Unlike the Little Boy and Fat Man fission bombs, about which a great deal of information is available in the open literature, virtually no information has been released publicly about the design and construction of thermonuclear weapons. Most of the information that is publicly available is based largely on speculation and supposition, gleaned from open sources and interviews and a few declassified documents.
In contrast to the uranium and plutonium atomic bombs, which depend for their energy on nuclear fission, in which heavy nuclei are broken into lighter components, the hydrogen bomb depends on the energy released during nuclear fusion, in which small light nuclei are fused together to produce larger heavier nuclei. Physicists had known for decades that the fusion process releases an enormous amount of energy. In 1920, astrophysicist Arthur Eddington was the first to suggest that nuclear fusion of hydrogen to make helium was the process that fueled the cores of stars, and this hypothesis was confirmed theoretically in 1928 by George Gamow and elaborated by Hans Bethe in 1939.
Since nuclear fusion required fantastically high temperatures (to allow the hydrogen nuclei to overcome their mutual electromagnetic repulsion and fuse together), it was presumed that no earthly process would be able to produce it. The atomic bomb changed that. In 1942, during a conference to discuss the theoretical aspects of the atomic bomb, Teller spoke about the possibility of making a “Super Bomb” – using the temperatures produced by a fission bomb to set off the even higher-power thermonuclear reaction in a supply of liquid deuterium (an isotope of hydrogen), producing a bomb potentially 1,000 times as powerful as the plutonium or uranium bomb. While the other Manhattan Project physicists dismissed the “Super” as impossible to develop in the time available, Teller focused almost exclusively on it, and carried out an amount of theoretical research on the concept. When the war ended, most of the Manhattan Project’s scientists left.
After the Soviet A-bomb test in 1949, however, Teller once again took up the cause of championing the Super, and the Atomic Energy Commission’s General Advisory Committee, made up of Manhattan Project veterans, was asked to prepare a report on whether a crash program should begin to produce a Super Bomb. Most Committee members were opposed to the idea. President Truman, however, decided to fund development of a workable thermonuclear weapon as rapidly as possible.
The inescapable first step in producing a hydrogen bomb is the production of an efficient fission bomb. By 1950, many advances had been made in the design and construction of plutonium implosion bombs. The hollow levitated composite core, the development of explosive lens systems using larger numbers of individual units (and thus thinner explosive layers), and the development of the pulsed neutron emitter (essentially a small nuclear ion accelerator) as an intiator, had led to fission weapons that were smaller and lighter than Fat Man, but could produce yields in the 100-kiloton range.
The first design offered for a thermonuclear weapon was the “classical Super” proposed by physicist Edward Teller back during the Manhattan Project. In this proposal, a fission bomb would be used to ignite fusion in a small container of tritium, and this reaction would in turn be used to ignite thermonuclear burning in a much larger supply of liquid deuterium. But there were serious questions about whether the fusion temperatures could be maintained long enough for any significant fusion to occur in the deuterium. Answering that question required complex mathematical calculations that could only be done by computer, and the crude electronic computers available at the time were incapable of the task. Theoretical work on the Super was put on hold.
In 1946, Teller proposed a new design of thermonuclear weapon, which he called “Alarm Clock”. In this proposal, the fission core would be surrounded by several alternating concentric layers of uranium and fusion fuel (it was expected that the compound lithium-6 deuteride would serve as fuel, since under neutron bombardment the lithium component would break into two tritium nuclei, which would then fuse with the deuterium). Upon implosion, the enormous temperatures inside the explosion would produce a fusion reaction in the fuel layers, and these fusion reactions would release fast neutrons that would fission the uranium layers, thus increasing the yield, while the surrounding heavy tamper would help hold everything together.
In 1949, Soviet physicist Andrei Sakharov, later to become a prominent political dissident but then in charge of the USSR’s thermonuclear program, came up with a similar idea that he called sloika, “layer cake”.
In 1947, Teller produced a new idea called a “Booster”, which would use fusion reactions to increase the yield from implosion fission bombs. In this design, a small amount of tritium and deuterium would be inserted into the center of the plutonium core. When the core fissioned, the high temperatures would spark a fusion reaction in the confined T-D, and this in turn would release a large shower of neutrons that would produce further fissions, thus significantly boosting the yield. Although not technically a “hydrogen bomb”, the Booster did provide a way of producing powerful nuclear explosions with small packages.
Schematic diagram of a boosted fission bomb. Not to scale.
By 1950, when President Truman ordered fullscale work on the hydrogen bomb, computer power had reached the point where the calculations could finally be done on Teller’s “classical Super”. Oddly, though, the computers were beaten to the punch by mathematician Stanislaus Ulam, who devised a simpler method of doing the calculations by hand – and discovered that the classical Super would require unrealistic amounts of tritium, and even then would not be able to support much of a thermonuclear reaction in a separate supply of deuterium. This was shortly afterwards confirmed by computer pioneer John von Neumann’s new machines.
That left only the Alarm Clock design, which was, to the American team, disappointing. The primary advantage of the Super was that it did not have a critical mass and was not compressed – once ignited, the thermonuclear reaction would presumably continue until it burned itself out. In theory, then, the Super could be of unlimited explosive power – bigger bombs could always be produced simply by adding more deuterium fuel. The Alarm Clock, however, was limited by the size that could be effectively imploded, and that set a maximum yield of about one megaton (one million tons of TNT).
So the Alarm Clock design was abandoned, and a series of tests called Greenhouse were contemplated to further study the fusion reaction. One of these tests, called Greenhouse Item, used the Booster concept to double the yield of a small implosion bomb to 45 kilotons. Another test, called Greenhouse George, was scheduled for May 1951. It was a test that used a fission bomb to ignite a small amount of tritium and deuterium that, rather than being contained inside the bomb as with the Booster and the Alarm Clock, was stored in a container called the Cylinder, that was separated from the fission bomb by a short channel inside the bomb casing. This setup was intended to allow more accurate measurements to be made of the fusion temperature and pressures. But shortly before the George shot, a theoretical breakthrough was made which increased its importance.
After finishing his mathematical calculations on the classical Super, Stanislaus Ulam turned his attention to improving the efficiency of fission weapons. In January 1951, he came up with the idea of placing a hollow tube of uranium or plutonium inside the bomb casing but outside the fission explosive lens assembly. The intense radiation pressures produced by the implosion trigger might, he thought, flood the bomb casing and momentarily produce enough pressure to squeeze the hollow tube into a solid rod, in effect imploding it into a critical mass which would then add to the yield. When Ulam told Teller about the idea, Teller put two and two together. If he were to replace Ulam’s plutonium tube with a separate container of fusion fuel, the same radiation pressure from the trigger explosion would compress and heat it, setting off a thermonuclear reaction. A few weeks later, Teller added the idea of placing Ulam’s hollow plutonium tube (now known as the “spark plug”) inside the fusion fuel, where it would be imploded by the radiation pressure and explode, increasing the efficiency of the thermonuclear fuel.
When the George test shot, which happened to be scheduled for a few months later, demonstrated that a fission bomb could indeed produce sufficient pressures outside its core to perform a secondary implosion in a separate second stage, the “staged radiation implosion” concept was adopted as the main theoretical method of H-bomb design. It became known as the Ulam-Teller design.
Plans were quickly made for a fullscale test of the Ulam-Teller concept, codenamed Ivy Mike, to be done in November 1952. Plans were also made for the fullscale industrial production of the lithium-6 deuteride compound that would be used as fusion fuel for the hydrogen bomb. In the meantime, since there was no lithium-6 available yet, the Mike device would use liquid deuterium boosted with tritium, instead. This required huge cryogenic and refrigeration equipment to keep the deuterium liquefied. As a result, Mike would not be a deliverable weapon, but simply a test device to prove the workability of the Ulam-Teller staged radiation-implosion design.
The Mike device contained two stages. The primary stage was a modified Mark 5 plutonium implosion bomb with a yield of about 45 kilotons. The secondary stage was a short distance away, separated by a hollow radiation channel. It contained a little over 250 gallons of liquid deuterium, held in a huge cryogenic pressure bottle called the Sausage. Inside the center of the Sausage was a tritium-boosted hollow spark plug of plutonium. The external casing of the Sausage was ten inches thick, and accounted for some 85% of the total weight of 82 tons. The finished Sausage was 20 feet long and almost 7 feet wide.
On November 1, 1952, the Mike device was detonated at Eniwetok atoll in the Pacific, producing a total yield of 10.4 megatons. It produced a fireball three miles wide, and left a crater 6240 feet wide and 164 feet deep.
After the Mike test, the Soviets moved quickly to develop the only workable thermonuclear design they had, Sakharov’s sloika. In August 1953, they detonated Joe-4, which produced an explosive yield of 400 kilotons, of which about 20% came from the thermonuclear reactions. It could be considered more as a highly boosted fission bomb rather than a “true” hydrogen fusion bomb. Its biggest advantage lay in the fact that, unlike the Mike device, which was a test apparatus that could not be delivered by bomber, the Joe-4 was a deliverable weapon system.
When sufficient quantities of lithium-deuteride compound became available, the US replaced its liquid-fueled Mike device with a solid-fueled version called the Shrimp. The Shrimp design was the final modification to the Ulam-Teller configuration, and became the basis for every American thermonuclear weapon since.
Schematic diagram of the Ulam-Teller configuration. Not to scale.
On March 1, 1954, the Shrimp design was tested at Bikini atoll (Operation Castle Bravo), using lithium-deuteride fuel that had been enriched to 40% lithium-6. The predicted yield was 6 megatons, but an unexpected reaction of the lithium-7 (which was fragmented into tritium and helium by the fast neutrons – the tritium then undergoing fusion) produced much more fusion fuel than expected and pushed the yield to 15 megatons.
Four weeks later, a version of the Shrimp known as the Runt was tested at Bikini (Operation Castle Romeo). The Runt used un-enriched natural lithium, with 7% lithium-6 and 93% lithium-7. Again the yield was much higher than expected – the Runt produced a yield of 11 megatons.
The Ulam-Teller configuration in action. Step 1. The exploding primary trigger floods the foam-filled radiation channels at the side of the bomb with x-rays, which are re-radiated and reflected to the secondary. Step 2. The outer surface of the secondary‘s pusher ablates, or boils away, and the resulting pressure crushes the entire assembly inwards, compressing the thermonuclear fuel and imploding the plutonium spark plug. Step 3. The spark plug detonates, igniting fusion in the lithium fuel. The pusher’s momentum helps maintain the fusion reaction for a few microseconds, and finally the uranium-238 in the pusher undergoes fission from the fast neutrons released by the fusion. The whole process takes a few millionths of a second.
In the Soviet Union, meanwhile, Andrei Sakharov had independently come up with the idea of radiation implosion, and designed a true thermonuclear weapon that operated on the same principles as the Ulam-Teller configuration. In November 1955, Sakharov’s design was successfully test-fired. Although designed for a 3 megaton yield, during the test the Russians replaced part of the bomb fuel and pusher with inert materials to reduce the yield to 1.5 megatons.
The first American thermonuclear weapon to be deployed in large numbers was the Mark 17, which measured 5 feet wide, 24 feet long, and weighed 21 tons. The yield was 15 megatons. The bomb could only be delivered by specially modified B-36 bombers, and was provided with a parachute to slow its descent long enough to give the bomber enough time to get out of range. Some 200 Mark 17 bombs were produced in 1954 and 1955. They were phased out in 1956 after the B-36 was replaced by the B-52, and smaller H-bombs became available.
By the mid-1950’s, rocket technology had reached the point where payloads as heavy as a nuclear weapon could be delivered by ballistic missile. The first missiles were intermediate range, capable of delivering nuclear warheads to ranges of roughly 1,000 miles. In 1957, the Soviets introduced the R-7 rocket, which could deliver a nuclear warhead to the United States. The first ICBM deployed by the US was the Atlas in 1959. By the 1960’s, both the USSR and the US possessed hundreds of ICBMs capable of delivering nuclear warheads onto each other’s territories.
As the Cold War continued through the 60’s, 70’s and 80’s, both sides built up massive numbers of weapons. The primary advance in nuclear weapons design during this time was miniaturization, which produced smaller, lighter warheads for missiles. Early ICBMs like the Titan II carried single warheads -- the Titan II used the W53 warhead with a 9 megaton yield. The Minuteman I missile, introduced in 1970, used a solid-fueled rocket which could be kept deployed inside a hardened underground silo for long periods without maintenance. It carried a single W-56 warhead with a 1.2 megaton yield.
In 1970, the Minuteman III missile was deployed, which carried a MIRV (Multiple Independently-targeted Re-entry Vehicle) system with three W-62 warheads, each with an explosive yield of 170 kilotons. In MIRV, the warheads are dispersed in flight by a carrier known as the “bus”, which allows each warhead to be sent to a different target. In the 1980’s, the MX “Peacekeeper” missile carried ten MIRV warheads of the W-87 type, each with a 300 kiloton yield.
Size comparison of several nuclear weapons (to scale).
(upper left) Little Boy – 15 kilotons.
(upper right) Fat Man – 20 kilotons.
(center) Mark 17 hydrogen bomb – 15 megatons.
(lower left) B-61 thermonuclear bomb – 340 kilotons.
(lower center) W-87 MX Peacekeeper warhead – 300 kilotons.
(lower right) W-80 cruise missile warhead – 200 kilotons.
The most compact missile warheads are designed to use spherical secondary stages, which are smaller and lighter than the original cylindrical designs. These are designed to use oblong radiation reflectors known as the “peanut”. Although the smaller amount of fusion fuel contained in the spherical secondary produced a lower yield, this was not viewed as a problem since the warheads themselves were more accurate, therefore being able to knock out their targets with a smaller explosive yield.
Schematic diagram of MIRV thermonuclear warhead. Not to scale.
With the development of smaller warheads that could be carried farther by missiles, and with the increasing accuracy of the missiles themselves, therefore, the primary military role of the ICBM began to change. Instead of destroying cities with megaton-range warheads, ICBMs were transformed into smaller warheads (most in the 100-300 kiloton range), whose primary purpose was to accurately knock out small hardened targets like underground command centers and enemy missile silos. Thus, the primary purpose of the nuclear missile system was to knock out other nuclear missile systems. The role of “city buster” was adopted by the air-delivered bomb. (Both the US and the USSR declared that they were targeting “urban industrial targets” rather than the cities themselves, but this was a diplomatic political fiction – both sides knew that any nuclear attack on any “urban industrial target” would inevitably and unavoidably destroy the entire surrounding city and everyone in it.)
Another reason for the shrinkage in explosive yield was that, militarily, there simply wasn’t much need for yields over one or two megatons, which were large enough to completely destroy any enemy city. But the Ulam-Teller staged design is, theoretically, unlimited in its yield. More powerful bombs can always be made simply by adding another stage of fusion fuel, and allowing the radiation from the previous stage to implode it. The explosive yield can therefore, in theory, be expanded to any arbitrary level.
In October 1961, the Soviets tested the concept of multi-stage weapons by detonating a three-stage H-bomb, codenamed “Ivan” but afterwards nicknamed the Tsar Bomba. The bomb was designed for an incredible 100-megaton yield, but during the actual test, the fissionable uranium-238 pushers in both fusion stages were replaced by lead, reducing the yield to 50 megatons – still large enough to make it the biggest nuclear explosion ever.
The Tsar Bomba was never deployed – it was intended more as a propaganda show of Soviet military power. In the United States, though, the multi-stage concept was utilized in the B-41 aerial bomb, which used three stages to produce a yield of 25 megatons, the highest of any American nuclear weapon. Some 500 B-41 bombs were produced between 1960 and 1962, and they remained in service until 1976.
The standard aerial nuclear bomb today is the B-61 gravity bomb. Measuring 12 feet long and 13 inches wide, the bomb weighs about 700 lbs, and can be carried by any nuclear-capable US or NATO aircraft. The explosive yield can be varied according to the mission, ranging from a low of 0.3 kilotons to a maximum of 340 kilotons. One low-yield variant, known as the “earth penetrator”, has a reinforced nosecone and a rocket booster that allows it to bury itself underground before exploding, to destroy underground bunkers and command centers. Although the yield of the “burrowing bomb” is probably less than 10 kilotons, its subterranean burst would excavate a large crater and toss out a huge amount of irradiated fallout.
By the height of the Cold War, in the mid 1980’s, the US and the Soviet Union had each deployed enough nuclear weapons to incinerate the entire planet several times over. This nuclear overkill prompted a large anti-nuclear peace movement which organized to bring about a “nuclear freeze”, which would end the increase in the number of warheads, and work for a reduction in the number of nuclear weapons. The nuclear arms race, however, ended only with the collapse of the Soviet Union in 1991.