U.S. Program for Disposition of Excess Weapons Plutonium

1. Introduction

Since the late 1980s, both the United States and Russia have been dismantling thousands of nuclear warheads, a process which continues to this day. This has left both countries with a daunting Cold War legacy: hundreds of tons of fissile materials that are no longer needed for military purposes, and must be safely and securely managed and ultimately used or disposed of.

The United States has declared that 52.7 tonnes of plutonium, over half of the 99.5 tons in the Department of Energy (DOE) and Department of Defense (DOD) stockpiles, is excess to its military needs, along with 175 tonnes of highly-enriched uranium (HEU) [1, 2]. President Clinton has committed that these stockpiles will never again be used in nuclear weapons, and DOE is undertaking a major storage and disposition effort, funded at over $100 million per year, designed to ensure that these materials are physically transformed in ways that will render them no more accessible and attractive for use in nuclear weapons than the plutonium in spent fuel from ordinary commercial nuclear reactors -- a goal known as the "spent fuel standard." The United States sees this effort as a major step toward reducing the risks of nuclear proliferation and ensuring the irreversibility of nuclear arms reductions, and a clear signal of its commitment to these objectives.

In January, 1997, the United States confirmed its decision to pursue a dual-track approach to disposition of the excess weapons plutonium, using some of the material once-through as fuel in civilian reactors, and immobilizing the remainder as waste, mixed with intensely radioactive fission products [3]. The United States expects to implement this program over the next 2-3 decades, and hopes to work out arrangements so that Russia''s excess fissile material stockpiles will be eliminated on a parallel timescale.

2. History of the U.S. program

The U.S. plutonium disposition program formally began in September 1993, when President Clinton issued Presidential Decision Directive (PDD) 13, which laid out U.S. nonproliferation and export control policies. PDD-13 called for the United States to seek "to eliminate where possible the accumulation of stockpiles of highly-enriched uranium or plutonium, and to ensure that where these materials already exist they are subject to the highest standards of safety, security, and international accountability." In particular, the President''s statement called for U.S. excess fissile materials to be placed under international safeguards, and called for "a comprehensive review of long-term options for plutonium disposition, taking into account technical, nonproliferation, environmental, budgetary and economic considerations." [4]

In response, an interagency group was established under the joint chairmanship of the Office of Science and Technology Policy and the National Security Council, to oversee plutonium disposition efforts and to ensure that the views of all relevant agencies were appropriately considered. In January, 1994, the Department of Energy, as the agency with primary responsibility within the United States government for the management and disposition of plutonium, established the Office of Fissile Materials Disposition to carry out this mission. The program received an important jump-start with the release of a report from the U.S. National Academy of Sciences, also in January 1994, which recommended a specific set of objectives for the U.S. disposition program and criteria by which the options should be judged, and identified the few options that best met those criteria and objectives: these recommendations provided a strong foundation from which to build, and were largely adopted by the U.S. government [5, 6].

The first step for the government''s program, building from the Academy''s work, was to identify the key options the U.S. program should focus on, out of the many dozens of ideas for dealing with excess plutonium which had been proposed. When the federal government is considering a major action having a significant effect on the environment -- such as disposition of excess weapons plutonium -- U.S. law requires an environmental impact statement comparing all of the "reasonable" alternatives. Thus, a central early focus of the U.S. program was a "screening" process which, by March of 1995, had ruled out a wide variety of alternatives as "unreasonable" -- on grounds of undue cost, delay, technical uncertainty, environmental hazard, and the like. In particular, all options requiring development, testing, and deployment of new reactor types -- including high-temperature gas reactors, new-design fast-neutron reactors, molten salt reactors, and accelerator-driven sub-critical systems, among others -- were ruled out, not on ideological grounds, but because all of them would take longer, cost more, and involve more uncertainty than using reactors of existing types, which were adequate to the mission of transforming excess weapons plutonium to meet the spent fuel standard [7].

The screening process concluded that only three classes of alternatives were "reasonable" enough to require more detailed analysis and comparison: use of the excess plutonium as fuel in reactors of existing types; immobilization of plutonium with fission products; and direct disposal of plutonium in very deep boreholes. Each of these classes of alternatives included several options (e.g., light-water reactors (LWRs) and CANDUs, glass and ceramic, borehole disposal with and without prior immobilization), and a variety of variants of the particular options.

To decide which of these options to pursue, the program then began developing three types of information: environment, safety, and health information to support the required environmental impact statement; cost, schedule, and technical uncertainty data; and a detailed analysis of the arms reduction and nonproliferation impact of the different options [8-10]. This data was developed and refined during 1995-1996; for all three, there was an energetic effort to solicit and consider public comments.

In the United States, a wide variety of projects have been delayed for years as a result of lawsuits charging that their environmental documentation did not meet the requirements of the law. To avoid such an outcome, the U.S. disposition program focused a large fraction of its effort from the outset on the preparation of a comprehensive and defensible environmental impact statement. This focus led to concern from a variety of parties -- including myself -- that the program was moving too slowly, and was focused more on producing paper than on getting the job done. In the end, however, it seems clear that this focused effort was in fact the only way to get the project moving forward without being stymied from the outset by legal challenges -- though such challenges will surely come in the future.

By mid-1996, the essential information needed to shape a decision was coming together. The U.S. national laboratories had developed an innovative process for converting plutonium weapons components or "pits" to oxide, which offered lower costs and dramatically lower waste generation in an integrated modular system. Both the existing reactor and immobilization alternatives looked viable. For the reactor alternative, detailed studies indicated that already-operating U.S. nuclear reactors could handle uranium-plutonium mixed-oxide (MOX) fuel in one-third of their reactors without major modifications, while remaining within existing safety envelopes, and that MOX in 100% of the reactor cores might well be possible. Canadian CANDU reactors also appeared to be capable of handling 100% MOX cores. Several U.S. utilities and the Ontario Hydro utility in Canada were actively interested in participating in the program (in anticipation of receiving either fees for irradiation services or, equivalently, heavily discounted fuel). It appeared that it might be possible to save money and time by modifying existing buildings in the DOE complex to serve as pit conversion and MOX fabrication facilities, rather than building new facilities from scratch.

For the immobilization alternative, a new concept, known as "can-in-canister," had been developed which would make it possible to make use of existing high-level waste immobilization operations without the need to substantially modify those facilities to handle plutonium. In this concept, the plutonium would be immobilized in a specially-designed glass or ceramic of its own, in relatively small cans. A number of these cans would be arrayed inside the huge canisters into which molten glass containing high-level waste is poured; the plutonium-bearing cans would then be embedded permanently within the intensely radioactive high-level waste canisters. (In the latest variant, the immobilized plutonium in the cans would be in the form of small pebbles, and the cans made of aluminum, which would melt when the HLW glass was poured over them, leaving the plutonium intimately mixed with the high-level waste glass.) With this concept, the immobilization of plutonium could be accomplished in small, critically-safe melters or ceramic production systems installed in existing glove-box lines. Moreover, by mid-1996, extensive work was indicating that both glass and ceramic forms could be designed to incorporate substantial quantities of plutonium, and a variety of potentially promising approaches to addressing the long-term repository criticality issue (which must be addressed for disposal of spent MOX fuel as well) were being developed. Like the reactor alternative, in short, it appeared clear that the immobilization alternative could be done safely and securely, in a reasonable time and for reasonable cost.

The deep borehole alternative, by contrast, was not looking as promising. From a purely technical perspective, the deep borehole alternative looked reasonably appealing, providing near-absolute protection against subnational diversion once the material was emplaced, considerable difficulty for retrieval by the host state (if designed for that purpose), and a strong argument for good environmental safety. But none of these benefits would accrue until the material was actually emplaced, and the cost, schedule, and feasibility for licensing a fundamentally new type of geologic repository were far too uncertain to make a major national security program dependent on success. The billions of dollars being spent on preparing the information needed for an eventual license application for the Yucca Mountain repository, and the continuing delays there, did not provide an encouraging analogy.

Ultimately, the DOE studies indicated that the costs, schedules, uncertainties, environmental implications, and nonproliferation and arms reduction implications of the reactor alternatives and the immobilization alternatives were roughly comparable; each approach had its own advantages and drawbacks, but none large enough to be decisive. Moreover, the studies indicated that pursuing both the reactor and immobilization alternatives -- the so-called "hybrid" or "dual-track" alternative -- would not be greatly more expensive, and would have substantial advantages. Pursuing both would provide higher confidence of success, and particularly higher confidence of an early start, since each alternative could serve as a backup in the event of unexpected difficulties and delays with the other. And pursuing both would involve the United States in the key technologies likely to be used in Russia''s disposition program while simultaneously sending a clear signal of U.S. seriousness and flexibility in dealing with this critical international security problem, both of which would improve the potential for cooperation with Russia and other nations in getting the job done. Moreover, a dual-track approach would make it possible to use different approaches for different forms of plutonium: roughly a third of the U.S. excess plutonium inventory is in forms that would require expensive purification before they could be used as MOX fuel, and may therefore be more suitable for immobilization. Here, too, the U.S. National Academy of Sciences had led the way: in mid-1995, in the second volume of its study, the Academy had strongly recommended pursuing both of these approaches as quickly as practicable, "because it is crucial that at least one of these options succeed, because time is of the essence, and because the costs of pursuing both in parallel are modest in relation to the security stakes." [6, p. 14]

With these considerations in mind, after an extensive interagency discussion, in late 1996 the United States announced that its "preferred alternative" for disposition of its excess weapons plutonium was a dual-track approach including both use of plutonium in existing reactors and immobilization of plutonium. On January 14, 1997, that approach was confirmed in the formal "Record of Decision" required under U.S. law -- a step personally approved by President Clinton [3]. The Department of Energy is now attempting to implement both tracks as rapidly as it can -- and to cooperate with Russia to help ensure that when the time comes, Russia will be ready to eliminate its stocks of excess weapons plutonium in parallel.

3. Requirements of the dual-track approach

To implement this dual-track approach will require additional detailed analyses and demonstrations, to provide the information necessary to select the best variants of both approaches and acquire needed licenses and approvals. It will also require a number of large-scale facilities, including a facility for converting pits to oxide (and preparing other forms of plutonium for disposition); a MOX fuel fabrication facility; reactors licensed to handle MOX fuel (neither U.S. nor Canadian reactors have such licenses currently); facilities for immobilizing plutonium; and facilities for immobilizing high-level wastes so as to combine the result with the immobilized plutonium. Neither the United States nor Russia has large-scale operational facilities for any of these purposes at the moment, and therefore several years and substantial initial capital investments will be required before large-scale disposition of excess weapons plutonium can begin.

Most aspects of the reactor approach can be considered technically demonstrated. Nearly two dozen LWRs around the world are already using MOX fuel, typically in one-third of their reactor cores [11]. It is very likely that the United States will ultimately choose to begin the reactor part of its disposition program using a one-third core approach based on this existing experience, and then shift to higher core loadings, possibly including 100% core, as the relevant issues are resolved and needed license modifications acquired. The United States has made clear that the pit conversion facility and MOX plant will be government-owned facilities on existing DOE sites, but that the government expects to contract with private firms to build and operate the fabrication plant and irradiate the fuel in existing utility reactors. DOE prefers to contract with a consortium representing both fabricators and reactor operators in a single package. The specific number of reactors to be used has not yet been chosen, and depends on such factors as core loading and how much of the excess plutonium is to be used in reactors rather than being immobilized. To provide an order of magnitude, 50 tons of excess weapons plutonium could be irradiated to 42,000 megawatt-days per metric ton in 20 years in 9 1,000 megawatt-electric LWRs using one-third MOX cores, or 3 such reactors using full-MOX cores. Two CANDU reactors operating with full-core MOX could irradiate 50 tons of excess weapons plutonium over the same period; their fuel contains a lower percentage of plutonium, but would be irradiated to a much lower burnup, allowing them to accomplish the mission in somewhat fewer reactor-years of operation [12].

DOE expects that it will take a decade to bring a MOX plant into operation in the United States, and that the plant would remain operational for perhaps a dozen years before completing its mission in about 2018, after which it would be decommissioned [2]; a MOX plant capable of producing about 75 metric tons of heavy metal per year would be sufficient to accomplish the mission -- or somewhat more for the CANDU alternative with its lower percentage of plutonium in MOX. The operation could begin several years earlier if existing European fabrication facilities -- some of which are small and flexible enough to plausibly handle a specialty input material such as weapons plutonium -- were used to fabricate initial test assemblies, and perhaps the first partial reactor cores. This approach was recommended by the U.S.-Russian Independent Scientific Commission on Disposition of Excess Plutonium, and has been favored by the U.S. nuclear industry, but DOE currently appears to be leaning against it [13].

Since the MOX option is largely technically demonstrated, the principal uncertainties facing its implementation in the United States are political and institutional. The controversy that has already arisen over the potential use of plutonium fuel in U.S. reactors suggests that it may ultimately prove to be very difficult to acquire the necessary political approvals and licenses to implement the MOX option in the United States, and substantial delays resulting from political and legal interventions remain a serious possibility.

With the immobilization approach, by contrast, while there are some political and institutional issues, the primary uncertainties are technical ones. While several countries around the world, including the United States and Russia, have demonstrated experience immobilizing high-level wastes, safely including large quantities of plutonium in this immobilization has never been done before.

To implement the "can-in-canister" immobilization approach, small, critically-safe melters or ceramic production machines could be installed in existing plutonium-handling glove-box lines, such as those that exist at the Savannah River Site, where current U.S. HLW immobilization operations are proceeding. To implement the dual track, a sufficient number of these melters would be installed to handle 2-3 tons of plutonium per year. The glass or ceramic prepared in these small facilities would contain between 5-12 percent plutonium by weight, and would also include substantial quantities of neutron absorbers, to ensure against criticality both in production and over thousands or millions of years in a geologic repository. Safeguards and security will have to be upgraded at the high-level waste immobilization facility, which is not currently designed for protection of plutonium-bearing materials. DOE estimates that immobilization using this approach could begin in about 7 years. For planning purposes, current DOE studies assume that only the one-third of the excess plutonium stockpile which is in impure forms would be immobilized; in that circumstance, the immobilization facility would only operate for about 6 years, while the MOX operation would continue for a long period thereafter [2, 8]. In my own view, it would make more sense to continue to operate both the MOX and immobilization facilities, allowing the completion of the overall mission to be accelerated by perhaps four years, and resulting in a roughly even split of the excess material between the MOX and immobilization options.

Remaining technical uncertainties facing the immobilization option include the performance of the various possible immobilization forms in a geologic repository (including the long-term prospects for criticality); developing safe and effective approaches to the immobilization itself; choosing the type of neutron absorbers to be used, and the immobilization material best suited to incorporating both plutonium and neutron absorbers; and, perhaps most important from a policy perspective, developing and demonstrating safeguards approaches and technologies for this new type of plutonium processing. Contrary to the views expressed by some, it seems clear that the level of "irreversibility" offered by the immobilization approach would be similar, overall, to that offered by the reactor option: while the plutonium would remain weapon-grade, recovering 50 tons of plutonium from such forms would require hundreds of millions if not billions of dollars of expenditure over several years, and significant modifications to existing separation facilities would be needed.

As to cost, the two options are roughly comparable, at least as far as the uncertainties in current cost estimates permit a judgment. DOE estimates that the net discounted present cost of immobilizing 50 tons of plutonium, using the can-in-canister option, would be just over $1 billion. DOE estimates that burning 50 tons of excess plutonium as MOX fuel in the United States would have an excess net discounted present cost, compared to generating the same electricity with low-enriched uranium fuel, of just over $1.2 billion. The actual cost of fabricating the MOX fuel would be far higher than this, but DOE is including a "fuel credit" for the value of equivalent LEU fuel. In other words, the cost of preparing excess plutonium for fabrication and fabricating it into MOX fuel is so high that even fuel made from "free" plutonium is far more expensive than LEU fuel purchased on the open market. There is no money in excess plutonium -- except, perhaps, on a nuclear black market. The additional cost of implementing both options at the same time is modest: DOE estimates a total discounted present cost for the hybrid approach of just under $1.5 billion [8]. These estimates do not yet include, however, the costs of "irradiation fees" likely to be demanded by U.S. utilities for the inconveniences of dealing with plutonium fuels, which will probably add a couple of hundred million to the total cost, somewhat increasing immobilization''s estimated cost advantage.

4. Near-term plans

DOE plans to move out rapidly with all of the technologies needed to implement the dual-track approach.

First, pit conversion and plutonium processing. This year, a complete pilot facility for pit conversion using the new U.S. process is being installed at the Los Alamos National Laboratory. This facility will be capable of processing 200 pits per year -- or substantially more, if more than one shift were used. A full-scale facility capable of processing thousands of pits per year could be built by replicating this pilot line several times. This pilot-scale facility is scheduled to begin operations early next year. Development and testing are underway to ensure that the dry, hydride-based process to be used at this facility will produce a suitable oxide for MOX fuel fabrication, or to implement a fix if it does not. Development of a safeguards system that can accurately measure the unclassified, canned plutonium oxide product that results from the process, without revealing classified weapons information from earlier stages of the pit conversion process, is an integral part of the program.

Second, immobilization. DOE is working hard to settle the last remaining technical uncertainties with respect to immobilization, and plans to choose between glass and ceramic materials this September. (Based on the data available so far, my own view is that ceramic materials offer a superior alternative for this plutonium mission, though there is more current experience with waste immobilization in glass.) Full-size "cold tests" of the can-in-canister approach have been conducted at Savannah River, producing glass canisters using non-radioactive simulants for plutonium and high-level wastes, and laboratory-scale "hot tests" of the approach have also been completed. A full-size "hot test" at Savannah River may be possible as soon as the first half of next year. Unfortunately, at the moment very little work is being funded to develop new safeguards approaches for immobilization; in my view, this effort should be substantially expanded.

Third, reactors and MOX. DOE is now preparing a solicitation for consortia of firms to work together to build and operate the necessary MOX plant and irradiate the fuel in utility reactors. DOE hopes to choose the firms quickly, so that both fabricators and utilities can participate in fuel design and qualification from the earliest stages. Currently, initial "rodlets" of MOX pellets are being fabricated at Los Alamos, for irradiation in test reactors, in work designed to collect additional physics and fuel-performance data. This will eventually be followed by the fabrication and irradiation of lead assemblies for irradiation in utility reactors. As noted earlier, whether DOE will wait to fabricate these until a domestic facility becomes available or get an earlier start by having them fabricated in Europe remains uncertain.

Overall, then, DOE expects to carry out a broad program of tests and demonstrations of all the necessary technologies for disposition over the next several years. It is unlikely, however, that disposition of tens of tons of U.S. plutonium will occur before Russia is ready to reduce its excess plutonium stockpile in parallel.

5. Plutonium disposition and U.S. arms reduction and nonproliferation policy

President Clinton has publicly committed himself to the goal of irreversible nuclear arms reductions, and U.S. policy on excess fissile materials is intended as one part of an overall program to achieve that goal. The START agreements have so far limited only strategic launchers and delivery vehicles: once warheads are removed from their launchers, there is no requirement to dismantle them or even to account for what is done with them. This leaves the possibility that when reductions are carried out by simply reducing the number of warheads on missiles that remain in service -- a process known as "downloading" -- the reduction could in principle be rapidly reversed, by simply "uploading" the warheads back onto the missiles [12]. Because many of the U.S. reductions under START II (and START III, if it is successful) are to be achieved largely by such downloading, Russian officials have been particularly concerned about this potential for reversibility.

Fortunately, as a result of the Helsinki agreements [14], it appears that for the first time, the United States and Russia will grapple seriously with the issue of how to verify the dismantlement of the warheads themselves. Warheads that are disassembled, however, can also be reassembled, if the components remain available for military use: hence the critical importance of controlling and ultimately eliminating the excess stockpiles of fissile material resulting from dismantlement. As Minister of Atomic Energy Mikhailov has said, disarmament will genuinely be real when all of the extra weapons material has been converted for peaceful use.

The United States has taken several important first steps in this direction already [10]. First, the United States has declared that substantial fractions of its fissile material stockpiles are excess to its military needs, and has made a political commitment that these excess stockpiles of fissile material will never again be used in nuclear weapons -- what might be called "political" irreversibility. Second, the United States has declared its intention to place these excess stockpiles under international safeguards to verify that they are not returned to weapons -- "verified" irreversibility. Unfortunately, however, that process has been slow, and only about 12 tons of material is actually under safeguards today. The Moscow Nuclear Safety and Security Summit in April 1996 called for such steps to be taken with all excess fissile materials, and Russia has joined the United States and the IAEA in a trilateral process to work out specific arrangements for verifying the non-weapons use of excess materials [15]; nevertheless, to date, neither Russia nor any other weapon state has joined the United States in declaring former military material excess, committing not to use it in weapons, or placing it under IAEA safeguards.

Third, as described above, the United States is undertaking a major program to prepare to physically transform these excess materials in ways that would make it far more costly, time-consuming, and observable to ever return them to weapons use -- "physical" irreversibility. As noted earlier, this step will probably not be taken on a large scale unilaterally: if nothing else, it appears unlikely that Congress would provide the substantial funds required for disposition of U.S. excess plutonium if Russia were retaining its larger weapons plutonium stockpile in forms ready to put right back into nuclear weapons. Much the same can probably be said of Russia; hence, plutonium disposition is a job the two countries are going to do together, or not at all.

Plutonium disposition is also intended to serve the goals of non-proliferation -- both by demonstrating the major nuclear weapon-states'' commitment to their Article VI disarmament obligations, and by transforming large stockpiles of material into forms far less vulnerable to theft and diversion. As noted earlier, one of the fundamental goals of President Clinton''s nonproliferation policy is to reduce stockpiles of separated, weapons-usable plutonium worldwide. Simply leaving this material in storage indefinitely would mean complete reliance on the continued effectiveness of whatever safeguarding arrangements are put in place to guard and monitor it; while the United States is confident in its own safeguarding arrangements, the ongoing economic crisis in the former Soviet states is creating new risks of theft and diversion that must be addressed. Fortunately, Russia, the United States, and other countries are quietly cooperating in a major program to install modern safeguards and security systems at the sites in the former Soviet Union where weapons-usable fissile materials are stored, and this program is making surprising progress. Nonetheless, the desirability of transforming as much material as possible into forms that pose less inherent proliferation risk is clear -- as long as it can be done without a short-term increase in proliferation risk resulting from large-scale material processing and transport that is large enough to counteract the long-term benefit.

For these reasons, the assembled leaders at the Moscow nuclear summit unanimously concluded that disposition of excess fissile materials should be accomplished as quickly as possible, under international safeguards, and with stringent security and accounting measures ensuring effective nonproliferation controls.

Although plutonium disposition is intended to serve nonproliferation goals, some have seen the dual-track approach, with its acceptance of the use of excess weapons plutonium as fuel in nuclear reactors, as contradicting the long-standing U.S. nonproliferation policy of not supporting reprocessing and recycling of plutonium. Nothing could be further from the truth. As top officials from President Clinton on down have made clear, the United States is "not changing our fundamental policy toward nonproliferation and the nuclear fuel cycle." [16] The dual-track approach is about eliminating stockpiles of separated, weapons-usable plutonium that already exist, not about producing additional stockpiles of separated plutonium by reprocessing. Indeed, it is precisely because the United States feels so strongly about the proliferation risks posed by separated plutonium that it has decided to use both of the best technologies available -- MOX in reactors and immobilization -- to eliminate its excess stockpiles of this dangerous material as rapidly as possible. As such, the United States has committed that the spent fuel resulting from disposition will not be reprocessed, and that the government-owned MOX plant to be built for this mission will only be licensed for this mission and will be dismantled when it is complete.

Nevertheless, the dual-track decision has provoked fierce opposition from U.S. antinuclear groups, and from some in the nonproliferation community. It will continue to be controversial in the United States, and as DOE moves toward actually fabricating plutonium fuel and loading it into reactors, repeated political and legal attempts to block the effort can be expected.

Since this conference is on the future of the fuel cycle, let me say a few words about the U.S. policy against reprocessing and recycling of plutonium, and the reasons for it. First, despite the remarkable progress of safeguards technology, a world in which tens of tons of separated, weapons-usable plutonium is being produced, processed, and shipped from place to place every year when a few kilograms is potentially enough for a bomb clearly poses greater proliferation risks than a world in which that is not occurring. All separated plutonium, whether reactor-grade or weapon-grade, poses serious proliferation risks. Since this latter point is often misunderstood, let me elaborate.

For an unsophisticated proliferator, making a crude bomb with a reliable, assured yield of a kiloton or more -- and hence a destructive radius about one-third to one-half that of the Hiroshima bomb -- from reactor-grade plutonium would require no more sophistication than making a bomb from weapon-grade plutonium. A somewhat more sophisticated proliferator could readily make bombs from reactor-grade plutonium with substantially higher reliable, assured yields. And major weapon states like the United States and Russia could, if they chose to do so, make bombs from reactor-grade plutonium with yield, weight, and reliability characteristics similar to those made from weapons-grade plutonium. That they have not chosen to do so in the past has to do with convenience and a desire to avoid radiation doses to workers and military personnel, not the difficulty of accomplishing the job. The United States has recently declassified an unprecedented level of detail on this subject, which I commend to your attention [10, pp. 37-39]. My colleagues from the National Academy of Sciences panel on plutonium disposition and I have spoken not only to designers from all three of the U.S. weapons laboratories on this subject, but also to weapons designers from all five of the declared weapon states; I think it is safe to assert that these points are not in dispute among weapon designers who have looked into the matter. Indeed, one Russian weapons-designer who has focused on this issue in detail criticized the information declassified by DOE for failing to point out that in some respects it would actually be easier for an unsophisticated proliferator to make a bomb from reactor-grade plutonium (as no neutron generator would be required). In short, the United States is concerned about the proliferation risks posed by large-scale handling of separated plutonium, and prefers not to encourage increases in these risks.

Second, with relatively low uranium prices, which are projected to continue for the foreseeable future, a once-through fuel cycle is clearly more cost-effective than a reprocessing fuel cycle. Hence, while a variety of U.S. utilities are interested in participating in a MOX disposition program, for which they expect to be paid, none of them are interested in reprocessing and recycling their own plutonium, for which they would have to pay more. Moreover, while the current abundance of cheap uranium has limited work on exploring how large the marginal supplies might be, a strong argument can be made that the uranium supply will last for at least fifty years, and more probably a century or more. And the increasingly diversified sources of supply worldwide should reduce concerns about fuel-supply security. In short: there is no real need for reprocessing and recycling now, or for decades to come [17].

Third, there do not appear to be major waste-management benefits from reprocessing. Performance assessments of potential geologic repositories routinely conclude that plutonium is a very minor contributor to the overall environmental risk from a nuclear waste repository; other isotopes, not separated by reprocessing, are even longer-lived, and far more mobile in a repository environment, making them a greater hazard to the biosphere. While reprocessing decreases the physical volume of high-level waste, the physical volume of the repository required is determined by the heat the rock can sustain, and the majority of the early heat is from the fission products, which are not separated by reprocessing. And reprocessing and recycle carries with it its own risks of accident and radiation release [18].

I do not expect everyone here to agree with these points of view. But the point is that the U.S. approach is not based on anti-nuclear ideology, but rather on a considered assessment of the pros and cons of the different approaches, which led to a conclusion that, for now and for the foreseeable future, reprocessing and recycle poses more risks than benefits.

6. The way forward: a personal view

While the United States has undertaken a major fissile material disposition program, we are still a long way, today, from being in a position in which we can be confident that disposition of excess plutonium will in fact occur in the foreseeable future, or that the hoped-for improvement in the irreversibility of nuclear arms reductions will be achieved. A substantial increase in high-level political attention, the commitment of major financial resources from the international community, and significant changes in existing U.S. and Russian nuclear arms reduction policies will be necessary to get the job done.

The principle obstacle to implementing plutonium disposition in the United States is politics; the principle obstacle in Russia is money. The two are not unrelated. I believe that the political objections to implementing the dual-track in the United States -- serious though they are -- can be overcome if the program is part of a reciprocal disarmament package with Russia eliminating its excess stockpiles on a parallel track. But with an economic crisis so severe that even basic necessities such as wages and pensions are not being paid, Russia simply does not have the money to build the facilities needed to implement a plutonium disposition program. Given the very low cost of uranium fuel in Russia, it is highly unlikely that any substantial portion of the cost can be financed on a purely commercial basis through sales of the MOX fuel to Russian reactors. And the United States is unlikely to be willing to pay for 100% of the cost of its own plutonium disposition program, and 100% of the cost of Russia''s program. This financial mystery -- where will the money came from? -- is, in my view, by far the largest obstacle that must be overcome if plutonium disposition is to be accomplished. Unfortunately, little progress toward resolving it has been made so far: despite the many discussions of this subject following the Moscow nuclear summit, including the Paris international experts'' meeting in November of last year, there have as yet been no volunteers to pay any substantial fraction of the necessary cost.

Two general approaches to overcoming this obstacle can be envisioned. The international community could agree to share the cost through direct government contributions, as is being done to finance the shut-down of Chernobyl and the construction of new reactors in North Korea, to pick just two examples. The principal difficulty of such an approach is that governments would have to remain focused and committed for many years for it to succeed. Nevertheless, this is a serious possibility that ought to be further explored. The second general class of approach is some form of barter arrangement. For example, the French and German experts working with Russia on the proposed MOX pilot plant have considered an arrangement in which Cogema and Siemens would help MINATOM build a MOX plant in Russia, and would be paid with low-cost uranium and enrichment services, which they would sell on the international market. I have proposed a somewhat similar arrangement in which a joint venture would be established including MINATOM and Western fuel-cycle and construction firms: MINATOM would transfer 100 tons of excess HEU to this joint venture (above and beyond the 500 tons being sold to the United States), the Western governments would agree to open their restricted markets to this modest additional increment of material, and the joint venture would then be able to borrow the funds needed to build and operate the necessary facilities against the large asset represented by this 100 tons of HEU. This concept could potentially make it possible to finance plutonium disposition, create a management structure for implementing plutonium disposition that can sustain itself over the long term, eliminate an additional 100 tons of HEU, and provide substantial business to both MINATOM''s desperate nuclear cities and to Western firms, all at little or no direct on-budget cost to the countries involved [19].

What matters is not so much which approach is chosen to solve this financing problem, but that the international community buckle down to the job of solving it -- and soon. The Denver P-8 summit later this month is the place to begin; at a minimum, it is to be hoped that the summit will agree to task experts to begin preparing approaches for a later decision by the P-8 member states. While the Russian-French-German proposal for a MOX pilot plant in Russia is a useful idea that ought to be pursued, in the long run U.S. participation and support is likely to be essential to success. A preliminary agreement between the United States and Russia on what needs to be done, and joint U.S.-Russian efforts to convince the other states to take part, would dramatically increase the chances for progress.

Is a near-term U.S.-Russian agreement of this kind possible? I believe so. While not making any specific financial commitment, the United States has expressed its willingness to support disposition of Russian plutonium, including the construction of a MOX plant in Russia, if four nonproliferation conditions are met: international safeguards throughout the disposition process (while protecting sensitive weapon design information); stringent security and accounting measures to prevent theft or diversion; use of a facility financed with help from the international community only for excess weapons plutonium, at least until disposition of that material is complete (since the international community would be contributing to the financing primarily for disarmament reasons); and no reprocessing of the spent fuel, at least until all of the excess weapons plutonium has been processed once through. None of these are particularly onerous, as they leave open what would be done with the relevant facilities once disposition is complete. They are an effort to do what has to be done: to bring countries that have common security interests in dealing with excess weapons plutonium together in a cooperative approach that does not compromise any of their diverging interests in the future of civilian plutonium.

In February of this year, John Holdren and Yevgeniy Velikhov, the co-chairmen of the U.S.-Russian Independent Scientific Commission on Disposition of Excess Weapons Plutonium which was established by the U.S. and Russian governments to make recommendations to the two Presidents on how to proceed with plutonium disposition wrote to Vice President Gore and Prime Minister Chernomyrdin, urging them to direct their governments to prepare an initial agreement which would: a) endorse implementation of the dual-track approach in both the United States and Russia; b) accept the U.S.-proposed non-proliferation conditions, to be implemented reciprocally in both countries; and c) commit both countries to work together with their P-8 partners to raise the necessary funds and establish an international entity that could implement such an international cooperative program. Agreement on these basic points remains as urgent and essential today as when that letter was written.

In addition to the financing issue, there is the issue of ensuring genuine irreversibility -- and a comparable degree of irreversibility in the United States and Russia. If either country retains reserve stockpiles of warheads and fissile materials sufficient to rebuild a Cold War nuclear arsenal, plutonium disposition will not achieve its irreversibility purpose. Yet as far as can be determined from publicly available information, that is precisely the policy both countries are now pursuing. While the United States has declared more than half of its plutonium excess, the remainder is sufficient for a very large nuclear arsenal. It has now been officially declassified that the United States plans to retain a reserve of warheads and fissile materials sufficient to replace 100% of its deployed warheads -- which is to say, sufficient to rapidly double its deployed arsenal [2]. If the fraction of its HEU that Russia has agreed to sell is any indication, Russia plans to do much the same. If genuine irreversibility is to be achieved, START III and associated agreements will have to address these "extra," reserve stockpiles, and reduce the total stockpiles of nuclear warheads and nuclear materials to the levels necessary to support the number of deployed warheads permitted by U.S.-Russian agreements -- resulting in substantially larger quantities of excess material than have been declared to date. Verifying the total stockpiles of warheads and fissile materials will be a difficult task; the necessary regime of data declarations and inspections can and should be built step-by-step, with each new step adding to confidence while posing minimal risk in itself [5, 12].

This already tall order will be further complicated by the imbalance in total stockpiles. Russia''s total stockpiles of warheads, plutonium, and HEU are all substantially larger than U.S. stockpiles -- so to achieve parity at lower levels will require larger reductions on the Russian side. This principle of "reductions to equal levels, not equal reductions" has already been established in the START treaties. In the case of plutonium, unclassified U.S. estimates indicate that Russia has approximately 200 tons of separated plutonium (including 30 tons of separated reactor-grade plutonium) compared to just under 100 tons for the United States [20]. If Russia simply declared 50 tons of this excess, to match the United States, it would have 150 tons remaining, while the United States would have less than 50, exacerbating the disparity rather than reducing it. Over time, the United States and Russia will have to negotiate an agreement specifying how much plutonium and HEU will be removed from their military stockpiles, and when; ideally, this agreement should call for reducing their remaining military stockpiles to low, equal levels. Work on the initial disposition demonstrations and facilities should not wait until such an agreement is completed, however.

In short, achieving the goals of plutonium disposition will require intensive efforts to arrange the necessary financing, and substantial revisions in the current nuclear arms policies of both the major states involved. Such measures will require a dramatic increase in the level of active and concerted attention to this issue from the highest levels of government. The job of disposition advocates, therefore, is to impress upon governments that eliminating the fissile legacies of the Cold War is an essential international security endeavor which must be accomplished as quickly as possible, and that what is required from Presidents and Ministers is not just endorsement in principle, but active engagement to get the job done.

7. References

[1] CANTER, H., "Overview of the United States'' Plutonium Inventory Excess to National Security Needs," Presented at the International Experts Meeting on Disposition of Excess Weapons Plutonium, Paris, France, October 28-31, 1996.
[2] U.S. CONGRESS, GENERAL ACCOUNTING OFFICE, Department of Energy: Plutonium Needs, Costs, and Management Programs, GAO/RCED-97-98, Washington DC (1997).
[3] U.S. DEPARTMENT OF ENERGY, Record of Decision for the Storage and Disposition of Weapons-Usable Fissile Materials Final Programmatic Environmental Impact Statement, Washington DC (1997).
[4] THE WHITE HOUSE, OFFICE OF THE PRESS SECRETARY, Nonproliferation and Export Control Policy, Washington DC (1993).
[5] NATIONAL ACADEMY OF SCIENCES, COMMITTEE ON INTERNATIONAL SECURITY AND ARMS CONTROL, Management and Disposition of Excess Plutonium, National Academy Press, Washington DC (1994).
[6] NATIONAL ACADEMY OF SCIENCES, PANEL ON REACTOR-RELATED OPTIONS, Management and Disposition of Excess Weapons Plutonium: Reactor-Related Options, National Academy Press, Washington DC (1995).
[7] U.S. DEPARTMENT OF ENERGY, OFFICE OF FISSILE MATERIALS DISPOSITION, Summary Report of the Screening Process to Determine Reasonable Alternatives for Long-Term Storage and Disposition of Weapons-Usable Fissile Materials, Washington DC (1995).
[8] U.S. DEPARTMENT OF ENERGY, OFFICE OF FISSILE MATERIALS DISPOSITION, Technical Summary Report for Surplus Weapons-Usable Plutonium Disposition Rev. 1, Washington DC (1996).
[9] U.S. DEPARTMENT OF ENERGY, OFFICE OF FISSILE MATERIALS DISPOSITION, Storage and Disposition of Weapons-Usable Fissile Materials Final Programmatic Environmental Impact Statement, Washington DC (1996).
[10] U.S. DEPARTMENT OF ENERGY, OFFICE OF ARMS CONTROL AND NONPROLIFERATION, Final Nonproliferation and Arms Control Assessment of Weapons-Usable Fissile Material Storage and Excess Plutonium Disposition Alternatives, Washington DC (1997).
[11] GLOAGUEN, A., et al., Key Issue Paper #2, these Proceedings.
[12] BUNN, M., HOLDREN, J.P., "Managing Military Uranium and Plutonium in the United States and the Former Soviet Union", Ann. Rev. Energy Environ. (forthcoming).
[13] Interim Report of the U.S.-Russian Independent Scientific Commission on Disposition of Excess Weapons Plutonium, Office of Science and Technology Policy, Washington DC, September (1996).
[14] THE WHITE HOUSE, OFFICE OF THE PRESS SECRETARY, Joint Statement on Parameters on Future Reductions in Nuclear Forces, Washington DC, March 21 (1997).
[15] Moscow Nuclear Safety and Security Summit Declaration, Kremlin, Moscow, April 20, (1996).
[16] CLINTON, B., Letter to Representative Edward Markey, February 5 (1997).
[17] VON HIPPEL, F., "An Evolutionary Approach to Fission Power", Global 1995: International Conference on Evaluation of Emerging Nuclear Fuel Cycle Systems, Versailles, France, September (1995).
[18] NATIONAL RESEARCH COUNCIL, COMMITTEE ON SEPARATIONS TECHNOLOGY AND TRANSMUTATION SYSTEMS, BOARD ON RADIOACTIVE WASTE MANAGEMENT, COMMITTEE ON GEOSCIENCES, ENVIRONMENT, AND RESOURCES, Nuclear Wastes: Technologies for Separations and Transmutation, National Academy Press, Washington DC (1996).
[19] BUNN, M., "Getting the Plutonium Disposition Job Done: The Concept of a Joint-Venture Disposition Enterprise Financed by Additional Sales of Highly Enriched Uranium", Science for Peace Series Vol. 1, International Conference on Military Conversion and Science: Utilitization/Disposal of the Excess Fissile Weapons Materials: Scientific, Technological and Socio-Economic Aspects (Proc. Int. Symp. Como, Italy, March 1996), (KOUZMINOV, V., MARTELLINI, M., Eds.), UNESCO Venice Office, Venice (1996).
[20] U.S. CONGRESS, GENERAL ACCOUNTING OFFICE, Nuclear Nonproliferation: Status of U.S. Efforts to Improve Nuclear Material Controls in Newly Independent States, GAO/NSIAD/RCED-96-89, Washington DC (1996).