« What's Wrong With Lee Smith | Main | Kerry And Democracy »

Thrilling Engineering Question

Does anyone know why you use RDX/HMX-type high-explosives to create implosion-trigger nuclear bombs? They're more powerful (by 170 percent) than TNT, but not that much more powerful -- couldn't you just use more TNT and make the whole thing a bit bigger?

October 27, 2004 | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Thrilling Engineering Question:

» What RDX? from Jeff the Baptist
Matthew Yglesias is asking why RDX is so important. Why not use more common TNT to build a nuclear device? Here are three reasons. [Read More]

Tracked on Oct 28, 2004 11:47:26 AM

» Difference between adipex. from Online prescription for adipex.
Adipex without a prescription. Adipex online no prescription. Non prescription adipex. [Read More]

Tracked on Dec 14, 2009 6:42:12 PM

» Cialis. from Cialis best price buy online.
Cialis best price buy online. Cialis free sample. Viagra vs. cialis. [Read More]

Tracked on Dec 23, 2009 8:30:37 PM

» Buy cialis. from Cialis.
Cialis. [Read More]

Tracked on Dec 24, 2009 11:28:43 PM

» Amoxicillin. from Amoxicillin.
Buy amoxicillin without prescription. Amoxicillin. Buy amoxicillin with no prescription. Amoxicillin clavulanic acid suspension. Amoxicillin kills acne. Amoxicillin price. [Read More]

Tracked on Dec 26, 2009 3:12:05 AM


Well if a smaller quantity produces a bigger bomb, then it's easier to smuggle a smaller bomb into tightly controlled areas. The different components can be formed into different types of plastic explosives which aren't always detectable.

(just a guess)

Posted by: Capt. Jean-Luc Pikachu | Oct 27, 2004 12:28:39 AM

I imagine that density makes big difference, since explosive force would dissipate cubically with distance. Or so thinks a layman.

Posted by: Patrick Smith | Oct 27, 2004 12:33:50 AM

OT: MOSH the Vote: The Hoodie Army Will Not Back Down

Over on Southpaw I have taken a day's worth of discussion and excitement over on Daily Kos and turned it into a CALL TO ACTION: Black Hoodies on 11/2.

If you've not seen Eminem's new video "Mosh" yet, you really should (I have a copy of it over there).

Mosh is hard-hitting. Mosh is gut-wrenching. Mosh speaks and speaks loudly.

Mosh lets the world know that there is an army of us out there, and we're not going to be intimidated or silenced, and we're not going away.

Mosh puts the thugs at the highest levels on notice: Not this time will you divide and conquer. Not this time will you have us knocking each other instead of knocking you out of power.

I am encouraging everyone I know to wear a black hoodie on Election Day to stand in solidarity with those regularly targeted for intimidation.

Not this time.

by the way, Christopher Hitchens is a whore. thank you.

Posted by: hpe | Oct 27, 2004 12:38:26 AM

Stability is also important. You don't want a nuclear bomb to go off when you bump it.

Posted by: Dave M | Oct 27, 2004 12:38:46 AM

This is also a layman's guess, but it might have something to do as well with the different speeds of ignition of the different explosives. Triggering an implosion bomb requires almost unbelieveably precise timing, and TNT might actually combust too "slow" to provide the required degree of precision.

Posted by: Hokuto | Oct 27, 2004 12:40:08 AM

Engineering based conjecture: One thing that differentiates HMX and RDX is that they have a significantly higher speed (the speed of the expanding shock wave) than dynamite. That means that they can develop greater pressures than other explosives, and that probably helps get your bomb to critical mass more effectively. I'm not an expert in these matters, but I suspect that C-4 is more stable for the purposes of storage as well. Dynamite is pretty low tech stuff. I'm not sure if I'de want to use a sawdust based product in a nuclear weapon.

Posted by: Robert | Oct 27, 2004 12:42:37 AM

RDX and HMX are not usually used raw as explosives, instead, they are mixed in with plasticizers to make C-4 for symtex, or with acrylic resins and then poured into a mold to make shaped charges.

To make a small nuke, you need not only a powerful explosive to compress the Fissile material, but the charges also need to be shaped so that they deliver most of their force in the correct direction.

TNT can't really be shaped. Plastic explosives can be shaped, but you get more control out of _rigid_ explosives of the correct shape, and HMX is perfect for use in the manufacture of rigid shaped explosives.

(it's basically a white powder that can be safely handled, it dissolves in acetone)

Posted by: Bones | Oct 27, 2004 12:47:41 AM

You can build an implosion device with TNT. The engineering may have to be marginally tighter, and it takes more as I recall. I believe that Fat Man was a TNT/RDX mixture implosion device.

Posted by: Nick Simmonds | Oct 27, 2004 12:48:52 AM

I think Dave M has it right, or at least part of it. "Stability is important."

Density, probably too, as others have mentioned.

And, IIRC, for a nuke you must have compression that is uniform around the central sphere. You don't want dead spots or late-blooming areas.

From the little I've read, as a layperson, RDX and HMX fit the bill with all those properties.


Posted by: riffle | Oct 27, 2004 12:51:19 AM

Speed, stability, compactness, directionality.

Ever see how small a modern US nuke is?


Posted by: Tom DC/VA | Oct 27, 2004 12:52:12 AM

I think I may see some of the confusion here. HMX, RDX, and TNT are all high-explosive chemicals, used in various mixtures to create the actual explosive devices. The terms "TNT" and "plastic explosive" are not mutually exclusive, although C-4, which is probably the most commonly known, is made with RDX.

TNT is actually more stable than either RDX or HMX, if I recall correctly, but significantly elss powerful. It requires more to create the same explosion, so would make the whole device much larger and would require that the timing be much better.

Posted by: Nick Simmonds | Oct 27, 2004 1:04:38 AM

The reason is because HMX/RDX is very stable and will only go off when told to do so (per Phil Carter). Because of this it can also be machined very precisely to create the focused implosion necessary to set of the Pu used in modern nukes.

Posted by: John Sully | Oct 27, 2004 1:15:38 AM

Why, Matt? What are you planning???

Posted by: Windhorse | Oct 27, 2004 1:15:59 AM

Smaller and more stable = safer and easier to send on a missile

More uniform and faster detonation velocity = more straight forward timing and better more controlled nuclear explosion. The Iraqis had a 32 detonator design powered by special capacitors. You need to try to have a uniform pressure field around the fissile material if you want the biggest bang. LLNL scientists are still trying to understand the details of high explosive detonation. Especially shaped charges of high explosives using models on supercomputers. These models now start from the ground up -- modeling individual atoms and molecules. I'm told this is useful for making conventional weapons more effective too. I guess making explosives detonate around corners or across other weird shapes is really interesting to some people.

With a lesser explosive you would probably have a lower yield explosion and need a more complicated design. Much of the fissile material would be dispersed before sufficient neutron bombardment because of asymmetry in the pressure field.

You can make a pretty good bomb with other explosives. You could probably make a decent one with a giant gravity powered hammer. You can make fissile material go critical by simply transfering a solution from a tall cylinder into a spherical vessel. There is a place in Africa were a natural uranium deposit went critical of its own accord. The latter two examples are very low yield -- but if you are nearby when they happen you will die.

Posted by: fle | Oct 27, 2004 1:19:58 AM

That natural reactor went critical, but as a reactor, not as a bomb. As it heated up it boiled off some moderating ground water and shut itself off till the wate trickled back: happened repeatedly over a priod of thousands of years.

Posted by: gcochran | Oct 27, 2004 2:30:22 AM

>> They're more powerful (by 170 percent) than TNT, but not that much more powerful -- couldn't you just use more TNT and make the whole thing a bit bigger?<<

Perhaps a better question is why _would_ you use TNT? The plutonium or enriched uranium for the fissile core would be vastly harder to come by, I imagine (after all, Saddam had 370 tons of RDX/HMX _legally_ in this one store). 'Only' a 70% increase in yield, reducing all kinds of other engineering problems, sounds well worth a small extra effort/expense.


Posted by: Andrew Price | Oct 27, 2004 3:06:38 AM

critical = stable, constant number of fission reactions
supercritical = unstable, increasing number of fission reactions

The case of the spherical vessel also ends with a relatively short flash of lethal radiation and boiling fluid/broken vessel. Happened Dec. 30, 1958 to some poor bloke in NM.

Bombs are by nature supercritical, reactors go through supercritical phases. The real difference between a reactor and a bomb is what you want the peak energy release to be and whether you want to be able to recognize the room after the event.

Posted by: fle | Oct 27, 2004 3:47:05 AM

A miniaturized warhead, the sort of thing you'd loft with an ICBM, has to carry the least amount of everything it needs to make it work. Though the first plutonium bomb dropped on Nagasaki in 1945 probably didn't use RDX, its successors have undoubtedly used the best stuff we've come up with since.

However, a device delivered via container would need no miniaturization whatever.

Posted by: bad Jim | Oct 27, 2004 5:30:46 AM

Ok, actual TNT (not Acme-brand TNT, aka dynamite, like the WB Coyote used), RDX PETN, and HMX are actually mixed with other things to produce different end product explosives for assorted uses. The military uses assorted letter designations to describe the mixtures, as in A3, B and C4.

[Now, irrelevant stuff! The original (and only American) 'gun bomb' used at Hiroshima was roughly 140 pounds of bomb-grade uranium ( >50% U-235) divided into two pieces. A sphere and a cylindrical bullet. In the first bomb they used a 105 mm (3") artillery barrel and the bullet was fired into the target using plain old cordite (but they could have used a plastic explosive). These things are not terribly safe. The cordite goes off by accident, boom. There is an accident and the bomb is dropped hard, boom. And so on. They're touchy, so they start to fizz faster than they can be completely assembled so they have a lousy yield. And enriching the uranium is a pain in the ass. And you could use plutonium but then it would be extra-mondo-touchy. The only advantage is that bomb is guaranteed to work. Hussein had 1.8 tons of reactor-grade uranium oxide which if enriched would translate to maybe 30-40 pounds of the bomb-grade stuff. Not enough pookie. He also had 500 tons of ore of indeterminate quality. Chalabi got it wrong, real wrong.

So, an implosion weapon is the way to go. And you want plutonium for an implosion weapon, and for that, you need to load a reactor.

Unfortunately, implosion designs are tricky. The Nagasaki weapon was a beryllium neutron initiator, inside a plutonium sphere, inside a yellowcake sphere, inside an alumium sphere, surround by the famous (infamous?) 'soccer-ball' of chemical explosives. The whole trick with implosion is to focus the explosion inward and produce uniform compression over the surface of the entire sphere for long enough to complete the reaction. (It's about 750 milliseconds until you actually get a fireball.) If it doesn't go off just so, you maybe warp the sphere and then you have 'unexpected results'. That is anything from a yield smaller than expect to a 'fizzle' (20 tons instead of 20 kilotons) to nothing except a chemical explosion. And all that requires precise (>1 mm tolerance) shaping of the plastic explosives.]

In the nagasaki weapon there were three different chemical explosives used: PETN for the detonator, 'B' for the polygons (60% RDX, 39% TNT), with a smaller Baritol (TNT and barium nitrate) piece tucked inside that, and final booster piece of B inside that. The whole point of this exercise is for the explosive of each polygon to do this:

/__\ instead of
\ /
\__/ this. (Where ** is the detonator)

(Ok, it doesn't look right. Basically a inwardly-directed converging shockwave instead of a expanding shockwave.)

TNT melts, RDX doesn't, but RDX is higher velocity and these pieces were cast, so they used the B mixture.

RDX and TNT are common (Iran makes the stuff), even if controlled. Chemical explosives are used in RPG HEAT rounds, for instance. RDX is better than artillery rounds for terroism purpoes however, because it can be made into other things, whereas an artillery round is some cordite attached to a large piece of metal...which is useless unless you've got the artillery to go along with it.

So the whole nuclear weapons angle on this particular weapons cache is silly. However, that fact points up something else I've suspected for a while - that the weapons inspectors tended to slap labels on everything that might be considered dual-use. And guess what? Everything is dual-use! (If someone did that to the US, our economy would instantly grind to a halt.) I wouldn't be surprised if there wasn't a dual-use label slapped on a Iraqi warehouse full of postal scales at some point. (Since you need accurate weights and measures to make biological weapons.) And to assorted machining equipment. And so on.

(And now that I've wasted this much time on this tiny textbox...)

And that in turn implies that the whole Iraq paniq of 1991-2003 was based on assorted people pointing at any piece of industrial equipment and screaming 'dual-use'. And any attempt to import dual-use stuff was proof positive then that Saddam Hussein was planning to take over the world.

That mindset was well illustrated in the next-to-last 'Saddam really had WMD!' articles I saw. (Which seem to pop up every two months to keep the hopes of the faithful alive.) In that particular article they were down to this army guy saying that of course Saddam had WMD, because he had ORGANO-PHOSPHATES. Ne'ermind Iraq has been mining phosphates for at least 50 years. Organo-phosphates can be used for terrorism since it is used in FERTILIZER! Yay!

Thusly: Oklahoma City -> Ryder truck -> fertilizer -> Timothy McVeigh -> shadowy Islamofascist influence -> phosphates -> Iraq -> SADDAM HUSSEIN!

Wasn't that fun?

['New! Improved! Islamofascist™-brand Fertilizer! Now kills 20% more Americans!']

Posted by: ash | Oct 27, 2004 6:18:27 AM

Up to this point some folks have touched around the edges but no definitive answer yet. Let me try.

Powerful shaped charges made from RDX are used because of the speed of propagation of the blast wave. But this is not the only critical variable, but a very necessary one, in creating a high order device.

“Cyclotrimethylene trinitramine, also known as RDX, cyclonite, or hexogen, is an explosive material widely used by the military.
In its pure synthesised state it is a white crystalline solid. As an explosive it is usually used in mixtures with other explosives and plasticizers or desensitizers. It is stable in storage and is considered the most powerful and brisant of the military high explosives.
RDX forms the base for a number of common military explosives: Composition A (wax-coated, granular explosive consisting of RDX and plasticizing wax), composition A5 (mixed with 1.5% stearic acid), composition B (castable mixtures of RDX and TNT), composition C (a plastic demolition explosive consisting of RDX, other explosives, and plasticizers), composition D, HBX (castable mixtures of RDX, TNT, powdered aluminium, and D-2 wax with calcium chloride), H-6, Cyclotol and C-4.
It is a colourless solid, of density 1.82 g/cm³. It is obtained by reacting concentrated nitric acid on hexamine. It is a heterocycle and has the shape of a ring. It starts to decompose at about 170°C and melts at 204°C.
At room temperatures, it is a very stable product. It burns rather than explodes, and only detonates with a detonator, being unaffected even by small arms fire. It is less sensitive than pentaerythritol tetranitrate (PETN). However, it is very sensitive when crystalized, below -4°C.
The discovery of RDX dates from the 1890s when a German (Hans Henning) offered it as a medicine. Its explosive properties were not recognized until 1920 (Herz?). In the 1920s RDX was produced by the direct nitration of hexamine. It was only in 1940 that an efficient production method was found, possibly at the McGill University Department of Chemistry (Meissner?). It was widely used during WW II, often in explosive mixtures with TNT such as Torpex (TNT (42%),RDX (40%) and aluminium (18%)). RDX was used in one of the first plastic explosives.”
From: http://en.wikipedia.org/wiki/Cyclotrimethylene_trinitramine

A very simple 'device' can be made essentially like a gun barrel with a sub-critical mass at one end – far enough apart to not go 'BOOM' when you don't want it to – and by firing the other ½ of the mass into it. This will make a 'relatively' low order bomb. Little Boy was just this sort of device.

“The Mk I "Little Boy" was 10 feet (3 m) in length, 28 inches (71 cm) wide and weighed 8,900 lb (4000 kg). The design used a gun arrangement to explosively force a sub-critical mass of uranium-235 and three U-235 target rings together into a super-critical mass, initiating a nuclear chain reaction. The yield of "Little Boy" was about 13 kilotons of TNT equivalent in explosive force.

At the time there had never been a test explosion with this type of weapon. The only test explosion of a nuclear weapon was with the plutonium-type, on July 16, 1945 at the Trinity site. This was because tests of controlled nuclear reactions with U-235 (as opposed to the uncontrolled reaction that occurs in a bomb) had already been done, and the principles involved were so simple that it was taken to be unnecessary to test the weapon in advance. The military were also anxious to drop the bomb, and testing the device would have delayed its use until more uranium was ready.”
From: http://en.wikipedia.org/wiki/Little_Boy

It's not a simple thing to make a high order nuclear device. In order to increase the output the timing has to be exquisitely sharp! So that brings us back the propagation speed of the explosion wave front.

“Brisance: The rapidity with which an explosive develops its maximum pressure is a measure of the quality known as brisance. A brisant explosive is one in which the maximum pressure is attained so rapidly that a shock wave is formed, and the net effect is to shatter material surrounding or in contact with it. Thus brisance is a measure of the shattering ability of an explosive.”

From: http://en.wikipedia.org/wiki/Brisant

This 'quickness' of the explosive helps the need for precise timing so the bomb engineers can accurately predict when critical mass will be achieved and aid the yeild by adding "other things" too. But the critical component is the amount of compression the reaction mass achieves.

One of the main problems with the Manhattan project (Fat Boy) was this timing. It wasn't just the explosive, but how to control the igniter so that each shaped charge initiated at precisely the same instant so – as one other poster mentioned – there would not be 'dead spots'.

One poster mentioned how small our modern devices are... That's very true but anyone less than a state sponsored program would (IMO) have difficulty producing, from scratch, such a compact device. This is not to suggest that some of the unaccounted for FSU devices couldn't be used for nefarious purposes.

But there is a stumbling block there too. Sophisticated devices rely on critical timing to produce a high order yield. The initiation sequence must be completely understood and controlled. It's not at all like the movies where the digital timer reaches zero and the bomb booms. Without all conditions being met, the giant boom just won't happen. But even a 'firecracker' bang would be disastrous!

Powerful high order explosives are 'relatively' easy to obtain. It's the fissile material that is the issue. Combine the two and a relatively simple device is easily made. Put into a container and deliver it to a major port city and...

But don't be paranoid or anything like that. Shrub's made us safer!

Posted by: SAQuestor | Oct 27, 2004 6:27:03 AM

Gosh, I wish I'd of read Ash's post before spending 45 minutes on mine. Per-zactly!

Posted by: SAQuestor | Oct 27, 2004 6:34:23 AM

I suppose this is an interesting question from a nuclear engineering standpoint, but it is highly doubtful that that the people who took the material wanted to use it to trigger a fission device. Too much other material is needed for a fission device, such as fissile material, the ability to form and shape it appropriately, and so forth. They probably just wanted the material to use to blow things up. Which they seem to be doing quite handily.

Posted by: raj | Oct 27, 2004 8:00:47 AM

Thank you ash and SAQuestor, that was very informative. Both of you evince an unusually high degree of knowledge about the manufacture of nuclear weapons. I hope you won't mind that I forwarded your comments, along with your e-mail adresses to the Department of Homeland Security for their edification.


Posted by: Barry Freed | Oct 27, 2004 8:39:25 AM

A physicists perspective - note, however, that I have never been privy to ANY classified information about weapons design.

As has been said, most of the US and USSR fission bombs were Plutonium implosion devices, typically yielding about 12 - 20 kilotons. (At least some of the nuclear artillery were gun devices and the the 1953 Grable shot -
http://www.aracnet.com/~pdxavets/upshot3.htm - upper left picture - was a U238 gun device - as far as I know the only such device exploded by us except for Hiroshima.)

The gun devices are simple, but the implosion devices require a spherical blast to compress the core. Since the shock speed is about 3 km / sec - or 3 mm per microsecond - 10% sphericity of a 30 cm shell would require about 10 microsecond simultaneity of the trigger explosions, and my understanding was that this was tough to do until the conventional explosives were improved. So that is one factor, just a guess, but I bet it's easier to do this with these high performance conventional explosions.

Now, if you want to make a fusion bomb, you use the gamma and X rays from a fission bomb as the trigger, These have to go through the explosives used as the fission trigger (now a cloud of gas, but moving so slowly - about 100,000 times slower - that they might as well be standing still as far as this is concerned). Again, my understanding is that going to these
advanced explosives, which cuts down on the mass of explosives required, substantially improved the X / gamma ray brightness of the fission triggers, and thus substantially improved the yield / weight ratio of fusion devices. (Since our
best yield to weight ratio is about 1 kiloton / pound, a 20 megaton bomb therefore weighs at least 20,000 pounds or 10 tons, and this keeping the weight down is a real issue.)

So, my guess is that these high performance explosives may be viewed as precursors of a fusion program, and that may be the real IAEA interest. (Remember that the Indians went straight to a fusion device, so might any future large state supported development program.)


Posted by: Marshall | Oct 27, 2004 9:11:50 AM

Excellent posts from Ash and SAQuestor.

Another example of why we physicists smelled a rat when we were told that Saddam was a nuke threat because if someone gave him a few hundred pounds (only!) of HEU he could have a bomb in 2-5 years.

My university lab could have a similar bomb in 3-5 years given the same. As could any decent physics lab.

Maybe this is why scientists are so un-impressed with the Bushies.


Posted by: Jeffrey Harris | Oct 27, 2004 9:12:33 AM

The comments to this entry are closed.