Schneier on Security
A blog covering security and security technology.
« Hushmail Turns Data Over to Government |
| Thoughts on the Security of qmail »
November 15, 2007
The Overblown Threat of Suitcase Nukes
From the AP:
...government experts and intelligence officials say such a threat gets vastly more attention than it deserves. These officials said a true suitcase nuke would be highly complex to produce, require significant upkeep and cost a small fortune.
Counterproliferation authorities do not completely rule out the possibility that these portable devices once existed. But they do not think the threat remains.
"The suitcase nuke is an exciting topic that really lends itself to movies," said Vahid Majidi, the assistant director of the FBI's Weapons of Mass Destruction Directorate. "No one has been able to truly identify the existence of these devices."
Interesting technical details in the article.
Posted on November 15, 2007 at 3:38 PM
• 35 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
Finally a story where the authorities are not overreacting to a movie-plot threat.
That it's possible to build a very small nuke is not news - The U.S. actually fielded 8" nuclear artillery shells, and I imagine those MIRVs we pointed at the Russians were each pretty compact. The fact that it costs billions to design and build one is a pretty big hurdle to their appropriation by terrorists.
I think it's reasonable to write this threat off as "possible, but unlikely." Shows like 24 really overplay this sort of risk.
This is the closest I can find to an actual "backpack nuke"; the core itself is on the order of 50 pounds, which is rather heavy for a single person to use.
"This is the closest I can find to an actual "backpack nuke"; the core itself is on the order of 50 pounds, which is rather heavy for a single person to use."
It would be easier to put it into a van and drive it into the downtown area of whatever city you were going to attack.
Weight really doesn't matter ... as long as it won't crack the suspension of the van.
Right now nukes are expensive. And how does a terrorists REALLY know that he's been sold a working nuke? Does he buy TWO and randomly set one off?
In the future, this might change. But right now nukes are not something you really need to worry about.
The government, at least in the case, has correctly put this in the unlikely category. Of course, if possible suicide bombers know it isn't considered a real risk, there is a small possibility they may give it more attention--but they still have to obtain a working bomb, so it would remain unlikely.
I would bet most of my money we won't see this kind of threat materialize in the foreseeable future, but I can still see the backlash against the government experts quoted here if it were to happen. Their statements here would be used as a hammer to beat them with, even though the statements about the risk being overblown are correct. (Sort of like how the risk of infant abductions are very small, but pity on the hospital CEO who states this fact if his hospital is the one in a million it happens to.) It's no wonder they don't always state the obvious.
@Arclight - forget the suitcase nukes, I want the cell phones from 24 that seem to operate *everywhere*
Clancy did the "drive up nuke" already.
The problem, of course, is that terrorists don't have to worry as much about experiencing premature detonation...
A working nuke, I believe, is less of a problem than a serious fizzle... or something that just delivers fall-out.
The manufacturing plant and industrial infrastructure required to make these also scales up, exponentially, with the sophistication and with the miniaturization of the warhead. Just procuring the plutonium by theft is not even close to being enough to make a suitcase nuke.
It's a little like imagining building your own vehicle. You might be able to knock together a go-cart in your back yard, with access to a junk yard. To build a street-legal vehicle, you'd need access to some serious industrial plant. To build an aircraft, you'd need to be a medium-sized corporation, or a small country. To build an orbital vehicle, you'd need to be a wealthy country.
Regular nukes are still "small country"-grade devices -- it's not really that plausible to imagine a bunch of guys in caves building one, without support of a rogue government. Backpack nukes are wealthy country devices. It's silly to even think of them as terrorist weapons. There are much better ways to stay scared.
@ Fred P: "...on the order of 50 pounds, which is rather heavy for a single person to use."
A few years back I was about 15-20Kg (33-44lbs) overweight, just shy of 40 years old, hadn't had anything that even looked like exercise in years and had had multiple hospital admissions for asthma in the past year. I still managed to walk the West Highland Way, 90 odd miles across mountains, in a week with a 40lb load in my rucksack. Fifty pounds would be a doddle for a fit young man. And yes, I was MUCH fitter at the end of it all.
This article is almost pure crap. Maybe zero to very few nukes were manufactured specifically in a suitcase form-factor, but the US built thousands of sub-200lb nuclear warheads that could stand in for a suitcase nuke for any terrorist plot you want to imagine - they'd perhaps just have to be wheeled luggage instead of carried in one hand. In the trunk of a car, they're an identical threat.
It is nice for the government to take a break from scaring us for a bit, but cleaning up stray nukes is the one thing where I wish they'd do a bit more!
One thing that seldom gets press is the maintenance requirements, and it's good to see that mentioned. While it's true that the huge 1940s nukes could sit on a shelf for ages and still be good to go, miniaturized nukes require extremely perishable components. Polonium, for instance, has to be replaced regularly because of its short half-life.
So it's not enough for a terrorist to somehow grab one of these nukes. It either needs to have been well-maintained, or the terrorist also needs to get his hands on technicians and spare parts.
Tritium has a half life of 13 years as well, not as bad as Polonium but still pretty perishable and hard to get.
Of course there's the SADM which was meant to be man portable but pretty much a big country weapon (Haven't seen them on eBay yet)
A airline suitcase containing Pb/Pt/U Could weigh > 300lb ... easily.
Pony nukes, barring the ever-present chance of technological surprise making them much easier to build, are a low probability terrorist threat. Between maintenance, permissive action links, the dangers of tampering with a captured weapon, and the difficulty of building one, it's pretty much a non starter.
Pony nukes are HEAVY. You could make a nuke the size of a large fire extinguisher, but you could not lift the resulting fire extinguisher. The lack of shielding also makes them much easier to detect (the real flaw in Clancy's novel.)
However, they must be classified as a high probability threat from the special operations perspective. I believe it was Heinlein who pointed out in the 1950s that mini-nukes posed the chance of making your hardened underground bunker complex into a colossal death trap. When one thinks of the creative ways in which a pony nuke could be used to cripple critical infrastructure worldwide . . . one wants to re-consider pissing off the Americans, British, Russians, Chinese, and so on.
Andrew: You do not have to detonate a working A-device to cause terror. It's enough if you deliver a credible threat that "the authorities manage to prevent" afterwards. The effects of the resulting wave of fear and paranoia will be comparable with the blast itself. Cf. liquid explosives plot, on steroids.
Grant: Little Boy, yes, that one is likely to be long-term storable. Fat Man requires the "urchin", the polonium initiator in the middle, which has to be fresh. More modern dial-a-yield boosted devices require tritium; though even without it the fizzle will look mightily impressive on the TV news.
Remember: it's not the yield that causes the primary damage; it's the news coverage.
A far more likely plot would be to load up a container with a few hundred kg of high explosive and then fill it with low level nuclear waste (of which there are 1000s of tons in poorly guarded sites.)
Then lose the container in the world shipping system.
Even if detected, disarming it would require clearing an area the size of a small city first.
The W54 compact nuke, which weighed over 20kg and would fit into a special (large) backpack, required years of development and multiple tests. These things are very difficult to build, even for the USA. And then - the yield of the thing is only about 2 to 4 times the yield of the home-made 1995 Oklahoma bomb, built by an idiot. For a terrorist, what's the point of even considering the "suitcase nuke"? Furthermore, a delivery van packed with conventional explosive is less conspicuous than a guy struggling under the burden of a bulky 20-kg backpack.
>Furthermore, a delivery van packed with conventional explosive is less conspicuous than a guy
>struggling under the burden of a bulky 20-kg backpack.
It really would depend on the target. If you want to destroy a building and can get a van close enough, you don't need a small portable nuke.
If, however, you want to sink the QE2 at sea, what would be more inconspicous than a woman with a really large suitcase? ;-)
This is not the first time this has come up on Bruce's blog,
The simplest type of nuke to build is the "gun type" that is you take two sub critical masses of your material (say a ball and a core) and either fire one into the other or at each other (making a three critical mass ball). Unfortunatly it has to be U235 or U233 due to the slow nature of the gun device (so slow that you can actually drop the slug through the ball without it going critical see "tickling the dragons tail" experiment).
It will go critical but the yield will be quite low (not that joe terorist cares about yield). To increase the yield you have a couple of basic choices include an initiator (golfball/urchin) or change the design so that the critical mass stays as a lump for longer. It also helps if the gun assembly is vacum sealed.
A more sensible device would be a three part with initiator with a heavy outer tamper. The problem being is that the timing is critical an extra half inch of cable to one of the detonators is going to effect the yield. Then there is getting detonators that will be time acurate (remember the Krytron switch solution anybody?)
Oh and you could make it a partial compression type as well by putting more explosives around the central core tamper and explode it just as the cores arrive which due to the difficulty of making lensed charges (think baritol etc http://web.umr.edu/~rogersda/... ) would sugest a cylinder arangment might be better...
Technicaly it is all doable, but you would need to do a lot of testing and need a very good machine shop
The question is what if it does not explode corectly and does not go critical? Well you could end up with the uranium burning and spreading it's self on the wind...
Any way there is a nice web site to lose yourself in at,
Have fun ;)
Read this last night but just noticed the humor of the title now.
@jack c lipton
And Frederick Forsyth did it before him in "The Fourth Protocol".
IIRC the objective wasn't to cause material damage per se but to make it seem that a real bomb had gone off accidentally at a USAF base in England and thus force a change in government.
@Carlo Graziani: "To build a street-legal vehicle, you'd need access to some serious industrial plant. To build an aircraft, you'd need to be a medium-sized corporation, or a small country."
A lot of dune buggy builders and EAA members will say otherwise.
Be careful about dismissing capabilities as being unreachable just because they are beyond your ken, others might prove you wrong.
"A lot of dune buggy builders and EAA members will say otherwise."
Yes they might.
But they would still be wrong.
Assembling a kit is not the same as producing the high grade steel yourself. And that was (I believe) his point.
"Assembling a kit is not the same as producing the high grade steel yourself. And that was (I believe) his point."
You don't need to use high grade steel to make an airplane, for eample, you can use wood, and fabric. You don't have to make it from ore... Thousands of people have built airplanes from the plans up with no kits. You buy engines on the market just like buying other things. Just like someone building a bomb would. This doesn't mean the point is wrong, but I also think the airplane example was bad given the thousands of counter examples for the last century.
"You buy engines on the market just like buying other things."
What are those engine made out of?
Assembling from a kit is still assembling from a kit.
"Just like someone building a bomb would."
We are not talking about "a bomb". We are talking about a nuke. Building a pipe bomb is different than building a nuke.
No one sells nuke parts (complete nuke, just add uranium). That was the point.
50 pounds is the limit for U235 nukes. If you use less U235, you need more explosives to compress it to two or three or four times the standard density. The more density, the less neutron leakage, the smaller the critical mass.
Initiators need to make neutrons. You can use light elements like berylium and alpha particles from something radioactive like a smoke detector, or you can use a capacitor to zeta pinch for neutrons. Since you won't get as many neutrons, you won't get as big a bang as a properly designed nuke.
As a practical matter, the Hiroshima bomb weighed a ton. You will get less yield. If you get a kiloton, be happy. Suitcase bombs were designed to blow up bridges and similar targets, not cities. Blowing up a dam, or a levee during a flood, or an operating nuclear reactor (full of radioactives), or a stadium, or an office building like the world trade center, or an important mine, or, well, something symbolic.
It's a lot easier to just get your hands on a thousand tons of explosives.
Now using the low yield fission bomb to make a higher yield fusion bomb is a whole different story.
@Brandioch, why not argue that building a plane out of wood requires growing trees and milling lumber? It's about the same. Converted VW and Mazda engines are flying in homebuilts.
Bringing it back to the nukes, you don't need to build an IC fab to make parts for the trigger circuits, they can be made using COTS parts. The only stuff that might be an issue is the "secret sauce" and even there it's a classic make vs. buy decision. Do you need to build an isotope separation facility (ala Iraq or North Korea) or can you just buy some materials on the black market? The reason for the concern is more the latter, how much confidence do you have that all the former Soviets are really to be trusted not to have lost some fissionable materials?
Point is, in the modern world you don't have to duplicate anything that's been done widely enough. Where did India and Pakistan get their fissionable materials? How confident can we be that they, and everybody else possessing significant quantities of fissionables, can be trusted to keep them under control, at all times, forever?
Without a satisfactory answer to those questions, we might as well go back to discussing growing trees for homebuilt ragwings.
one other thing about homebuilt aircraft. kits are common, and plans can be bought, but some builders design their own plans from a blank sheet of paper.
I've seen all of the above flying.
Likewise, it may be that only one in ten million could design even a dirty nuke. That is still greatly different from dismissing it as being impossible!
"Now using the low yield fission bomb to make a higher yield fusion bomb is a whole different story."
To get a "low yield fission bomb" you usually need a small nuke to start it...
But as you noted the difference between low and high yield is realy just a "simple" engineering job by comparison to building a reasonable yield nuke (but still not quite a file and Kraft knife in the garage job). Unlike making a nuke a knowledgable post grad could (/should be able to) draw up working plans...
"The only stuff that might be an issue is the "secret sauce" and even there it's a classic make vs. buy decision."
What are you talking about? That's the WHOLE POINT.
Anyone with enough fissionable material ... and able to smack it together fast enough ... will produce a fission "bomb".
WHERE are you going to get that amount of fissionable material?
"The reason for the concern is more the latter, how much confidence do you have that all the former Soviets are really to be trusted not to have lost some fissionable materials?"
There are many old master paintings that have been lost. We know this.
Now, if someone comes up to you offering to sell you an original, never before known, Renoir ... Just because something is not accounted for does NOT mean that it is available to you for purchase.
"Point is, in the modern world you don't have to duplicate anything that's been done widely enough."
And by "widely enough" you mean "you don't have to duplicate" it.
Nice circular logic there.
"Where did India and Pakistan get their fissionable materials?"
ZOMG!!! You've found the flaw in my position!!!
Duh, they have uranium mines in their countries. Their countries are situated on top of uranium deposits. They dig the uranium up.
As was stated earlier in this thread, that requires the resources of a NATION.
And it requires a LOT of uranium.
"How confident can we be that they, and everybody else possessing significant quantities of fissionables, can be trusted to keep them under control, at all times, forever?"
Ummmm, since the half-life of those materials is far, Far, FAR less than "forever", the answer is 100% positive. The materials will be inert long before "forever" arrives.
"Without a satisfactory answer to those questions, ..."
Again, because you do not accept the answers, the answers are not "satisfactory" to you. Nice circular logic.
You watch too much TV.
Terrorists with nukes are not something to be concerned about. They cost too much. They require too much expertise. They require too much technology. They require a NATION to build / maintain / support.
Ops, I made a glaring mistake in my above post (anybody apart from me notice it?)....
Moral never cut-n-paste and post whilst to tired to see what you have done...
To get a "low yield fission bomb" you usually need a small nuke to start it...
It should have been,
To get a "low yield fusion bomb" you usually need a small nuke to start it...
The two points I was trying to make in the post were,
1) A fusion bomb no matter what is's yield usually needs a small fission bomb to get it going.
2) The engineering required to turn a low yield Fusion device to a high yield Fusion device is considerably simpler than getting the design of the small fission device right.
Ho humm now just to read things again...
Back in 1978 (roughly) there was much flapping of arms over an article in New Scientist (I think) when a post grad. in the UK published a "design" for a DIY nuke using two sub-critical masses of plutonium, half balls, that where brought together using gravity in a vertical pipe.
simple gun design.
At the time I cant recall if the design was shown to not work although I can now imagine the issue of holding the two halves together for enough time is pretty major, even with an explosive boost and neutron reflectors.
The question I now have, when an analysis is done of this design (I'm assuming your average PC can do the simulation these days) does it come slightly close to more than than a fizzle?
From what I recall at the time it was stated to be only infeasible due to access to the Plutonium.
With hind site it seems like a good movie plot set in Wall street but that's about all.
If you want information on nuclear proliferation and "suitcase nukes", I suggest you look at blogs like Arms Control Wonk.
Schneier.com is a personal website. Opinions expressed are not necessarily those of BT.