Thursday, October 23, 2014

Teller Ulam - Rube Goldberg Meets Doomsday

Surprisingly, the inner workings of an atomic bomb are not terribly complicated. There's the 'gun design', where a 'bullet' of highly enriched uranium is fired with conventional high explosive into a matching 'target'. On high velocity impact, the Uranium becomes a single, critical mass and a fission explosion happens, because physics. Sure, adding some neutrons with a polonium/beryllium initiator is nice, but assembling a weapon like this isn't rocket science. And just like that, you lose a city like Hiroshima. Alternatively, if you can get your hands on 20 kilograms of Plutonium-239, the design is even simpler. You cast the plutonium in a sphere, surround it with high explosives, set fast detonators all around the sphere, with timers configured so it all goes boom at the same instant. The plutonium is compressed to critical mass, and once again you get fission, a few grams of matter is converted to energy and, once again, you lose a city, this time Nagasaki.

What you may not realize is that nobody uses these designs anymore. Atomic bombs are SO fifties. Even as the rubble in Japan was still cooling, physicists were looking for a more powerful, energetic weapon. If fission is good, then adding a little fusion to the mix must be even better. Virtually every nuclear warhead deployed today is a thermonuclear device patterned after the original "Teller - Ulam" device. And good lord, it's the scientific equivalent of a madman's ramblings on full yellow legal tablets. Except, to the best of our knowledge, the damn thing seems to work.

The details? Where to start? Teller Ulam is a multistage weapon. It starts with a conventional explosive that initiates a fission explosion. And that's where things get dicey. The fission explosion, even if it is a 'boosted' fission explosion, isn't enough. So the goal is to use that initial fission bomb to create a fusion reaction. The challenge is that from the moment that fission is initiated, you have only milliseconds to create your fusion reaction before the whole thing is vaporized. You are essentially creating a complex chain reaction inside an atomic bomb in the process of detonating.

The idea is that you use a typical first generation atomic bomb (as crudely described above) to focus a blast of X-Rays on a 'secondary' - a target that, under the right conditions will initiate nuclear fusion. You focus the x-rays, you reflect them with a 'tamper' and you enhance them by surrounding the secondary with a fissionable shell. If all goes according to design, for a very brief moment you get nothing short of a miniature star - a hydrogen fusion reaction. Of course, to understand the engineering challenge this represents, you have to go back to the very first step. The whole process was kicked off by detonating an atomic bomb. And everything that happens afterward, the whole complex set of steps and processes, all happens inside the casing of the warhead in the microseconds before that initial atomic blast vaporizes the whole kit, kat and kaboodle.

The part that nobody outside of the nuclear weapons community is clear on is precisely the mechanism by which the secondary is compressed enough to begin a fission reaction. We know that the X Rays from the primary are somehow 'focused' across the 'interstage' section of the weapon, and by some process (there are several competing theories, any or all of which may be in use), those X Rays initiate a fission reaction in the secondary. The most likely method is one where a very secret formula 'foam' surrounds the secondary, and the X Rays from the primary convert the foam into a highly energetic plasma that compresses the secondary, which fissions and compresses the Lithium Deuteride fuel to thermonuclear temperatures, initiating the fusion reaction.

The term 'hydrogen bomb' is a bit of misnomer. It's true that the entire device is designed to produce hydrogen fusion, but most of the energy released is from good old fission. The fusion reaction's primary purpose is to add enough energy to greatly enhance the efficiency and output of the fission reactions. The original bomb designs were less than ten percent efficient - that is, they needed to contain a large amount of fissile material so that getting 10% of it to fission before the whole thing was vaporized would result in a large enough yield.  Now, despite the unlikely complexity of the modern thermonuclear device, you get much more energy out of a given amount of fuel than you did before. Which is ironic, because the modern warheads such as the W-88 are designed to have a maximum yield of just a few hundred kilotons, while the earlier weapons often yielded 10 megatons or more.

There's a few lessons to take away from all this. First, even though the first Teller - Ulam device was tested in 1952, and despite it's rather unlikely design, nobody has figured out how to improve on it or replace it. There's also a lesson here about letting the physics and math dictate the engineering. In order to make something work in the real world, you sometimes might have to compromise on the most elegant applied science to get the desired result. But most of all, it's kind of bizarre to consider that human culture could be ended for all time by these goofy, Rube Goldbergian bombs.


  1. Hey mikey, a little off topic, but you might know.

    There are scientific instruments that can detect gamma radiation from 13,000,000,000 light years away.

    Now if I was one of the myriad alphabet soup intelligence agencies my government so loves and I pointed one of those instruments at the Earth would I be able to detect every nuclear warhead on the planet?
    (My understanding is that uranium bombs produce less gamma than plutonium bombs, but still.)

  2. It might work - the gamma radiation from the early universe is MUCH more energetic - some of the most powerful EM radiation EVER. But even if it did, the only reason you'd care about the location of the warheads is if you were planning a massive first strike, and nobody wants to do that. Besides, with the exception of the ones that are intentionally hidden (SSBNs for example), everybody knows where each nation keeps much of the stockpile. The security and readiness requirements kind of limit the options.

  3. I'm more worried about the ones we don't know about (or lost track of), or the precursor or dismantled assemblies.
    The old rouge nuke or dirty bomb bugaboo.

  4. And for what it's worth, your commenting box does not seem to like Firefox.

  5. Excellent article.