start fissioning until supercritical mass is assembled. Nuclear reactions happen in microseconds, so if your bomb starts exploding before it’s supercritical it’ll fizzle (consequently, “two hemispheres” design is actually not optimal to achieve rapid reactivity increase).
Implosion-type bombs on the other hand are magnitudes more complex. There is no way it can be assembled without state-level expenses, even if one has detailed designs.
Stephen "DarkSyde" Andrew says
Those are interestng points, but is it really necessary to use the “umm, you have misconceptions” smart ass retort? I could easily do the same thing to your response, go into detail on some over simplifiction you made and pretend you are labolring under “umm, several misconceptions.” :)
cyberax says
Sorry, I was writing this before I had my morning coffee and my writing style suffers a lot when I’m not caffeinated enough.
We’ve actually computed critical masses for bomb-type devices back when I was at university. I’ve even computed profiles of reactivity for various types of assembly: two hemispheres, cylindrical core of a sphere and cylindrical core of a cylinder. It can even be done analytically.
But certainly, feel free to totally demolish my post! I really love to argue :)
Stephen "DarkSyde" Andrew says
Ahh, my fault, I was in a shitty mood this morning :)
unbound says
Perhaps a minor point, but enriched U-235 is not easy to come by either (I’m assuming we are focusing on highly enriched U-235 since we are talking bombs). It’s rather expensive to produce and, to my knowledge, only used in military reactors (other than potentially bombs) where it is more expensive to open up a ship to replace the fuel rods.
Commercial nuclear plants use LEU, and I’m pretty sure the concentrations are too low to be useful in making a bomb (again, not easy to enrich it further…even Iraq with all its oil money was not getting it easily with their centrifuges). So obtaining from that source will have limited use.
I’m with Stephen on this…the easiest method (and most realistic) to go down this path is to steal the components. Defunding the nuclear security resources is really not a good idea.
cyberax says
Availability of U-235 is certainly not a “minor point”. It’s the reason terrorists have not yet exploded a nuclear bomb.
I’m speaking about a theoretical case when terrorists somehow obtain weapons-grade U-235. It’ll be BAD.