Sleepwalking With the Bomb

by John C. Wohlstetter

  The good news is that it is very hard to make bombs; the bad news is that it is not impossible. Let’s look at uranium and plutonium and see why this is so.

  Uranium and Fission

  All uranium atoms have 92 positively charged protons at their nucleus (with 92 almost weightless negatively charged electrons orbiting that nucleus). But all uranium atoms are not the same. Though there is no way to tell one from another chemically, different isotopes of uranium have different numbers of neutrons, the proton’s neutrally-charged companion. Most atoms in a vein of natural uranium ore have 146 neutrons (for a total mass, protons plus neutrons, of 238). But a very few of them—less than 1 percent—have three fewer neutrons, and this “U-235” is extremely important for our story.

  U-238 is fissionable, but not readily so. U-235, on the other hand, is fissile—its nucleus is easily split, creating two smaller nuclei, but more importantly, releasing energy plus two or three free neutrons. In a small enough space, those neutrons can each enter other uranium nuclei, splitting them to release more energy and neutrons, and so on, in a chain reaction. A critical mass of uranium-235—roughly 100–115 pounds in metal form and smaller than a soccer ball—will start a self-sustaining chain reaction on its own. If the same object is sufficiently compressed, it can become supercritical, dangerously increasing the rate of the chain reaction.

  But the vast majority of uranium ore is U-238 and cannot emit neutrons rapidly enough to support a chain reaction. The solution for anyone seeking that reaction is for high-speed centrifuges to spin uranium that has been processed from ore to a powdered form called yellowcake, so that the marginally heavier U-238 molecules move to the bottom of the spinning cylinder, separating out from the precious, infinitesimally lighter U-235 that stays on top of the centrifuge. This process—of removing U-238 from U-235—is called uranium enrichment.

  In order to generate an uncontrolled, supercritical chain reaction in uranium (a nuclear explosion), a would-be bomb maker must: 1. sufficiently enrich the uranium, 2. compress it ultra-rapidly into a supercritical mass, and 3. set it in an explosion-friendly physical shape.

  For example, the Hiroshima bomb used uranium enriched to 80 percent U-235. Within the bomb, half the uranium was fired—by a miniature version of a World War II warship’s naval gun—into the other half, causing a supercritical mass to form and detonate in microseconds (millionths of a second).

  Supercritical chain reactions in uranium typically at least double with each fission. Think of the parable about the king who offers a peasant serial doublings of wheat stalks on a chessboard—one stalk of wheat on square one, two on square two, etc. Before reaching 64 doublings the kingdom goes broke; the final squares are never covered, as there is no wheat left with which to do so. The difference in the nuclear case is that doublings go past the 64th square—to the 84th. Exponential progressions look like the famed “hockey stick” curve, one that accelerates at an ever-increasing rate with each doubling.

  In the 84-doubling sequence not uncommon in a fission weapon, after 70 doublings only 1 percent of the energy will have been released. After 80 doublings only 5 percent will have been released, and after 83 doublings only 50 percent. North Korea’s early tests fell far short of the Hiroshima bomb in yield. A primitive weapon releases far less energy than a well-engineered one.

  Commercial fuel is not sufficiently enriched to attain supercriticality, but failure to control the reaction or a failure in the cooling system can lead to an uncontrolled chain reaction and a “meltdown” in which the reactor fuel in the core overheats and melts into the floor. This is a highly radioactive event, and highly dangerous to those exposed to the intense doses of radiation (few in number, if the containment vessel protecting the reactor remains intact). A runaway chain reaction cannot generate a nuclear explosion but in water-cooled designs can cause a hydrogen explosion from the reaction of steam with core-surrounding cladding, as happened in the 1986 Chernobyl nuclear accident in Ukraine and at several reactors in the March 2011 nuclear meltdowns in Japan.

  The Simple Arithmetic of Nuclear Proliferation

  At first glance it seems a huge leap for a nuclear proliferator state to get from 3.5 percent, low-enriched, commercial uranium fuel for a power reactor all the way up to 93 percent, highly enriched, weapons-grade uranium fuel for a bomb. But simple arithmetic gives a counterintuitive result: commercial-grade fuel is perilously close to weapons-grade fuel.

  Recall that significantly less than 1 percent of mined uranium is fissile—the less-desirable isotope makes up 139 out of every 140 uranium atoms. Commercial-grade fuel requires a minimum of 3.5 percent U-235, which means that 1 out of every 28 atoms must be U-235. Thus, for every 139 U-238 atoms, 112 of them must be removed. Now you can run a commercial nuclear reactor. To appreciate how close you already are to having nuclear weapons fuel, at this stage you have done 80 percent of the isotopic separation needed to build a full weapons-grade bomb of the kind in the U.S. arsenal.

  The next important step is 20 percent enriched fuel—four atoms of U-238 for every U-235 atom—that can run a medical research reactor. Reaching this step requires taking away a total of 135 out of the 139 U-238 atoms that originally accompanied each atom of U-235. At this stage, you have done 97 percent of the isotopic separation work needed to make a full weapons-grade nuclear bomb.

  Put another way:

  Natural Uranium Ore. In nature, there are 139 atoms of nonfissile U-238 for every one atom of fissile U-235.

  Commercial Nuclear Reactor Fuel. Remove 112 U-238 atoms, leaving 27 U-238 atoms and 1 U-235 atom. This makes 3.5 percent enriched uranium.

  Medical Research Reactor Fuel. Remove 23 U-238 atoms, leaving 4 U-238 atoms and 1 U-235 atom. This makes 20 percent enriched uranium.

  Weapons Grade Uranium. Remove 4 U-238 atoms, leaving 1 U-235 atom. This leaves 1 U-235 atom, 100 percent enriched uranium. U.S. weapons grade uranium fuel is 93.5 percent enriched uranium.

  Such separation work is by far the hardest part of the total work needed to assemble a nuclear device. One conservative estimate for Iran in early 2012 showed how enrichment times accelerate with higher levels of enrichment:

  Start with 14,000 kilograms (15 tons) of natural unenriched uranium ore.

  It takes 331 days to enrich to 1,400 kilograms of 3.5 percent commercial-grade fuel.

  It takes 37 days to enrich the 1,400 kilograms of commercial fuel to make 116 kilograms of 19.75 percent medical research– grade fuel.

  It takes only 8 days to enrich 116 kilograms of 19.75 percent enriched fuel to make 15 kilograms (33 pounds) of 90 percent enriched uranium for weapons-grade fuel, enough to make a single Hiroshima-size bomb.

  Other expert calculations assume a six-fold progression through the three stages, but let us assume ten-fold, to be conservative. In round figures apply two rules of thumb:

  10-10-10 for the three tenfold stages of material shrinkage listed above, from uranium ore to medical-grade to weapons-grade.

  11-1-1 for the three time periods: 11 months for commercial reactor fuel, then 1 month more for medical reactor fuel for research, then 1 week for weapons-grade fuel for a bomb.

  Thankfully, putting together the vast, industrial-scale infrastructure needed to enrich uranium via these methods is extremely difficult; no terrorist is going to do this in a garage or on a back lawn with presently available methods.

  To these 11-1-1 and 10-10-10 rounding rules noted above we can add one more number each, to complete the sequences. Adding another 1 to the first sequence tells us that once all components needed for a bomb are in place it takes about one day to assemble them into an operational bomb. Adding a final 10 to the 10-10-10 sequence captures the difference between the minimum amount needed for a crude uranium bomb a terrorist can use (roughly 60 kilograms—the amount used in the Hiroshima bomb), and the minimum amount needed for a highly sophisticated plutonium bomb that a first-rank nuclear state can use to optimize its nuclear arsenal (roughly 6 kilograms).

 
Some specialized reactors run on fuel enriched beyond commercial grade. Nuclear-powered submarines and surface ships actually run on weapons-grade fuel, because they must provide very high power in a very small space. Such fuel, if diverted, could make fuel for a nuclear weapon. (A submarine or surface-ship reactor, though running on weapons-grade fuel, cannot generate a nuclear explosion, for want of the necessary physical configuration and compression.)

  Now the bad—very bad—news: You do not need a full U.S. weapons-grade enriched bomb to get a nuclear explosion. Less than 20 percent enriched uranium suffices. In 1962 the United States tested a uranium bomb at its Nevada underground test site, and obtained a nuclear explosion with fuel enriched somewhat short of 20 percent (the exact figure remains classified). It was, in the parlance, suboptimal. If detonated in a city, such a bomb would cause less devastation and kill fewer people than a full U.S.-grade enriched bomb. But its destructive power could still be immense. The 1,336-pound (two-thirds of a ton) conventional truck bomb that exploded in a garage of the World Trade Center in 1993, had it been more carefully placed a few of yards away, would have toppled one tower into the other, killing many tens of thousands. The much bigger 1995 Oklahoma City bomb, which destroyed a large federal building and killed 168 people, used two and a half tons of conventional explosive. A “puny” A-bomb (like that detonated in North Korea’s 2006 plutonium test, for example) could easily be equivalent to a few hundred tons of high explosive.

  Plutonium, Fission, and Fusion

  So much for uranium, the fuel of choice for proliferators. But what about plutonium? Plutonium barely exists naturally—the young American nuclear chemist, Glenn Seaborg, found it by making it from U-238,25 and every day more accumulates in the spent fuel collected from nuclear reactors. The U-238 in nuclear reactors will catch a neutron, and instead of fissioning, become an extremely unstable atom with 239 neutrons and protons. In a series of transmutations (changes in chemical composition), this U-239 naturally becomes fissile plutonium-239, the most common modern fuel for nuclear weapons.

  How a reactor is designed and run determines how readily and conveniently it creates that plutonium-239. The reactor the Iraqis built in the late 1970s was to run on weapons-grade fuel and was made to maximize plutonium production. Israel understood this perfectly well, and hence destroyed it in 1981, before it was fueled, to avoid scattering radioactive material for miles upon bombing it. Proliferation expert Henry Sokolski writes that a light-water reactor rated at a tenth the size of a commercial plant can be run so as to produce dozens of pounds of plutonium in a year. This is more than enough to fuel several nuclear bombs.

  Because a reactor can produce plutonium, a terrorist might think of stealing nuclear waste to obtain it. But plutonium is just one component of some forms of nuclear waste, and most plutonium in nuclear waste is not fissile. The longer the newly made Pu-239 sits in a reactor, the longer the neutron-capture process goes on, producing heavier, less controllable, forms of plutonium.26 These soon outnumber fissile Pu-239, and are hard to separate from it. This problem can be avoided by replacing fuel rods before they absorb too many neutrons.

  Weapons-grade plutonium is a more efficient bomb fuel than weapons-grade uranium, and thus offers more explosive power per pound. The actual amount of plutonium converted into energy released by plutonium-239 nuclei that fissioned inside the core of the Nagasaki bomb was about one gram—one-third the weight of a penny. Einstein’s E = mc2 equation explains this. The released mass (m) is infinitesimally small—less than a thousandth of the mass that fissioned, as most of what fissioned careened around in search of other nuclei to split; the remainder was converted into and released as kinetic, thermal, and radiation energy. But the “c2” represents the square of the free-space speed of light in kilometers per second, a huge multiplier that explains the vast energy liberated from an infinitesimally tiny nucleus. Applying this to every atom whose nucleus is split in a nuclear detonation yields a vast release of energy in various forms.

  But Pu-239 is much harder to make into a nuclear bomb. It must be placed in a special configuration, far more complex than that for a uranium bomb. The Manhattan Project scientists were so certain a gun-trigger design would work with uranium that they did not even test it—uranium was in short supply and they needed it to create plutonium for the Trinity test and then the Nagasaki bomb.

  A plutonium detonation occurs in about a nanosecond (a billionth of a second), a thousand times faster than a uranium detonation. To make sure as much of the plutonium as possible fissioned, the Trinity and Nagasaki bombs were “implosion” devices. A complicated arrangement of 32 symmetrically spaced conventional explosives surrounded those bombs’ plutonium cores. Thirty-two lenses converted the shock waves from convex to concave, to compress the plutonium core extremely rapidly and symmetrically. A timing discrepancy among the implosion lenses of one-millionth of a second reduces symmetry and can create a dud; a timing discrepancy of 10 microseconds—10 millionths of a second—is enough to create a partial dud. In essence, plutonium bombs require super-speed, super-symmetry, and super-small compression.

  For a nuclear weapons state seeking to use missiles to carry nuclear warheads, plutonium is the fuel of choice, because it provides more yield per pound, and thus is more suitable for small warheads. It is very unlikely that terrorists would be able to build a plutonium fission device on their own, due to the extreme sophistication involved.

  And it is even harder to master the deep subtleties of a hydrogen bomb. This requires a conventional explosive to trigger an atomic bomb, whose radiated thermal energy then compresses the plutonium core so rapidly and compactly as to fuse hydrogen atoms and generate a thermonuclear explosion.

  Terrorist Bombs and Military Bombs

  In the parlance of nuclear proliferation there are three significant nouns commonly added after the adjective nuclear: “capability,” “device,” and “weapon.” A nuclear capability means the ability to make a nuclear device or weapon. A nuclear device is the kind of weapon we have been worrying about since 9/11, a bomb too large to be delivered by traditional military means, but which can be put into a van, truck, or shipping container. A nuclear weapon denotes a bomb compact and light enough to fit into a missile warhead, or the business end of a bomb or artillery shell.

  A nuclear device is the kind of crude weapon a terrorist would use. Once the necessary amount of enriched uranium is in hand, a crude terror device can be easily assembled. How much is needed is design dependent: for the simplest device more is needed than for a sophisticated bomb. A nuclear weapon is the kind of bomb we worried about during the Cold War and is what proliferators like North Korea and Iran are working on. What remains for them is to achieve the requisite miniaturization required to place a nuclear bomb inside a missile nose cone.

  But a nuclear state need not have a weapon to aid terrorists. A device will suffice. For terrorists, a uranium-fueled atom bomb with a gun trigger is the preferred route. But even this is not duck soup for an individual, and not just any design suffices.27 The damage a crude nuclear device—in proliferation parlance, an “Improvised Nuclear Device” or IND—can inflict if detonated in the nation’s capital was recently assessed by the Federal Emergency Management Agency. A 10-kiloton blast—about 70 percent of the explosive yield of the Hiroshima bomb—would extinguish nearly all human life and obliterate most structures within a half-mile. Glass would be shattered out to 10 miles, radioactive dust would spread at least 20 miles. Severe structural damage would reach 1.5 miles from ground zero, with many casualties due to blast shockwave and thermal effects. Out to nearly 5 miles there would be light structural damage, plus numerous casualties from radioactive particulate fallout. In all, the study estimated 45,000 fatalities and 323,000 injured. This would overwhelm local medical facilities. (The 9/11 jetliner explosions yielded about 1/10 of a kiloton.)

  To go nuclear using currently available methods, a terrorist organization needs the help of a state.28 This is especially true o
f the so-called “suitcase nuke” attack scenario. A truly man-portable nuclear weapon requires highly advanced miniaturization of components, to make it compact and light enough to be carried. Russia developed an atomic demolition munition (ADM) the size of a footlocker, not designed to be carried by one person. Tales of Russian “loose nukes” floating around are dubious; and they likely wouldn’t work now anyway, as their core elements decay over time and eventually are no longer fissile.

  With this understanding of what is required to build a nuclear weapon, and a sense of the thin line between development of commercial nuclear power and the development of weapons, we turn to the Indian subcontinent.

  __________________

  23. As explained below, a “bomb” is either a military “weapon” or a terrorist “device.”

  24. In their 2007 book Foxbats Over Dimona: The Soviets’ Nuclear Gamble in the Six-Day War, authors Isabella Ginor and Gideon Remez offer evidence that during the June 1967 conflict the Soviets planned to invade Israel and bomb Israel’s nuclear reactor at Dimona, where Israel makes nuclear fuel for its weapons. But Israel’s rapid destruction of Egyptian, Syrian, and Jordanian forces preempted Soviet plans. The Soviets moved faster in 1973, so it took a U.S. threat to stop them.

  25. This was in 1941, 11 years after the discovery of Pluto, and over a century and a half after a German apothecary and chemist, Martin Klaproth, discovered uranium in 1786 and named it after the planet Uranus, newly discovered that same year. U-238 decays—transmutes itself by releasing energy—in 23 minutes to neptunium-239, named after the planet Neptune; Np-239 then decays in 2.3 days to plutonium, P-239, named after Pluto.

  26. These include plutonium isotopes Pu-240, Pu-241, and Pu-242. The complex physics and chemistry of how they interact with Pu-239 are beyond this book’s scope.

  27. In 1976 an American high school student named John Aristotle Phillips sketched a design that Manhattan Project physicist Freeman Dyson judged might possibly work. Of course design alone does not a weapon make. As a kid I liked to sketch F-104 Starfighters, which I thought the coolest looking supersonic aircraft (dubbed the “Flying Coffin,” it was not always thought so cool by pilots who had to fly it). But sketching it did not mean I could build one.