Click for next page ( 12


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 11
Clarifying the Spent-Fuel Standard We begin with a review of the definition and application of the spent- fuel standard in the initial CISAC study, before turning to issues raised subsequently and the challenge of making the standard easier to apply. THE STANDARD AS ORIGINALLY CONCEIVED CISAC's original formulation (CISAC, 1994, p. 34) held that "Options for the long-term disposition of weapons plutonium should seek to meet a 'spent-fuel standard' that is, to make this plutonium roughly as inac- cessible for weapons use as the much larger and growing stock of pluto- nium in civilian spent fuel." What was meant by the "inaccessibility" of plutonium in spent fuel was elaborated In a two-page box later in the same volume (CISAC, 1994, pp. 150-151) and further clarified ~ a passage In the successor volume (CISAC, 1995, p. 73) staking that the spent-fuel standard does not imply a specific combination of radiation barrier, isotopic mix- ture, and degree of dilution of plutonium. Rather, it describes a condi- tion in which weapons plutonium has become roughly as difficult to acquire, process, and use in nuclear weapons as it would be to use plu- tonium in commercial spent fuel for this purpose. The rationale for the spent-fuel standard is, first, that the bulk, composition, and ionizing- radiation field of spent fuel pose very appreciable barriers to the theft or diversion of this material and extraction of contained plutonium for use in nuclear weapons and, second, that the existence in the world of many hundreds of tons of civilian plutonium in spent fuel means that there 11

OCR for page 11
12 SPENT-FUEL STANDARD FOR DISPOSITION OF EXCESS WEAPON PLUTONIUM would be little security gain from special efforts to eliminate the weapons plutonium, or to render it much less accessible even than the plutonium in spent fuel, unless society were prepared to take the same approach with the global stock of civilian plutonium. The terms "accessible" and "inaccessible" in short formulations of the spent-fuel standard, then, referred to the ease or difficulty of acquiring, processing, and using in weapons the plutonium that is embedded in typical spent fuel. A footnote to the above passage added this further important point about the time dimension of the standard Concerning the spent-fuel standard, we are aware that the accessibility of plutonium in commercial spent fuel is quite variable and increases with time as the fission-product radioactivity that provides the principal barrier to processing of the material for weapons use decays. An appro- priate interpretation of what sort of spent fuel constitutes the standard follows from consideration of the situation that will exist at the time in the future when most of the surplus weapons plutonium at issue here is being processed for final disposition, say, between 2000 and 2030. There is likely to exist, in that period, upwards of 1,000 tons of civilian plutonium in spent fuel, ranging in age from freshly discharged to several decades old. If the inaccessibility of weapons plutonium is made comparable to that of civilian plutonium In the middle of this age distributionthat is, civilian plutonium In spent fuel 20-30 years old the existence of the weapons plutonium In this form would not markedly increase the secu- rity risks already associated with the civilian spent fuel. Dependence on intrinsic properties only The CISAC reports also stressed that meeting the spent-fuel standard depends only on the intrinsic properties of the final plutonium form asso- ciated with a disposition option. "Intrinsic" means, in this context, the properties of the smallest plutonium-containing item that could be removed from an interim or final repository for the dispositioned form, or from a vehicle transporting plutonium in this form to such a repository, without a degree of physical processing likely to be impractical for any- body but the host state itself. (By "physical processing" we mean cutting, blasting, melting, dissolution, and the like. The determination of what would be "impractical" must take into account the amount of time likely to be available before the authorities discover the attempt and intervene.) In the case of ordinary spent fuel itself, we would consider the rel- evant item to be the fuel assemblyan item removable intact from the reactor, or spent-fuel storage pool, or shipping task, but not further sub- dividable without a substantial amount of cutting (made more difficult, of course, by the radiation field associated with the item.) We do not include

OCR for page 11
CLARIFYING THE SPENT-FUEL STANDARD 13 the casks in which ordinary spent fuel would be shipped or stored as part of the 'intrinsic' barriers associated with such fuel, because the lid of such a cask can be removed by cutting or blasting In a matter of a few m~nutes.9 Nor do we count as part of the 'intrinsic' barriers the other types of engi- neered and institutional barriers that may surround spent-fuel assemblies or other plutonium forms, including vaults, buildings, fences, alarms, guard forces, and so on. It is) of course, the combination of intrinsic properties with additional engineered and institutional barriers that governs the overall prolifera- tion resistance of a dispositioned plutonium form. The spent-fuel stan- dard was not developed to describe overall proliferation resistance, but only to describe the contribution to overall proliferation resistance that should appropriately be sought from the intrinsic properties of the final plutonium form. The original CISAC formulations about this standard were intended to make clear that it should be regarded as (a) a necessary but not a sufficient criterion for adequate overall proliferation resistance of the final plutonium form and (b) a ceiling as well as a floor on what is worth achieving In this intr~nsic-property contribution to proliferation resistance. Because these important points seem not to have been made entirely clear (or not to have been entirely accepted!),~ we revisit them here. We believe the spent-fuel standard is a necessary condition for meeting convincingly the criterion that the existence of dispositioned plutonium should not constitute a sigruficant addi- tion to the security risks posed by plutonium in ordinary spent fuel (a form In which much more plutonium resides than In the military stockpiles). This is so in part, we think, because addi- tional engineered and institutional barriers may not have as high a degree of reliability (or demonstrability of reliability) as Me intrinsic 9Approaches in which ordinary storage and shipping casts and/or their contents have been modified to make the contents substantially more difficult to extract as might be done to try to compensate for barriers to plutonium recovery from the items inside that were lower than those for ordinary spent-fuel assemblies would need to be analyzed on a case-by-case basis. This would entail initially considering the entire object to be the "item" whose intrinsic resistance to attack must be assessed, and ultimately reaching a conclusion, based on analysis and comparison, as to whether this item's degree of resis- tance to attack, together with the properties of its contents, constitute compliance with the spent-fuel standard. That is just what has been done in this report for the can-in-canister immobilization option and the CANFLEX variant of the CANDU MOX option. 10See, e.g., Leonard W. Gray and Thomas H. Gould, Jr., Immobilization Team Comments on Interim Report of NAS Panel Review of Spent-Fuel Standard for Disposition of Excess Weapons Plutonium, Lawrence Livermore National Laboratory Report PIP-99-152, 28 October 1999. c

OCR for page 11
14 SPENT-FUEL STANDARD FOR DISPOSITION OF EXCESS WEAPON PLUTONIUM barriers do (and indeed are certainly less reliable in Russia than in the United States), and in part because the imposition of intrinsic barriers sends a much stronger signal about the intention of the possessor state with respect to irreversibility of arms reductions than does the imposition of engineered and institutional barriers that, in many circumstances, would hardly impede the possessor state's recovery of the plutonium at all. At the same time, the spent-fuel standard is not sufficient because, as the original CISAC reports stressed, the intrinsic barriers to acquiring, processing, and using in weapons the plutonium embed- ded in typical spent fuel are not high enough for this material to be considered adequately "self protecting." Thus additional engi- neered and institutional barriers are appropriate for this material and for other plutonium forms with intrinsic barriers comparable to those of typical spent fuel. Indeed, society should plan to increase these engineered and institutional barriers against the weapons use of spent fuel and comparable material over time (including, eventually, by emplacement of the material in a monitored geologic repository), as the technological capacity to handle and reprocess this material becomes more commonplace and the radiation bar- rier to handling it becomes less daunting. The spent-fuel standard is a ceiling as well as a floor on what is worth achieving in the degree of proliferation resistance conferred by the intrinsic properties of dispositioned weapons plutonium. Achieving this much would eliminate the excess proliferation haz- ard represented by the weapons plutonium in comparison with the "background" hazard represented by the much larger stocks of civilian plutonium embedded in spent fuel. Spending addi- tional time and money to bring the intrinsic-property proliferation resistance of dispositioned weapons plutonium to a higher level than that of plutonium in typical spent fuel would not signifi- cantly reduce proliferation risks overall. Indeed, incurring delays in disposition in order to reach a higher standard would add to those risks. Intrinsic characteristics, we repeat, are only a part of adequate secu- rity. But they are an important part. That is why CISAC defined a spent- fuel standard, and why CISAC and we have emphasized that material that does not meet this standard, based on its intrinsic properties, cannot be made to meet the standard by increasing the engineered and institu- tional safeguards that are applied. If the spent-fuel standard is deemed to be satisfiable based on such engineered and institutional barriers as vaults and alarms and guards alone no matter what the characteristics of the .

OCR for page 11
CLARIFYING THE SPENT-FUEL STANDARD 15 plutonium form inside then one could assert that pure plutonium ingots or even intact plutonium "pits" (nuclear-weapon cores) meet the spent- fuel standard, as long as the vault is strong enough, the alarm sensitive enough, the guards competent enough. By reductio ad adsurdum, this dem- onstrates the need for a criterion based on intrinsic properties alone. After all, no matter what engineered and institutional safeguards were applied, storing plutonium ingots or pits indefinitely in Russia would not be regarded by the United States as an adequate approach to the risks of reincorporation of the material into new Russian nuclear weapons or its theft for incorporation into someone else's weaponsnor would this approach in either Russia or the United States be deemed, by others, an adequate indication of good intentions. Further qualifications on the application of the standard The 1994 and 1995 CISAC reports defining and elaborating the spent- fuel standard emphasized several further disclaimers about its applica- tion. We reiterate them here and associate ourselves with them. . . First, not only is a judgment on intrinsic properties of the final plutonium form insufficient (even though necessary) for conclud- ing that the risks associated with the final form are sufficiently small, but consideration of the final form and the protection af- forded it is not sufficient for reaching a judgment about the overall resistance of a disposition method to re-use of the plutonium in weapons. Resistance to acquisition and weapons re-use of the plu- tonium at earlier stages of the disposition process must also be taken into account. Typically, pursuit of increased resistance to proliferation in the final plutonium form entails additional han- dling and processing steps that add to proliferation risk. A judg- ment must be made that the gain at the end warrants the loss along the way. Second, actual resistance to acquisition and weapons re-use of the plutonium is not the only criterion for judging a disposition method satisfactory. Demonstrability of and perceptions about resistance are also important, as are timing, safety characteristics, environmental hazards, economics, tractability of institutional and regulatory requirements, domestic and international political accept- ability, and influences on the proliferation resistance of nuclear- energy systems not directly involved in the disposition effort (which influences may be positive or negatively ). |lSee, e.g., CISAC, 1995, p. 256.

OCR for page 11
16 SPENT-FUEL STANDARD FOR DISPOSITION OF EXCESS WEAPON PLUTONIUM . Finally, getting to final disposition of excess weapons plutonium is not the only important part of managing the hazards of excess nuclear weapons and nuclear materials in the post-Cold-War world. The initial CISAC study and many others on this topic have emphasized the importance of de-activating, consolidating, ~nvento- rying, and dismantling excess weapons; consolidating and invento- rying weapons-usable nuclear materials; storing and protecting all nuclear-weapon components and directly weapons-usable nuclear materials with the degree of diligence appropriate to intact nuclear weapons; blending down highly enriched uranium to levels not directly usable in weapons; subjecting all of these activities to a high degree of bilateral (U.S.-Russian) and eventually international monitoring and transparency; and increasing the attention given to improving the resistance of civilian nuclear-energy systems to the diversion of weapons-usable materials. Defining and imple- menUng standards for disposition of excess weapons plutonium is important, but it is not a substitute for and should not distract attention from these other steps. Application of the standard to final plutonium forms in the initial CISAC study The first volume of the CISAC plutonium study (1994) concluded that the two most promising plutonium-disposition options for meeting the spent-fuel standard and other disposition desiderata In a timely way were (1) fabrication of weapons plutonium into MOX fuel for once-through use in selected civilian power reactors of currently operating types and (2) immobilization of weapons plutonium by vitrification together with 12Besides the CISAC reports cited in Note 1, see, e.g., Frank van Hippel, "Fissile material security in the post-Cold-War world," Physics Today, June 1995, pp. 26-30; Graham Allison, Owen Cole, Richard Falkenrath, and Steven Miller, Avoiding Nuclear Anarchy: Containing the Threat of Loose Russian Nuclear Weapons and Fissile Material, Cambridge, MA: MIT Press, 1995; Matthew Bunn and John P. Holdren, "Managing military uranium and plutonium in the United States and the former Soviet Union," Annual Review of Energy and the Environ- ment, vol. 22, 1997, pp. 403~86; U.S.-Russian Independent Scientific Commission on Pluto- nium Disposition, Final Report, Washington, DC: Office of Science and Technology Policy, Executive Office of the President of the United States, September 1997; Committee on Dual- Use Technologies Export Controls and Materials Protection, Control, and Accounting, National Research Council, Proliferation Concerns: U.S. Efforts to Help Contain Nuclear and Other Dangerous Materials and Technologies in the Former Soviet Union, Washington, DC: National Academy Press, 1997; and Committee on International Security and Arms Con- trol, National Academy of Sciences, The Future of U.S. Nuclear Weapons Policy, Washington, DC: National Academy Press, 1997.

OCR for page 11
CLARIFYING THE SPENT-FUEL STANDARD 17 high-level radioactive wastes In glass logs of the approximate size and composition already selected for use in immobilizing high-level defense wastes at the Savannah River site of the U.S. nuclear-weapons-production complex. The second CISAC volume (1995) went beyond the "most prom- ising" characterization to state flatly that current-reactor options using light-water or CANDU reactors and the then-envisioned heavy-log/ vitrification-with-wastes option would both be able, if implemented, to meet the spent-fuel standard (CISAC, 1995, p. 10~3 With respect to security of the final plutonium forms, the current-reactor options obviously meet the spent-fuel standard, and the Panel judges that the vitrification option meets this standard also. The plutonium In the spent fuel assembly would be of lower isotopic quality for weapon purposes than the still weapons-grade plutonium In the glass log, but since nuclear weapons could be made even with the spent fuel plutonium this difference is not decisive. Under typical assumptions, Me radiological barrier presented by glass logs would be about three times smaller than that presented by a fuel assembly (but still very high), and the mass of a glass log~ontaining, coincidentally, about the same amount of pluto- nium as a fuel assembly- would be about three times greater. The diffi- culty of separating the accompanying materials would be roughly com- parable In the two cases. This conclusion, In which a MOX spent-fuel form containing about twice as much plutonium as typical spent fuel and a vitrified waste form with weapon-pluton~um isotopics were both deemed to meet the spent- fuel standard, underlined CISAC's view that the standard should be understood to mean "roughly" as resistant to acquisition and use In weapons as is plutonium In typical spent fuel, not necessarily identical to typical spent fuel (which would then itself require more precise deft tion) In each characteristic that matters. Questions about the Spent-Fuel Standard The ambiguity inherent in judging whether a weapons-plutonium- disposition form meets a standard of "roughly" equivalent to typical spent 13The indicated comparison was between a 660-kg pressurized-water-reactor fuel assem- bly, irradiated to 40,000 megawatt-days per metric ton of heavy metal, and a 2,200-kg glass log of the type foreseen for production at Savannah River, containing 20 weight percent defense high-level wastes and 1.3 weight percent weapons plutonium mixed with the glass. Radiation doses from both were computed at the surface of the objects, 30 years after fuel- discharge and log production, respectively.

OCR for page 11
18 SPENT-FUEL STANDARD FOR DISPOSITION OF EXCESS WEAPON PLUTONIUM fuel in resistance to acquisition and use of the plutonium for weapons has naturally given rise to questions about whether particular forms meet the standard or not, as well as to calls for greater precision in the specification of the standard for use In making these determinations. Some have ques- tioned whether any plutonium form in which the isotopic composition of the plutonium is that of weapons plutonium should be judged to meet the spent-fuel standard. Others have wondered by how much the plutonium concentration in MOX should be allowed to exceed the value typical for spent fuel arising from once-through use of low-enriched uranium before such MOX is deemed out of compliance with the standard. Still others have questioned at what point the combination of smaller fuel-assembly size and lower radiation barrier associated with CANDU fuel at the burnups typical for this reactor type would disqualify such fuel under the standard. And some have expressed worries that an overly strict interpre- tation of the spent-fuel standard in any or all of these cases could lead to degrees of delay in moving ahead with plutonium disposition, in the United States or Russia, that would increase proliferation dangers overalls Of particular concern to DOE and others interested in current U.S. plutonium-disposition plans is whether DOE's current design for the final plutonium form in the immobilization track in the dual-track option can reasonably be deemed to meet the spent-fuel standard. This design was developed subsequent to the 1995 CISAC report's determination that the then-current vitrification-with-wastes immobilization option and the once-through MOX option both meet the spent-fuel standard. In the new variantcalled the "can-in-canister" approach plutonium oxide is incor- porated in ceramic pucks that themselves contain no fission products; the pucks are stacked in an array of cans suspended on a frame in a large steel canister; and molten borosilicate glass, bearing fission products, is poured into the canister to solidify around the cans and thus contain them in a massive, highly radioactive glass log. In the immobilization approach previously considered by CISAC, by contrast, the plutonium oxide would have been added directly to the fission-product-bearing molten glass with the aim of creating a more-or-less homogenous mixture of plutonium and fission products in the resulting highly radioactive glass log. 14Publications raising the questions mentioned in this paragraph are cited, and prelimi- nary responses to the questions are provided, in John P. Holdren, John F. Ahearne, Richard L. Garwin, Wolfgang K. H. Panofsky, John J. Taylor, and Matthew Bunn, "Excess weapons plutonium: how to reduce a clear and present danger," Arms Control Today, November/ December 1996, pp. 3-9. Virtually all of these questions were posed also in the briefings and public comment sessions arranged in connection with the meetings of this Panel (see Appen- dix B).

OCR for page 11
CLARIFYING THE SPENT-FUEL STANDARD 19 The can-in-canister approach was chosen by DOE in preference to the homogeneous plutonium-in-glass approach for several reasons.~5 It had become apparent that designing, testing, and implementing modifica- tions to the Savannah River melter and the composition of its glass- required in order to enable addition of adequate quantities of plutonium directly to the melt while observing criticality constraints would be technically difficult, costly, and likely to substantially set back the time- table for the already scheduled high-level-waste immobilization program at the Savannah River site. In particular, a change in glass composition from the original borosilicate glass to a lanthan~de borosilicate glass would have been necessary to achieve the desired plutonium loading, but the processing temperature needed for the new composition (around 1475C) was too high to allow incorporation of the cesium needed to provide the radiation barrier. (Cesium volatilizes above 1200C.) It might also have been necessary to reduce the log size In order to maintain criticality mar- gins, which not only would have entailed a new melter design but also would have reduced the resistance of individual logs to theft. Switching from glass to a homogeneous ceramic incorporating plutonium and cesium would entail producing all of this ceramic by hot isostatic pressing in hot cells, a considerable complication compared to the cold-press-and- s~nter method, In glove boxes, which can be used if the ceramic contains plutonium but no fission products. DOE's choice of the heterogeneous can-~n-can~ster approach allowed staying with the original glass composition to contain the fission prod- ucts, while gaining the improved performance of ceramic as the pluto- n~um-conta~n~ng material (including greater durability under repository conditions and greater ease of nondestructive assay for verification pur- poses) and avoiding criticality concerns attendant on adding multiple critical masses of plutonium to 1,700 kilograms of molten glass and fission products at a time. And leaving fission products out of the plutonium- bear~ng ceramic pucks In the can-~n-can~ster approach allowed for lower manufacturing costs than would be entailed if the pucks themselves con- ta~ned strong gamma-ray emitters The most difficult question about the can-in-canister approach's meet- ~ng the spent-fuel standard is whether extraction of the plutonium from i5See, e.g., Office of Fissile Materials, Department of Energy, Technical Summary Reportfor Surplus Weapons-Usable Plutonium Disposition, Rev. 0, Washington, DC: Department of En- ergy, 1996; M. J. Plodinec et al., "Survey of glass plutonium contents and poison selection," in Plutonium Stabilization and Immobilization Workshop, Washington, DC: Department of Energy, 1995, pp. 229-239; and Leonard Gray and Malvyn McKibben, An Analysis of Pluto- nium Immobilization Versus the "Spent Fuel Standard," Lawrence Livermore National Labora- tory Report POP-98-073, August 1998.

OCR for page 11
20 SPENT-FUEL STANDARD FOR DISPOSITION OF EXCESS WEAPON PLUTONIUM the fission products in the heterogeneous pucks-in-glass arrangement is significantly easier than extracting plutonium from spent fuel. The corre- sponding questions about the MOX option's meeting the spent-fuel stan- dard relate to whether the high residual plutonium concentration in spent light-water reactor (LWR) MOX or the relatively low mass and radiation field associated with spent CANDU MOX fuel assemblies would make these plutonium forms significantly more proliferation prone than typical spent fuel from lightly enriched uranium (LEU)-fueled LWRs. As prepa- ration for addressing these questions, we proceed first to elaborate some ingredients of a systematic approach to applying the spent-fuel standard. A SYSTEMATIC APPROACH TO CONSIDERING COMPLIANCE WITH THE STANDARD Until now there has been no simple formula that can be mechanisti- cally applied to determine whether the final plutonium form resulting from a disposition process is sufficiently close to typical spent fuel in the array of characteristics governing resistance to acquisition, processing, and use in weapons of the contained plutonium that it can be deemed to meet the spent-fuel standard. In the current study, we considered whether such a formula could usefully be constructed. We concluded that doing so is very difficult; neither are we convinced that it would even be desirable. Many characteristics are germane; the importance of these character- istics relative to one another may vary with the type of threat that is deemed most important at a given time and place; the range of variation with respect to the relevant characteristics is substantial within the array of fuel types, degrees of irradiation, and ages since discharge in the global spent-fuel inventory; a final disposition form's departures from typical spent-fuel characteristics in the direction of lower resistance to prolifera- tion in some respects may be offset by departures in the direction of higher resistance in other respects; and the benefit of trying to narrow a given "gap" between a characteristic of a final disposition form and the corresponding characteristic of typical spent fuel must be weighed against the delays and other increases of in-process proliferation risks that may result from this effort. In so complex a space of possibilities, it seems to us, the considered judgment of experienced people in answering the ques- tion, "How close to spent fuel is close enough?" will continue to be diffi- cult to replace with a mechanistic formula. We do think, however, that the needed judgments can usefully be informed by systematic comparison of the relevant quantitative and quali- tative characteristics of candidate final plutonium forms, against those of typical spent fuel, in a matrix format that groups the characteristics by the

OCR for page 11
CLARIFYING THE SPENT-FUEL STANDARD 21 kinds of barrier against proliferation they confer and that indicates the relative importance of these different barriers against the main categories of proliferation threat. We employ such an approach here. Interactions of threats and barriers The three main classes of proliferation threats to which intrinsic bar- riers provided by final plutonium forms are germane are as follows: (1) "Host-nation breakout" means that the country legitimately hold- ing the dispositioned plutonium elects to recover it for re-use in its nuclear arsenal. This is likely to entail large amounts of plutonium (from several hundred to thousands of kilograms), no physical limitations on access to the dispositioned plutonium forms and the ability to transport them, high technical capabilities for sepa- rating the plutonium and employing it to make sophisticated nuclear weapons, high performance requirements for the weapons, and concerns with detection of the effort while it is underway ranging from negligible in the case of overt breakout to very sub- stantial in the case that breakout is intended to be clandestine. (2) "Theft for proliferant state" means that members of a subnational group and/or agents of a proliferant state including, potentially, facility insiders steal the material by stealth or force and transfer it to the state for use in nuclear weapons. Much smaller amounts of material are germane here (tens to perhaps one or two hundred kilograms); physical barriers to access and transport are impor- tant, in the context of limits on the time and technological capaci- ties available to the thieves for dealing with these barriers; the technical capacities of the state receiving the material for process- ing it and employing it in nuclear weapons are likely to be moder- ately high albeit lower than in the "host-nation breakout" case; the performance requirements for the resulting weapons are likely to be moderate; and concerns with detection would be high in the theft and transport stages before the material is on the territory of the proliferant state and moderate to high thereafter. (3) "Theft for subnational group" means that a subnational group steals the material by stealth or force and either tries to use it to make nuclear weapons itself or transfers it to another subnational group for this purpose. In this case the quantity of material of interest can be as small as one bomb's worth; the situation with respect to physical barriers to access and transport in relation to limits on the time and technological capacities available to the thieves are the same as in the "theft for proliferant state" category;

OCR for page 11
22 SPENT-FUEL STANDARD FOR DISPOSITION OF EXCESS WEAPON PLUTONIUM the technical capacities available for processing and employing the stolen material in nuclear weapons are likely to be less than in the "proliferant state" case; the performance requirements for the resulting weapons are likely to be low; and concerns with detec- tion would be high at all stages of the effort. The indicated differences in the characteristics of these three types of threat give rise to differences in the relative importance of the various intrinsic characteristics of final plutonium forms as barriers against the threats. We summarize our judgments on the interaction of threats and intrin- sic barriers in Table 1, which arranges the characteristics of final pluto- nium forms according to the barriers these characteristics provide at dif- ferent steps in the proliferation chain and indicates the relative importance of these barriers against the three classes of proliferation threats. The relative-importance ratings reflect a combination of the needs/capabili- ties of the threat groups with the nature of the barriers. We choose a scale of only four ratings- zero, low, moderate, and high to reflect distinc- tions in relative importance without implying more precision than the complexity and judgmental character of these considerations permit. The term "item" as used in Table 1, refers to the smallest embodiment of the final plutonium form that could be removed from a storage facility or transport operation without a degree of on-site physical processing (cutting, blasting, melting, dissolution, and so on) likely to be impractical for anybody but the host state itself. The term "technical difficulty" in- cludes requirements for manpower and specialized knowledge, skills, and equipment, as well as an allowance for the amount of time likely to be required to complete a task with a given level of resources. The detectability of an activity, which is an important barrier in cases where concealment is important to the proliferators, depends on resource and time requirements for the activity and on other signatures (e.g., ther- mal, chemical, nuclear) arising from the interaction of the intrinsic prop- erties of the dispositioned plutonium form with the operations being car- ried out on it. Detectability also depends on the capabilities deployed to achieve detection. This underlines that, although Table 1 is intended to address the intrinsic properties of final plutonium forms and not the char- acteristics of the engineered and institutional protections supplementing these, there are interactions between intrinsic properties and the engi- neered and institutional protections (as, for example, in the relation be- tween intrinsic properties related to detectability and the monitoring sys- tems in place to achieve detection).

OCR for page 11
CLARIFYING THE SPENT-FUEL STANDARD TABLE 1 Intr~nsic-barrier and threat characterization for final plutonium forms 23 Importance of barrier against the threat Barrier Host-nation Theft for a breakout proliferant state Theft for a subnational group Barriers to acquisition of the Pu from its storage site Mass and bulk of itema Zero to lowb (low) concentration of Zero to lowb Pu in item Radiation hazard to Low acquirers Technical difficulty of partly separating Pu from bulk components of item on sitea Thermal, chemical, and nuclear signatures aiding detection Barriers to separation of the Pu from diluents and fission products Technical difficulty of disassembly Technical difficulty of dissolution and separation Quantity of material to be processed Hazards to separators Signatures aiding detection Moderate High Moderate Zero to lowb High Moderate High Moderate High Zero to Moderate to Moderate to moderateb'C highc highc Low Low Low to moderates Low Zero to moderates Barriers to use of the separated Pu in nuclear weapons Deviation of isotopic composition from "weapons grade" Low to moderate Moderate to high Moderate to high Moderate Moderate to highC'd Moderate High High Moderate Highc Moderate Moderate Low a Barrier relates both to technical difficulty and detectability, which are themselves related (see text). b Importance depends on whether breakout is open or clandestine. c Importance depends on sensor capabilities. d Importance depends on degree of proliferant state concern with detection.

OCR for page 11
24 SPENT-FUEL STANDARD FOR DISPOSITION OF EXCESS WEAPON PLUTONIUM Explanations of judgments in Table 1 The first set of intrinsic barriers comprises those that impede the acquisition of the plutonium i.e., removal of the plutonium-bearing item from its place of storage or transport- including barriers to processing the item before it is removed in order to extract the plutonium from it or otherwise simplify its removal. In this category: . ~ , The mass and bulk of an individual item would be barriers against the threat of host-nation breakout only in the case where the breakout was intended to be clandestine, in which case the item size might be expected to have some effect on the detectability of operations to remove the items from storage or divert them in transport. (We judge this to be of low overall importance in light of the relative ease with which a host country could probably overcome it.) If breakout was open, item size would be of no consequence to a host state (which would be well equipped to handle items of any mass and bulk). In the cases of theft for a proliferant state or a sub- national group, the barriers posed by mass and bulb to ready removal of an item are more important we rate them "moderate" because of their effect on the character of the equipment needed to accomplish the theft (which affects, to some degree, the resources the thieves would need and the chance of their operation's being detected). The concentration of plutonium in the item is a barrier the lower the concentration the higher the barrier insofar as it affects the total mass of material (and thus the number of items) that must be acquired in order to obtain a given quantity of plutonium. As with item size, and for the same reasons, this factor would be of no consequence at the material-acquisition stage (although of some consequence at the processing stage, about which more below) to a state engaged in open breakout, and of only low consequence to a state engaged in clandestine breakout. But we believe it is of high importance in relation to theft for a proliferant state or a subnational group, because concentration even more than indi- vidual item size determines the scale of the entire theft operation (personnel and equipment), directly affecting both the resources the thieves would need to mobilize, the time required for the acqui- sition operation, and the chances of their being detected and thwarted in the course of it. The radiation hazard to the acquirers of the plutonium (as opposed to the radiation hazard to the processors, which is treated below) would be of low but not zero importance as a barrier to host-

OCR for page 11
CLARIFYING THE SPENT-FUEL STANDARD . 25 nation breakout; such a state would be well equipped to minimize this hazard with shielding and remote-handling equipment. This barrier would be greater against theft for a proliferant state or a subnational group, but we rate it as "moderate" in importance rather than "high" for two reasons: first, even the highest radiation fields associated with spent fuel and other plutonium-disposition forms would not produce immediately incapacitating doses if the thieves took modest precautions; and, second, many potential thieves (and their bosses) might not give high priority to the avoid- ance of the kinds of doses that would be involved (either out of ignorance or out of willingness to bear the risk or impose it on someone elsein exchange for expected high reward). The technical difficulty of partly separating the plutoniumfrom the bulk components of the item on site would be of no importance in the case of open breakout by a host nation, which would face no difficulty in transporting the intact items to a processing site of its choice. The barrier would be a bit higher if the host-nation breakout was intended to be clandestine, since transporting the intact items to a processing site might be at least somewhat easier for other coun- tries to detect than transporting more concentrated forms of pluto- nium would be. In the case of theft for a proliferant state or for a subnational group, however, it would be a great advantage for the thieves if the quantity and/or radioactivity of the material that needed to be removed from the site of the theft could be signifi- cantly reduced by operations that could be effected at the site without greatly prolonging the thieves' stay there or otherwise increasing the chance of their being detected in the act. This would ease substantially the thieves' subsequent problems of transport and concealment of storage and processing. Thus we rate the barriers against this as being of "high" importance. Thermal, chemical, and nuclear signatures that would aid detection dur- ing the course of a they and subsequent transport and storage would be of no importance to a host nation engaged in open breakout. In the event the breakout was intended to be clandestine, however, such signatures could significantly affect the chance that other countries would detect the activity; thus we consider this barrier of "moderate" importance in this case (the highest of any of the barriers to host-nation breakout at the plutonium-acquisition stage). Sensitivity to detection during theft and subsequent trans- port and storage would be even greater in the cases of theft for a proliferant state or a subnational group, so we rate the barrier as "moderate to high" in these cases.

OCR for page 11
26 SPENT-FUEL STANDARD FOR DISPOSITION OF EXCESS WEAPON PLUTONIUM The second set of intrinsic barriers listed in Table 1 relate to the work of separating weapons-usable plutonium from the structure, diluents, and fission products accompanying the plutonium in the form in which it was acquired (that is, either its final dispositioned form or something derived from that by processing undertaken at the site of acqui- sition, as discussed above). In the case of host-state breakout, this activity could take place either clandestinely or openly, using old facilities or new ones constructed openly or clandestinely for the breakout purpose. In the case of theft for a proliferant state, this processing could be accomplished by the thieves before transfer of the material to the state, in which case it would likely be done on the territory of the state from which the material was stolen or, after smuggling it across one or more borders, on the terri- tory of a third state. Or the thieves might manage to transfer the stolen items themselves to the proliferant state, whereupon the latter would do the subsequent processing in facilities on its own territory (in which case these could, again, be either open or clandestine, but more likely the latter). In the case of theft for a subnational group, this processing would most likely be in clandestine facilities, on whatever territory. The intrinsic barriers against these activities and the bases for our judgments about their relative importance are as follows: . The technical difficulty of mechanical disassembly of the plutonium- containing items would be a barrier of only low importance in the context of host-nation breakout, inasmuch as such nations would have facilities adequate to handle this rather easily for any imagin- able disposition form. It would also be of low importance to a proliferant state that is conducting this processing itself, since the technology for this mechanical disassembly step is not very demanding. If the processing were being done by the thieves before transferring the plutonium to the proliferant state, how- ever, this barrier would be of moderate importance, as it would be in the case where a subnational group was the final recipient, because the relevant technologies/facilities would be harder for a subnational group to acquire anal use (and hide) than for a state to do so. The technical difficulty of dissolution of the plutonium-containing com- pounds and chemical separation of the plutoniumfrom the other elements present is, like the technical difficulty of mechanical disassembly, a barrier of low importance to a host nation (although not zero, insofar as the differences could be great enough to motivate the choice of one plutonium source over another if they were equally attractiveor equally difficult in other respects). We judge the importance of this barrier to be "moderate to high" in the case of

OCR for page 11
CLARIFYING THE SPENT-FUEL STANDARD . . 27 theft for a proliferant state (depending on whether the processing is being done by the state or by the thieves themselves) and "high" in the case of theft for a subnational group. Although the relevant technologies for at least some disposition forms are well described in open literature, they are fundamentally more demanding for small states and subnational groups than are the mechanical- disassembly technologies. The quantity of material to be processed (which of course is related to the plutonium concentration, discussed separately above as a bar- rier to initial acquisition as opposed to a barrier to separation) is a barrier of low importance in the case of open host-nation breakout (although not of zero importance, because it increases time and cost in some combination). It is of moderate importance in the case of clandestine host-nation breakout, because its effect on the scale of the operation increases the chance of detection. We judge the importance of this barrier to be "moderate to high" in the case of theft for a proliferant state, depending on who is doing the processing and, in the event it is being done by the proliferant state, depending on the importance attached to concealment and on the sophistication of the facilities available to the particular state. The radiation, criticality, and toxic hazards during the separation pro- cess are barriers of only low importance in the case of host-nation breakout, because these nations have ample facilities and experi- ence for minimizing these risks. Radiation and criticality are more important barriers in the cases of theft for a proliferant state or for a subnational group, because protection against these hazards during processing requires the development (and in some cases the concealment) of facilities and capabilities that the processing entities did not possess before. (Still, we do not rate these barriers "high" for proliferant-state processors because the needed capa- bilities are well within the means of most states, and we do not rate them "high" for subnational-group processors, even though their capabilities would generally be less than those of states, because such groups are likely to be willing to assume higher risks in these categories than states are.) Toxic hazards are not likely to be great enough to constitute more than a low barrier in any of the cases. Detectability of processing operations may be based on the scale of the required operations (including floor space, electrical power, spe- cialized supplies, and the duration of the activities) and on chemi- cal, nuclear, and thermal signatures from the specific operations involved. (Dissolution and separation of plutonium, for example,

OCR for page 11
28 SPENT-FUEL STANDARD FOR DISPOSITION OF EXCESS WEAPON PLUTONIUM can release effluents derived from the solvents involved, which are detectable remotely through technologies such as LIDAR (LIght Detection And Ranging), as well as releasing radionuclides that are detectable by various means. Chemical and radioactive "taggants" chosen for detectability may be deliberately added to disposition forms to raise this barrier. Infrared interrogation and observation of infrared emissions moreover, can determine whether or not known processing facilities are operating.) Of course, the importance of the detectability barriers depends on whether the activities are clandestine or open; it depends on whether discovery would necessarily be fatal to the enterprise (as it almost certainly would In the case of processing by a subnational group, might be In the case of processing by a proliferant state, and probably would not be In the case of clandestine host-state breakout); and it depends as well on the state of the sensor capa- bilities In relation to the strength of the signatures. These consider- ations In combination lead us to rate the detectability barriers as "zero to moderate" for the case of host-nation breakout, "moder- ate to high" In the case of theft for a proliferant state, and "high" for the case of the theft for a subnational group. The last set of intrinsic barriers addressed In Table 1 are those against the utilization of the plutonium that the proliferators are able to separate for the fabrication of functional nuclear weapons. If it is assumed that proliferators In all categories will ultimately be capable of obtaining rea- sonably pure plutonium metal starting from the dispositioned forms as we believe to be the cas~then the main intrinsic barriers In this category are those associated with deviation of the plutonium's isotopic composi- tion from "weapons grade." The isotopic composition of the plutonium in the spent fuel is com- pared with that of weapons-grade plutonium in Table 2. The indicated differences lead to a neutron background nearly 7 times higher in the spent-fuel plutonium than In weapons-grade plutonium, a heat genera- tion rate about 6 times larger, and a surface gamma-ray dose about 16 times higher. These differences would produce additional difficulties for those who might choose to design, manufacture, and deploy nuclear weapons made from typical spent-fuel plutonium rather than from 16The unshielded surface gamma ray dose from reactor-grade plutonium is in the range of 20 rem/hour (see, e.g., CISAC, 1995, p. 270~. This may be compared with the short-term dose that would be associated with a 50 percent chance of death within 30 days from acute radiation syndrome, which is in the range of 500 rem.

OCR for page 11
CLARIFYING THE SPENT-FUEL STANDARD TABLE 2 Isotopic composition of plutonium in typical LWR spent fuel versus that in weapons-grade plutonium 29 Isotope Type of plutonium Pu-238 Pu-239 Pu-240 Pu-241 Pu-242 Am-241 Typical spent-fuel plutonium from light-water reactors Weapons-grade plutonium 1.3% 60.3% 24.3% 0.01% 93.8% 5.8% 5.6% 5.0% 3.5% 0.13% 0.02% 0.22% Source: CISAC, 1995, p. 45. weapons-grade plutonium difficulties that account for the historical preference of nuclear-weapon states for using weapons-grade material. But, as emphasized In the previous CISAC plutonium reports and In other unclassified but authoritative studies, the differences do not preclude the design and construction of effective nuclear weapons from typical spent- fuel plutonium, at all levels of sophistications We rate the barrier posed by isotopic deviations from weapons grade as "moderate" In importance for host-nation breakout In Table 1 mainly because recovery of weapons-grade plutonium from dispositioned forms would permit production of weapons from existing designs without new nuclear-explosive tests, whereas use of plutonium of different isotopic compositions would be likely to entail design modifications and, even if not, would probably require new nuclear-explosive tests to confirm that the change in isotopic composition had not unacceptably degraded per- formance. In the case of theft for a proliferant state we rate the barrier likewise as "moderate" In importance: such a state would probably prefer to avoid if possible the burdens posed by isotopic deviations for design, fabrication, and maintenance of nuclear weapons, but it would also prob- ably have the capabilities to cope with these burdens In ways that achieved a level of weapon performance adequate for a proliferant state's initial purposes. We rate importance of the isotopic barrier as "low" in the case 17See CISAC (1994, pp 29-33), CISAC (1995, pp. 43-46), and Department of Energy, Non- proliferation and Arms Control Assessment of Weapons-Usable Fissile Material Storage and Excess Plutonium Disposition Alternatives, Washington, DC: Department of Energy, January 1997, pp. 37-39.

OCR for page 11
30 SPENT-FUEL STANDARD FOR DISPOSITION OF EXCESS WEAPON PLUTONIUM of theft for a subnational group because, although the weapon-related capabilities of such a group would probably be lower than those of a proliferant state, the subnational group would be likely to be much less concerned about deviations from ideal performance inasmuch as a lower-than-expected yield would still mean an explosive force more than adequate for the likely purposes of such a group and probably less concerned about radiation exposures to those making and handling the weapons. Relative importance of threat categories For purposes of deciding which characteristics of dispositioned plu- tonium forms are most germane to a determination of compliance with the spent-fuel standard, it might be thought useful to ask which of the three categories of threat is deemed to be of greatest concern. It is our view, however, that the answer to this question is likely to vary with time and with other circumstances. For present purposes, therefore, we give equal weight to the three threat categories. It is to be emphasized that none of these three classes of threat to dispositioned plutonium will pose much danger of actually being carried out until a time In the future when sources of plutonium in much more convenient forms for proliferators have been considerably diminished compared to their abundance today. The countries of greatest potential concern in terms of host-nation breakout, for example, are Russia and the United States, which will have the largest quantities of dispositioned plu- tonium; but both countries are likely to retain, for some time to come, such large quantities of deployed and reserve nuclear weapons and reserve nuclear material, compared to any imaginable need, that it is difficult to envision any incentive for them to want to recover plutonium from the amounts they have declared excess and eligible for disposition. With respect to the ''theft for proliferant state" and "theft for subnational group" threats, various military and civilian stocks of already separated plutonium and highly enriched uranium are likely to remain more attrac- tive targets for proliferators than spent fuel or dispositioned plutonium forms would be for some years to come. It is, nonetheless, important to move forward now with plutonium disposition- and, in that connection, important to determine the compli- ance of candidate approaches with the spent-fuel standard both because disposition of excess plutonium is a process that will require decades under the best of circumstances (during which time it may be hoped that the stocks of warheads, separated plutonium, and highly enriched ura- nium will have been greatly reduced) and because, as the 1994 and 1995 CISAC plutonium reports emphasized, the barriers provided by pluto-

OCR for page 11
CLARIFYING THE SPENT-FUEL STANDARD 31 nium disposition against host-state breakout have arms-control and non- proliferation value through the signals they send (between the host states and to the rest of the world) about the intended irreversibility of nuclear arms reductions. With these disclaimers, we conclude from the ratings in Table 1 that the characteristics that should receive the most weight in the determina- tion of a disposition form's compliance with the spent-fuel standard are as follows. (1) With respect to barriers to acquisition of the plutonium from its storage site: (a) the concentration of plutonium in the items that could be stolen, (b) the technical difficulty of partly separating the plutonium from the bulkier components of the item on site, and (c) the strength of the aids to detection of the items provided by their thermal, chemical, and nuclear signatures. (2) With respect to barriers to subsequent separation of the plutonium from diluents and fission products: (a) the quantity of material that needs to be processed to obtain a weapon's worth of pluto- nium, (b) the technical difficulty of dissolution of the plutonium, (c) the technical difficulty of chemical separation of the plutonium from solution, and (d) the size of the aids to detection of these activities provided by their thermal, chemical, and nuclear signa- tures and the scale of the needed facilities. Characteristics deserving somewhat smaller but still significant weight in the determination of compliance with the spent-fuel standard are the mass and bulk of the items that would need to be removed from the storage site, the radiation and criticality hazards associated with acqui- sition and processing of the material, and the deviation of the plutonium's isotopic composition from "weapons grade." 3: s