The power of fission and fusion can be unleashed in various ways to cause devastating explosions. The 13 and 21 kiloton explosions over Hiroshima and Nagasaki in August 1945 burned both cities to the ground, killing over 200,000 people instantly. Yet nuclear weapon States went on to develop far more destructive weapons that dwarf the power of these simple fission weapons. The reduced size and weight of these more advanced weapons also makes them much easier to deliver than earlier types.

At the height of the Cold War, thousands of U.S. and Soviet ballistic missiles on high alert were capable of delivering up to 10 independently targeted warheads at a time, each one twenty times more powerful than the Hiroshima bomb. Though decades of arms control agreements have slowly reduced the size of their arsenals, nuclear weapon States still possess the capability to destroy each other many times over.

The largest nuclear explosion in history was the 1961 Soviet “Tsar Bomba” test, which measured more than 50 megatons (3,800 times more powerful than the Hiroshima bomb). To generate an explosion of this magnitude using dynamite, it would require 50 billion kilograms (over 110 billion pounds) of TNT, which is more than the weight of all the cargo that has passed through London’s Heathrow airport in the past 40 years. Expressed in volume, this would amount to 18 blocks of TNT each as large as the Empire State building.

Countries have ultimately made the decision to test nuclear devices for a number of reasons—both technical and political (World Overview: Why test?). Though basic knowledge of nuclear weapons design has become public information, the challenging task of actually building functional devices is based on trial and error. Only through nuclear testing can a country confirm that its design and engineering has been successful and gain insights for potential further development. As scientists become confident that weapons will function reliably according to their specifications, militaries are able to incorporate them into their strategic doctrines. The CTBT’s ban on nuclear testing will greatly obstruct these types of developments, and the Treaty’s verification regime will assure that a nuclear test anywhere on the planet is detected.

Nuclear fission is a process in which a neutron collides with an atom’s nucleus, splitting the atom into two smaller atoms and releasing a significant amount of energy. Every collision also releases more neutrons, which in a critical mass of fissile material will sustain a chain reaction of fission. By manipulating the size and speed of the chain reaction, nuclear fission can be exploited for power generation or alternatively, for weapons of mass destruction.

HEU is defined as uranium with a concentration of at least 20 percent U-235; however, at 20 percent enrichment, 400kg of material are needed for a bare critical mass, more than is practical for a weapon. “Weapons-grade” uranium (also used in submarine and icebreaker propulsion) is defined as uranium enriched to at least 90 percent U-235. By contrast, low enriched uranium (LEU), which is used as fuel in the majority of nuclear power plants, generally contains only 3 to 5 percent U-235. The same technology used to make LEU for peaceful purposes can be configured to make HEU for weapons.

The gun-type weapon is produced through a rather simple process in which one mass of U-235 is ‘shot’ into another by conventional explosives, creating a critical mass. The impact generates more neutrons, ensuring a fission chain reaction.The gun-type nuclear explosion is the most inefficient in terms of burning up the fissile material; only about 1.4 percent of the HEU in the Hiroshima bomb actually fissioned.

Yet a large amount of fissile material is required to ensure that a nuclear chain reaction will take place. Therefore, gun-type weapons will necessarily be heavier and bulkier than other types of nuclear weapons. While this suggests that States seeking strategic nuclear weapons would look to more advanced designs, the simplicity of a gun-type device may be attractive to terrorists. A weapon of this type is too large to be mounted on a long-range missile, but it could be dropped from a plane or delivered in a truck or a shipping container.

In terms of testing, no nuclear components need to be verified with the gun-type design; only the conventional components must be tested. Although the United States was able to prove that the chain reaction would work, it did not conduct an actual nuclear test of this design before it was used to bomb Hiroshima. South Africa built seven nuclear weapons using this design, but verifiably dismantled them in the 1990s.

As explained above, creating a sufficient amount of HEU is the most challenging step of building this type of nuclear weapon, requiring raw materials, expertise, infrastructure, and massive amounts of energy. Because producing fissile material is so difficult and costly, it is highly unlikely that terrorist organizations would be capable of taking this route to acquiring HEU. It is more likely that they would steal it. Thus, it is crucial that countries around the world secure all existing stockpiles to reduce the risk of theft.

The world’s first nuclear explosion was an implosion device that used plutonium for fissile material. The Manhattan Project scientists who designed the device dubbed it the "gadget". It was detonated successfully on 16 July 1945 near Alamagordo, New Mexico, in what was called the Trinity test. A weapon of the same design was used a few weeks later in the attack on Nagasaki, Japan, on 9 August 1945.

Unlike uranium, plutonium does not occur naturally on Earth; it must be created in a nuclear reactor. It is a byproduct of all power reactors, but in order to be used as fissile material, it must be chemically separated from the rest of the highly radioactive waste. This is a costly and hazardous process requiring specialized knowledge, facilities, and equipment. It is thus highly improbable that terrorist organizations would learn to master this technique and more likely that they would steal the plutonium.

Therefore, it is crucial that countries around the world secure all existing stockpiles to reduce the risk of theft.It takes a smaller amount of plutonium than HEU to achieve a self-sustaining chain reaction of nuclear fission. However, plutonium’s physical properties are such that a gun-type device cannot combine two separate masses fast enough to achieve this critical mass.

For this reason, a nuclear explosion using plutonium actually begins as an implosion that relies on a sophisticated arrangement of high explosive lenses that must fire inwards simultaneously from all directions towards a plutonium pit.

In order to test an implosion device, the compression of the pit must be uniform and needs to be tested quickly enough to avoid a premature nuclear explosion, a so-called fizzle. Before or after this particular moment, the conditions are not right for sustaining the chain reaction until most of the fissile material has been consumed. This is a far more challenging engineering problem than building a gun-type device, but experts at the Nuclear Control Institute warn that it could still be accomplished by a small group of people with the right training and experience if they have access to plutonium.

For an instant, the plutonium is compressed to a high density, making the mass critical. However, simply compressing the plutonium to critical mass does not ensure that a nuclear chain reaction will begin. For this to happen, high energy neutrons need to be available at the moment of compression. Since relying solely on natural decay of plutonium is too risky, the certainty provided by a neutron initiator is needed.

The implosion design can also be used with HEU, allowing a smaller device to achieve the same yield as a gun-type device. China’s first nuclear test in 1964 was a 22 kiloton HEU implosion device named “596.” After the first Gulf War, UN inspectors discovered that Iraq was attempting to build an implosion bomb using HEU.

Fusion reactions power the sun and the stars. The fusion of deuterium and tritium, both heavy isotopes of hydrogen, releases energy as well as a neutron with seven times more energy than a fission neutron. Fusion’s energy output per kilogram of source material is much higher than that of fission.

Fusion can be used inside a fission explosion to improve the efficiency of the weapon (boosting), or a large amount of fusion fuel can be triggered separately (thermonuclear weapon). The fusion of deuterium and tritium is initiated by the extremely high temperatures and radiation that result from fission.

(Fusion) boosted fission weapons

When fusion takes place in a fission weapon, the high-energy neutrons are more likely to collide with fissile atoms. Those higher-energy collisions release even more neutrons than simple fission, speeding up the chain reaction. This enables more material to fission before the device blows itself apart.

Boosted weapons are typically implosion devices with deuterium and tritium gas introduced into the hollow pit in the centre of the fissile pit. As fission begins, the high temperature causes fusion, and the high-energy neutrons released by fusion accelerate the fission chain reaction. Boosting has led to a hundred-fold increase in the efficiency of fission weapons since 1945, and it plays a role in nearly every nuclear weapon deployed today. Pakistan reportedly tested a boosted fission weapon in 1998.

Thermonuclear weapons (also called hydrogen bombs)

Thermonuclear bombs yield explosions in the megaton range, that is, orders of magnitude more powerful than the atomic bombs described above. The standard “Teller -Ulam design” uses a fission primary to trigger a powerful fusion secondary. The x-rays released from the primary explosion compress and ignite the secondary. The massive release of high-energy neutrons from the fusion reactions in the secondary causes even the U-238 tamper (not fissile) to fission, allowing for massive explosive yields. Using these design principles, an arbitrary number of additional stages could theoretically be added to create a “doomsday device” that would dwarf even the Tsar Bomba. However, the main threat posed by thermonuclear weapons is their ability to pack huge amounts of explosive power into small, light-weight packages that can be delivered by missiles.

The first U.S. thermonuclear test was “Ivy Mike” in 1952, which was carried out on the Enewetak atoll in the Marshall Islands. The Soviet Union responded with “Joe 19” in 1955, a bomb which was air-dropped over the Semipalatinsk test site. Britain and France have both tested devices with yields exceeding two megatons, Britain in 1957 at Christmas Island and France in 1968 at Fangataufa Atoll. China first tested a thermonuclear weapon at its Lop Nur site in 1967, less than three years after its initial atomic test.

According to the Nuclear Threat Initiative, Israel may possess boosted and possibly even thermonuclear weapons, but it has never confirmed this nor has it, to public knowledge, tested any explosive nuclear device. India claimed that its Shakti 1 test in 1998 was a successful reduced-yield thermonuclear bomb, though many outside experts remain skeptical.

Mastering thermonuclear weapons technology makes it possible for countries to miniaturize their weapons, allowing for flexibility in delivery and yield. U.S. warheads such as the B83 can be adjusted on the battlefield for either a low sub-kiloton yield (primary only) or higher yields up to 1.2 megatons. Many of the missile-delivered warheads currently deployed by the nuclear weapon States are light-weight thermonuclear weapons that fall in the 100-300 kiloton range. Other types of nuclear weapons not believed to be currently deployed include neutron bombs; and cobalt bombs.

If a country or sub-state actor chooses to build nuclear weapons, its design choices will be constrained by the type of fissile material that it can obtain. Today only the five Nuclear Non-Proliferation Treaty (NPT) nuclear weapon States and the countries outside the NPT have access to unsafeguarded nuclear reactors and plutonium reprocessing plants, or uranium enrichment facilities, which are needed to create new fissile material for weapons. However, several countries have at various points initiated secret programmes in an effort to develop fissile material production capability. As of 2007, there were approximately 500 tons of separated plutonium and 1400 tons of HEU stockpiled worldwide.

As the mass of plutonium needed for starting a chain reaction is less than that of HEU, most countries have opted for plutonium as chemical separation (reprocessing) is easier and less energy intensive than isotope separation (enrichment). Also, plutonium allows for building lighter and more compact warheads that can be adapted to longer-range missiles. The first United States weapons were different types because HEU and plutonium production were both in very early stages and material was limited. The Soviet Union, United Kingdom, and France all tested plutonium implosion devices because of the availability of plutonium and the limited strategic value of gun-type weapons. The Soviet Union initially had a parallel program to develop a gun-type device but scrapped it as soon as its implosion test proved successful.

China’s first nuclear test in 1964 used an HEU implosion device because the country had not yet developed the means to produce plutonium.

Israel has a plutonium-producing reactor at Dimona as well as a reprocessing plant, which the International Panel on Fissile Materials estimates can handle about 40 to100 tons of spent fuel each year.

South Africa possesses significant uranium reserves and it pioneered a new uranium enrichment method, giving it a route to the bomb; its nuclear devices, which have been verifiably dismantled, all used HEU.

India’s 1974 nuclear explosion used plutonium extracted from its CIRUS heavy water reactor, supplied for energy purposes only (This misuse led to sanctions on India and the creation of the Nuclear Suppliers Group). Although it has now developed centrifuge enrichment capability, India claims that all HEU is intended for naval propulsion. To cope with its shortage of uranium reserves, India hopes to develop reactors that breed plutonium and to move eventually to a thorium fuel cycle.

Pakistan drew on the expertise of A. Q. Khan; and his procurement network to develop centrifuge enrichment capability. Since Pakistan did not initially have plutonium production capability, it is assumed that HEU was used for the tests it conducted in 1998. However, Pakistan’s plutonium production reactor at Khushab came online in 1998. These heavy water reactors will produce plutonium and tritium, key elements for thermonuclear weapons.

The Democratic People's Republic of Korea (DPRK) separated plutonium for weapons from its Yongbyong reactor after announcing its withdrawal from the NPT in 2002.