All Things Nuclear, Part I

This is the first post in a series. This took a significant amount of time to write (most of the research had been done previously), so don’t expect these to come with a low delta between posts.

We’re going to start with The Bomb, we’ll get to Nuclear Power in subsequent posts.

First, an important disclaimer. I am not a nuclear physicist, nor have I worked with actual nuclear weapons in any capacity. This is obviously not expert testimony, this is merely a summary of publicly available information together with some of my own assumptions regarding capabilities. Obviously anyone “in the know” can find errors below, if you have any additional information that can be provided and you’re not legally or ethically forbidden from providing it, please do so in the comments. I’m offering this because a few people have asked for it, and because I believe that most of the public is under-informed regarding the scope of this subject and its potential impact.

All of the below information is declassified. Wikipedia links are provided for a lot of the below, but interested members of the public can also read the text of the Nuclear Non-Proliferation Treaty, the Strategic Offensive Reductions Treaty (SORT), and New START, which is the current arms treaty in place between the U.S. and Russia.

One can assume that it is possible that nobody is following these treaties entirely, but the one glaring difficulty with nuclear weapons is that it is impossible to test these turkeys without lots of immediate, recognizable side effects that are easily measured by the scientific community. You can’t set them off in the atmosphere, under ground, under water, or even in outer space without someone, somewhere, being able to know within a pretty small delta that you just triggered a nuclear device. You can’t enrich uranium without the proper equipment (and lots of it), a lot of time, and a supply of uranium… all of which is not precisely easy to do without someone noticing. (edited to add) Density Duck points out that if one includes modeling in “nuclear testing”, then sufficiently advanced nuclear-equipped nations can indeed continue testing without setting off nuclear bombs. This has merit, as a point: certainly the U.S., Russia, the UK, France, and China have enough empirical data to continue nuclear development via modeling. It is possible that both Pakistan and India cut this mustard. However, the NNSA (National Nuclear Security Administration) has argued that modeling is insufficient to provide for the long term viability for the U.S. nuclear weapon arsenal. Since they have a vested interest in supporting actual testing, take this with your own evaluation.(/edit)

Since these are the sorts of devices that (a) nations want to have but typically not use and (b) nobody wants to play around with building by the seat-of-your-pants, one of my assumptions is that it is significantly unlikely that anyone has greatly surpassed the capabilities that are reflected by their test programs.

Stealing plans wouldn’t be enough. You have to build the thing, and you won’t really know if you’ve done it right until you use it or test it.

Takeaway Summary

If you’re of a mind to look at the below with a TL/DR: unless we go to war with China, it is pretty unlikely in this writer’s opinion that a nuclear device in excess of “ones of kilotons” will be detonated by any agency, terrorist or state or otherwise, in a hostile action. The practical difficulty of acquiring a megaton device is quite high (all hyperbole to the contrary, Russia is not handing out working RT-2PM Topol warheads to random nations), and the traceability of all known nuclear weapons is also quite high (so if Russia *did* sell a working RT-2PM Topol warhead and someone set the thing off, it can be determined that the source of the nuclear weapon was Russia. This does not jibe with a strategy of Mutually Assured Destruction).

Even the potential insecurity of Pakistan’s nuclear arsenal is not quite as worrisome as you might currently surmise, as they have a small enough number of warheads to notice rather quickly if one goes missing.

More below the fold…

Yield and Scale

Before we go any farther, it is immensely important for everyone to understand the differences in scale for these devices.

The largest conventional explosion on record was either the Minor Scale test at the White Sands Missile Range, or the British Bang, depending upon whether or not you regard the Guinness Book as authoritative. At Minor Scale, the U.S. military detonated 4.8 kilotons of ANFO (the same explosive you see the Mythbusters play around with occasionally, albeit a much, much bigger pile of the stuff). This rated about a 4.2 kT explosion. At the British Bang, several thousand tons of conventional explosives left over from WWII were disposed of in the simplest manner possible. For the easily distracted, here’s a list of other non-nuclear explosions.

At the short end, you have these little guys, about as small as one can make a nuclear device. The Davy Crockett had a yield between 10 to 20 tons of TNT; somewhere between two and eight times as powerful as the bomb Timothy McVeigh used in the Oklahoma City bombing. They were designed for infantrymen to use as bazooka shells, which gives you an idea of exactly how small they had to be to not vaporize the trooper launching the thing. For comparison, the largest conventional bomb currently used in the U.S. arsenal is the Massive Ordinance Air Blast bomb, which clocks in at 11 tons in yield. Note that things at this end are considerably smaller than the Minor Scale explosion, above.

At the biggest end, you have the Tsar Bomba, the largest nuclear bomb ever detonated, with a yield of 50 megatons of TNT. The Tsar Bomba was too big to be mounted on an ICBM, weighing in at a massive 60,000 lbs. Practically speaking, no device of this scale was ever going to be used against a military target; it would simply be impossible to build enough of them to equip a suitable number of bombers that could penetrate a target nation’s air defenses.

In between, you have all of the more practical weapons that have been built on a large scale. Most of these devices fall in the 1-10 megaton range. China, for example, has never detonated a nuclear device in excess of 4 megatons… at least, not as far as I’ve been able to determine from the public record. South Africa only ever built gun-type nuclear devices and never detonated a thermonuclear device. India has tested thermonuclear devices, but nothing in the megaton range. Pakistan’s nuclear program has not been conclusively established to be thermonuclear; they claim a yield of 30-35 kT but estimates from seismologists studying the below-ground detonations rate them at around an upper bound of 12 kT. France has tested devices in the megaton range (2.6 mT), as has the UK (3 mT). The North Koreans have detonated a device that has been estimated to be between 0.55 kT and 12 kT.

Almost all of the current arsenal of U.S. nuclear devices is composed of W87 warheads mounted on Minuteman III missiles with a yield of ~475 kT, B61 bombs that range up to a theoretical 340 kT (modified versions of these are the “bunker buster” nukes you may have read about in the news in the last decade), and the larger B83 bombs that range up to 1.2 mT, and the W80 cruise missile warheads that have between a 5 and 150 kT yield. (edited to add) Peter points out in the comments that I didn’t include sub-launched nukes in this list. That’s correct, an oversight, but the W88 nukes are 475 kT, for the record (/edit). Random movie trivia: the two bombs stolen by John Travolta in Broken Arrow were B83 bombs. Russian warheads that are still available by treaty fall into a similar range. The list of operative nuclear devices is public. As you can see from those figures, the combined throw-weight of all of the devices in mT is far exceeded by the number of devices themselves, indicating by basic algebra that the devices themselves are on average less than a megaton in explosive yield.

To see what your actual effect radius is for various nuclear weapon yields, this web app is handy. You can choose your yield and location (Los Angeles), and move the blast radius around on a satellite map, to see what the impact would be in practice.

A ground-burst 1 kT weapon set off in Los Angeles would be devestating, but as you can see a nuclear explosion of that size at LAX wouldn’t do much to the rest of the city, or indeed anything even as close as Compton (although fallout would still be a concern). Indeed, even up to a 10kT range, the overall damage (in economic terms) would probably be much less than the Katrina or Galveston hurricanes. The human cost in lives lost would be dramatically higher, of course, but probably not on the scale to dwarf, say, the Haiti earthquake. A 1 mT device, on the other hand, detonated by an airplane at a few thousand feet over downtown Los Angeles would level a good portion of the city and set a nearly uncontrollable firestorm off that would probably engulf most of the basin.

Engineering

There are five types of nuclear devices that may be used as weapons. More reading on nuclear bomb engineering here.

The Hiroshima bomb was a gun-type nuclear device where a conventional explosive propells a fissile plug (Uranium 235 in this particular case) into a target of the same material. The explosive effect of the Little Boy bomb was about 13 kilotons of TNT, or 13kT – putting it just about 3 times bigger than the “Minor Scale” test with conventional explosives, above. These devices have a fairly low yield, and require a decent amount of enriched uranium to make; however, they’re fairly stupid-simple and thus have a low engineering barrier to entry – given enough nuclear fuel, just about any reasonably cautious entrepenuer could build one in his backyard using existing tools and equipment. In addition to being used in bombs, the gun-type nuclear device was a model for some U.S. nuclear artillery shells. There were three types of these shells: the W9, which was retired in 1957; the W19, which was retired in 1963; and the W33 which was retired in 1992.

The second type of device, an implosion device, crushes a subcritical sphere of fissile material until its density increases to the point where the device hits critical mass and (in technical terms) goes kablooie. These have a higher yeild than their gun-type cousins, but require much more in the way of mechanical precision to manufacture, as both the sphere and the explosives that induce the implosion have to be engineered properly for the device to explode with efficiency. If you machine the sphere improperly or place the explosives incorrectly, only part of the sphere will achieve enough density to go critical, and the rest of the nuclear fuel will be ejected by the resulting explosion before it can contribute to the overall energy of the device. There were nuclear artillery shells made with implosion warheads, in addition to bombs and missile warheads. The Fat Man bomb dropped on Nagasaki had a yeild of 21kT, for reference.

The third type of device is a higher engineered version of the second, which roughly doubles yield.

The fourth is the big daddy, a Teller-Ulam thermonuclear warhead, wherein a nuclear device is itself used to trigger a second-stage explosion. This is actually something of a misnomer, as you have a chemical explosion with a conventional explosive triggering a nuclear explosion which in turn triggers a thermonuclear explosion – so technically they’re three-stage explosions, but I digress. France, the UK, China, the U.S., Russia, and India have all demonstrated this capability, neither South Africa, Pakistan, or North Korea has done so. Israel, of course, is an unknown quantity.

The fifth is the dirty bomb; a conventional explosive surrounded by nuclear material, which will eject that material to cover a wide area, or a badly-engineered bomb of the first or second type that fizzles, or fails to reach full critical mass. The tricky part here is getting enough “boom” to spread out your radioactive material with the greatest area of effect, and carrying around enough radioactive material to spread around to have a major effect.

This stuff is, after all, hazardous to your health.

Use and Risk

Now, let’s talk about how that collection of things that go “boom” can translate into something actually going off in a populated area.

There are basically four types of nuclear attacks that can occur. The first is a large, technologically advanced, thermonuclear hydrogen bomb mounted on a missile. The second is the same, but in bomb form. The third would be an improvised nuclear bomb, probably made by cannibalizing one of the previous two; most likely not a hydrogen bomb but “just” a nuclear device. The last would be a dirty bomb of some sort; something that takes radioactive material (probably just waste) and uses a conventional explosive to disburse it.

The first and second (if it is sufficiently airborne when it goes off) gives you the full range of effects: the fireball, the blast wave, a decent amount of fallout, and the EMP. This Wikipedia page lists the range of effects from nuclear blasts.

These are the sort that are stuck at the end of a ICBM or delivered by a strategic bomber, pretty much only available via five major governments: surplus/stolen from the U.S., China, France, the UK, or the former USSR. India and Pakistan couldn’t duplicate this effect on U.S. soil using an ICBM (they don’t have the range), but theoretically could detonate a device like this from a conventional aircraft.

Currently the U.S. has only one major member of its arsenal, the Minuteman III. The Minuteman II and the Peacekeeper (MX) missiles have been retired. The Minuteman III that are still in service have a W78 warhead that has a yield of about 340KT. The W62 warheads that they replaced had a yield of about 170KT. The W62s were retired in 2010, so it’s possible (but of course unlikely) that one could be misplaced.

You can see a whole list of nuclear weapons on wikipedia, along with country of origin. Following the links will give you the yield range, size of the warhead, etc.

This is the Federation of American Scientists bible on nuclear effects, showing what would happen if a nuclear device was detonated over Detroit or Leningrad. They describe the effects of both a ground and an air burst in thorough detail.

Excerpts from here :

Surface Burst. A surface burst is an explosion in which a weapon is detonated on or slightly above the surface of the earth so that the fireball actually touches the land or water surface. Under these conditions, the area affected by blast, thermal radiation, and initial nuclear radiation will be less extensive than for an air burst of similar yield, except in the region of ground zero where destruction is concentrated. In contrast with air bursts, local fallout can be a hazard over a much larger downwind area than that which is affected by blast and thermal radiation.

High Altitude Burst. A high altitude burst is one in which the weapon is exploded at such an altitude (above 30 km) that initial soft x-rays generated by the detonation dissipate energy as heat in a much larger volume of air molecules. There the fireball is much larger and expands much more rapidly. The ionizing radiation from the high altitude burst can travel for hundreds of miles before being absorbed. Significant ionization of the upper atmosphere (ionosphere) can occur. Severe disruption in communications can occur following high altitude bursts. They also lead to generation of an intense electromagnetic pulse (EMP) which can significantly degrade performance of or destroy sophisticated electronic equipment. There are no known biological effects of EMP; however, indirect effects may result from failure of critical medical equipment.

A high altitude nuclear weapon will have an EMP effect over an **extremely** large area. Starfish Prime was a nuclear test in 1962 that detonated a relatively small nuclear weapon (1.4 MT) at 250 miles altitude, in space. The EMP messed up electronics in Hawaii, almost 900 miles away from the detonation point. Note that this sort of a detonation is nearly impossible for anyone without space capabilities. Potential nuclear terrorist angle for a movie coming out in 2025 -> nuclear terrorists hijack Virgin Galactic flight! My idea, you saw it here first, Hollywood!

Who launches the attack is important in this calculus, too. A terrorist organization is probably going to want a ground burst, for maximum photo op. A state actor, on the other hand, would probably want to detonate a high-altitude device to affect as much of the U.S. as possible via EMP. Regardless of what the attacker wants, of course, there are practical considerations. Detonating a device on an aircraft is probably very traceable to country of origin. A missile, of course, is right out if you don’t want people to know for certain who fired the attack. Backpack nukes (cannibalized nuclear triggers off of thermonuclear devices, a-la the movie Peacemaker or something of that nature) are man-portable but anything bigger than that is probably easily traceable to a point of origin.

Removing a trigger from a thermonuclear device and detonating it as a weapon itself is highly inefficient, as the triggers from thermonuclear devices are designed to give off most of their energy effect as soft X-rays (to trigger the secondary bomb), not kinetic energy. It would certainly produce a boom, but even with a moderate amount of research I’ve been unable to find out how big of a boom it would be.  NOTE: there is a severe engineering difficulty here.  Nuclear weapons are designed to be tamper- and fail-safe for a large number of readily apparent reasons.  Contrary to the movies, removing a trigger from a thermonuclear warhead would almost certainly turn it into an inactive brick, so you’d have to remove the fissile material and re-engineer the bomb itself.  This is not a trivial project. A nuclear physicist could probably crunch the appropriate numbers for us. Since the device isn’t designed for maximum efficiency, however, it’s certainly safe to say that tens of kilotons would be on the very, very high end of possible yield. The Castle Bravo explosion at Bikini Atoll, the first hydrogen bomb test by the U.S., occurred in 1954. The declassified info on U.S. nuclear bomb tests leads me to believe that the nuclear triggers for the Castle Bravo test were probably developed during Operation Ranger, where the yields ranged from 0.5 kT to 22 kT. This is bolstered by the fact that Ranger’s outputs were considerably smaller than the outputs of Operation Sandstone, its predecessor.

Addendums

Here’s Trinity, the first device detonated during the Manhattan Project. A absolute worst-case homemade “terrorist bomb” would probably look something like this, if it actually went off. A more practical terrorist bomb would have a significantly lower yield.