Unmaking the Bomb: A Fissile Material Approach to Nuclear Disarmament and Nonproliferation

Harold A. Feiveson, Alexander Glaser, Zia Mian and Frank von Hippel

Adapted from the introduction to Unmaking the Bomb: A Fissile Material Approach to Nuclear Disarmament and Nonproliferation (MIT Press, 2014)


It is seven decades since the first nuclear explosion. On 16 July 1945, in the Alamogordo Desert in southern New Mexico, the United States tested the plutonium bomb that it exploded 24 days later over Nagasaki. A bomb of much simpler design made from highly enriched uranium (HEU) was used against Hiroshima on 6 August 1945.

Announcing the atomic bombing of Hiroshima, President Harry Truman made public the enormous effort involved in making the fissile materials for new bombs:1 
“We now have two great plants and many lesser works devoted to the production of atomic power [fissile materials]. Employment during peak construction numbered 125,000 and over 65,000 individuals are even now engaged in operating the plants. Many have worked there for two and a half years. Few know what they have been producing.”

Since then, the technologies for uranium enrichment and plutonium separation pioneered by the United States have been mastered by eight other weapon states and also many other states, some of whom have considered but decided not to build nuclear weapons. Only one state (South Africa) has made and then renounced these weapons.

During the Cold War, the weapon states collectively produced for weapons over 2,000 tons of HEU and about 250 tons of separated plutonium. The number of nuclear warheads peaked at over 65,000. Warhead stockpiles have fallen, but 25 years after the Cold War’s end there remain about 10,000 operational warheads and components for many more. About 90 percent belong to Russia and the United States. The United Kingdom, France, China, Israel, India, Pakistan, and North Korea (in historical order) have about 1,000 warheads between them.

The fissile material problem is larger than this, however. Despite nuclear arsenal reductions, the global stockpile of fissile material as of 2014 was about 1,845 tons of HEU and plutonium. The ongoing dismantlement of tens of thousands of warheads has left large national stockpiles of excess fissile materials that have to be rigorously secured, and also, if not eliminated, could be used for weapons again.

Also, more HEU and plutonium was produced than was used for weapons. Hundreds of naval-propulsion reactors and research reactors are fueled with HEU, and more plutonium has been separated for civilian purposes than for weapons. All of this material is weapon-usable. There is in fact enough fissile material in the world today for about 200,000 simple fission weapons.

We wrote our book, Unmaking the Bomb, to provide a roadmap for practical policy initiatives to cap, reduce and eventually eliminate the global stockpile of weapon-usable fissile material in the world. It builds on analysis and reports prepared for the International Panel on Fissile Materials (IPFM). The Panel’s reports and the book offer a fissile material perspective on how to enable deep reductions of nuclear warheads, make nuclear disarmament more difficult to reverse, raise the barriers to nuclear weapon proliferation, and prevent possible nuclear-weapon acquisition by terrorist groups.

How the Nuclear World Emerged

The fact that the U-235 nucleus can be fissioned, releasing tens of millions of times more energy than the same weight of chemical explosives, was discovered in December 1938 in Germany, just before World War II. In March 1940, fearing a German atomic bomb, two refugee physicists at Birmingham University in England wrote a technical memo alerting the British government that an explosive nuclear fission chain reaction might be possible in a mass of nearly pure U-235 and allow for the production of a “super-bomb.”2

Uranium-235 makes up only 0.7% of natural uranium. The remainder is U-238 plus trace amounts of U-234 from the alpha decay of U-238. The two physicists noted, however, that “effective methods for the separation of isotopes have been developed recently” that could allow U-235 separation from natural uranium on a sufficiently large scale to permit construction of an atomic bomb. Their memo eventually galvanized the establishment of the U.S. nuclear-weapon program.

While the U.S. effort to design the atomic bomb, led by Robert Oppenheimer at Los Alamos, has captured most attention from historians, the largest investment of resources and people in the “Manhattan Project” was the effort in Oak Ridge, Tennessee to enrich uranium in U-235. Three different techniques were developed, but the one the U.S. adopted at the end of the war and used during the Cold War was gaseous diffusion. Gaseous uranium hexafluoride is pumped through thousands of porous barriers with the stream becoming slightly more enriched in U-235 at each stage because the molecules carrying the lighter U-235 atoms pass through the barriers more quickly than U-238. In 1941, a second material able to undergo a fission chain reaction was discovered, plutonium-239. Pu-239 has a half-life of 24,000 years and does not exist in significant concentrations in nature. It can be produced, however, by the capture of neutrons by uranium-238 nuclei in a nuclear reactor and the rapid beta decay of the resulting U-239 nuclei to neptunium-239 and then to Pu-239.

To produce plutonium for weapons, three high-power reactors were set up at the Hanford Site in central Washington State in late 1944 and early 1945. According to General Leslie Groves, the man in charge of the Manhattan Project, an isolated location was chosen for the production reactors because “no one knew what might happen, if anything, when a chain reaction was attempted in a large reactor.” One fear was “some unknown and unanticipated factor” might lead a reactor “to explode and throw out great quantities of highly radioactive materials into the atmosphere.”3

The plutonium has to be separated from the highly radioactive fission products in neutron-irradiated uranium. This is done remotely behind thick concrete radiation shielding in a chemical “reprocessing” plant. Once the plutonium is separated, it can be handled relatively easily in a “glovebox” which protects workers from inhaling carcinogenic plutonium oxide particles.

The Hiroshima bomb contained about 60 kilograms of HEU, while the Nagasaki bomb contained only about 6 kg of plutonium. This was because the Manhattan project developed two different types of nuclear-weapon designs. A simple “gun-type” design was developed for HEU. A much more difficult but efficient “implosion” design was developed for plutonium after it was discovered that the slow gun-type assembly would not work for plutonium. Pu-240 was being produced along with the Pu-239 and neutrons from Pu-240’s high rate of spontaneous fission would prematurely initiate a chain reaction.

Modern thermonuclear weapons (“hydrogen” bombs), also pioneered by the United States, use a “primary” fission explosion to trigger a much more powerful “secondary” fusion-fission explosion. These weapons generally contain an average of about 3 kg of plutonium in the “pit” of the fission primary and 15–25 kg of highly enriched uranium in the thermonuclear secondary.

The Soviet Union patterned its first fissile material production facilities, and its first weapon design, on those of the United States. Later, in the early 1960s, the Soviet Union broke new ground by shifting to gas-centrifuge technology for uranium enrichment. Uranium hexafluoride gas is spun at high speed inside a long cylinder so that the molecules carrying the heavier U-238 atoms are pressed more tightly against the wall. Combined with a circulation of the gas along the centrifuge rotor, this effect can be used to extract slightly enriched and depleted uranium streams from the machine. By connecting many such centrifuges in series and parallel, uranium can be enriched to any desired enriched level, from the 3–5% U-235 used in light-water reactor fuel to “weapon-grade” containing more than 90% U-235. Modern commercial uranium enrichment relies on this technology.

Britain’s nuclear weapon program was led by physicists who had participated in the U.S. wartime program. France followed Britain in its choice of technologies and scale. Many of China’s nuclear experts were trained in the Soviet Union, which also provided expert advisors and designs for fissile material production facilities, although the Soviet experts were withdrawn before China’s uranium enrichment and plutonium production facilities were completed. These five states are now recognized as nuclear-weapon states under the 1968 Non-proliferation Treaty (NPT).

Today, four additional states also have nuclear weapons. Israel received a complete plutonium production complex from France. India received assistance from the United States and Canada in building a plutonium production complex, nominally for a plutonium breeder reactor development program. Pakistan clandestinely purchased key technologies, components and materials from Europe’s gas-centrifuge production complex and received the design of a tested warhead from China. North Korea used the published design of a 1950s UK plutonium-production reactor and obtained gas-centrifuge technology from Pakistan. South Africa in the 1980s produced six nuclear weapons employing HEU. In 1990-91, it dismantled the weapons, placed the recovered HEU under international safeguards, and joined the NPT as a non-nuclear-weapon state.

Global Stocks

Global stocks of HEU

Figure 1. Global stocks of HEU and separated plutonium in metric tons, by category, as of 2014.4 Also shown are their weapon-equivalents (using an average of 3 kg of weapon-grade plutonium, 5 kg of reactor-grade plutonium, 15 kg of highly enriched uranium per warhead). The global stockpile of fissile material is now more than 200,000 weapon-equivalents.

There is considerable uncertainty about global fissile material stockpiles—only the United States and the United Kingdom have made public declarations of their inventories of military fissile materials. Estimates for the other nuclear weapon states carry uncertainties of 20-40 %.5

An estimated 940 tons of HEU and 140 tons of plutonium remain available for weapon purposes—mostly in Russia and the United States. Another almost 900 tons of HEU and 80 tons of plutonium have been declared excess for weapons use, of which over 660 tons of HEU have been eliminated by downblending it to low-enriched uranium for power-reactor fuel. The United States has allocated about 150 tons of its excess weapons HEU as a reserve for its military naval propulsion reactors.

France, Japan, Russia and the United Kingdom have declared a combined total of 260 tons of separated civilian plutonium as of the end of 2013, and about 70 tons of HEU are dedicated to civilian research reactor fuel. Starting after President Eisenhower’s Atoms for Peace speech in 1953, the United States and Soviet Union distributed HEU-fueled research reactors to 30 non-weapon states during the 1950s and 1960s. Many of these reactors have been shut down, and the United States and Russia have sought to convert the others to low-enriched uranium fuel.

The Nuclear Weapon–Nuclear Energy Link

The U.S. Atoms for Peace initiative also led to the founding of the International Atomic Energy Agency in 1957 with a mandate both to promote peaceful uses of nuclear technology and to monitor nuclear materials in non-weapon states to assure that they were not diverted to weapons uses. This approach to managing the proliferation risks of civilian nuclear energy programs was codified in the NPT.

Today, there are two principal civilian nuclear “fuel cycles.” The United States and most of the 30 or so countries with nuclear power plants use natural or low-enriched uranium (LEU) fuel containing 3–5% uranium-235 “once-through.” The discharged spent fuel is stored pending final disposal. This fuel system has the critical nonproliferation advantage that weapon-usable fissile material is nowhere easily accessible. LEU, defined as uranium containing less than 20% U-235 cannot be used for weapons without further enrichment and the plutonium in the spent fuel is not separated. There is no weapon-useable fissile material in such a system.  

If a country acquires an enrichment plant to produce LEU fuel, however, the plant could be converted rapidly to produce weapon-grade uranium. This possibility has been at the heart of international concern about Iran’s uranium enrichment program. The proliferation danger of national enrichment and reprocessing programs was recognized at least as early as 1946 when the U.S. proposed international control.6 Today, multinational control may be more realistic.7

A small number of countries have chosen a second nuclear fuel system, in which plutonium is separated for use as a reactor fuel. From the earliest days of the nuclear era, interest in civilian reprocessing was driven by the dream of breeder reactors that would produce more fissile material than they consumed, typically by converting uranium-238 into plutonium. Efforts at breeder reactor commercialization by over half a dozen countries largely failed, despite five decades of research, development and demonstration projects and a combined cost in excess of $100 billion.

The United States, United Kingdom, and Germany abandoned their breeder reactor efforts in the 1980s and 1990s, and France and Japan postponed theirs. Only Russia and India—joined in 2010 by China at a pilot scale—now separate plutonium with the intention of using it as fuel for prototype plutonium breeder reactors.

Despite the breeder dream having faded, France, Japan and the United Kingdom continued reprocessing. France and Japan decided to mix their separated plutonium with depleted uranium in mixed-oxide (MOX) plutonium-uranium fuel for existing light water reactors. This fuel cycle is much more costly than the once-through fuel cycle and also complicates radioactive waste disposal and in both countries the future of reprocessing is being debated. In 2012, with its nuclear utility refusing to renew its reprocessing contracts, the UK decided to end reprocessing when it completed existing contracts.

HEU (defined as uranium containing 20% or more U-235) in naval fuel cycles also is a security risk. The single largest illicit diversion of fissile material thus far may have been the several hundred kilograms of weapon-grade uranium that were secretly transferred in the 1960s from the NUMEC naval fuel fabrication facility in the United States to Israel with the cooperation of the plant’s owner.8 In 1993, a much smaller amount of HEU submarine fuel was stolen from a Russian storage facility.9 This incident helped focus attention on the need to secure Russian nuclear materials after the collapse of the Soviet Union.

The United States, Russia, the United Kingdom, and India fuel nuclear submarines with HEU, and the United States and Russia also operate HEU fueled ships. France, however, fuels its submarines and nuclear-powered aircraft carrier with LEU. It is believed that China also uses LEU fuel. Brazil, which is planning to be the first non-weapon state to have nuclear-powered submarines, also has chosen to use LEU. If the countries that use HEU fuel converted to LEU, much more military HEU could be eliminated.

Two hopeful signs are that Russia has recently developed an LEU-fueled reactor to power its future nuclear-powered icebreakers, and the U.S. Congress has become interested in the possibility of funding an R&D Program to develop LEU fuel for future U.S. nuclear-powered submarines and aircraft carriers.10

Eliminating Fissile Materials

The most effective and enduring way to deal with the dangers from fissile materials is to stop producing them and dispose of them as irreversibly as possible. This will require new policy and technology initiatives.

A Fissile Material Cutoff Treaty (FMCT) to end the production of fissile material for weapons was first proposed in the 1950s by the United States as a means to cap the relatively smaller Soviet fissile material stockpile and was rejected. The idea was revived at the end of the Cold War and has been under consideration at the United Nations Conference on Disarmament in Geneva since 1993. In recent years, with international attention focused on fighting Islamist militants, Pakistan has successfully stalled the start of talks as a way to buy time to build up its fissile material stockpile.

Under an FMCT, the IAEA would monitor enrichment and reprocessing plants to determine that any enriched uranium and separated plutonium produced is not used for weapons. If it were agreed, the storage and use of pre-existing fissile materials that the weapon states have declared excess for all military purposes too could be monitored.11 If the use of plutonium and HEU fuels were abandoned, an FMCT could be broadened further into a ban on the production and use of fissile materials for any purpose. This last option would have the largest nonproliferation impact.

The disposal of HEU is relatively straightforward. It is blended with natural or slightly enriched uranium to produce LEU that can be used as reactor fuel. Russia and the United States together have down-blended over 650 tons of highly enriched uranium. Given the size of current arsenals, much more HEU could be declared excess by Russia and the United States and sent for down-blending.

The disposal of plutonium has proven more costly and complicated, and has made little progress since the United States and Russia concluded a Plutonium Management and Disposition Agreement in 2000 that committed each party to dispose of at least 34 tons of weapon-grade plutonium. Russia plans to use its excess separated plutonium in prototype breeder reactors. The United States planned to fabricate most of its excess plutonium into MOX fuel for use in light-water reactors but the MOX fuel fabrication facility that the Department of Energy has been building in South Carolina became so costly that the Obama Administration decided that it is “unaffordable.”

There are less costly options for plutonium disposal. These include mixing the plutonium into the concentrated fission-product wastes from which it was originally separated as those wastes are embedded in glass for disposal in a deep geological repository. Another option is dilution and immobilization of the plutonium in a durable matrix and disposal in 3–5 kilometer deep boreholes.12


International efforts to reduce nuclear weapon stocks and to prevent proliferation and nuclear terrorism have been operating largely in parallel. The fissile material perspective presented here provides a common basis for these efforts. If we are to reduce the threat from nuclear weapons, we must deal with the dangers posed by the production, stockpiling, and use of fissile materials. Unmaking the bomb requires eliminating the fissile materials that make nuclear weapons possible.

Confidence in and verification of nuclear disarmament will be far easier in a world where there is no production or use of separated plutonium or highly enriched uranium and where fissile material stocks have been eliminated. Together, these efforts would make it more difficult and more time-consuming for any country to make fissile materials for nuclear weapons and would make it easier for the international community to detect and respond to what would be a clear threat to international peace and security.

Harold A. Feiveson, Alexander Glaser, Zia Mian and Frank von Hippel
Program on Science and Global Security, Princeton University


1. White House Press Release on Hiroshima, Statement by the President of the United States, August 6, 1945.

2. Otto Frisch and Rudolph Peierls “On the Construction of a Super-bomb’ based on a Nuclear Chain Reaction in Uranium”, Memorandum, March 1940.

3. Quoted in Richard Rhodes, The Making of the Atomic Bomb (New York: Simon & Schuster, 1986), p. 496.

4. International Panel on Fissile Materials, Global Fissile Material Report 2015.

5. For details on the published estimates used, see International Panel on Fissile Materials, Global Fissile Material Report 2010: Balancing the Books—Production and Stocks, http://fissilematerials.org/library/gfmr10.pdf

6. U.S. State Department, A Report on the International Control of Atomic Energy, 1946, www.fissilematerials.org/library/ach46.pdf

7. Alexander Glaser, Zia Mian and Frank von Hippel, “After the Iran nuclear deal: Multinational enrichment,” Science, 19 June 2015.

8. Victor Gilinsky and Roger J. Mattson, “Did Israel steal bomb-grade uranium from the United States?, Bulletin of the Atomic Scientists, 17 April 2014.

9. Oleg Bukharin and William Potter, “Potatoes were guarded better,” Bulletin of the Atomic Scientists, May/June 1995, pp. 46-50.

10. H.R.1735, National Defense Authorization Act for Fiscal Year 2016, Section 3142, Research And Development of Advanced Naval Nuclear Fuel System Based on Low-Enriched Uranium.”

11. International Panel on Fissile Materials, Global Fissile Material Report 2008: Scope and Verification of a Fissile Material (Cutoff) Treaty, www.fissilematerials.org/library/gfmr08.pdf

12. Frank von Hippel and Gordon MacKerron, Alternatives to MOX: Direct-disposal Options for Stockpiles of Separated Plutonium, International Panel on Fissile Materials, 2015, www.fissilematerials.org/library/rr13.pdf.

These contributions have not been peer-refereed. They represent solely the view(s) of the author(s) and not necessarily the view of APS.