Propositions & Reports

Can Reactor Grade Plutonium
Produce Nuclear Fission Weapons?

May 2001

Council for Nuclear Fuel Cycle
Institute for Energy Economics


There have frequent discussions on whether plutonium extracted from the spent fuel burnt at nuclear power plants can be used to create nuclear weapons. However, those who ask the question and those who answer them know nothing of nuclear weapons. One who has not actually made nuclear weapons cannot possibly answer this question with a clear "yes" or "no."

Otto Hahn, who discovered the nuclear fission phenomenon, Enrico Fermi, who created the world's first nuclear reactor, and many other scientists have contributed greatly to human beings providing the blessing of incalculable potential of nuclear energy. Unfortunately, the World War II urged politicians to use this new energy to create weapons of mass-destructions. After the war, this nuclear fission energy began to be utilized to generate electricity, and it supports a stable energy supply. It is also the best energy source today with which to combat global warming.

Peaceful uses of nuclear energy is a revolutionary technology that can realize the effect of energy use 1 million times larger than that of fossil fuels. However, it is estimated that, without the use of plutonium, this energy supply will last for only 70 more years. The Bush Administration announced in last May that they would positively promote the construction of nuclear power plants in order to cope with issues such as energy supply and electric power in the U.S., and the global environmental issues as well. However, if the U.S. continues to just use uranium and not recycle plutonium, the world's energy supply will last for even shorter than 70 years.

The Council for Nuclear Fuel Cycle (CNFC) was established in order for many people to understand the necessity for the nuclear fuel cycle to efficiently utilize the enormous nuclear energy. However, without the use of plutonium, it is impossible to take full advantage of nuclear energy as a revolutionary technology. For this reason, we have thought that we could clarify whether it is possible to create nuclear weapons from the plutonium extracted from the spent fuel burnt in nuclear power plants (light-water reactors).

Japan is a non-nuclear weapons state. We, therefore, are unable to make a concrete verification of whether it is possible to create nuclear weapons from reactor grade plutonium. Accordingly we decided to conduct our investigation on whether it was possible to create "practical weapons" (not just nuclear explosion devices) from reactor grade plutonium through technical documents published in the U.S. and exchange of opinions with experts to clarify comparison of the compositions of weapon and reactor grade plutonium and other technical issues. We consigned this investigation to the Institute for Energy Economics (President: Ryukichi Imai). The following report is the result of this investigation.

May 2001, CNFC

About This Paper

This is an abbreviated (and translated) version of a report prepared by the Institute for Energy Economics (President: Prof. Dr. Ryukichi Imai), Tokyo for the Council for Nuclear Fuel Cycle regarding the major dividing line between peaceful and military nuclear technology. Fission based nuclear explosives are made from weapon grade plutonium (Wpu), which is extracted from metallic natural uranium fuel (in thin Aluminum or Magnesium coating) burned up to 2,000 or 3,000 MWD/t in graphite or heavy water moderated reactors built for that special purpose. Such reactors are called "production reactors." Plutonium is separated from the burned fuel and typically contains more than 93% of fissionable Pu-239. 3,000 MWD/t burn up is too low to produce meaningful amount of electricity (or other power) so that such Wpu-production reactors are not a part of peaceful nuclear power industry. On the other hand, there are many nuclear power reactors, which produces more than a million kilowatt electricity. These power reactors, if they are Pressurized or Boiling Water Reactors, use Zirconium-clad ceramic fuel and burn fuel up to 40,000 MWD/t or 50,000 MWD/t. That much power extraction from fuel makes electricity production economical. On the other hand Plutonium extracted from such fuel (Rpu) has been exposed in the neutron atmosphere so long that isotopic compositions become very complicated with Pu-238, 239, 240, 241 and 242, they are considered not suitable as weapons material.

Debate has been taking place for a long time and throughout the world that whether it is possible to use Rpu into making nuclear weapons. If it can be so converted, this constitutes the strong link between peaceful electricity generation and nuclear weapon production, because the same material may be employed for the two different purposes. For a long time, it had been automatically assumed that Rpu are unfit for the bombs, and this was the basis of accepted wisdom. The Nuclear Non-Proliferation Treaty (NPT) was written broadly on this assumption, until in 1977 when the United States reversed the assumption and proposed the world-wide prohibition of power reactor fuel reprocessing (Plutonium extraction) as a major proliferation risk. The debate has continued for quite sometime about the feasibility of Rpu based nuclear weapons. Contrary to many popular misunderstanding the Democratic People's Republic of Korea (North Korea), Iraq, or any of the so-called "rogue states" represent possible cases of Rpu based nuclear weapons. As will be discussed below Rpu has so many technical problems in order to become fissionable source for nuclear explosion and so far as known everybody avoided taking such a path.

Summary and Conclusion of This Report

Since the arguments for and against Rpu weapons are very complicated and technical, while obviously highly sensitive as information for public discussion, not every knowledgeable scientists and engineers are fully aware of all the details of the problems involved. While searching for the answer this time, the author is grateful to many of his friends including Richard Garwin of IBM, John Holdren of Harvard University, Takehiko Mukaiyama and Hideo Kuori of Japan Atomic Energy Research Institute for trying to help him understand the essence of the problem as well as possible solutions. While the author alone is responsible for the conclusion as stated in this section, he is convinced that the world nuclear community in general has considered this problem very seriously and thought about the problem. A list of major documents that contributed to preparing this report has been attached at the end of this report.

One important thing that came to his notice is that in the public domain:

1) There is no scientific or engineering discussion about design, construction and so forth of fission nuclear weapon that leads any definitive conclusion about plutonium based fission weapons, either with Wpu or Rpu.

2) Almost all documents available through internet etc. leads to one document, by J. Carson Mark, "Explosive Properties of Reactor-Grade Plutonium" Science and Global Security, 1993 Volume 4, pp.111-128 (Director, Theoretical Division, Los Alamos National Laboratory, 1947-1972)

3) Mark's paper, however, is not scientific or technical discussion about weapons design or about weapon fabrication. The paper is a mathematical model treatment of fission chain reaction by implosion process, and discusses the probabilities of achieving various levels of design yield (in this case, taking Trinity 1945 as a model, the design yield is assumed to be 20 kt.) By changing such design parameters as neutron source and/or speed of implosion, Mark calculates the relative probabilities of achieving Nominal Yield, above 5 kt, above 1 kt, fizzle to 1 kt (see table 3 and 4 below). His calculations are based on nominal parameters, which J. Robert Oppenheimer had communicated to Leslie R. Graves prior to the Trinity experiment. Oppenheimer only gave his estimates of probabilities and did not discuss how he arrived at such numbers (at least Mark did not discuss how these numbers might have been arrived at).

4) It is clearly stated that Mark's discussion represented very much simplified mathematical model, which only proved that different levels of explosive yield are possible with different design parameters, including, according to Mark, use of reactor grade plutonium. No quantative analysis is involved, while the calculations did not provide positive details that reactor grade plutonium can be fabricated into a meaningful and practical nuclear weapon.

The only conclusion that can be derived from this discussion is that it is theoretically possible to use reactor grade plutonium for the weapons purposes, but there is no positive evidence that a meaningful arsenal can be realized through such processes. An excerpt from a statement by the American Academy of Science would provide the feel of the level of possibilities and uncertainties. It is also known that Lawrence Livermore Laboratory (LLL) has fabricated a fission device in 1962 using British origin Rpu, and the author had been given a confidential briefing in Washington D.C. from Robert W. Selden of LLL in 1977. However, no detailed information about isotopic composition, yield, expected vs. achieved yield was provided, and it was not meaningful scientific information.

Data from Carson Mark's Paper

Figure 1 gives isotopic composition of Pu-238, 239, 240, 241 and 242 as function of fuel burn-up. At 5 MWD/kg burn-up Pu-239 is about 90% and Pu-240 about 10% (Wpu) while at 40,000 MWD/kg Pu-239 is less than 60% and Pu-240 more than 20% (Rpu).

Table 1 names different isotopic composition of Plutonium, such as weapon grade and reactor grade.

Table 2 describes problems with different Pu isotopes. Pu-238 gives large decay heat and thus warms up the fabricated device. Pu-240 has large spontaneous fission neutrons and thus can lead to premature chain reaction. Pu-241 in 14.4 years half-life becomes Americium (Am)-241, a strong gamma source and makes handling difficult.

Table 3 and Table 4 are calculated probability of achieving different levels of yield with changes in neutron source and speed of implosion. As discussed earlier, the numbers represent relative and indicative possibilities based on a crude mathematical model.

Figure 1: Plutonium isotope composition as a function of fuel exposure in a pressurized-water reactor, upon dscharge.

Table 1: Approximate isotopic composition of various grades of plutonium.

Grade Isotope
Pu-238 Pu-239 Pu-240 Pu-241(a) Pu-242
Super-grade .98 .02
Weapons-grade(b) .00012 .938 .058 .0035 .00022
Reactor-grade(c) .013 .603 .243 .091 .050
MOX-grade(d) .019 .404 .321 .178 .078
FBR blanket(e) .96 .04
  1. Pu-241 plus Am-241.
  2. N. J. Micholas, K. L. Coop and R. J. Estep, Capability and Limitation Study of DDT Passive-Active Neutron Waste Assay Instrument (Los Alamos National Laboratory, LA-12237-MS, 1992).
  3. Plutonium recovered from low-enriched uranium pressurized-water reactor fuel that has released 33 megawatt-days/kg fission energy and has been stored for ten years prior to reprocessing (Plutonium Fuel: An Assessment (Paris:OECD/NEA, 1989) Table 12A).
  4. Plutonium recovered from 3.64% fissile plutonium MOX fuel produced from reactor-grade plutonium and which has released 33 MWd/kg fission energy and has been stored for ten years prior to reprocessing (Plutonium Fuel: An Assessment(Paris:OECD/NEA, 1989) Table 12A).
  5. FBR=Fast-neutron plutonium Breeder Reactor.

Table 2: Various properties of plutonium isotopes (and americium-241).

Isotope Half-life(a) Bare critical mass Spontaneous
fission neutrons
Decay heat
years kg, Alpha-phase (gm-sec)-1 watts kg-1
Pu-238 87.7 10 2.6x103 560
Pu-239 24,100 10 22x10-3 1.9
Pu-240 6,560 40 0.91x103 6.8
Pu-241 14.4 10 49x10-3 4.2
Pu-242 376,000 100 1.7x103 0.1
Am-241 430 100 1.2 114
  1. By Alpha-decay, except Pu-241, which is by Beta-decay to Am-241.

Table 3: Probability (based on Oppenheimer's letter) of achieving indicated yields in the assembly system used at Trinity with neutron sources of various sizes.

Neutron source
(multiple of Trinity)
(20 kilotons)
above 5 kt above 1 kt fizzle to 1 kt
Trinity .88 .94 .98 .02
10x .28 .54 .82 .18
20x .08 .29 .67 .33
30x .02 .16 .55 .45
40x .006 .08 .45 .55

Table 4: Probability (based on Oppenheimer's letter) of achieving indicated yields in an assembly system twice as rapid as Trinity with neutron sources of various sizes.

Neutron source
(multiple of Trinity)
(20 kilotons)
above 5 kt above 1 kt fizzle to 1 kt
Trinity .94 .97 .99 .01
10x .54 .74 .90 .10
20x .28 .54 .82 .18
30x .16 .40 .74 .26
40x .08 .30 .67 .33

Statement by the American Academy of Science

The following excerpts from "Reactor-Grade and Weapons-grade Plutonium in Nuclear Explosion" rely heavily on Carson Mark's paper. Some of the comments are reflection of subjective judgments rather than the technical comments the author has tried to communicate. It seems that non-proliferation and the rogue states mentality have sometimes taken over the better scientific judgments and have led to making unsupported statements.

Virtually any combination of plutonium isotopes can be used to make a nuclear weapon.

Not all combinations, however, are equally convenient or efficient. The most common isotope, Pu-239, is produced when the most common isotope of uranium, U-238, absorbs a neutron. It is this plutonium isotope that is most useful in making nuclear weapons, and it is produced in varying quantities in virtually all operating nuclear reactors.

As fuel in a reactor is exposed to longer and longer periods of neutron irradiation, higher isotopes of plutonium build up as some of the plutonium absorb additional neutrons, creating Pu-240, Pu-241, and so on. Pu-238 also builds up from a chain of neutron absorptions and radioactive decays starting from U-235. Because of the preference for relatively pure Pu-239 for weapons purposes, when a reactor is used specifically for creating weapons plutonium, the fuel rods are removed and the plutonium is separated from them after relatively brief irradiation. The resulting "weapon grade" plutonium is typically about 93% Pu-239. Such brief irradiation is quite inefficient for power production, so in power reactors the fuel is left in the reactor much longer, resulting in a mix that includes more of the higher isotopes of plutonium.

Use of reactor-grade plutonium complicates bomb design for several reasons. First and most important, Pu-240 has a high rate of spontaneous fission, meaning that the plutonium in the device will continually produce many background neutrons. Second, the isotope Pu-238 decays relatively rapidly, thereby significantly increasing the rate of heat generation in the material. Third, the isotope Americium-241 (which results from the 14-year half-life decay of Pu-241) emits highly penetrating gamma rays, increasing the radioactive exposure of any personnel handling the material.

In a nuclear explosive using plutonium, the plutonium core is initially "sub-critical", meaning that it cannot sustain a chain reaction. Chemical high explosives are used to compress the plutonium to higher than normal density. In a well-designed nuclear explosive using weapon grade plutonium, a pulse of neutron is released to start this chain reaction at the optimal moment, but there is some chance that a background neutron from spontaneous fission of Pu-240 will set off the reaction prematurely. With reactor-grade plutonium, the probability of such "pre-initiation" is very large. Pre-initiation can substantially reduce the explosive yield, since the weapon may blow itself apart and thereby cut short the chain reaction that releases the energy. Calculations demonstrate, however, that even if pre-initiation occurs at the worst possible moment, the explosive yield of even a relatively simple device similar to the Nagasaki bomb would be of the order of one or a few kilotons. While this yield is referred to as the "fizzle yield," a 1-kiloton bomb would still have a radius of destruction roughly one-third that of the Hiroshima weapon, making it a potentially fearsome explosive. With a more sophisticated design, weapons could be built with reactor-grade plutonium that would be assured of having higher yields.

Dealing with the second problem with reactor-grade plutonium, the heat generated by Pu-238 and Pu-240, requires careful management of the heat in the device. Means to address this problem include providing channels to conduct the heat from the plutonium through the insulating explosive surrounding the core, or delaying assembly of the device until a few minutes before it is to be used.

In short, it would be quite possible for a potential proliferator to make a nuclear explosive from reactor-grade plutonium using a simple design that would be assured of having yield in the range of one to a few kilotons, and more using an advanced design. Theft of separated plutonium whether weapon grade or reactor-grade, would pose a grave security risk.


  • "Plutonium," Nuclear Issues Briefing Paper 18, February 1999, Appendix November 1999
  • J. Carson Mark, "Explosive Properties of Reactor-Grade Plutonium," Science and Global Security, 1993, Volume 4, pp.111-128 (Director, Theoretical Division, Los Alamos National Laboratory, 1947-1972)
  • Robert Serber, "Introduction Courses" given in April 1943 in connection with the starting of Los Alamos Project and printed in the Los Alamos Primer, unclassified in 1963
  • Committee on International Security and Arms Control, National Academy of Sciences, "Management and Disposition of Excess Weapons Plutonium," National Academy Press, Washington D.C. 1994
  • Richard L. Garwin, IBM Fellow Emeritus, "Letter to Ryukichi Imai" Via E-Mail, November 9, 2000
  • "Nuclear Weapons Frequently Asked Questions" Version 2.24, February 20, 1999
    • Section 1.0 Types of Nuclear Weapons
    • Section 2.0 Introduction to Nuclear Weapon Physics and Design
    • Section 4.0 Engineering and Design of Nuclear Weapons
  • Richard L. Garwin, "Reactor-Grade Plutonium Can be Used to Make Powerful and Reliable Nuclear Weapons: Separated plutonium in the fuel cycle must be protected as if it were nuclear weapons." August 1998
  • Richard L. Garwin, "Maintaining Nuclear Weapons Safe and Reliable Under a CTBT" What Types of Weapons Can Be Developed Without Nuclear Explosions?" May 31, 2000
  • John P. Holdren and Ryukichi Imai, "Letters Exchanged December 2000"
  • Andre Gsponer and Jean-Pierre Hurni, Technical Report "Fourth Generation Nuclear Weapons," International Network of Engineers and Scientists Against Proliferation, Seventh edition, September 2000