[LAU] OT: metal, money, changes, bleg

[Fons Adriaensen]

On Mon, Apr 04, 2011 at 01:04:11PM +0200, Jörn Nettingsmeier wrote:

>> So, just having enough Pu-239 in one location is definitely,
>> categorically not going to be able to cause a nuclear explosion? Not
>> even a little one that might cause a rapid escalation?
>
> no. and to see how hard it actually is: there were news of a botched  
> nuclear test in north korea a while ago, where international observers  
> found evidence of an explosion (by satellite or seismic measurements)  
> that was vastly less than what was anticipated. such a situation is  
> called a "fizzle", and what happens is that you bring the two halves of  
> sub-critical material together too slowly or in the wrong way, and the  
> chain reaction then heats it up and tears the material apart again  
> before the entire mass has started to react.

Yes. Keeping the thing together for long enough is the key
to a full nuclear explosion. And it's hard and not going to
happen by accident: the energy generated when a reaction start
to sustain itself destroys the conditions for it to continue.

Criticality != explosion. All reactors are critical when they
produce power. There is even evidence that controlled sustained
fission reactions have occured in nature. That's not possible
any more as the natural uranium on our planet has decayed too
much to become critical in any quantity.

> when you pile up too much plutonium or uranium, criticality accidents  
> can occur, but there is no possibility of an explosion even remotely  
> comparable to a fission bomb. which is not necessarily a good thing:  
> iiuc, a good clean atomic bomb test can be more "environmentally  
> friendly" than a botched one, because a fizzle explosion distributes the  
> material over a large area and could even lead to greater contamination  
> than the fallout of a full explosion.

Most bombs are actually very dirty and only use part of the fissionable
material they contain - it's easier to make them that way. 

-- 
FA