Marvin Miller

Security Studies Program & Department of Nuclear Engineering, Massachusetts Institute of Technology

presented to seminar at
Nuclear Control Institute
March 27, 2002


Under what circumstances would Japan seek to acquire nuclear weapons?

What political/legal route to weapons might it take?; in particular, what are the safeguards implications of a decision to withdraw from the NPT and utilize transferred nuclear technology?

What technical options would Japan have to produce fissile material for weapons; in particular, could it use its stocks of reactor-grade plutonium (RGPu) to make reliable, high-yield weapons, and is this option more attractive than others ?

What are the implications of the above for US nonproliferation policy? 



1.   Selig S. Harrison, "Japan and Nuclear Weapons", in Japan's Nuclear Future: The Plutonium Debate and East Asian Security, S. S. Harrison, Ed., Carnegie Endowment for International Peace, Washington, DC, 1996, pp. 7-44.

2.   Yuri Kase, "The Costs and Benefits of Japan's Nuclearization: An Insight into the 1968/70 Internal Report", The Nonproliferation Review, Summer 2001, pp. 55-68.

3.   Lillian Hoddeson et al., Critical Assembly: A Technical History of Los Alamos during the Oppenheimer Years, 1943-1945, Cambridge University Press, 1993.

4.    Lorna Arnold, Britain and the H-Bomb, Palgrave, New York, 2001. 




The official policy of the post World War II Japanese government has always been to disavow any intention to acquire nuclear weapons. The stated rationale is that Japan, as the only victim of nuclear attack, is especially sensitive to the immense human suffering caused by their use, and that it is protected from nuclear threats by other countries by the US nuclear umbrella. Moreover, Article 9 of the Japanese Constitution renounces war as a sovereign right of the nation, and prohibits the maintenance of military forces as well as other war potential.

Nevertheless, the intensive and protracted debate in Japan over whether and on what terms it should sign and ratify the Nuclear Non-Proliferation Treaty (NPT) - it signed in 1970 and ratified in 1977 - reflected the fact that some Japanese leaders wanted to keep a nuclear weapons option open. Indeed, in 1967 the Prime Minister E. Sato commissioned a secret study to assess the costs and benefits of nuclearization from the technical, economic, strategic, political, and diplomatic perspectives.

More recently, speculation about possible Japanese nuclearization has increased due to several factors including: (1) China's assertive political and military posture towards Japan; (2) North Korea's nuclear and missile development program; and (3) Japan's strong commitment to a closed nuclear fuel cycle and its growing surplus of reactor-grade plutonium.

On the other hand, a nuclear-armed Japan would not be in China's security interests, and the US-Japan alliance remains firm, especially after its upgrade in 1996. While the US has encouraged Japan to help maintain the US nuclear deterrent in the Pacific, it has not encouraged Japan to acquire nuclear weapons of its own. Such encouragement would violate US obligations under Article 1 of the NPT, and would deal a severe, if not fatal blow to the global nonproliferation regime.

Thus, it would take a geopolitical sea change" for Japan to overcome its "nuclear allergy". However, if Japan decides to go nuclear, what options would it have, politically and technically? 


Politically, Japan could legally withdraw from the NPT by giving three months advance notice, in accordance with Article X, par. 1, instead of violating the treaty by acquiring nuclear weapons while still a non-weapons state party to it. But the agreements for nuclear cooperation that Japan has with its major nuclear suppliers, the US, France, and Canada, all provide for "back-up" safeguards on transferred materials and technology in the event that Japan withdraws from the NPT.

In particular, the Agreement for Cooperation Between Japan and the US Concerning Peaceful Uses of Nuclear Energy (November 4, 1987) requires both back-up safeguards on and the right to require the return of "any material, nuclear material, equipment or components transferred [by the US to Japan or vice versa] pursuant to this agreement or any special fissionable material produced through the use of such items if either party terminates or materially violates a safeguards agreement with the Agency [i.e., the IAEA]." (Article 9, par. 2 and Article 12, par. 1).

Thus, Japan would have to violate these agreements in order to use such materials and technology for nuclear weapons. Moreover, if Japan left the NPT the nuclear suppliers would be obligated both by domestic legislation and by the Nuclear Supplier's Group (NSG) guidelines to terminate further transfers of materials and technology, e.g., the natural uranium that would be needed as feed for production reactors and centrifuge plants. [This underscores the utility of both uranium from seawater and breeder reactors as routes to both nuclear energy security and nuclear weapons.]

Whether these constraints would count for much if Japan decided that it needed nuclear weapons is open to question. Thus, it's worth considering its 



Besides the direct use of reactor-grade plutonium (RGPu), Japan has several other options to obtain plutonium and weapons-grade uranium (WGU) [Ref. Harrison, pp. 18, 19]:

1. Producing weapons-grade plutonium (WGPu) in power reactors by shutting them down more frequently for refueling, thus reducing the fuel burnup, and increasing the Pu-239 content of the contained plutonium;

2. Upgrading RGPu to WGPu, or producing WGU from natural or low-enriched uranium (LEU) by a laser separation process; 

3. Using its commercial centrifuge plant to produce WGU instead of LEU;

4. Separating the plutonium produced in the natural uranium blankets of its Monju experimental breeder reactor which has a higher Pu-239 content than WGPu;

5. Producing WGPu in a production reactor; and one option not considered by Harrison:

6. Using the stocks of WGPU and WGU - several hundred kg of each - in its fast critical assemblies.

Arguably, one or more of these options, e.g., (3) or (6), might be more attractive than using RGPu, depending on how many weapons would be required in what time frame, and the perceived need for testing. However, in the following we consider whether Japan could build reliable, high-yield weapons using RGPu. According to the latest unclassified US DOE guidance on the subject of using RGPu in weapons:

The degree to which the obstacles to the use of RGPu can be overcome depends on the sophistication of the state or group attempting to produce a nuclear weapon. At the lowest level of sophistication, a potential proliferating state or sub-national group using designs and technologies no more sophisticated than those used in first generation nuclear weapons could build a nuclear weapon from RGPu that would give an assured, reliable yield of one or a few kilotons, and a probable yield much greater than that. At the other end of the spectrum, advanced nuclear weapons states such as the US and Russia, using modern designs, could produce weapons from RGPu having reliable explosive yields, weight, and other characteristics generally comparable to those of weapons made from WGPu...Proliferating states using designs of intermediate sophistication could produce weapons with assured yields substantially higher than the kiloton-range possible with a simple, first generation nuclear device.

So the question is: where is Japan on the weapons sophistication spectrum?; also how much can one say/has been said/is worth saying on an unclassified basis with regard to how the problems involved in using RGPu can be minimized using designs more advanced than the "Fat Boy" implosion bomb ? [Fuchs' sketch]

In my judgment, Japan is at least at the intermediate point, and most probably at the high end of the weapon's capability spectrum. By this I do not mean to imply that Japan has a clandestine nuclear weapons program under way - I have no knowledge of such an effort - rather that the competence of their scientists in related applications indicates that they could make advanced weapons using RGPu if the political decision is made to go ahead. Although design considerations relevant to the use of RGPu in weapons are still classified, much has already been said on the subject in the open literature, e.g.,

1. After the discovery of the high rate of spontaneous fission of Pu-240 by E. Segre in Spring 1944, the scientists at Los Alamos realized that only the implosion design would work for plutonium, and then only if high-purity plutonium was used and/or it was in the form of a shell rather than a solid pit. While "Fat Boy" did use high-purity plutonium, " the hollow implosion device, as conceived then, could not have been built with the wartime technology of high explosive and fissile material", and a solid pit configuration, the "Christy gadget", was chosen instead. (Ref. L. Hoddeson et al., p. 446, footnote .# 9).

2. In 1956, British scientists realized that the plutonium primaries in their thermonuclear designs were vulnerable to predetonation from external neutron sources. "They came to a startling conclusion - that there was a high risk that British nuclear weapons could be disabled by an intense flash of radiation from a nearby nuclear detonation, that is, from a defensive enemy warhead launched for this purpose. Fission weapons with plutonium cores were particularly vulnerable because of plutonium's susceptibility to pre-detonation... To make an immune warhead required knowledge of certain techniques imperfectly understood at Aldermaston at the time. One was boosting, the addition of thermonuclear material to a fission bomb in order to enhance the efficiency of the fission process and so increase the yield." (Ref: L. Arnold, Britain and the H-Bomb, Palgrave, New York, 2001, p. 177.)

Without going into details, a study of Japanese work in such areas as high-explosive technology, inertial fusion, and production and handling of hydrogen isotopes leads me to the conclusion that they are capable of solving the problems involved in using RGPu in weapons, specifically predetonation. In sum, I would agree with the following assessment by Richard Garwin: 

As for the more sophisticated designer, it is my judgment that not only the five nuclear weapons states but also the nuclear weapon establishments of India, Pakistan, and Israel are capable of converting reactor grade plutonium into nuclear weapons that have similar yield to those made with weapons grade plutonium; they would not suffer at all from the high neutron generation rate of reactor grade plutonium

And add Japan and other advanced non-nuclear weapons states such as Germany to his list. 


The question then is not whether Japan could build reliable, high yield, low weight weapons using RGPu if it decided to go nuclear, but would it, considering its other possible options. In my judgment, the nuclear weapon states, particularly the US, should strive to keep Japan as far as possible from the need to seriously consider this issue. This will require leadership by example in devaluing the role of nuclear weapons, while working to lessen regional tensions, in the Middle East, South and North Asia. At the same time, safeguards and physical security on all existing weapons-useable materials, including RGPu, need to be upgraded, and their stocks decreased.