THE FUTURE OF IMMOBILIZATION UNDER THE

U.S.-RUSSIAN PLUTONIUM DISPOSITION AGREEMENT

Edwin S. Lyman

Nuclear Control Institute

1000 Connecticut Avenue, NW Suite 410

Washington, DC 20036 USA (202) 822-6594

Paper presented at the 42^nd Annual Meeting of the Institute of Nuclear Materials Management, Indian Wells, CA, July 18, 2001

ABSTRACT

In September 2000, the U.S. and Russia signed an agreement committing each nation to disposing of 34 metric tons (MT) of surplus weapons-grade plutonium (WG-Pu) at an initial rate of 2 MT per year, commencing in 2007. According to the agreement, on the U.S. side, 25.6 MT of WG-Pu would be fabricated into mixed-oxide (MOX) fuel and irradiated, and 8.4 MT would be immobilized in a ceramic form and disposed of in canisters containing vitrified high-level radioactive wastes (HLW) at the Defense Waste Processing Facility (DWPF) at the Savannah River Site (SRS). On the Russian side, MOX fuel use (in both light-water and fast reactors) was the only designated option. The proposed utilization in the U.S. of both MOX and "can-in-canister" (CIC) immobilization in the Agreement conformed to the U.S. policy that a "dual-track" be instituted to ensure the availability of at least one disposition method.

However, the Bush Administration has decided to indefinitely suspend development of immobilization in the U.S. and has ordered the dismantlement of the Plutonium Ceramification Test Facility at Lawrence Livermore National Laboratory (LLNL), which was to have tested the process for incorporating WG-Pu into ceramic pucks. Unless reversed by Congress, this foolish decision will effectively terminate development of a promising plutonium disposition technology with significant benefits from the standpoint of cost and non-proliferation, and will probably result in a significant delay in implementation of the U.S.-Russian agreement.

INTRODUCTION

In January 1997, DOE announced that it had decided to pursue a "dual track" policy for the disposition of approximately 50 metric tons (MT) of plutonium produced for weapons programs that had been declared excess to military needs. The two tracks of the dual track refer to two different approaches for converting separated plutonium into a dilute and highly radioactive form that is harder to steal and more difficult to return to weapons.

Under one approach, plutonium would be used to produce MOX fuel assemblies, which would then be loaded and irradiated in U.S. commercial nuclear reactors, displacing some or all of the low-enriched uranium oxide (LEU) fuel assemblies the reactors currently use. DOE initially planned to utilize this option for up to 33 MT of weapons-grade plutonium (WG-Pu). In 1999, a contract was awarded to the Duke Cogema Stone & Webster (DCS) consortium to build a MOX plant at the Savannah River Site (SRS) and to irradiate the MOX fuel at Duke Power's Catawba and McGuire stations near Charlotte, NC (as well as at Dominion Resources North Anna station in Virginia, which has since been withdrawn from the program).

Under the other approach, known as "can-in-canister" immobilization (CIC), plutonium would be incorporated into chemically stable ceramic discs at a new facility to be constructed at SRS. These discs would in turn be embedded in canisters of "vitrified" (glassified) high-level radioactive waste (VHLW) at the Defense Waste Processing Facility (DWPF) at SRS. DOE initially proposed using CIC to dispose of approximately 17 MT of excess plutonium in impure forms that would not be suitable for MOX fabrication without substantial chemical processing.

One of the chief goals of the plutonium disposition program is to render separated plutonium as inaccessible as the plutonium in commercial spent nuclear fuel, thereby meeting the "spent fuel standard" defined by the National Academy of Sciences (NAS).^[1] This would mitigate the unique proliferation risks associated with separated plutonium and allow a relaxation of the physical protection and safeguards requirements for the material. Both MOX irradiation and immobilization were judged to be roughly comparable means of meeting the spent fuel standard. However, DOE decided to pursue both tracks for a number of reasons, including the need for a backup strategy in case one track did not succeed.

One of the chief objectives of the U.S. disposition program is to engage Russia in a bilateral process to address the severe proliferation risk associated with unsecured fissile materials throughout the Russian nuclear complex. The issue of plutonium disposition has been addressed through a series of bilateral and multilateral summits, working groups and joint agreements since 1993, culminating in the September 2000 signing of a U.S.-Russian agreement on the management and disposition of excess plutonium ("the Agreement").^[2] The Agreement contains specific timetables for execution of a bilateral program to dispose of 34 metric tons (MT) of WG-Pu on each side, commencing in 2007, at an initial rate of 2 MT WG-Pu per year. (Because Russia intends to blend WG-Pu with about 12% RG-Pu to conceal its isotopic composition, this corresponds on the Russian side to a disposition rate of about 2.2 MT of total Pu per year). The agreement also calls for rapid completion of a plan for expanding the disposition rate to 4 MT (4.5 MT total Pu for Russia) per year.

Consistent with the dual-track policy, the U.S. committed in the Agreement to dispose of 25.6 MT of WG-Pu --- mostly from dismantled warheads --- as MOX, with the remaining 8.4 MT of impure WG-Pu materials to be disposed of through CIC immobilization. On the Russian side, however, the final agreement contains commitments to dispose of MOX both in LWRs and liquid-metal fast reactors (LMRs), but has no commitment to dispose of Russian WG-Pu through immobilization. While one may think that this was due to Minatoms long-standing opposition to Pu immobilization, in fact this was the fault of U.S. negotiators. In the months of negotiation preceding the final agreement, Russia did offer approximately 1 MT of WG-Pu contained in a low-assay sludge for immobilization, but the offer was rebuffed by the U.S. side, which insisted that only high-assay materials be covered by the agreement. This was a tactical mistake, because a Russian commitment to pursue immobilization technology and build an immobilization plant would itself have been a far more important achievement than a specific commitment to immobilize high-assay material. That might have happened eventually, if difficulties later were to arise in implementation of the MOX track.

The position of DOE at the time the agreement was concluded was that it contained a tacit commitment for Russia to develop an immobilization process for treatment of plutonium-bearing wastes from the pit conversion and MOX fabrication processes. This interpretation was critical for U.S. attempts to secure a financial contribution from the government of Germany for the Russian plutonium disposition program, as Germany has taken the position that it will only contribute funds for immobilization in Russia. However, it is unclear whether the Russian side also accepted DOE's interpretation.

After assuming power in 2001, the Bush Administration ordered a National Security Council review of all DOE (now NNSA) non-proliferation programs, including the plutonium disposition program. Although this review is apparently still in progress, the likely outcome with respect to immobilization has already been anticipated with the order in March 2001 by NNSA headquarters to immediately suspend all immobilization activities. Although characterized by NNSA as a temporary suspension for budget reasons, the suspension plan approved by NNSA will make later reconstitution of the program extremely difficult to accomplish, since it includes dismantlement of the almost-completed
Plutonium Ceramification Test Facility at LLNL. The high cost of dismantling the facility required NNSA to reprogram FY 01 money into the immobilization account for this purpose.

Since the U.S. had been the only nation seriously pursuing immobilization, the future of the technology is therefore in doubt. This will ultimately result in increased costs and further delays in the timetables in the U.S.-Russian agreement, which will likely have to be substantially revised.

WHY IMMOBILIZE?

The suspension of the Pu immobilization program by the Bush Administration is a very unfortunate and unwise decision. Immobilization technology holds considerable promise as a plutonium disposition alternative superior to MOX with regard to cost, safety and non-proliferation.

For example, immobilization is much better suited to the goal of consolidation and minimization of transport of fissile materials in Russia. The WG-Pu storage facility being constructed with U.S. funds at the Mayak facility in Ozersk is in the vicinity of the Mayak high-level waste vitrification plant that could be used as a source of fission products for CIC immobilization. Therefore, if a ceramification plant were build at Mayak, plutonium disposition could take place at the same site as the facility where the bulk of the plutonium will be consolidated and stored.

In contrast, the MOX plan in Russia proposed by Minatom would involve three different sites for MOX fuel fabrication (a pilot line for LMR fuel at Mayak, a vibropak line for LMR fuel at Dimitrovgrad and a new facility for both LWR and LMR fuel at Zheleznogorsk), and a minimum of four reactor sites (the Balakovo and Kalinin VVER-1000 stations near Saratov and Tver, respectively, and the BOR-60 and BN-600 LMRs near Dimitrovgrad and Yekaterinburg, respectively). Moreover, an increase of the disposition rate from 2 MT Pu/yr to 4 MT/yr would require enlisting additional reactor capacity in countries in Eastern or Western Europe. Compared to the requirements for immobilization, this network would involve thousands of additional shipments of separated plutonium and unirradiated MOX fuel over vast distances, as well as a greater number of Category I facilities requiring the most stringent safeguards and physical protection. In addition, the immobilization process would most likely be easier to safeguard than MOX fabrication and irradiation, since it requires fewer steps and would generate fewer Pu-rich scrap and waste streams (for instance, pellet grinding will not be necessary).

For these reasons alone, it is apparent that the immobilization option should receive greater support from observers concerned about the vulnerability of fissile materials in Russia. Even MOX supporters must realize that immobilization could provide an important additional mechanism for expanding disposition capacity to 4 MT/yr without flooding the Western European market with subsidized MOX fuel from Russia, which is the likely alternative.

Despite these advantages, the immobilization program has been vehemently opposed by MOX promoters in government and industry, both in the West and in Russia. This opposition has been spearheaded by commercial interests in Western Europe and Japan who support MOX for self-serving reasons. The argument for opposing immobilization most often cited is a disdain for "throwing away" plutonium without utilizing its energy content. However, data continues to accumulate indicating that throwing away excess WG-Pu is actually cheaper than utilizing it as reactor fuel. A principal reason for the MD budget crisis cited by NNSA when it suspended the immobilization program is the fact that NNSA's revised 2001 estimates for the MOX program capital and operating costs were greater by 50% (or $1 billion) than DOEs previous estimates, whereas the estimate for the immobilization program has remained stable.

Based on the cost escalation of the MOX program, one can estimate that the life-cycle cost to dispose of the declared U.S. WG-Pu surplus (38.2 MT) via immobilization would be approximately $2 billion less than disposing of the surplus as MOX fuel. This difference takes into account the fact that the MOX facilities would have to be expanded to accommodate the additional equipment necessary for processing the diverse plutonium feed streams that otherwise would have been immobilized. These cost savings alone would be sufficient to fund a very large chunk of the Russian WG-Pu disposition program! Needless to say, NNSA is reluctant to release the data supporting these estimates --- as it completely undermines the case for suspending immobilization and pursuing MOX --- and is continuing to withhold the latest MOX cost figures from Congress.

NNSA is currently conducting a study to evaluate whether existing facilities --- the SRS F-Canyon in particular --- could be used in lieu of the Pit Disassembly and Conversion Facility (PDCF), the aqueous "polishing" line at the MOX fabrication facility, and the feed preparation equipment needed for impure Pu previously destined for immobilization. If NNSA decides to go this route, there will be a negative impact on the international transparency and verifiability of the U.S. plutonium disposition program. For example, if a warhead transparency regime is eventually established under the Trilateral Initiative, use of F-Canyon facilities to convert weapon pits into plutonium oxide would disrupt the IAEA's ability to maintain continuity of observation of items as they are transferred from storage and processed into oxide.

OBSTACLES TO IMMOBILIZATION

Even before the immobilization program was suspended, it faced numerous obstacles. First, the CIC immobilization approach has been heavily criticized by some observers, who argue that it does not meet the spent fuel standard, citing a report by Sandia National Laboratories that raises the possibility that the inner cans containing plutonium could be cleanly separated from the fission product glass by means of a detonation using shaped charges. Although supporters of this view have been unable to propose a credible scenario by which terrorists would be able to gain access to a storage site, use explosives to blow apart a canister, disperse highly radioactive glass fragments and escape without injury or detection, a National Academy of Scientists (NAS) panel examining this question concluded that more information was necessary to determine whether CIC met the spent fuel standard and recommended an experimental program "to clarify [immobilization's] resistance against on-site attack."^[3] Such a program, which would entail high-explosive tests on mock canisters, would increase the cost of the immobilization option and would be likely to cause delays, especially if NNSA were to decide to modify the concept. However, it is not clear that the benefits of such tests would be justified by the costs. CIC produces a waste form that is highly resistant to theft and unauthorized recovery of plutonium.

A more serious issue is related to the fact that the success of CIC is dependent on the operation of the DWPF, a plant with a troubled history. DWPF is currently unable to vitrify the highly radioactive, cesium-bearing salts in the SRS waste tanks, because the proposed method for concentrating the salts, in-tank precipitation, was acknowledged to be a failure and was abandoned in 1998. To date, DWPF has only produced canisters containing "sludge" from the waste tanks, which has low levels of cesium. Although DOE has initiated a process to develop an alternate technology for waste salt processing, it is unlikely to be available until after 2010.

This is a potential problem for immobilization because the Agreement establishes a radiation barrier requirement for the Pu-bearing disposal forms of 1 sievert (Sv) per hour at one meter (i.e. the IAEA standard for "self-protection") for 30 years after production, whereas the "sludge-only" canisters have dose rates two orders of magnitude below this value. Since according to the Agreement, the immobilization plant is slated to begin operation by 2008, the first several years' worth of immobilized plutonium cans would have to be stored for several years until HLW salt processing were available. Although the MOX track is technically capable of disposing of 2 MT/yr with the four available U.S. reactors, there is no margin for error without exceeding the MOX design parameters (such as the maximum MOX core loading of 40%).

However, this problem may be solved with a new source of cesium that may become available at SRS in the near future. SRS is currently receiving aluminum-based HEU-bearing material test reactor (MTR) spent fuel from foreign and domestic sources and will continue to do so until 2035. To stabilize this spent fuel for long-term storage and geologic disposal without reprocessing it, DOE is developing a process known as "melt-and-dilute" (M-D).^[4] Although originally a Clinton Administration initiative, the Bush Administration has requested funding for M-D in FY 2002. The schedule anticipates startup of a test facility in FY 02 and a full-scale facility in FY 08; the program is currently ahead of schedule.

The M-D process involves melting MTR spent fuel in a furnace together with depleted uranium at around 850C to form stable metallic ingots with uranium enrichments below 20%. It has been observed that under M-D process conditions, a significant amount of the cesium contained in the spent fuel is volatilized and can be adsorbed in an off-gas system utilizing zeolite filter beds. In experiments, after about 90 minutes at temperature, about 21% of the cesium in the spent fuel was measured to have been volatilized and retained in the zeolite filters.^[5]

One of the disposition options for the spent filter waste stream is to crush the zeolite and send it to DWPF for vitrification. The Cs-bearing zeolite could be added to sludge-only canisters or mixed with clean glass to provide an adequate dose rate for immobilization. According to the estimated spent fuel inventories presented in the SRS Spent Nuclear Fuel Management Final Environmental Impact Statement (SNF EIS), there will a considerable amount of Cs-137 contained in the assemblies that will be treated by M-D (about 35 MCi).^[6] Most of this inventory will have been received at SRS by 2008.

For the baseline immobilization plant throughput (1.3 MT/yr, or 46 canisters per year at 28 kg Pu per canister), the annual Cs-137 requirement would be about 0.55 MCi per year, based on an estimated minimum requirement of 12 kCi per canister to meet the dose-rate requirement specified in the U.S.-Russian Agreement.^[7] Using the data in the SRS SNF EIS, this could be comfortably accommodated by the planned throughput of the M-D plant (about 1/7 of the total inventory per year), which would make available beginning in FY 2008 about 1 MCi of Cs-137 annually, assuming that 20% of the Cs is trapped in the zeolite filters. By providing an interim source of cesium, this approach can allow Pu immobilization to proceed in accordance with the Agreement timetable until the HLW salt treatment process is up and running.

A number of questions would have to be addressed to confirm the feasibility of this approach, including verification that the necessary quantity of Cs could be obtained. If more Cs is required, it is possible that the M-D process could be modified to increase the amount of Cs evolved from the melt. Self-protection (i.e. Cs retention) will not be required for the M-D waste ingots because the uranium they contain will be diluted to an enrichment below 20%.

It is also conceivable that if immobilization were eventually implemented in Russia that a similar approach could be utilized there, if sufficient fission products from HLW were not available in a usable form. There is a large quantity of Russian-origin MTR spent fuel throughout the world, and the U.S. and Russia have held discussions in the past about initiating a spent fuel takeback program in Russia similar to the U.S. program. Introduction of M-D technology would be a viable alternative to reprocessing at Mayak for this fuel if it were returned to Russia.

THE FUTURE OF PLUTONIUM DISPOSITION

Although most of the G-8 nations have embraced the MOX approach for plutonium disposition in Russia on a rhetorical level, they have not been as eager to provide funding for the project. It now appears certain that the goal of raising the required $1 billion in investment costs by the G-8 Genoa Summit in July 2001 has fallen short by several hundred million dollars, and as a result the subject of plutonium disposition may not even be on the July 2001 meeting agenda of the G-8 heads of state.

Part of the reason for the shortfall is the policy of Germany, which is only willing to provide a financial contribution for immobilization. Another reason is that U.S. involvement in plutonium disposition in Russia continues to undergo a policy review by the White House, creating a climate of uncertainty around the effort. Consequently, many observers now believe that the profile of the program will be downgraded in the international community and will almost certainly experience significant delays.

Meanwhile, the suspension of the immobilization program in the U.S. will eliminate consideration of a technological alternative that could well have proven more cost-effective and feasible for both the U.S. and Russia. Although Germany is now a prominent supporter of immobilization, it is unlikely to assume a more prominent role as a developer of the technology in the wake of the suspension of the U.S. program.

Ultimately, these events may result in a renewed international focus on the most urgent problem, which is to provide adequate security for the large inventories of fissile materials in Russia that remain in a vulnerable condition. The emphasis of the international community on establishing a MOX infrastructure in Russia has proven to be a distraction from this more fundamental goal.

Endnotes