Managing Military Uranium and Plutonium in the United States and the Former Soviet Union: Monitored Reductions in Nuclear-Weapon and Fissile-Material Stockpiles

MONITORED REDUCTIONS IN NUCLEAR-WEAPON AND FISSILE-MATERIAL STOCKPILES

Bunn, Matthew, and John P. Holdren. "Managing Military Uranium and Plutonium in the United States and the Former Soviet Union." Annual Review of Energy and the Environment 22 (1997): 403-486.

While past arms-control agreements have focused primarily on limiting missiles and launchers, the objectives of both irreversible nuclear arms reductions and reduced risk of nuclear theft call for the next generation of agreements to focus in addition on controlling nuclear weapons themselves and the fissile materials needed to make them. President Clinton and Russian President Boris Yeltsin have repeatedly agreed to pursue the goal of "transparency and irreversibility" of nuclear arms reductions, and at their summit in Helsinki, Finland, on March 21, 1997, the two Presidents agreed on a framework for a future START III agreement that included a call for transparency measures relating to "strategic nuclear warhead inventories and the destruction of strategic nuclear warheads," along with other measures to help ensure irreversibility and prevent "a rapid increase in the number of warheads." Transparency measures for tactical nuclear warheads and fissile material stockpiles are to be considered as separate issues (7). The United States has indicated, however, that it will not begin negotiations of a START III agreement until Russia's parliament ratifies the START II treaty; hence, the two governments have not yet begun to flesh out what the broad language of the Helsinki framework might mean in practice. To fully achieve the objectives of transparency and irreversibility of nuclear arms reductions and to lay the foundation for deep nuclear arms reductions in the future will ultimately require a comprehensive transparency and monitoring regime— but such a regime can be built step by step over time.

Outline of a Comprehensive Regime

The NAS report recommends that the United States work to reach agreement with Russia on a broad, reciprocal regime to include (8) (a) declarations of stockpiles of nuclear weapons and all fissile materials, (b) cooperative measures to clarify and confirm those declarations (including physical access to production facilities and production records for fissile materials), (c) an agreed, monitored halt to additions to these stockpiles of warheads and fissile materials for weapons, and (d ) agreed, monitored net reductions in these stockpiles.

Monitoring of warhead dismantlement and commitment of excess fissile materials to non-weapons use or disposal, initially under bilateral and later under international safeguards, would be integral parts of this regime, as would some form of monitoring of whatever warhead assembly continues. Both the NAS study and the Bilateral Commission reports recommend that the United States and Russia agree to reduce to small, roughly equal remaining numbers of nuclear warheads and small, roughly equal remaining levels of plutonium and HEU in military stockpiles. All other plutonium and HEU, in this approach, would be placed under bilateral and ultimately international monitoring to ensure that it was used only for non-explosive purposes.

Such a regime would do a great deal to build confidence in the size and management of each side's nuclear stockpiles and the progress of nuclear arms reductions— and the information exchanged and the site visits conducted would provide useful additional information to support cooperative MPC&A efforts (though bilateral or international monitoring itself could only detect, not prevent, thefts of material). Creation of this broad regime could be approached step by step, with each step adding to security while posing little risk. While such a regime could never be rigorously verified, in the sense of absolutely confirming that a few dozen nuclear weapons or a few tons of fissile material had not been hidden away somewhere, these measures would be mutually reinforcing, building increasing confidence over time that the information exchanged was accurate and that the goals of the regime were being met. With a sufficiently inclusive approach, the difficulty of falsifying the broad range of information exchanged in a consistent way, so as to hide a stockpile large enough to be strategically significant, could be made reasonably high. The NAS report recommends that this regime ultimately be internationalized, so that all states would declare their holdings of fissile material and all the declared weapon states would declare their holdings of nuclear weapons.

While such a regime could ultimately provide a very high degree of transparency in the management of nuclear weapons and materials, nuclear arms reductions can never be truly irreversible; both the United States and Russia will retain the knowledge and the industrial strength required to rebuild large nuclear arsenals should a national decision be taken that it is a high priority to do so. Nevertheless, a series of steps is possible that could each add to the cost, delay, and observability involved in rebuilding a large arsenal, and thereby greatly reduce the probability of such a decision being made.

Removing warheads from missiles (the primary measure required for the United States under START II) would take only weeks or months to reverse. Dismantling the missiles required to launch those warheads, and the launchers for those missiles (required for certain types of missiles under the START agreements) greatly increases the cost and time required to rebuild. Dismantling the warheads themselves, not yet required under START I or START II, also provides a significant increment of increased cost and delay in rebuilding (particularly useful if the missiles and launchers that might carry those warheads have not been dismantled). The United States, for example, has been dismantling warheads at maximum capacity for roughly a decade, and the time required to reassemble them would be comparable (unless the existing facility for this purpose began working multiple shifts, which might cut the time required by 50% or more); substantially expanding the available capacity for warhead assembly would take nearly as long. If the plutonium and HEU warhead components are dismantled, then refabrication of these components would be necessary before the warheads could be reassembled, and this too would take time and cost money; today, the only operational US facility for fabricating plutonium weapons components is Technical Area 55 at the Los Alamos National Laboratory, which can only fabricate a few hundred components a year.

Ultimately, however, warheads and components that are disassembled can be reassembled— as long as the fissile material needed to do so remains available. While a degree of political and legal "irreversibility," as well as transparency, can be achieved through committing these materials to civilian use or disposal and placing them under bilateral or international monitoring (as well as through monitored commitments not to reverse other steps of this process), if circumstances changed, materials could always be removed from such arrangements. Thus, a fundamental long-term step toward "physical irreversibility" is to transform the excess stockpiles of plutonium and HEU into forms that would require re-enrichment or reprocessing to return them to weapons; recovering, for example, 50 tons of plutonium that had been transformed into spent reactor fuel or immobilized with highly radioactive fission products would require years of operation of large reprocessing facilities, at a cost likely to total billions of dollars (for a discussion of making these distinctions among political, legal, and physical irreversibility in more detail see 36). Moving forward as quickly as practicable in accomplishing such a transformation would also send the world an important signal of the United States' and Russia's intentions not to reverse their ongoing reductions. Even after such transformation, however, either the United States or Russia could reverse the process should they choose to devote enough time and resources to the task; indeed, even if the fissile material were eliminated completely (for example by shooting it into outer space), either nation could eventually produce new fissile material from which to rebuild a large nuclear arsenal. Of course, if sufficient warheads or fissile materials are retained in reserve to rebuild a Cold War strategic arsenal, then programs to eliminate other warheads and fissile materials will not truly achieve the irreversibility objective.

The remainder of this section focuses primarily on US and Russian efforts to build transparency in the management of fissile materials and the dismantlement of nuclear warheads; the question of transforming excess fissile material stockpiles into forms less readily usable in weapons is addressed in detail in a subsequent section.

For years before the Helsinki framework was agreed upon, the United States and Russia had been discussing a variety of elements of a transparency regime. During 1994-1995, the two countries reached several important agreements in principle, but none has been implemented— in large part because of concerns on the Russian side about revealing sensitive nuclear information. Unfortunately, little attention has been paid to date to examining how financial and other incentives could be structured to convince Russian officials that it is in their nation's interest to reverse five decades of Cold War nuclear secrecy and participate in a broad program to monitor and reduce stockpiles of warheads and fissile materials.23

A broad range of efforts is underway in this area, including formal bilateral US-Russian transparency negotiations; negotiations concerning international, rather than bilateral, monitoring; and alternative, less formal transparency approaches. We now discuss the status of each of these areas in turn.

Bilateral Transparency Negotiations

INSPECTING PLUTONIUM AND HEU FROM DISMANTLED WEAPONS . . . In March 1994, in the first major US-Russian agreement in this area, Secretary of Energy Hazel O'Leary and Russian Minister of Atomic Energy Victor Mikhailov agreed to establish a regime of mutual inspections to confirm the inventories of plutonium and HEU removed from dismantled nuclear weapons. This initiative eventually came to be called Mutual Reciprocal Inspections (MRI).24 Since 1994, US and Russian experts have carried out a number of joint experiments and come close to agreeing on the specific types of measurements that would be used to confirm that an inspected canister contained a plutonium weapon component; a less intrusive regime is proposed for inspections of HEU components. Such inspections have not yet been implemented, however, because doing so would require an agreement providing the legal basis for exchanging limited types of classified nuclear information (see below).

EXCHANGING NUCLEAR STOCKPILE DATA . . . At their September 1994 summit, President Clinton and President Yeltsin agreed that for the first time ever, the two sides would exchange "detailed information" on "aggregate stockpiles of nuclear warheads, on stocks of fissile materials and on their safety and security" (5). This data exchange has never been implemented either, for the same reason. (Both the United States and Russia still consider the size of their nuclear weapon stockpiles to be classified information.)

German Foreign Minister Klaus Kinkel and others have recommended the establishment of an international nuclear weapons register containing declarations from each nuclear weapon state concerning its stockpiles of nuclear weapons (and in some formulations, stockpiles of plutonium and HEU as well). The weapon states, however, have not been willing to seriously consider this proposal (17, 64, 65; for a similar 1992 proposal from Andrei Kozyrev, then Russia's Foreign Minister, see 66).

CONFIRMING WARHEAD DISMANTLEMENT AND FISSILE MATERIAL DATA . . . In their May 1995 summit statement, Presidents Clinton and Yeltsin reaffirmed that their two governments would negotiate agreements on MRI and the stockpile data exchange and also called for a third agreement on "other cooperative measures, as necessary to enhance confidence in the reciprocal declarations on fissile material stockpiles." More vaguely, the two Presidents agreed to "examine and seek to define" possibilities for "intergovernmental arrangements to extend cooperation to further phases of the process of eliminating nuclear weapons." These modest additional steps forward were based on measures included in a comprehensive US transparency proposal tabled in December 1994, which called for detailed data exchanges on all warhead and fissile material stockpiles and a range of measures (including on-site inspections) to confirm the data exchanged and help confirm the dismantlement of nuclear weapons.25 The latter objective came to be referred to in the US government as tracing the "chain of custody" of nuclear weapons, from storage sites to dismantlement sites to the fissile components resulting from dismantlement. With neither the MRI concept nor the data exchanges moving forward, no negotiations on these additional subjects were pursued after the May 1995 summit statement, and as of mid-1997, no official US-Russian discussions of how to implement either these past commitments or the broader measures called for in the Helsinki framework had yet begun. Various approaches to confirming the dismantlement of warheads have been proposed. They range from simply monitoring the buildup of plutonium warhead components, or "pits", in storage and assuming that these did in fact come from warhead dismantlement to establishing monitoring systems surrounding the warhead disassembly facilities that would count the warheads and pits going in and out of the facilities (identifying them as such by a variety of possible means, particularly checking of certain radioactive "signatures") (for discussions see 8, 28-30, 67).

LAYING THE LEGAL BASIS FOR CLASSIFIED NUCLEAR EXCHANGES . . . Both the US and Russian legal systems impose stringent requirements for protecting classified information related to nuclear weapons. In 1994, Congress amended the Atomic Energy Act to provide legal authority to negotiate an "Agreement for Cooperation" with Russia that would provide the legal basis for exchanging classified nuclear information (known under the Act as "restricted data") for nonproliferation and arms control purposes. The two sides began negotiating such an agreement in 1995 and had it nearly completed by late 1995, but at that time the Russian government called off further talks pending a "policy review," and the talks have never resumed.

This should perhaps have been expected. Breaking down the barriers of five decades of Cold War secrecy continues to be a very difficult task in the United States, with its established democracy, and could be expected to be even more difficult in a system emerging from communism. The broad Agreement for Cooperation that was being negotiated in 1995 would have allowed (though not required) the exchange of a broad range of nuclear information related to both warheads and fissile materials and therefore brought into play the interests of a wide range of competing agencies in Russia, including the Ministry of Defense, the Federal Security Service (successor to the KGB), the Ministry of Foreign Affairs, and others, in addition to MINATOM itself. Fundamentally, it appears that in late 1995-1996, with Yeltsin's future uncertain (first with elections upcoming and then with his health in doubt), Russian officials concluded that there would be considerable risk and little benefit in signing a document making it possible to open Russia's nuclear secrets to the United States. Whether this will now change remains to be seen.

As Russian officials have frequently suggested, a step-by-step approach may be needed, narrowing the range of issues initially tackled. One idea suggested privately by some Russian officials is to reach an Agreement for Cooperation limited to only one very specific initiative— such as inspections of plutonium from dismantled weapons— which would not raise the whole range of concerns brought up by the broader proposed agreement. As further steps were agreed to, the language of such an initial agreement could be adapted for new specific agreements on those subjects as well.

NUNN-LUGAR STORAGE SITE TRANSPARENCY . . . As noted above, the United States and Russia have agreed in principle that transparency measures will be applied at the storage facility being built with US assistance. No significant progress has been made, however, in determining what those measures will be— a process complicated by the question of how, if at all, they would differ from the bilateral transparency measures for storage facilities contemplated under the MRI approach or from the international measures contemplated in discussions of placing excess material under IAEA verification (described below). The overlap of three different proposed transparency regimes at this one facility— unilateral Nunn-Lugar transparency, bilateral MRI measures, and international IAEA verification— has complicated negotiations considerably, and the US government's inability, through mid-1997, to provide a clear answer to Russian concerns about how these measures would interrelate and what reciprocal measures would be imposed on US excess weapons components in storage has provoked significant suspicion on the Russian side.

HEU PURCHASE AGREEMENT TRANSPARENCY . . . The one area where the United States and Russia have actually agreed on and are implementing specific transparency measures is the one area where large sums of money are involved— the US purchase of HEU from Russian weapons, blended down to LEU. Transparency measures have been established to provide the United States high confidence that the LEU it is purchasing in fact came from HEU, and at least modest assurance that the HEU came from weapons. These measures are also designed to give Russia confidence that the LEU, once in the United States, is used only for peaceful purposes.26

International Monitoring of Excess Material

Some progress is being made toward placing excess material under international monitoring to verify that it is never again returned to weapons— a step recommended in both the NAS and Bilateral Commission reports— but as with bilateral transparency efforts, the process has been painfully slow.

In September 1993, President Clinton announced that the United States would make its excess fissile material eligible for IAEA safeguards, to assure the world that these materials would never again be used for nuclear weapons. To date, the United States has placed only 12 tons out of more than 225 tons of excess material (10 tons of HEU and 2 tons of plutonium) under IAEA safeguards; classification issues, budget constraints, and safety concerns related to monitoring material in radioactive facilities have slowed progress. Several tens of tons of additional material, however, are now being made available for IAEA verification.

At the Moscow Nuclear Safety and Security Summit, the assembled P-8 leaders agreed that excess fissile material should be placed under international safeguards as soon as it is practicable to do so (68). On that occasion, Russian President Yeltsin made a commitment to place the storage facility being built at Mayak, which will hold an estimated 40 tons of plutonium and a much larger amount of HEU, under IAEA safeguards.

Traditional safeguards, however, cannot be applied to material in the form of weapons components— a category that includes the material to be stored at Mayak and much of the US excess fissile material— without revealing classified information that could contribute to nuclear proliferation. Neither the United States nor Russia currently has operational industrial-scale facilities for converting metallic plutonium pits to unclassified forms such as plutonium oxide. Thus, if these materials are to be placed under international monitoring in the near term, a modified verification approach that provides confidence without revealing sensitive weapon design information will be required.27

A consensus is developing among the IAEA, the United States, and Russia that this effort represents a fundamentally new mission verifying disarmament in heavily-armed nuclear-weapon states rather than verifying nonproliferation in non-nuclear-weapon states— and that therefore new terminology ("verification" rather than "safeguards") and new approaches should be used. In September 1996, the United States, Russia, and the IAEA established a trilateral forum to discuss the broad range of issues related to placing excess materials under IAEA verification— including not only the technical issues related to classified materials but also questions such as who will pay for the substantial costs incurred by the IAEA in implementing such monitoring; how the commitment to keep these materials under safeguards can be made irreversible when both the US and Russian "voluntary offer" safeguards agreements with the IAEA allow them to remove material from safeguards at any time; and how much effort should be placed on monitoring different types of material (weapons components, metals, oxides, and difficult-to-measure impure forms and residues). Little progress has been made in this forum to date, however. Indeed, as of mid-1997, disputes continued over even the basic purpose of this trilateral initiative (should the monitoring be intended to ensure that the material is never returned to weapons, and if so, should it not continue after the material leaves the storage facility?) and its scope (should the effort cover only the weapons components in storage at Mayak, similar US excess weapons components as well, or broader categories of material excess to military needs?).

Such safeguards on excess fissile material could ultimately cost millions of dollars a year to implement, a cost small by security standards but large in the context of the IAEA budget. The NAS report recommends that the United States and Russia (possibly with some contributions from other members of the international community interested in verified disarmament) pay the IAEA's costs for this mission. More broadly, to carry out its expanded, post-Gulf War missions, along with the new missions of monitoring excess nuclear material and monitoring a global fissile material production cutoff, the IAEA's safeguards budget will need to be substantially increased and other steps taken to strengthen the IAEA's ability to carry out its expanding responsibilities. The IAEA Board of Governors' approval, in May 1997, of a Model Protocol to states' safeguards agreements, which would vastly expand IAEA access to a broad range of sites and information, is a welcome and important step in the long-term effort to strengthen the global safeguards system— but the proof of the effectiveness of the new approaches will be in their implementation (69).

Informal Approaches to Transparency Objectives

Given the slow pace of formal negotiations, a variety of more informal efforts are being made to pursue transparency objectives, including unilateral openness measures and lab-to-lab transparency technology development initiatives.

UNILATERAL OPENNESS INITIATIVES . . . The United States in particular has taken major unilateral initiatives to increase the openness of its nuclear activities, including those related to fissile materials. The objective of these measures was not only to help provide the basis for monitored reductions in warhead and fissile material stockpiles but more broadly to provide the US and world publics with the information needed for democratic decision-making on the future of these nuclear activities. The declassified information ranges from details of past radiation experiments on humans to the number of US nuclear tests. For present purposes, the most crucial information concerns the size, characteristics, and locations of the stockpiles of nuclear weapons, plutonium, and HEU. Although information about the size and locations of US nuclear weapon stockpiles remains classified, detailed data on US warhead dismantlement rates and plutonium stockpiles have been released. Data on the US HEU stockpile are being prepared for release (43). In addition, visits to many nuclear facilities by the public and by Russian representatives have been permitted for the first time. The NAS report recommends that additional information be declassified, including the number of nuclear weapons and the average amount of plutonium in each weapon pit— both important to building an overall reductions regime (8).

While Russia has not yet matched all of these initiatives, a significant increase in openness is apparent. Russia has allowed unprecedented visits to many formerly secret nuclear sites (while keeping the most sensitive areas off-limits) and has declassified full information on past Russian nuclear testing, paralleling the information the United States released earlier on its own nuclear testing program. Information on the size, locations, and characteristics of Russia's stockpiles of warheads and fissile materials remains classified, however.

Several other countries are also increasing transparency with respect to their nuclear stockpiles. In particular, most major states engaged in separating and recycling civilian plutonium are now annually publishing figures on their stockpiles, and discussions among nine of these states, including the United States and Russia and all the states most extensively involved in civilian plutonium recycling, recently led to general agreement— now being considered by the central governments of those states— on a set of guidelines intended to provide a generally consistent level of transparency in civilian plutonium management (for a brief discussion see 17).

LAB-TO-LAB TRANSPARENCY TECHNOLOGY DEVELOPMENT . . . Building on the successful model of the MPC&A program— in which laboratory experts working directly together succeeded in demonstrating technology, building trust, and establishing a constituency to expand similar programs, eventually leading to new government-to-government agreements— US and Russian laboratories have begun a modest program to jointly develop and demonstrate transparency technologies. The first initiative in this effort was a demonstration of technology for remote monitoring— using video cameras and similar technologies to monitor material in storage without on-site inspectors. Equipment was hooked up to monitor HEU in storage at the Kurchatov Institute in Moscow and at Argonne National Laboratory-West in Idaho, and the images and data were uplinked via satellite. Additional remote monitoring efforts, including other facilities, are now being planned. This approach, based on technology developed to improve the cost effectiveness of IAEA safeguards, offers a potentially promising method of confirming that material is secure and not being tampered with or removed, without revealing classified information and without the substantial cost of permanent inspector presence.

A newer initiative involves US and Russian laboratories cooperating to identify and demonstrate technologies for verifying the actual dismantlement of nuclear warheads while minimizing any compromise of sensitive information. This is a modest-scale effort so far (slated to receive approximately $1 million in the current fiscal year) but holds enormous potential for the future.

The Bilateral Commission's reports review the status of these various projects and recommend a reinvigorated transparency effort (11, 12). Specifically, the reports suggest that the United States and Russia

1. resume negotiations with the objective of completing agreements during 1997 providing the legal basis for exchanges of classified nuclear information and the framework for implementing the agreed stockpile data exchanges;
   
2. accelerate work on developing measures to verify warhead dismantlement and to inspect plutonium and HEU components arising from dismantlement, with the goal of conducting initial demonstration experiments during 1997 and having a full bilateral regime in place during 1998;
   
3. develop a credible approach for international verification of material in classified forms by the end of 1997 and begin implementing it in 1998; and
   
4. make tens of tons of additional material eligible for international verification during 1997.

Unfortunately, however, despite the Presidential-level agreement on the Helsinki framework, as of mid-1997 few experts in either the US or Russian governments were optimistic about achieving rapid progress in these transparency initiatives.

23 If, for example, financial assistance were offered for actual dismantlement of nuclear weapons in return for some means of confirming that dismantlement (as envisioned in the original Nunn-Lugar legislation but never implemented), and reciprocal confirmation of US dismantlement were permitted, this could provide an attractive package.
24 This name suffers from two drawbacks: the apparent redundancy of "mutual" and "reciprocal" and the fact that the initiative actually covers only a few of the many areas where mutual inspections might occur in the future.
25 The most detailed unclassified description of this proposal is in (39). The proposal was quite sweeping, offering to open all fissile material sites to reciprocal visits except those containing intact warheads and naval fuel. It was not as sweeping as the regime recommended in the NAS report, however, as it did not include perimeter-portal monitoring of dismantlement facilities to verify the dismantlement of nuclear weapons and it did not include "nuclear archaeology" measures designed to compare the physical state of fissile material production facilities to the data exchanged in order to build confidence in the accuracy of that data and increase the difficulty of providing false data without detection on a scale large enough to be strategically significant (8).
26 At the facility where the material is blended, the United States will have regular access to the three pipes in the "Y" joint where the blending occurs: One carries 90% enriched uranium hexafluoride, one carries 1.5% enriched uranium hexafluoride used to blend down the HEU, and the third carries the merged blend, at approximately 4.4% enrichment. (Slightly enriched material rather than natural or depleted uranium is being used for the blending to further dilute undesirable isotopes in the HEU, such as U-234.) The United States will also make several visits each year to the facilities where HEU metal weapon components are cut into metal shavings and converted to oxide; during these visits, the United States has the opportunity to take rough measurements of the U-235 enrichment of the weapons components in containers and of the resulting metal shavings and oxide in containers, to tag and seal containers being readied for shipment to the blending facility, and to review records of these activities that take place when US inspectors are not present. Russian inspectors have a similar permanent presence at the US enrichment facility where the LEU is received and processed and similar frequent visits to the US fabrication facilities where the material is fabricated into reactor fuel (DOE officials, personal communications).
27 The NAS report concluded that adequate safeguards could be provided without compromising sensitive weapon-design information; it proposed permitting inspectors to confirm the total amount of plutonium in a large number of containers (thus revealing the average amount of plutonium per pit, which it proposed be declassified). A variety of other technical approaches to this task have also been considered. One potentially promising idea under active consideration is to develop "templates" of the radiation signatures of given types of weapon components. A detection machine could then be built that would tell international inspectors only whether the object inspected matched the template. The validity of the templates might be certified to the IAEA bilaterally by the United States and Russia (if an agreement to exchange necessary classified information between these two states can be reached).