[NCI Logo]


Codifying the Phase-Out of Bomb-Grade Fuel for Research Reactors

by Alan J. Kuperman [1]

Prepared for a Symposium

"The Scope of a Fissile Material Convention"

United Nations Institute for Disarmament Research

and the Oxford Research Group

Geneva, Switzerland

August 29, 1996

UPDATED -- October 9, 1998


Because of the availability of basic nuclear-weapons design information in the open literature and even on the Internet, the main obstacle to fabrication of a nuclear weapon today is the acquisition of sufficient weapons-usable fissile material -- plutonium or highly enriched uranium (HEU). The degree to which a fissile material convention can prevent the spread of nuclear weapons, therefore, hinges on its ability to limit the production of, and access to, such materials. A convention that prohibits only the un-safeguarded production of weapons-usable fissile materials, but allows unlimited production and use of such materials under safeguards, falls short on two grounds: it permits continued production of weapons-usable material ostensibly for civil purposes that could later be quickly converted by states into weapons; and it permits continued civil commerce in fissile materials, perpetuating the risk of their acquisition by terrorist groups for weapons.

Civil commerce in HEU presents a particular concern because such material has a low background radiation level, making it easier to handle and fabricate into nuclear weapons. Indeed, as Manhattan Project physicist Luis Alvarez wrote in his memoirs:

With modern weapons-grade uranium, the background neutron rate is so low that terrorists, if they had such material, would have a good chance of setting off a high-yield explosion simply by dropping one half of the material onto the other half. Most people seem unaware that if separate HEU is at hand it's a trivial job to set off a nuclear explosion . . . even a high school kid could make a bomb in short order. [2]

Moreover, civil HEU has historically been used as a fuel in nuclear research reactors, often located on university campuses that lack the physical security measures employed at many nuclear powerplants and government weapons facilities. The threat posed by continued civil commerce in HEU was recently underscored dramatically by the disclosure that Iraq, in 1990, diverted bomb-grade uranium fuel from safeguarded research reactors for a crash program to build nuclear-weapon components.

Fortunately, an international cooperative effort, starting in the late 1970s, has made great strides in reducing civil commerce in HEU by converting reactors to non-weapons usable, low-enriched uranium (LEU) fuels, and by eschewing new HEU-fueled reactors. This effort, known as the Reduced Enrichment for Research and Test Reactors (RERTR) program, has laid the groundwork for the total phase-out of civil commerce in HEU for research reactors. This progress, however, has recently been endangered by Germany's proposal to build a new, large research reactor fueled with HEU -- the 20-megawatt FRM-II at the Technical University-Munich -- which would be the first such reactor outside of Libya, China, or the former Soviet Union constructed to use HEU since establishment of an international consensus supporting the RERTR program in 1980.

To preserve the progress of the last two decades, the international community should appeal to Germany to abandon its proposed HEU design, other states should complete the conversion of their operating reactors, and the fissile material convention should be expanded to codify the international consensus against new HEU-fueled reactors.


In the late 1970s the international community belatedly came to the realization that the fuel used in many nuclear research reactors -- bomb-grade, highly enriched uranium -- could be stolen or diverted for nuclear weapons by nations or terrorists. The RERTR program was established in 1978 to develop substitute fuel of higher-density, low enriched uranium (LEU), which is not suitable for weapons. As the substitute fuels were developed, existing reactors would be converted to LEU and new reactors would be designed to use LEU. The RERTR program has proved remarkably successful, facilitating the conversion of dozens of reactors worldwide from bomb-grade to non-weapons-usable fuel and sharply reducing international commerce in HEU.

Outside the United States, some 42 research reactors with power of at least 1 megawatt were built that originally relied on U.S.-supplied HEU fuel. To date, 38 either have converted to LEU, are in the process of converting, or have no further need for fuel-- which has enabled a sharp decline in U.S. HEU exports. Since the United States has historically been the major exporter of HEU for civilian use, and in recent years the sole exporter, this translates into a sharp reduction in total international commerce in bomb-grade uranium.

*Note: Does not include 1994 export to France of surplus Fort St. Vrain HEU fuel, since this material must be blended down to LEU as a condition of its export.

**Note: 1998 figure to date is zero, but does not include possible export of 3 kgs. of HEU to Canada for target material for radioisotope production. A license for this export was granted by the U.S. NRC on June 8, 1998.

Sources: 1993 Nuclear Regulatory Commission report to Congress pursuant to the Energy Policy Act of 1992 and subsequent NRC export data.

In addition, the United States has taken steps to reduce its own use of highly enriched uranium. In 1986, the U.S. Nuclear Regulatory Commission ordered the conversion of all licensed, domestic research reactors. Of the 25 such reactors operating at the time, eight have converted, another seven are in the process, and six have ceased operation. [3] The U.S. Department of Energy also has initiated studies on the feasibility of converting its own research reactors, which are not licensed by the U.S. NRC. Overall, 28 Western reactors already have completed conversion from HEU to LEU. [4]

In recent years, the United States has entered into agreements with Russia and China to work on conversion of research reactors operating in, and supplied by, these countries. The United States also is developing a system for production of molybdenum-99 for medical isotopes using LEU rather than HEU targets, which currently constitute the main civilian demand for HEU other than reactor fuel. The fuel and target development work is led by the U.S. Argonne National Laboratory.

The key to the RERTR program's success has been two core tenets: universality and spent-fuel return. [5] Universality has meant three things: 1) those reactors that can convert to existing LEU fuel must do so; 2) for remaining reactors, advanced fuel will be developed, to which they must convert when it is successfully qualified; and 3) no new reactors will be constructed to use HEU fuel. Reactor operators have been willing to convert to non-weapons-usable fuel -- and to accept the economic and performance penalties of doing so -- because the universality principle guaranteed that they would not be put at a competitive disadvantage with respect to neutron research, medical-isotope production, or other reactor activities.

In keeping with this principle, at least 11 large new research reactors (with power at least 1 MW) constructed since 1980 have been designed to use LEU fuel, including a 20-megawatt reactor in Japan, 30-megawatt reactors in South Korea and Indonesia, and two U.S. research reactors, all of which have been successfully completed. (See accompanying chart.) In addition, both China and France have now joined the international consensus, designing their next-generation, high-power research reactors to use LEU fuel. [6] In the same vein, in 1995, the United States abandoned plans for a new HEU-fueled research reactor, the Advanced Neutron Source, despite the pleas of U.S. neutron researchers. The Clinton Administration stated at the time that it made this decision at least partly because the bomb-grade fuel presented "a non-proliferation policy concern." [7] The German government is now presented with the same decision on the fuel for the FRM-II.


New Reactors Avoid Dangers of HEU


The FRM-II Does Not Require HEU

Technical studies by Argonne National Laboratory conclude that the FRM-II can be re-designed to produce equivalent experimental performance using LEU fuel, and would actually enjoy a slightly increased fuel-cycle length. [8] This redesign can be achieved using LEU fuel that exists today and has been qualified since 1988 -- without additional fuel development. Advanced fuel development has also been restarted at Argonne, which may well enable qualification of even higher density fuels in the future. However, there is no need to await such fuel development in order to convert the FRM-II design to LEU today. Conversion to LEU need not entail substantial delays in completion of the reactor or major increase in cost. [9] The reactor could be redesigned to use already-qualified LEU fuel within a year, and re-licensing this new design might require another year. The new design would require a slightly larger reactor core, so that any already completed work constraining the size of the core would have to be replaced.

The FRM-II's conversion to LEU would not in any way affect the "scientific position" of Germany, since the reactor's experimental performance would be essentially the same (and slightly better with regard to fuel-cycle length). Not a single experiment has been identified that would be precluded by conversion to LEU. An LEU-fueled FRM-II also would be no less safe overall than the current HEU design. Indeed, it is the unprecedented, high-density HEU fuel that has failed to undergo irradiation safety tests, whereas the alternate LEU fuel has been used safely around the world for years. [10] The Nuclear Control Institute does not oppose neutron research, per se, and would not oppose the FRM-II if redesigned to use LEU. Ironically, Germany's scientific position is more likely to be adversely affected by recently disclosed reductions in planned German contributions to, and participation in, international scientific initiatives, including the CERN particle accelerator and the Laue-Langevin Institute HFR reactor in Grenoble, a research reactor with significantly higher neutron flux than the planned FRM-II. [11]

The FRM-II Undermines Non-Proliferation Efforts

If built to use HEU, the German FRM-II would undermine international nonproliferation efforts in numerous ways:

1. A Ton of HEU Required for the FRM-II -- The FRM-II is projected to require 40 kilograms of HEU annually, amounting to 1.2 metric tons over its 30-year life. Thus, the reactor single-handedly would be responsible for the introduction of dozens of bombs' worth of HEU into international commerce.

2. Russia Becomes New HEU Supplier -- If the FRM-II uses HEU, the only possible long-term source of supply is Russia. Traditionally, the United States supplied HEU to European reactors, but this was curtailed under the RERTR program and then formally halted by the Schumer Amendment to the Energy Policy Act of 1992. Existing stocks of previously exported, U.S.-origin HEU within Euratom are insufficient to supply European research reactors that continue to require HEU in addition to the FRM-II over its 30-year life. The press has recently reported a pact on HEU supply between Russia and Germany. [12] Such a deal would undermine years of international efforts to control Russia's vast stocks of bomb-grade uranium, send a message that reactor operators could flout the RERTR program, and encourage Russia to look for new customers for its HEU. The best hope for preventing bomb-grade uranium from falling into the wrong hands is for Russia and the West to maintain united opposition to the export of HEU to any facility that refuses to cooperate with the RERTR program.

3. Existing Reactors Will Abandon RERTR -- Outside of China, Libya, and the former Soviet Union, the FRM-II would be the first research reactor (with power of at least 1 megawatt) built to use bomb-grade fuel since an international consensus supporting the RERTR program was reached in 1980. If this should happen, it would undermine the program's core principle of universality and could well lead to rejection of the RERTR program by operators of existing reactors. They would justifiably ask why they should absorb the expense and inconvenience of converting their reactors to LEU when a new reactor -- that could be built to use LEU -- is instead being built to use bomb-grade fuel?

Notes: Research reactors of at least one megawatt. Construction start date for two Russian reactors that went critical in 1983 and 1984 is uncertain; included in graph during '75-'79 period.

Sources: Nuclear Research Reactors in the World, December 1994, IAEA, Vienna.

4. New Reactors Will Demand HEU -- Once Germany breaks the taboo against new reactors using HEU, it will set a precedent and send a message that modern reactors require and are entitled to use HEU. In the future, other countries will likewise demand the right to use HEU in new reactors and will turn to Russia for supply of the fuel. Indeed, a key German official has confirmed that Siemens views the FRM-II as a prototype for future exports.  [13] The international nonproliferation regime, which emphasizes the principle of non-discrimination, will be sorely tested if it rejects the demands of other countries for HEU-fueled reactors while permitting Germany to proceed with the current design of the FRM-II.

5. Risks of Proliferation and Terrorism Will Rise with HEU Commerce -- If the FRM-II uses HEU, Russia begins to supply it, existing reactors renew their demand for it, and new reactors are designed to use it -- international commerce in bomb-grade uranium would expand rapidly, and nearly two decades of progress by the RERTR program would be severely undermined. As large volumes of HEU again began to be shipped around the globe, they would become vulnerable to theft and diversion by increasingly sophisticated and violent terrorists and renegade states. Most disturbing, if HEU were obtained, construction of a nuclear weapon would be relatively straightforward, as indicated by the Alvarez quote cited above.

Other Threats to the RERTR Program

The FRM-II is one of a number of recent developments that threaten to undermine international support for the RERTR program. The other principal threats to RERTR are:

• Reactors Refusing to Convert -- Several Western reactor operators are violating the principle of universality by refusing to convert their reactors even though suitable LEU fuel is available. Outside of the United States, there are now only two -- South Africa's Safari I reactor and Germany's FRJ-2 (at KFA Jeulich) -- because the operator of the EU's HFR Petten reactor recently indicated his intention to convert to LEU. The FRJ-2 is perhaps of lesser concern because it reportedly will close in the near future without requiring further shipments of HEU fuel. In the United States, there is only one remaining violator, the Department of Energy's Brookhaven Medical reactor (BMRR) reactor, since the Omega West reactor at Los Alamos and the Tower Shielding reactor at Oak Ridge have been shut down. [14] Proceeding with the current FRM-II design would reinforce the intransigence of the operators of these foreign and domestic reactors and undercut international leverage to compel their conversion.

• Advanced Fuel Development -- Several other reactors have not converted because they apparently cannot convert to LEU fuels of the density thus far qualified. Outside the United States, there are three: France's HFR and Orphee, and Belgium's BR-2. In the United States, there are six: the Department of Energy's Advanced Test Reactor (at INEL), High Flux Isotope Reactor (at ORNL) and High Flux Beam Reactor (at BNL), the Department of Commerce's NIST reactor, and the university reactors at the Massachusetts Institute of Technology (MIT) and University of Missouri - Columbia. (However, the HFBR is now shut down and is not expected to re-open.) In 1989, the United States suspended development of advanced, high-density (>4.8g/cc) LEU fuels for such high-power research reactors, undermining the principle of universality by permitting them to continue using HEU fuel. However, in 1996, fuel development was re-started and has begun to show promising results. While much of the U.S. government's rhetorical emphasis has focused on fuel development for Chinese and Russian reactors, it recently clarified that the program also has application to development of fuels for the remaining reactors in the United States and Europe, estimated to require approximately five years. In addition, the Department of Energy recently carried out feasibility studies on converting most of its own remaining HEU-fueled reactors. [15] Work is also progressing on developing alternative LEU targets for production of molybdenum-99 for medical isotopes, which should enable elimination of this other remaining segment of civil HEU commerce. [16]

• Spent Fuel Take-Back -- In 1986, the United States suspended the acceptance of U.S.-origin spent fuel from foreign research reactors, despite longstanding commitments to take back this fuel. As the policy remained lapsed for nine years, with renewal uncertain, reactor operators faced the prospects of paying higher costs for reprocessing of their spent fuel in Europe, removing an incentive for cooperation with the RERTR program. Such reprocessing also would have perpetuated the HEU fuel cycle and forced some operators to continue using HEU fuel, because there is no reprocessing line in Europe for LEU research reactor fuel. However, these problems were resolved in 1996 by U.S. issuance of a final environmental impact statement and record of decision, renewing the take-back program. As a result, the FRM-II now stands as the primary obstacle to fulfillment of the RERTR program's mission.



Status of Conversions

(not including Chinese- and Russian-supplied reactors)

* U.S. university reactors are being converted even if they had low-power (less than 1 MW) and lifetime HEU cores that did not require fresh fuel. This is in recognition of the extreme vulnerability of university reactors to theft, due to traditionally lax security on most campuses. Other low-power reactors in the United States and elsewhere are not now planned for conversion under the RERTR program, because they do not require fresh shipments of HEU.

1 DOE's ATR (INEL) and HFIR (ORNL), the Department of Commerce's NIST reactor, and the university reactors at MIT and University of Missouri - Columbia. A sixth reactor, the HFBR (BNL), is shut down due to a tritium leak and not expected to re-start; it is counted in the chart as "shutting down."

2 DOE's BMRR (BNL). Two other facilities, the Omega West (LANL) and Tower Shielding (ORNL) reactors, have recently been shut down and are not included in the total of 23 U.S. research reactors.

3 France's HFR and Orphee, and Belgium's BR-2.

4 Germany's FRJ-2 and South Africa's Safari I. The FRJ-2 may shut in the next few years and reportedly does not require further shipments of HEU. The Petten reactor, operated by the EU's JRC, has been removed from this category because it has indicated its intention to convert.


The RERTR program is one of the unsung heroes of the International Atomic Energy Agency and Nuclear Non-Proliferation Treaty regimes. Since 1978, the program has made great progress in reducing HEU commerce. If the international community provides its full support, the RERTR program can within the decade fulfill its goal of eliminating entirely civil commerce in bomb-grade uranium for research reactors. However, if Germany and other leading nuclear industrial states continue the recent trend of withholding their full cooperation from the program, it could soon collapse, resulting in a resurgence of HEU commerce.

The fissile material convention presents an ideal opportunity to lock in the gains of the RERTR program, by broadening the proposed convention to codify both the phase out of HEU commerce and the international consensus against new HEU-fueled reactors. This important step will be obstructed, however, if a new Western reactor is constructed to use HEU. Thus, how Germany proceeds on the FRM-II will have far-reaching consequences for the effectiveness of a fissile material convention, as well as for global efforts to combat nuclear proliferation and nuclear terrorism.


1. Alan J. Kuperman is a senior policy analyst for the Nuclear Control Institute, a research fellow at the Brookings Institution, and a PhD candidate at the Massachusetts Institute of Technology.

2. Luis Alvarez, Adventures of a Physicist (Basic Books, 1987), p. 125.

3. "Non-Power reactor HEU to LEU Conversion Programs," chart, U.S. Nuclear Regulatory Commission, updated September 29, 1998, by Theodore Michaels, U.S. NRC.  The remaining four reactors are the university reactors at MIT and U. of Missouri-Columbia, the Department of Commerce's NIST reactor, and a private General Electric reactor.  The first three cannot use the LEU fuels so far developed.   The last is low-power (100 kw) and does not require any fresh fuel.

4. A. Travelli, "Progress of the U.S. RERTR Program," presented at 2nd International Topical Meeting on Research Reactor Fuel Management, Bruges, Belgium, March 29-31, 1998.

5. The guarantee of spent fuel return (for both LEU and HEU fuel) is based on four grounds: 1) Reducing the vulnerability of spent HEU fuel to theft or diversion; 2) Abiding by longstanding U.S. commitments; 3) Inducing cooperation with the RERTR program; and 4) Avoiding an additional, perverse penalty for conversion to LEU -- i.e., losing the guarantee of spent fuel return.

6. Shi Yongkang et al., "The China Advanced Research Reactor Project," and Yuan Luzheng et al., "Preliminary Study of Core Characteristics for the Scheduled CARR," presented at the Fifth Meeting of the Asian Symposium on Research Reactors, Taejon, Korea, May 29-31, 1996.  A. Ballagny, "The Jules Horowitz Reactor: A new test reactor for fuels and materials," presented at the 1997 International Meeting on Reduced Enrichment for Research and Test Reactors, Jackson Hole, Wyoming, October 5-10, 1997.

7. "DOE Facts: A New Neutron Source for the Nation," U.S. Department of Energy, February 1995, p. 1.

8. N.A. Hannan and J.E. Matos, "Fluxes at Experimental Facilities in HEU and LEU Designs for the FRM-II," presented at the 1997 International Meeting on RERTR, Jackson Hole, Wyoming, October 5-10, 1997.

9. The director of the FRM-II project, Dr. Klaus Boening, is quoted estimating that a re-design to 32 MW would cost 50-100 million DM.  (Jeanne Rubner, "Warnung vor deutschem Sonderweg," Suddeutsche Zeitung, April 7, 1998, p. 10)  In light of the fact that he opposes such conversion, this is unlikely an under-estimate.  The projected total construction cost of the current design is at least 720 million DM.  Thus, if we accept Boening's estimate, the extra cost imposed by conversion represents no more than a 7 to 14 percent marginal increase.  Such a level of additional cost traditionally has been accepted by states and reactor operators as a necessary and acceptable trade-off for sustaining the nonproliferation and anti-terrorism benefits of LEU fuel and the RERTR program.

10. See letter urging that such standard safety tests be conducted prior to any further work on FRM-II, from Nuclear Control Institute to Thomas Goppel, Bavarian Minister for Environment and Development, March 29, 1998.  Goppel denied the request.

11. Malcolm W. Browne, "German Budget Trims Put Research at Risk," The New York Times, August 6, 1996.

12. Alison Abbott, "Germany owns up to weapons-grade uranium deal with Russia," Nature, June 11, 1998.  See also, Veronika Romanenkova, "Russia to supply uranium for German reactor," ITAR-TASS News Agency, June 9, 1998.  See also, Nuclear News, July 1998, p. 84.  An inter-government agreement was signed in Bonn on June 8, and a final contract is projected to be ready for signature in autumn 1998.  The agreement is to export "1.2 tons of HEU," sufficient for FRM-II's "entire service life."

13. Juergen Grosskreutz, Bavarian Minister for Education and Science, at session of Committee for Research Policy, Bavarian Landtag (State Parliament), July 10, 1996.

14. Two other U.S. facilities, the Department of Energy's High Flux Beam Reactor (at BNL) and the Department of Commerce's NIST reactor, have been determined unable to convert to existing LEU fuels, and thus cannot accurately be characterized as "refusing" to convert at this time.  They are included in the next section.

15. See C.L. Trybus, et al., "Design and Fabrication of High Density Uranium Dispersion Fuels," presented at the 1997 International Meeting on RERTR, Jackson Hole,Wyoming, October 5-10, 1997.  In 1996, DOE clarified that advanced fuel development will have applications in Europe and the United States (see letter from Jon B. Wolfsthal to Argonne National Laboratory, April 16, 1996), and directed Argonne to study the potential for converting remaining HEU-fueled DOE reactors to LEU fuel.  Such studies have already been completed for three of DOE's four remaining high-power reactors not yet planned for conversion.  See J.E. Matos, "LEU Conversion Status of U.S. Research Reactors," RERTR Program, Argonne National Laboratory, September 1997.

16.  G.F. Vandegrift, et al., "Progress in Chemical Processing of LEU Targets for Mo-99 Production," presented at the 1997 International Meeting on RERTR, Jackson Hole, Wyoming, October 5-10, 1997.

[What's New] What's New      [HEU Page] HEU Page     [Home Page]Home Page