?

Log in

No account? Create an account
The difficulty of Making Nuclear Bombs - AdrianG — LiveJournal [APOD]
September 12th, 2004
07:26 pm

[Link]

Previous Entry Share Next Entry
The difficulty of Making Nuclear Bombs

(10 comments | Leave a comment)

Comments
 
From:timtad
Date:September 13th, 2004 01:57 am (UTC)
(Link)
The first time you invent something from scratch, it can be incredibly difficult. But once the cat is out of the bag (the knowledge that it can be done), there are only a couple of really hard problems to solve and a small team of engineers and physicists could design and build such a device. Many of the original problems were the result of lack of computing power. Some of the remaining problems include modeling the implosion characteristics* and producing the correct high explosive ‘wrapper’ and obtaining and working with the nuclear materials themselves. There are several high tech and specialized components to be procured also. It’s also simpler to use an obsolete ‘gun assembly’ design.

The difficulty lies not so much in the actual fabrication, but in all the preparation required, setting up the right shop, obtaining the material, testing some sub assemblies and keeping the team together and secret from the eyes of the international community.

If I could steal or buy the nuclear material, was willing to forego testing, and did not have to operate in secret, it would not take that much effort or money (a few tens of millions). Of course my design would have an somewhat unpredictable yield.

So in essence you are correct that the limiting factors are secrecy and detection during your design and development cycle, rather then design hurdles. The design hurdles are just not that high.


* the computing horsepower now exists on any basic PC, but the software would have to be stolen or written from scratch.
[User Picture]
From:adriang
Date:September 13th, 2004 03:04 am (UTC)
(Link)
I think you are right about the gun device being simpler, but I also think you are underestimating the number of problems that must be solved, even with the gun device. As examples:
  • the initiator must be designed.

  • Purification of the initiator materials is also complicated, and getting the fissionable material from an external source doesn't solve the problem of aquiring initiator material

  • the timing of setting off the gun and the initiator must be determined.

  • critical mass is dependent on the density and geometry of the nuclear material. I remember that plutonium has different densities depended on how it is purified. Is the same true for uranium? What will be the exact shape of the combined mass of nuclear material be when the gun has finished bringing them together? Will the nuclear material be compressed at all by the gun? That would also affect critical mass.

  • Uranium is a difficult metal to work with? How is it to be shaped properly?

In addition to all these points, I think gun devices are harder to make small.

It easy to say that something isn't very hard. But for this kind of device, I think it is still very difficult to solve all the problems required for putting together a good design. Certainly there are more than a couple of problems to solve.

Adrian

From:timtad
Date:September 13th, 2004 03:26 am (UTC)
(Link)
I'm impressed.

My reference to specialized components was an oblique way of bringing up the initiator, one of the hardest things to build, and you are exactly right about initiator timing being critical for pre and post detonation reasons. Also, you are correct in that the gun design cannot be made small on the same scale as an implosion design because the gun does not really compress the uranium, it just assembles a 'normal density' critical mass. It's an inefficient design from a 'use of nuclear materials' point of view. I believe (but I could be wrong) that both Uranium and Plutonium are cast in 'near net' shape and then machined. None of these is an insurmountable problem, just challenging in multiple different fields. I agree with you.
[User Picture]
From:polytwo
Date:December 19th, 2004 09:13 pm (UTC)

Here's a cool link, if you don't already have it...

(Link)
http://nuclearweaponarchive.org/index.html

I have taken to studying this type of nuclear physics area a lot over the years, both due to my (life long) geeky scientific curiosity, and my desire to be able to have educated guesses about proliferation like you are discussing here.

I just friended AdrienG back and look forward to reading you. Any friend of Malabar... It is my usual practice to friend anybody back unless casual inspection of their journal scares me, heh!

Best wishes...
[User Picture]
From:adriang
Date:December 19th, 2004 09:30 pm (UTC)

Re: Here's a cool link, if you don't already have it...

(Link)
That looks like a pretty interesting web site. Thanks for the link.

Adrian
From:timtad
Date:December 20th, 2004 01:51 am (UTC)

Re: Here's a cool link, if you don't already have it...

(Link)
Thanks for the link
From:timtad
Date:December 20th, 2004 04:06 am (UTC)

Re: Here's a cool link, if you don't already have it...

(Link)
You have an interesting journal, I'm going to lurk a while.

May I suggest for your reading pleasure...
"The Symbolic Species: The Co-Evolution of Language and the Brain" by Terrence Deacon
[User Picture]
From:polytwo
Date:December 20th, 2004 05:07 pm (UTC)

Re: Here's a cool link, if you don't already have it...

(Link)
I googled it and it looks very interesting... Thanks for the reference.

I really get a look at how important language is for social well being when I work with kids with language problems, that's for sure!

Have a great day.
My Resume Powered by LiveJournal.com