4
Protection, Control, and Accountability of Direct-Use Material

THE PROLIFERATION RISK FROM LEAKAGE OF DIRECT-USE MATERIAL

Acquiring direct-use material—separated plutonium or unirradiated highly enriched uranium (HEU)—is a principal technical barrier for any nation or group seeking to develop nuclear weapons.1 Several kilograms of plutonium or several times that amount of HEU are sufficient to make a nuclear weapon, with the quantity depending on the composition of the material, type of weapon, and sophistication of the design.2 Estimates put the current inventory of direct-use material in the former Soviet Union (FSU) at about 200 tons of plutonium and about 1,200 tons of HEU, much of which is not incorporated into nuclear weapons.3 Almost all of this material is in Russia. Other FSU states have much smaller

1  

Many other commodities and technologies are also required to construct a weapon, but most of these items probably can be more readily obtained than direct-use material.

2  

For example, the exact amount of HEU depends on its level of enrichment. Much more HEU would be needed for a device if the material were only enriched to the level of 20 percent rather than 90 percent. Nonetheless, it is possible to fabricate a nuclear explosive device with 20 percent material. With natural uranium (0.7 percent) or with common reactor fuel (3 to 4 percent), a complex process would be required to reach an enrichment level for use in a weapon. The United States requires the highest level of security protection at buildings classified as Category I, that is, where 2 or more kilograms of plutonium or 5 or more kilograms of any type of HEU are located.

3  

General Accounting Office, "Nuclear Proliferation: Status of U.S. Efforts to Improve Nuclear Material Controls in Newly Independent States," GAO/NSLAD/RCED-96-89 (Washington, D.C.: General Accounting Office, March 1996), p. 3. Estimates vary on the amount and enrichment level of HEU in the FSU.

Throughout this report all references to tons are to metric tons. One metric ton is 2,205 pounds.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 52
--> 4 Protection, Control, and Accountability of Direct-Use Material THE PROLIFERATION RISK FROM LEAKAGE OF DIRECT-USE MATERIAL Acquiring direct-use material—separated plutonium or unirradiated highly enriched uranium (HEU)—is a principal technical barrier for any nation or group seeking to develop nuclear weapons.1 Several kilograms of plutonium or several times that amount of HEU are sufficient to make a nuclear weapon, with the quantity depending on the composition of the material, type of weapon, and sophistication of the design.2 Estimates put the current inventory of direct-use material in the former Soviet Union (FSU) at about 200 tons of plutonium and about 1,200 tons of HEU, much of which is not incorporated into nuclear weapons.3 Almost all of this material is in Russia. Other FSU states have much smaller 1   Many other commodities and technologies are also required to construct a weapon, but most of these items probably can be more readily obtained than direct-use material. 2   For example, the exact amount of HEU depends on its level of enrichment. Much more HEU would be needed for a device if the material were only enriched to the level of 20 percent rather than 90 percent. Nonetheless, it is possible to fabricate a nuclear explosive device with 20 percent material. With natural uranium (0.7 percent) or with common reactor fuel (3 to 4 percent), a complex process would be required to reach an enrichment level for use in a weapon. The United States requires the highest level of security protection at buildings classified as Category I, that is, where 2 or more kilograms of plutonium or 5 or more kilograms of any type of HEU are located. 3   General Accounting Office, "Nuclear Proliferation: Status of U.S. Efforts to Improve Nuclear Material Controls in Newly Independent States," GAO/NSLAD/RCED-96-89 (Washington, D.C.: General Accounting Office, March 1996), p. 3. Estimates vary on the amount and enrichment level of HEU in the FSU. Throughout this report all references to tons are to metric tons. One metric ton is 2,205 pounds.

OCR for page 52
--> stocks, totaling less than one-half ton, but these quantities are still significant.4 Since the breakup of the Soviet Union, there is growing concern in the West that this material is increasingly vulnerable to theft or diversion. There is an urgent need to improve controls over direct-use material in the countries of the FSU. Acquiring substantial quantities of either HEU or plutonium could greatly simplify the efforts of a nation to obtain or augment a nuclear weapons capability. Kilogram quantities of HEU could be used by groups with relatively limited technical capability to construct a crude but effective nuclear device. And terrorists could disperse into the environment modest amounts of plutonium or other radioactive materials, which could cause substantial damage and societal disruption. In Russia direct-use material is found in many forms at a variety of military and civilian facilities5 that fall under the jurisdiction of several ministries, agencies, and institutes. For its purposes, the U.S. Department of Energy (DOE) considers that the material is distributed among five "sectors":6 Nuclear weapons, which are largely under the custody of the Ministry of Defense (MOD). These weapons, which are not included in the scope of this study, are currently deployed or stored at fewer than 100 sites, down from over 500 in the late 1980s.7 Material in the Ministry of Atomic Energy (MINATOM) defense complex, such as the weapons design institutes at Arzamas-16 and Chelyabinsk-70. Like DOE in the United States, MINATOM is responsible for production, assembly, and disassembly of nuclear warheads. An estimated 2,000+ warheads are being dismantled each year as a result of U.S.-Russian arms reduction agreements. This sector thus has large amounts of direct-use material and its inventories are growing. One recent study estimates that 15 tons of plutonium and 45 4   Estimate based on committee discussions with officials at selected FSU institutes. Latvia, Uzbekistan, and Georgia also have small quantities of HEU. U.S. bilateral programs with these countries are beyond the scope of this report. 5   In this report the term "facility" is used to denote a collection of buildings and/or structures that serve a common purpose. A facility may contain more than one building, and in some cases two or more facilities may be grouped at one site, such as Tomsk-7, which has at least six. 6   "Unified US-Russian Plan for Cooperation on Nuclear Materials Protection, Control, and Accounting (MPC&A) Between the Department of Energy Laboratories and the Institutes and Enterprises of the Ministry of Atomic Energy (MINATOM) Nuclear Defense Complex." Department of Energy, September 1, 1995, pp. 7-8. 7   John Deutch, Director of Central Intelligence, "The Threat of Nuclear Diversion," testimony to the Permanent Subcommittee on Investigations of the Senate Committee on Government Affairs, March 20, 1996, p. 8. The Soviet Union had withdrawn its tactical nuclear weapons from Eastern Europe by 1991, and, as a result of agreements reached in 1994, Russia has become the heir to all nuclear weapons on the territory of the FSU. All nuclear warheads were transferred to Russia from Kazakstan in 1995 and from Ukraine and Belarus in 1996.

OCR for page 52
--> tons of HEU are now being transferred annually from MOD to MINATOM custody.8 Material in the MINATOM civilian institutions, such as the Institute of Physics and Power Engineering in Obninsk, the plutonium reprocessing facility at Mayak, and the Luch Scientific Production Association in Podolsk.9 Many of these facilities are devoted to research and development on nuclear power reactors, along with producing power reactor fuel and other nuclear materials for civilian applications. Although the amounts of direct-use material in this sector are smaller than in the defense sector, the quantities are still significant as a proliferation risk. Material at civilian research facilities outside MINATOM, such as the Kurchatov Institute of Atomic Energy, the Dubna Joint Institute of Nuclear Research, and the Moscow Engineering Physics Institute. In most cases the amounts of material at these facilities are relatively small, but Kurchatov has several tons. Material for marine propulsion applications in submarines, surface ships, and civilian icebreakers. The facilities have stocks of HEU at various levels of enrichment. Outside Russia, all known stocks of direct-use material are in civilian facilities. Thus, in the four categories outside the Russian MOD, direct-use materials are stored at an estimated 80 to 100 facilities in the FSU.10 Both HEU and plutonium represent serious proliferation risks. HEU is of particular concern because, unlike plutonium, it can be used in a simple gun-type device.11 HEU can be blended down into low-enriched uranium (LEU) and used as fuel in nuclear power reactors.12 The United States has adopted this HEU-to-LEU approach, agreeing in 1992 to buy 500 tons of HEU from dismantled Russian weapons which is being converted to LEU. During the Soviet Union era, the security over all direct-use material was not in question, reflecting the formidable police power of the state and the loyalty of 8   G. Allison, O. Cote, R. Falkenrath, and S. Miller, Avoiding Nuclear Anarchy: Containing the Threat of Loose Russian Nuclear Weapons and Fissile Materials (Cambridge. Mass.: The MIT Press, 1996), p. 21. 9   Unlike the United States, which now maintains a strict separation between military and commercial nuclear activities, in the Soviet Union the same facilities sometimes performed both kinds of work. Commercial nuclear research thus also takes place in facilities that are usually counted as part of the MINATOM defense sector. 10   "Report on Control and Accountability of Materials Related to Weapons of Mass Destruction in the Former Soviet Union," Department of Defense, June 1, 1995, p. 2. The estimate of 80 to 100 facilities is used for both Russia and the entire FSU in different reports. 11   J. Carson Mark, "Explosive Properties of Reactor Grade Plutonium." Science and Global Security, vol. 4, 1993, pp. 111-128. 12   In contrast, converting plutonium to a form that is unsuitable for weapons use is an economically less attractive and technically more challenging task.

OCR for page 52
--> the managers, soldiers, and workers in the Soviet nuclear complex. The security system focused on controlling people—relying on "closed borders, closed cities, a controlled society, and extensive surveillance of personnel by the KGB."13 The very existence of many closed cities was considered a state secret, and they were surrounded by perimeter fences and numerous guard posts. "The physical protection programs [relied] more on manpower than on technical systems. The security system ultimately depended on a responsible, competent, and well-disciplined establishment, and well treated and loyal personnel."14 The civilian portion of the Soviet nuclear complex was also subject to strict security, although not as exacting, and again it focused heavily on controlling personnel. The Soviets maintained accounting systems for material in MOD and MINATOM weapons inventories to complement the system of personnel security and physical protection, but the accounting efforts apparently did not extend beyond the maintenance of paper records that were not always complete or easily accessible. There is also anecdotal evidence that some facility managers maintained stocks of material off the books to ensure that quotas for producing material would be met. In addition, as in the nuclear complex in the United States, the nature of the process of producing and handling direct-use materials results in uncertainty about the exact quantities of material that a facility actually possesses. Measurements are particularly difficult when material is in process, held up in pipes and vessels, or otherwise inaccessible.15 According to DOE, there is no central inventory of all direct-use material in Russia.16 Although individual facilities do have inventory data, much of the information is reported to be incomplete or inadequate. Inventories of shipments of materials between facilities may also be complicated by differences in measuring instruments and sampling procedures among facilities. Not surprisingly, the best data are available on high-quality materials, while information on scrap, residues, or materials in process is generally poor. The U.S. intelligence community believes that "the Russians may not know where all their material is located."17 13   John P. Holdren, Chairman, Panel on U.S.-FSU Cooperation to Protect. Control, and Account for Weapons-Usable Nuclear Materials, testimony to a joint hearing of the Permanent Subcommittee on Investigations, Senate Committee on Governmental Affairs, and Subcommittee on Europe. Senate Foreign Relations Committee, August 23, 1995. 14   David Osias, National Intelligence Officer for Strategic Programs, "Security of Nuclear Weapons and Weapons-Usable Material in FSU," testimony to the Senate Foreign Relations Committee. August 22, 1995. 15   For example, in February 1996 DOE released estimates of cumulative inventory differences in the United States of 2.8 metric tons of plutonium over a 50-year period (U.S. Department of Energy. Plutonium: The First 50 Years, p. 52). 16   DOE responses to the committee's questions, October 23, 1995. 17   John Deutch, Director of Central Intelligence, "The Threat of Nuclear Diversion," testimony to the Permanent Subcommittee on Investigations of the Senate Committee on Government Affairs. March 20, 1996.

OCR for page 52
--> Upon the breakup of the Soviet Union, the non-Russian states started essentially from scratch to create modern systems to protect their newly inherited nuclear materials. In some cases, senior officials of new ministries initially were not aware of the existence or amount of direct-use material in their keeping. In addition, and particularly in the case of Kazakstan, much of the experience in MPC&A was tied to Soviet requirements generated in Moscow. Responsibility for implementing MPC&A systems was largely in the hands of specialists from Russia, some of whom returned to Russia in the early 1990s. While some elements of the old physical protection systems (e.g., fences, alarms), even if inadequate, could be transferred to the new owners and operators of facilities, control and accountability systems at both the facility and the national levels had to be constructed. More generally, the future of the FSU nuclear research and production complex is uncertain. The end of the Cold War and the prospect of significant nuclear arms reductions have reduced the roles for many facilities. MINATOM has instructed its defense-related and civilian facilities to become more self-supporting and find other sources of income beyond government funds. Some laboratories and storage buildings are deteriorating. There is little or no money for maintenance, purchase of equipment, or, sometimes, payment of salaries. Many workers are leaving to pursue other jobs. In such circumstances, the vulnerabilities of the security systems present temptations for disaffected or desperate workers and reduce the prospects that thefts will be detected. In March 1996, John Deutch, director of the Central Intelligence Agency, testified that a comprehensive examination revealed that none of these (non-MOD) facilities in Russia or other newly independent states had adequate safeguards or security measures by international standards for weapons-usable material.... The chilling reality is that nuclear materials and technologies are more accessible now than at any other time in history—due primarily to the dissolution of the former Soviet Union and the region's worsening economic conditions.18 By most accounts, the nuclear weapons in the MOD system remain under good control, although concern for their security is growing among western governments. The farther one moves from intact warheads, in MOD and especially in MINATOM and in the rest of the FSU, however, the more vulnerable the material appears to be.19 18   Ibid. 19   Ibid. As noted, the MOD complex is outside the scope of this study. Deutch testified that: "We believe the likelihood of the loss of a nuclear weapon is still slight today. But the threat from within the Russian military and a deteriorating economy mean that this judgment could change rapidly."

OCR for page 52
--> To date, despite hundreds of reports, there are only a few known cases of thefts and/or illicit exports of direct-use material.20 The cases all involved quantities of material far smaller than what would be required to make a nuclear weapon. And none of the thefts was from a nuclear weapons storage facility. Nonetheless, after initial reluctance to acknowledge the shortcomings in MPC&A, Russian officials have now called for improvements in the systems to protect direct-use and other types of nuclear materials.21 COMPONENTS OF MPC&A MPC&A should be one of the core elements of a national system of safeguards and security that the international community expects all countries possessing nuclear materials to establish. Such systems are designed to protect the material against theft or diversion and to detect such events if they occur. Briefly, Physical protection systems are "designed to detect any unauthorized penetration of barriers and portals, and to respond with immediate investigation and use of force if necessary."22 The systems should delay perpetrators long enough for guards and, if necessary, additional forces to respond. Physical protection measures are generally the most visible and pervasive components of a safeguards systems. Fences, multiple barriers to entry, limited access points, alarms, and motion detectors are all examples of elements of a physical protection system in addition to guards. Material control systems are designed to "prevent unauthorized movement of special nuclear materials and to detect promptly the theft or diversion of the material should it occur."23 These systems may include portal monitors and other devices to control egress from storage sites; authorized flow paths, storage locations, and secure containers for material; and seals and identification codes that make it possible to readily verify the location and condition of material. Good material control may also assist with physical protection. Material accountability systems are designed "to ensure that all material of interest is accounted for, or to measure the loss of any, and to provide information for follow-up investigation, within error limits imposed by the process and 20   In a meeting with the committee during its visit to Moscow in May 1996, a senior MINATOM official said that there had been 23 cases of thefts and attempted thefts of nuclear-related materials from MINATOM facilities during 1993 and 1994 but that only three were successful. According to the same official, there had been no thefts or attempted thefts in 1995 and 1996. 21   No theft of direct-use material is known to have occurred in 1995 or 1996. but it is not known if this is the result of improved security, more skillful thefts, or reluctance of officials to disclose thefts. 22   National Research Council, Material Control and Accounting in the Department of Energy's Nuclear Fuel Complex (Washington, D.C.: National Academy Press, 1989), p. 38. 23   Ibid., p. 41.

OCR for page 52
--> by instruments."24 These include both traditional inventory systems and an array of equipment to measure the types and quantities of material in a given area. Integral to all three of the above systems is personnel reliability, which includes "security screening, indoctrination, training, and some personnel records functions. . . ."25 It may also include both procedures ensuring that no single worker is left alone in sensitive areas (the two-man rule) and limits on access to certain facilities, which can be part of material control procedures as well. The design of an MPC&A system rests on fundamental principles of "graded safeguards" and "defense in depth" against a spectrum of threats. The principle of graded safeguards reflects the belief that the effort and resources devoted to improving MPC&A must be commensurate with the particular risks to material at any given facility. Defense in depth incorporates redundant and diverse layers of defense to increase the difficulty of penetration and to guarantee that the failure of any single layer will not result in a major loss.26 In accordance with these principles, MPC&A systems should be designed to protect against a range of threats. The threats may be external—for example, break-ins or attacks by individuals or groups such as terrorists, or internal, such as thefts by one or more employees with access to the material. Insiders might also work with an outside group so that a facility could be subject to a combined internal and external threat. Table 1.4 in the Executive Summary outlines the basic features of an MPC&A system. During the time of the Soviet Union, external threats were the primary concern because officials were confident of their control over personnel.27 During the committee's visits to Russia, Ukraine, Kazakstan, and Belarus, a number of officials and facility managers remarked that they still see the largest threat as coming from outsiders, either from terrorists intent on theft or sabotage or, in the worst case, from large-scale civil unrest. Western experts and many in the FSU, however, consider the internal threat from employees who believe there is a market for nuclear material to be underestimated. The known thefts and attempted thefts to date have been carried out by insiders acting alone or in small groups. An effective MPC&A system must be able to cope with a range of threats, with each potential threat triggering an appropriate level of protection. 24   Ibid., p. 42. 25   Ibid., p. 38. 26   DOE answers to committee questions, October 23, 1995. 27   "Set up under Soviet rule for a strictly regimented closed society worried only about external threats, the security often amounts to little more than barbed wire fences and armed guards, providing scant protection against insiders and their accomplices who hope to get rich by smuggling out nuclear materials for sale on the black market . . ." (testimony of Lawrence Gershwin, Central Intelligence Agency, before the House Committee on Appropriations, DOD Appropriations for 1993. Part 5, May 6, 1992, p. 498).

OCR for page 52
--> Identification of relevant and appropriate threat scenarios is the first step in the creation of an effective system, particularly its physical protection components. The scenarios provide the "Design Basis Threat" for assessing, correcting, and monitoring the vulnerabilities of a facility. A critical aspect of this process is ensuring the security of information about vulnerabilities. Such details, which are considered classified information by governments, would provide an "instruction book" for terrorists. Control and accounting systems for direct-use material tend to be more generic and less affected by specific threat scenarios, although some material control equipment may also serve specific physical protection roles and the systems certainly contribute to the overall physical protection of the material. All nations in possession of nuclear materials have some form of national MPC&A system, although national programs and standards vary significantly. The parts of the system related to material control and accountability are also subject to international standards as part of a country's nonproliferation obligations. But these "safeguards" are intended to thwart undisclosed diversion by the state itself rather than prevent theft and to provide "timely warning" that such diversion has occurred. The standards provide a baseline against which to measure national material control and accountability systems; and FSU governments seem to find it politically desirable to accept and work toward internationally accepted norms. INTERNATIONAL CONTEXT FOR MPC&A SYSTEMS National MPC&A programs take place in the context of a number of international treaties and agreements designed to control the proliferation risk posed by direct-use materials while still permitting peaceful nuclear activities. The nuclear NonProliferation Treaty (NPT) divides nations into three groups: five nuclear weapons states (NWSs—China, France, Great Britain, Russia, and the United States), the remaining nonnuclear weapons states (NNWSs), and non-signatories.28 NNWSs that are parties to the NPT are required to accept full-scope safeguards by the International Atomic Energy Agency (IAEA). "Full-scope" safeguards apply to "all nuclear materials in all peaceful nuclear activities within their territory or under their control." They consist of "a system of procedures involving material control and accountancy, containment and surveillance, and verification (including on-site inspections at declared facilities) that are implemented through agreements between the IAEA and individual countries." NNWSs that are not parties to the NPT may also enter into safeguards agreements with the 28   Three other countries, India, Pakistan, and Israel, are generally considered to have acquired nuclear weapons capability, but none has acknowledged this publicly. None of the three is a party to the NPT. A fourth, South Africa, acquired a few nuclear weapons but abandoned its program and destroyed its stockpile in the early 1990s. It became a party to the NPT in 1991.

OCR for page 52
--> IAEA, and most have done so for at least some of their facilities, largely as a result of pressure by nuclear suppliers.29 By contrast, the NWSs are not required to accept IAEA safeguards, although all five have signed "voluntary offer" agreements under which they have agreed to subject some of their facilities to such safeguards. Currently, no facilities in Russia are under IAEA safeguards, although there are continuing discussions of possible monitoring for portions of the disposition of the excess fissile materials resulting from arms reduction agreements. In addition, Soviet/Russian specialists have worked for the IAEA and thus have acquired knowledge about and experience with the international system. The other successor states with nuclear facilities and material, as NNWS parties to the NPT, are in various stages of developing safeguards agreements with the IAEA and implementing the associated accountability requirements to cover their entire nuclear programs, which are limited to peaceful activities. They also have cooperative programs with the United States and several other governments to assist in this process. Kazakstan, for example, is working closely with the IAEA in creating its MPC&A system, in part because it wants to establish its credentials as a responsible nuclear trading partner so that it can continue to export uranium and beryllium. Unlike material control and accountability systems that are required for NNWSs by international agreements and that are aimed at detecting diversions of nuclear material by governments, the responsibility for physical protection of nuclear materials "rests entirely with the government of the State."30 Recognizing, however, that "it is not a matter of indifference to other States whether and to what extent that responsibility [for national physical protection] is being fulfilled,"31 the IAEA periodically issues updated guidelines for countries to follow in implementing their internal MPC&A programs. But the guidelines are purely advisory. The only relevant international accord concerning the physical protection of nuclear materials against theft—the 1980 Convention on the Physical Protection of Nuclear Material—covers nuclear material for peaceful purposes while in international transport. However, it has neither verification nor enforcement provisions.32 Thus, efforts to improve physical protection do not have a strong international standard against which to measure goals or progress. The United States and other industrialized states have detailed regulatory provisions establishing standards for MPC&A. These, together with the IAEA's advisory 29   Office of Technology Assessment, Nuclear Safeguards and the International Atomic Energy Agency (Washington, D.C.: U.S. Government Printing Office, 1995), p. 27. 30   Hans Blix, "Preface," in The Physical Protection of Nuclear Material, INFCIRC/225/Rev. 3, (IAEA, Vienna, December 1989). 31   Ibid. 32   U.S. Arms Control and Disarmament Agency, Arms Control and Disarmament Agreements: Texts and Histories of Negotiations (Washington, D.C.: U.S. ACDA. 1990), pp. 301-313.

OCR for page 52
--> guidelines, form a de facto benchmark for assessing the adequacy of controls in the Soviet successor states. Today, Russia, Ukraine, Belarus, and Kazakstan are undertaking efforts to improve protection of direct-use material at their principal facilities. But the task is enormous, and none of the countries has adequate funds or personnel to do the task on its own. Given the proliferation risks these materials represent, western and other governments have found it in their own national interests to offer assistance. The United States has undertaken a significant program of cooperation with Russia, as well as with Belarus, Kazakstan, and Ukraine, to bring the level of security for their direct-use materials up to international standards. The IAEA and the G-7 governments are also engaged in the effort, although on a much smaller scale. SCOPE AND OBJECTIVES OF U.S. COLLABORATION WITH RUSSIA, UKRAINE, BELARUS, AND KAZAKSTAN Russia U.S.-Russian discussions on MPC&A cooperation began in late 1991, shortly after the U.S. Congress passed the Nunn-Lugar legislation. The primary focus of the Nunn-Lugar legislation and the activities of the U.S. Department of Defense (DOD) with Russia during the first years of cooperation was dismantlement, transportation, storage, and safeguarding of weapons themselves. But DOD recognized the need for MPC&A upgrades as well and in 1993 finally signed an agreement with MINATOM on developing national MPC&A systems and improving controls over civilian nuclear material. In 1994, prompted by its success in scientific collaborations with Russian institutes in other areas, DOE initiated additional efforts to expand collaboration to MPC&A. The principal U.S. effort in Russia now consists of two complementary programs administered by DOE: government-to-government, which was funded by DOD under the Cooperative Threat Reduction (CTR) program through fiscal year 1995 and thereafter by DOE; and lab-to-lab, initiated by DOE in 1994 and funded with DOE resources from the outset. Under a third related program on regulatory support, the U.S. Nuclear Regulatory Commission and DOE are assisting Russia in developing and implementing a stronger national regulatory structure for MPC&A. Coordination of U.S. involvement in these programs is provided by an interagency group, established in 1994 and headed by the National Security Council. The government-to-government program in Russia is based on the 1993 U.S.-Russian agreement and several subsequent implementing agreements and amendments. Those agreements identify the facilities that will participate and establish the roles and responsibilities of the participating organizations. But the agreements did not immediately dispel fears and suspicions over

OCR for page 52
--> motivations that had built up over many decades. While desiring collaboration, MINATOM nonetheless was reluctant to provide information and access to facilities that the United States considered critical for cooperation.33 Other factors delaying implementation were the needs to establish a complicated array of bilateral agreements that could serve as the formal frameworks for activities in Russia, to adjust organizational responsibilities and interagency procedures in the U.S. Government, and to adapt DOD financial regulations and procurement procedures to these unique bilateral programs. "Buy America" requirements, and the attendant lengthy procurement process, also delayed the program. These conditions were to be met at a time when political turmoil abounded in Russia, with repercussions in Washington—problems that also would arise in the other successor states of concern. During the first few years, the effort was also complicated by the fact that, although the program was carried out by DOE, it was funded and managed by DOD. Progress was slow at the beginning, and there were few concrete results for the first two years. In part to circumvent the difficulties of implementation in the government-to-government program, and also to take advantage of the potential to build trust more readily through direct contacts among scientists who share common knowledge of and appreciation for nuclear security issues, DOE's national laboratories and their counterpart Russian institutes initiated the lab-to-lab approach for MPC&A collaboration. Contacts between scientists from U.S. and Soviet weapons laboratories had begun in the waning days of the Cold War. Those contacts increased in 1992 and 1993 with reciprocal visits and discussions of possibilities for pursuing direct collaboration. In April 1994, DOE approved a proposal from Los Alamos National Laboratory to extend the successful U.S.-Russian scientific collaborations to include joint work on MPC&A. Los Alamos and Arzamas-16 signed the first umbrella contract laying out the administrative, financial, and legal arrangements necessary to expedite the implementation of subsequent specific program-oriented contracts. Five other U.S. laboratories (Sandia National Laboratories, Lawrence Livermore National Laboratory, Oak Ridge National Laboratory, Pacific Northwest National Laboratory, and Brookhaven National Laboratory) soon joined in the lab-to-lab efforts. Each activity is conducted with DOE approval and in coordination with the U.S. Department of State, but the lab-to-lab program has offered more flexibility and has been spared some of the administrative hurdles to implementation that the government-to-government program encountered. Today, the two programs are proceeding on parallel tracks and are in many respects indistinguishable. 33   In a separate but related development, Presidents Yeltsin and Clinton agreed at their September 1994 summit to an exchange of information about aggregate stockpiles of nuclear weapons, stocks of fissile materials, and their safety and security. The negotiations to implement that exchange have not yet been completed. Certainly, the information provided by such an exchange would be a useful confidence-building measure for efforts to improve MPC&A in Russia and should also provide information relevant to a national inventory of direct-use material.

OCR for page 52
--> unique position to help ensure that such upgrades are given high priority and installed in a prompt and effective manner. Recommendation: Continue to fund MPC&A efforts in the FSU at least at the level of fiscal year 1996 for several more years and be prepared to increase funding should particularly important high-impact opportunities arise. Continued funding will allow the U.S. and FSU governments to carry on cooperation at the technical level at important facilities. The present level of funding is satisfactory for the tasks recommended in this report, which can be achieved with a U.S. work force of about the same size as is currently engaged in the joint programs. In principle, additional funds could be used to increase the rate of implementation, but whether Russia and the other successor states will be able to absorb and use effectively the recent rapid increase in U.S. funding remains to be demonstrated. Another limitation on the size of the program in Russia, of course, is the readiness of MINATOM to expand efforts. If previously closed facilities unexpectedly become open for cooperative efforts, DOE should request additional funds. 2. Indigenize MPC&A Capabilities Finding: Once U.S. funding ends, the cooperating governments must be committed and able to assume full responsibility for funding and maintaining upgraded MPC&A systems. The challenge is great, as economic shortfalls even for basic program support limit the domestic funds available for MPC&A. Many nuclear facilities are in poor or deteriorating condition, and the economic situation in the FSU makes any dramatic improvement unlikely in the near term. Shortages of funding for facility maintenance and basic program support make it difficult to give high priority to internal funds for MPC&A. In some cases, funding is not available even for installation and maintenance of equipment provided by others. Ministries, institutes, and individuals will need access to income streams that will permit them to continue their efforts in the long term. DOE's approach to MPC&A projects has appropriately stressed their collaborative nature and the mutual benefits from increased security. Although keeping direct-use material out of countries of proliferation concern and terrorist groups is vital to U.S. national security, MPC&A is clearly a national responsibility; and U.S. assistance should develop and strengthen indigenous MPC&A capabilities. Officials and specialists in the four successor countries will understandably undertake "cooperative activities" on sensitive issues such as MPC&A much more readily than they will accept "assistance." It becomes even more important to define the program in terms of cooperative activities, looking to the day when U.S. funding scales down and the FSU governments and facility managers are expected to assume responsibility for MPC&A.

OCR for page 52
--> Recommendation: Continue to emphasize the importance of MPC&A as a nonproliferation imperative at the highest political levels in the FSU. Many officials in the region do not attach great importance to proliferation as an international security threat. Steps in MPC&A may sometimes be taken more to satisfy the U.S. and other western governments than out of serious concern over proliferation. Senior U.S. officials should continue to emphasize in interactions with their FSU counterparts the commonality of U.S. and FSU interests in the nonproliferation sphere. In addition, to help ensure that the United States is not alone in pressing the countries on the importance of nonproliferation, the United States must maintain diplomatic priority and appropriate support for the IAEA and other international approaches, particularly initiatives of the G-7 governments. Recommendation: Prior to initiating MPC&A projects at specific facilities, obtain assurances at both the ministry and the institute levels that the upgrade programs will be sustained after improvements have been made. Financial incentives, such as support for related research activities, should be considered as a means to stimulate long-term commitments. The U.S. program does not require institutes or governments to provide any assurance that they will sustain MPC&A upgrades after U.S. funding ends. In some facilities MPC&A equipment and technology might be used for other purposes after the cooperative efforts are completed. At the Kiev Institute for Nuclear Research, for example, the main interest of the staff is to restart the reactor, which they hope to do in 1997. But the dire financial situation of the institute suggests that computers and some other equipment provided by the United States for MPC&A could be diverted to support the reactor research program. At the Institute of Nuclear Power Engineering in Belarus, physical protection upgrades have been completed, but there is no assurance that the government or institute will be able to maintain the newly installed system. Consequently, efforts should be made to assure that both ministries and facilities have the incentives to implement a sustained MPC&A program. Recommendation: Involve institute personnel to the fullest extent possible in determining how to use available funds for upgrades. The U.S. program has sought to include working-level personnel in the FSU institutes in decisions on equipment needs, specifications, and installation. However, the strict decision-making hierarchy and centralization of authority in most institutes can make this difficult. Developing ways to maximize the involvement of working-level personnel in key decisions on MPC&A upgrades should encourage them to assume responsibility for the upgrades. Recommendation: Give greater emphasis to near-term training of local specialists. Developing indigenous capabilities in all aspects of MPC&A sys-

OCR for page 52
--> tems is a key to having the FSU governments assume full responsibility for their installation and maintenance. The U.S. program should continue supporting the establishment of an MPC&A training facility at the Methodological Training Center in Obninsk, which will be managed by Russian specialists and provide an important means to "train the trainers" for future maintenance of the upgraded MPC&A systems. The European Union, through Euratom, is also contributing to the center, where training courses began in October 1995 and are gradually expanding. Effective courses were held in the spring and summer of 1996 on the overall design of MPC&A systems and on specific problems of software engineering for the design of control and accountability systems. As part of the effort to create a "safeguards culture" as well as to increase technical skills, coursework included studies of the international problems of proliferation and international cooperation to prevent it. The United States should also incorporate appropriate training programs at an early stage of bilateral activities at each facility where joint MPC&A programs are under way so as to ensure rapid transition from assistance to cooperation at these facilities. Recommendation: Reward those institutes that are making good progress in upgrading MPC&A systems by giving them preference for participation in other U.S.-financed cooperative programs. DOE and other U.S. government agencies have an array of cooperative programs under way in the FSU in addition to the work on MPC&A. DOE's Initiatives for Proliferation Prevention (formerly the Industrial Partnering Program), for example, uses U.S. laboratories as intermediaries to facilitate U.S. investments at FSU institutes. This and other cooperative programs involve work with the same facilities where DOE is helping to upgrade MPC&A. It should be a relatively easy task to give priority for such cooperation to institutes whose MPC&A performance is particularly strong. Recommendation: Encourage the establishment of new income streams that can provide adequate financial support for MPC&A programs in the long term, such as earmarking for MPC&A programs a portion of the revenues from Russian sales of HEU. Developing sustainable domestic funding sources in the FSU for any activity presents a significant challenge in the current economic climate of the region. The most feasible approach to ensure funding for MPC&A in the near term may be to use revenues that are becoming available to Russia or the other FSU countries from closely related activities, particularly other U.S. or western programs to reduce the risk posed by direct-use materials. The proceeds from the U.S. purchase of 500 tons of HEU from dismantled Russian nuclear weapons is an example of a source that might be tapped. Recommendation: Rely increasingly on domestically produced and locally available equipment for physical protection, detection, analysis, and

OCR for page 52
--> related MPC&A tasks. The purchase and use of FSU equipment and technology can encourage local commitments to the program; stimulate a supply industry; and reduce acquisition, transportation, and maintenance costs. One example of successful use of Russian technology is the reliance on equipment produced by Eleron, a MINATOM enterprise that specializes in R&D and production of physical protection equipment. Eleron provides important specialized services, operation and maintenance, and training for security personnel. This practice should be expanded to the extent that appropriate equipment is available. Officials in Ukraine expressed concern to the committee that they have not been consulted on equipment purchases and that equipment comparable to that purchased from U.S. firms could be bought at a lower cost in Ukraine or Europe (saving transportation costs). Because the Ukrainians will have to operate as well as maintain the equipment after the U.S. program ends, using locally available equipment whenever possible is clearly a wise choice. 3. Simplify the Problem Finding: The amount of material and its dispersion in many buildings at many facilities increase the cost and complexity of MPC&A upgrades, as well as the risk of diversion. There are a number of cases in which upgrades are under way for selected buildings or for caches of material, while comparably important buildings or stocks of material at the same facility are not being addressed. The challenge of controlling small amounts of direct-use material located in hundreds of buildings, including many in a poor state of repair, seems overwhelming. If the amount of material and/or the number of storage areas could be substantially reduced, the time and costs involved in installing and maintaining MPC&A systems also could be significantly decreased. Some initiatives in this direction will require strong efforts. The possession of direct-use material is viewed as essential to participation in some of the most cherished aspects of the FSU nuclear program. Thus, if the material is removed, the raison d'etre of the facility may suffer as a consequence. Holding direct-use material may also confer status on a facility or a laboratory director that would not otherwise be available. And if the material is removed, the flow of U.S. funds for MPC&A also could cease. Nonetheless, the security benefits of consolidation are significant enough that efforts to overcome these obstacles are a priority, particularly at facilities where there may be redundant stocks of especially proliferation-sensitive material. One possible approach would be to allow an institute to maintain ownership of material that is stored elsewhere. In any case, the proliferation benefits of consolidation will not be achieved if consolidation will deprive the institutes of funds they would otherwise expect to receive. Creative use of some of the incentives recommended in this chapter may be necessary to elicit support for consolidation. Related to the specific consolidation measures recommended below but out-

OCR for page 52
--> side the scope of this study is the issue of continued production of plutonium in Russia. The United States is trying to encourage an end to plutonium production at the reactors in Krasnoyarsk and Tomsk. Also, closing the nuclear fuel rod reprocessing plant at the Mayak complex, where almost 30 tons of separated reactor-grade plutonium is already stored and more is being generated, has been strongly advocated by some U.S. specialists. Future use of the small stocks of direct-use material outside Russia is uncertain at best. Retaining these stocks even in a limited number of locations requires significant MPC&A expenditures and continued vigilance against the possibility of theft. Recommendation: In Russia, encourage consolidation of direct-use material in fewer buildings, at fewer facilities, and at fewer sites. Where economically feasible, consolidating the material could significantly simplify the task facing the cooperative program at a number of sites. Russian officials and facility managers recognize the problems posed by widespread stocks of direct-use material, and DOE has encouraged recent consolidations at a number of facilities. At Obninsk, for example, plans are in place to consolidate the current 30 material balance areas, all of which need improved MPC&A.46 The original plan was to consolidate the material to two major "nuclear islands," one for the critical assemblies and one for the institute's other activities. That approach has proved more costly than was anticipated. Obninsk officials now estimate that the task will take three to four years to complete and that they may have to settle for some smaller islands because of the diversified nature of the research activities. DOE states that it will encourage consolidation when it can but generally regards this approach as taking advantage of targets of opportunity rather than as a priority at the present time. Recommendation: Take steps to encourage the removal of all HEU at research facilities outside Russia, including the purchase of HEU when appropriate. One of the largest amounts of non-Russian direct-use material is the HEU at the Kharkiv Institute in Ukraine, where approximately 70 kilograms is stored. Purchase of this material should be considered. Some of the proceeds of the sale might be earmarked for MPC&A upgrades at the facility. The price would be a matter of negotiation, but there are precedents in both the ongoing U.S. purchase of 500 tons of HEU and the earlier purchase of almost 600 kilo- 46   Material balance areas are separate parts of a facility in whose boundaries reliable inventories of nuclear materials can be established and material flows in or out can be monitored (Office of Technology Assessment, Nuclear Safeguards and the International Atomic Energy Agency, U.S. GPO, Washington, D.C., 1995, p. 114).

OCR for page 52
--> grams of HEU from Kazakstan in Project Sapphire.47 Similar purchases should be considered for HEU at other FSU facilities outside Russia not encompassed by other recommendations. Recommendation: For research reactors outside Russia where important and adequately financed research programs are planned in the foreseeable future, support conversion of the reactors so that they can use LEU instead of HEU. U.S. policy has long favored conversion of research reactors using HEU to reactors using LEU as a nonproliferation measure. The United States has previously provided direct support to a number of countries for such conversions. Conversion assistance would include replacement of existing stocks of HEU fuel with LEU fuel and assistance to modify the reactors. The United States should not become involved in indefinitely subsidizing the operation of these reactors, but for several facilities conversion assistance offers the best near-term prospect for returning to active research while also reducing proliferation problems. 4. Minimize the Opportunities to Bypass MPC&A Systems Finding: If a national MPC&A program is to be effective, all relevant organizations and all sources of direct-use material must be addressed. Large stocks of direct-use material are located at some Russian facilities that have not yet become active participants in the bilateral program. Also, there is uncertainty among both Russian and American specialists as to the precise amounts of direct-use material present at many facilities, given the history of maintaining stocks of material "off the books" and the weaknesses in past inventories. As more stringent MPC&A systems are being installed at the facility, temptations to hold material outside these systems may arise. During a period of political and economic turmoil and expanded criminal activities, efforts to remove material from the MPC&A systems as a first step to subsequent diversions may emerge. At the same time, an important oversight agency in Russia, the State Nuclear Regulatory Committee (Gosatomnadzor or GAN), suffers from a shortage of well-trained inspectors, qualified staff, and necessary analytical and related equipment. Also, GAN's administrative authority in areas related to military activities is very uncertain. Recommendation: Ensure that all stocks of direct-use material are encompassed in the program, including icebreaker nuclear fuel, supplies at naval facilities, and off-specification and scrap material. Some institutions with responsibility for direct-use material have been more reluctant than others to 47   Avoiding Nuclear Anarchy, pp. 102-106.

OCR for page 52
--> open their facilities to U.S. specialists. For example, the U.S. collaborative program was not able to include fuel used by the Russian icebreaker fleet until 1996. In this case, both the Russian and the U.S. navies were concerned about the precedents that openness might create. In addition, scrap and off-specification materials from weapons production, some of which is direct-use, are not yet included in cooperative MPC&A programs. Recommendation: Encourage rapid development of a comprehensive national material control and accounting system in Russia and the prompt incorporation of all existing direct-use material into that system. All relevant agencies in Russia, including GAN, have agreed on the concept of such a system. The initial activity would be managed by Atominform, a MINATOM institute, and individual facilities are undertaking their own inventories, using a standard approach. The project is expected to proceed slowly, however. DOE is providing financial support and equipment but should elevate the priority it gives to a national system. Recommendation: In Russia, increase support of GAN as an important independent agency by assisting it in developing MPC&A methodologies, training inspectors, obtaining staff support from research institutions, and procuring necessary equipment for MPC&A inspections. The involvement of a competent independent regulatory agency will significantly bolster long-term MPC&A program development and maintenance and increase confidence that diversions would be detected in a timely manner. Such an organization will help deter attempts to elude MPC&A systems at facilities. At the moment, GAN has the potential to become such an agency in Russia but needs clarification of its administrative authority in military-related areas, enhanced technical capability, and more effective relationships with other government agencies. Recommendation: Encourage a system of incentives, possibly including monetary rewards, that will stimulate participants in MPC&A programs to report promptly to the central authorities any irregularities in the implementation of MPC&A systems. An important part of developing an MPC&A culture in the FSU is conveying to workers and managers the importance of immediately reporting any indication of theft or diversion. Financial rewards for whistle blowers might be particularly effective in view of the economic problems there. Recommendation: Emphasize the importance of developing a culture among MPC&A specialists that does not tolerate shortcuts or exceptions in implementing MPC&A systems. Under the Soviet system, workers were generally expected to subjugate their views and actions to their superiors, even if that meant breaking rules. To be effective, MPC&A upgrades must take place in a

OCR for page 52
--> management atmosphere that stresses individual responsibility and places a high value on full implementation of rules and regulations. The U.S. cooperative program should continue to stress such non-technical aspects of MPC&A. 5. Enhance the Program Finding: A number of enhancements would increase the effectiveness of U.S. efforts. The threats of theft and diversion in the FSU differ significantly from threats in the United States. In addition to the general economic and crime situations, which may create specific threats, there are differences among the facilities that affect their susceptibility to loss of material. Many buildings where direct-use material is stored are in poor repair, long perimeters with inadequate protection characterize some sites where material is located, and old accounting systems of dubious reliability are used at some facilities. Some local specialists are not prepared to use sophisticated technologies effectively. Modest immediate enhancements at a large number of facilities may be more important than major investments at a limited number of storage locations. A second area of concern is the vulnerability of direct-use material during transport, which to date has not been a priority for the bilateral programs. Relatively large quantities of material move on a regular basis among the sites that comprise different parts of the MINATOM complex, and much of the hauling is done with ordinary trucks or vans rather than special armored vehicles. At Electrostal, for example, officials acknowledge that truck transport is a weak link of the MPC&A system. In particular, the accounting process to track material as it moves from one storage site or one facility to another may not be sufficiently developed and implemented. In addition, the continued isolation of some facilities where MPC&A upgrades are needed limits opportunities for specialists at one facility to learn from the experiences of their colleagues at other facilities. Finally, several agencies are usually involved in providing security for direct-use material, including responding to incidents and alarms. The Ministry of Interior and the Federal Intelligence Service do not appear to be seriously involved in designing MPC&A upgrades, which is the responsibility of specialists of MINATOM or other concerned research organizations. Recommendation: Emphasize MPC&A approaches that respond to threat scenarios that are appropriate for the FSU, recognizing that they may differ from the threat scenarios used in the United States. One of the sensitive issues in cooperating with Russia and other states in the design of MPC&A upgrades is the choice of threat scenarios that are the basis for the physical protection systems. The current U.S. approach is based largely on U.S. threat scenarios. A number of officials told the committee that the threat of sabotage is a matter of more serious concern in the FSU than the United States has recog-

OCR for page 52
--> nized. The U.S. approach is intended to improve protection against theft or diversion, not against an attack intended to cause damage or contamination. A good U.S.-style MPC&A system would certainly impede a would-be saboteur from gaining access to a facility, but may not protect vulnerable areas inside the facility that are sabotage targets. This difference in focus requires prompt attention. Recommendation: Recognize that in the near term it may be necessary to install systems that fall short of internationally accepted standards in anticipation of subsequent refinements. In this regard, use appropriate MPC&A measures, whether they involve high-tech or low-tech approaches. The physical protection systems being recommended by U.S. specialists reflect how a U.S. facility would respond to such problems. Comprehensive sophisticated systems may not be the most cost-effective use of funds or the fastest means of establishing protection, however. For example, replacing wax seals with modern tamper-indicating devices could quickly provide an enhanced level of protection at many storage sites. Replacing flimsy warehouse doors with sturdy doors and strong padlocks can be done quickly. Other low-tech investments in consolidation, vaults, and fencing might be appropriate first steps at many facilities. Also, hand-held radiation detectors could provide an interim step in providing some protection before the installation of portal monitoring equipment is completed at all facilities. Recommendation: In Russia, give greater attention to MPC&A of direct-use material during transport within and between facilities. At an early stage, the DOD-MOD program addressed the vulnerability of railroad cars used to transport nuclear warheads, and this remains a priority in the cooperative effort on weapons control and accountability. But the DOE MPC&A program began to address transportation problems only in June 1996, when an agreement between MINATOM and DOE provided the framework for a new cooperative effort. Yet, as previously noted, transportation vulnerabilities were readily acknowledged by Russian officials to committee members. Recommendation: Promote greater communication and cooperation among ministries and facilities involved in MPC&A in each of the countries where bilateral programs are being implemented. Exchange of information among relevant officials at the facilities concerning approaches and successes will improve efficiency and the rate of progress of the overall MPC&A effort. The benefits of increased communication were evident at the week-long Conference on NonProliferation and Safeguards of Nuclear Material held at the Kurchatov Institute in May 1996 and cosponsored by DOE, where officials from numerous Russian institutes and U.S. laboratories discussed MPC&A strategies and results. DOE is planning to co-sponsor with MINATOM a major interna-

OCR for page 52
--> tional conference in 1997 on MPC&A that will bring together specialists from across Russia and the rest of the FSU. Recommendation: In Russia, encourage more active involvement of the Ministry of Interior in the planning, testing, and implementation of physical security systems. An important component of the physical protection system is the "response team"—the guards or police trained to respond if an incident occurs. Public and private facilities in the United States place great emphasis on the readiness of these teams, include them in the development of vulnerability assessments, and conduct regular exercises to test the effectiveness of the response. In Russia, the Ministry of Interior both supplies the guards for the perimeter of the sites and is responsible for responses. The division of labor is similar in the other three countries of concern. But the involvement of the Ministries of Interior in the new MPC&A systems being installed with U.S. cooperation is uncertain. Russian officials expressed concern about designing physical protection systems without the full participation of the agency charged with a major role in implementing them. Similarly, officials at the Kharkiv Institute of Physics and Technology in Ukraine expressed concern about whether troops stationed across the city would be able to respond in time to be of real assistance. To date, DOE has believed that the sensitivity of vulnerability assessments and threat scenarios would make the Ministries of Interior reluctant to cooperate openly with the U.S. agencies, and therefore DOE has not sought regular involvement from the ministries. In a few cases the cooperative programs have taken advantage of opportunities at individual facilities where cooperation could be established, but this piecemeal approach is not sufficient. In Russia, the success of the cooperation with MINATOM and other independent institutes suggests that, provided financial incentives are available, senior officers from the Ministry of Interior might be willing to participate more actively in efforts to improve physical protection. AREAS FOR ADDITIONAL STUDY Consideration of MPC&A activities touches on many related areas of national security importance. Some of the areas beyond the scope of this report that deserve further study include the following: Physical protection, control, and accountability of nuclear weapons in Russia. The bilateral program of cooperation between DOD and MOD is now beginning to focus more sharply on protection of nuclear weapons throughout the Russian military complex. Some of the approaches used in the MPC&A bilateral program may be relevant to these new efforts. Nuclear smuggling. The interests and capabilities of organized crime to penetrate the nuclear establishments of the countries of the region need additional

OCR for page 52
--> attention. While the bilateral MPC&A programs should address some aspects of this issue, the U.S. intelligence services have not participated to the fullest possible extent and probably could provide additional perspectives on the problem. Radiological weapons. The feasibility of terrorists acquiring radioactive material (e.g., spent fuel rods) and disseminating it through high-explosive weaponry or other means is of increasing concern. More serious assessment of this type of threat might suggest refinements in the overall strategy for MPC&A programs. Sabotage at nuclear facilities. Deterring saboteurs from penetrating nuclear reactor sites or other sites containing dangerous materials and defining means to counter such penetrations are rapidly becoming important dimensions of international crime prevention. The MPC&A programs should provide useful insights on how to approach this problem. Reducing the inventory of direct-use materials. The U.S. and Russian governments are implementing a program for American purchases of HEU and supporting studies of alternatives for permanent disposition of plutonium. These efforts should be vigorously pursued, since the smaller the inventory the less difficult the MPC&A problem. None of the likely disposition options could begin implementation in less than 10 years, however, so protecting the material remains an urgent security problem.