[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Electromagnetic Pulse



At 11:19 PM -0800 10/4/96, jim bell wrote:
>At 08:46 PM 10/4/96 -0800, Timothy C. May wrote:
>
>>
>>The key is that the effects of near-ground-level bursts are _extremely_
>>localized. Shocking so, no pun intended. The largest bomb in the U.S.
>>arsenal, believed to be 20 MT, might leave a crater several miles in
>>diameter, but would hardly be felt 30 miles away. Certainly almost no
>>electronic devices would be damaged, except if close to the blast center.
>>
>>--Tim May
>
>What about the neutrons?  What is the effect of a neutron flux on silicon
>devices?

This is starting to get far afield from CP, so I'll make this my last
comment on this thread.

Neutrons have low crossection for interacting with silicon (percentage that
interact per unit length...for the few microns of active surface of a chip,
a tiny, tiny fraction interact). Charts and tables are availabel in the
literature for the effects of various fluences on various devices, cf. the
December issue of any year of "IEEE Transactions on Nuclear Science." Or do
a Web search on "single event upset" and/or "soft error."

As it happens, I did a series of calculations about the effects of a
neutron fluence (n/cm^2) on various types of silicon devices, doped various
ways. (The doping is crucial, because the primary effect is a neutron
interacting with a boron atom. Boron is one of the dopants used in some
types of devices. In small amounts, of course.) In rare cases a neutron
interacts with a boron and yields an alpha particle, which has known
effects in some devices.

(Quantification of the effect depends on fluence of neutrons, their energy
(normal vs. "thermal"), the device details, the doping concentration, the
sensititivity to alphas, etc.)

And then there's the inverse-square law. Even if their were sufficient
neutron fluences near the blast center to affect devices, there would be
almost no effect a few klicks away.

("Total dose" failure, as opposed to single-event upset, or soft error,
requires fluences in very high levels. As it turns out--in case anyone is
thinking of neutron bombs!--the fluences, or doses, needed to make people
sicken and die are much lower than what will kill a modern device--partly
due to the details mentioned above, partly due to the very thin active
layer, partly due to a bunch of other factors.)

--Tim May

"The government announcement is disastrous," said Jim Bidzos,.."We warned IBM
that the National Security Agency would try to twist their technology."
[NYT, 1996-10-02]
We got computers, we're tapping phone lines, I know that that ain't allowed.
---------:---------:---------:---------:---------:---------:---------:----
Timothy C. May              | Crypto Anarchy: encryption, digital money,
[email protected]  408-728-0152 | anonymous networks, digital pseudonyms, zero
W.A.S.T.E.: Corralitos, CA  | knowledge, reputations, information markets,
Higher Power: 2^1,257,787-1 | black markets, collapse of governments.
"National borders aren't even speed bumps on the information superhighway."