DOE REPROCESSING POLICY AND
THE IRREVERSIBILITY OF PLUTONIUM DISPOSITION
Edwin S. Lyman
Nuclear Control Institute
1000 Connecticut Avenue, NW Suite 804
Washington, DC 20036
The U.S. Department of Energy (DOE) is currently conducting an evaluation of nuclear materials throughout the weapons complex to determine if any may require the use of the F- and H-Canyons at the Savannah River Site to prepare them for interim storage and/or final disposal. Although DOEs current policy is to phase out operation of the canyons over the next several years, a number of possible campaigns have been identified which, if implemented in full, could delay shutdown until 2035.
This is undesirable, because prolonged canyon operation will interfere with efforts to negotiate verifiable agreements on halting the production of fissile materials for nuclear weapons and disposing of stockpiles of weapons plutonium no longer needed for nuclear arsenals. Therefore, there is an urgent need to resolve uncertainties in baseline stabilization plans for DOE-owned nuclear materials and develop alternatives, if possible, so that the option of canyon processing will be no longer required.
The negotiation of a multilateral treaty on the cessation of production of fissile materials for nuclear weapons is a key non-proliferation objective of the Clinton Administration, but little progress has been made since talks began in 1994. Some arms control analysts argue that institution of bilateral transparency measures on fissile material storage and production facilities in the U.S. and Russia could serve as an important first step toward achieving this challenging goal. 
Although both the U.S. and Russia have declared that they are no longer producing plutonium for nuclear weapons, the two countries continue to operate military reprocessing plants, primarily for nuclear waste management. In the U.S., the F- and H-Canyons at the Savannah River Site (SRS) are being used to stabilize a variety of residual materials in the Department of Energy (DOE) inventory. At the Seversk (Tomsk-7) and Zheleznogorsk (Krasnoyarsk-26) sites in Russia, reprocessing plants separate plutonium from the bulky and corrosion-prone spent fuel discharged from the three military production reactors which remain in operation to provide heat and electricity for the surrounding communities. In addition, Russia operates the "Mayak" plant at Ozersk (Chelyabinsk-65), which reprocesses both civilian and naval spent fuel.
The U.S. has stated that all plutonium-239 or highly-enriched uranium (HEU) separated as a result of stabilization programs will be designated as "excess" (prohibited from use for nuclear explosive purposes) and ultimately will be subject to International Atomic Energy Agency (IAEA) control.  Russia has stated that it is no longer producing plutonium metal for the military and has committed to submitting all weapons-grade plutonium separated in 1997 or thereafter to U.S. monitoring. 
While monitoring of declared plutonium stockpiles is important, a meaningful fissile cutoff agreement must also include provisions for verifying that fissile materials are not being clandestinely produced for weapons. At a minimum, these provisions should include the application of IAEA-type safeguards to all reprocessing facilities. However, applying international safeguards to old reprocessing plants that have been operated as military facilities in the past (and occasionally may still have classified missions) will be a costly and difficult endeavor. Because these plants were not designed to accommodate international safeguards, the IAEA may not be able to carry out its duties effectively without access to classified information or routine entry into high-radiation areas. Therefore, the task of verifying a moratorium on plutonium production for weapons would be much easier if these plants were simply shut down.
By closing old military reprocessing plants and submitting them to IAEA monitoring, the U.S. and Russia could also help to strengthen the existing non-proliferation regime. As the IAEA attempts to improve its monitoring and verification capabilities through the "strengthened safeguards" program (formerly known as "93+2"), additional unilateral gestures on the part of nuclear weapon states can help to offset the growing disparities in obligations between nuclear and non-nuclear parties to the Non-Proliferation Treaty (NPT), a frequent source of tension in international arms control discussions.
Both the U.S. and Russia are planning to embark on costly programs to dispose of plutonium that has become excess to military requirements. The conversion of excess weapons plutonium to the "spent fuel standard"  will only provide a significant barrier to rapid host-state breakout if access to large-scale reprocessing plants is precluded. Again, this can best be achieved by shutdown and decommissioning of facilities that are capable of carrying out reprocessing of spent fuel or equivalent materials (e.g. plutonium immobilized in vitrified high-level radioactive waste). Otherwise, it will be more difficult for the U.S. and Russia to convince each other, and the international community, that their plutonium disposition programs are effectively irreversible.
At the present time, there is no definite timetable for shutdown of the reprocessing plants either in the U.S. or in Russia. Under the DOE "phased canyon strategy," both SRS canyons will cease to operate by 2005.  However, a significant number of additional campaigns have been proposed, which if implemented in full could delay shutdown by six years or longer. In the most extreme case, H-Canyon could operate to reprocess domestic research reactor (DRR) fuel until 2035, at which time DRR shipments to SRS will cease.  DOE has no plans to permit international monitoring of these operations.
In Russia, the recently completed agreement with the U.S. on converting the cores of the production reactors to a high-burnup fuel cycle allows the reprocessing plants at Seversk and Zheleznogorsk to continue operating without application of bilateral safeguards. Although non-intrusive and approximate accountancy measures will be applied to the reactors themselves, transparency measures at the reprocessing plants will consist only of an annual unverified declaration.  This will not eliminate the risk that the plants could be used for reversal of plutonium disposition or for reprocessing of spent fuel from clandestine production reactors. In addition, Russia has never offered to place the Mayak reprocessing plant under safeguards, although it has accumulated a stockpile of over 30 tonnes of weapons-usable plutonium.
In both the U.S. and in Russia, there is significant pressure to keep reprocessing plants operating or in operational condition for an open-ended period. Supporters of this policy range from those who wish to preserve employment at contracting defense facilities (and perhaps envision a future commercial role for the canyons) to conservatives who wish to maintain a capability for weapons plutonium production. However, for the reasons pointed out above, this policy conflicts with arms control goals and interferes with increased bilateral stability. The U.S. and other Western nations considering large financial investments in a plutonium disposition program in Russia should also consider whether such expenditures are worthwhile as long as Russia continues to operate unsafeguarded reprocessing plants.
Obstacles to Canyon Shutdown
Although the defense missions of the SRS canyons have largely ended, DOE has identified a number of other tasks that may be carried out using canyon facilities. Over the past several years, the canyons have been used to stabilize SRS materials identified as posing immediate health and safety risks, such as F-Canyon plutonium solutions and corroded Mark-31B production reactor targets. Now that many of the most critical campaigns have been completed, DOE is considering use of the SRS canyons both for stabilizing off-site at-risk materials and for reprocessing materials throughout the DOE complex that do not pose near-term health and safety risks.
In November 1997, DOE proposed that three types of materials now in storage at the Rocky Flats Environmental Technology Site (RFETS) --- plutonium fluoride residues, sand, slag and crucible (SS&C) and scrub alloy --- be shipped to SRS for reprocessing.  DOE estimates that executing these proposed missions would delay the shutdown of F-Canyon from FY 1998 to mid-FY 2000 and would not affect the projected H-Canyon shutdown date (now estimated to be 2005). 
Also in 1997, Secretary Pena requested a "complex-wide evaluation to validate there is no additional material that could be readied for final disposition only by use of the canyon facilities." This study, which is known as the Processing Needs Assessment (PNA), was completed in March 1998. The PNA recommended that DOE consider six additional categories of materials, including Hanford Single Pass Reactor spent fuel and RFETS classified plutonium metal parts, for processing in SRS canyon facilities, and estimated these campaigns would extend the H-Canyon lifetime by one to two years and the F-Canyon lifetime by less than one year. 
In addition, the PNA identified a larger number of classes of materials within the DOE complex for which the current stabilization baselines, primary backup approaches and/or final disposition pathways were uncertain, and concluded that "...a processing capability (not necessarily a canyon) [may be required] for several years past 2004." The PNA referred to these materials as "potential" candidates for canyon processing. Plutonium residues in this category include RFETS impure plutonium oxides (<50% Pu) and pyrochemical salts, with projected campaigns of 3.5 years in H-Canyon and 3 years in F-Canyon, respectively. Roughly 2-2.5 tonnes of plutonium would be separated during these two campaigns. DOE spent fuels in this category include the Hanford N-Reactor spent fuel inventory, which contains 4 tonnes of plutonium and is projected to require a 4-5-year campaign in (a suitably modified) F-Canyon.
Materials that are currently under consideration for SRS canyon processing can be roughly divided into three categories. These are, in decreasing order of priority:
(a) Materials that pose safety risks but may not have a proven stabilization alternative available by the "need date" dictated by DOEs current cleanup schedule, such as RFETS sand, slag and crucible (SS&C);
(b) Materials that can be easily stabilized, but which may require further processing to be brought into strict compliance with the current long-term storage standard (DOE-STD-3013-96), such as RFETS impure plutonium oxides (<50% Pu);
(c) Materials that can be easily stabilized (or are already acceptable for long-term storage), but for security and/or safety reasons may require further processing prior to ultimate disposition, such as RFETS pyrochemical salts and SRS Al-clad HEU spent fuel.
DOEs current strategy for nuclear material stabilization is driven by the commitments it made in response to the concerns expressed in the Defense Nuclear Facilities Safety Board (DNFSB) Recommendation 94-1.  In the "DNFSB 94-1 Implementation Plan" (94-1 IP), DOE committed to stabilization and repackaging of "high-[safety] risk" residues by the end of 1997 (a deadline that has already been missed), and all lower-risk materials by 2002.  This schedule then determined the "need dates" by which time a particular stabilization technology would have to be fully demonstrated. However, because of delays in the development of baseline technologies, as well as outstanding technical issues with regard to storage and disposal requirements, there is at present a high risk that "need dates" will not be met for many material categories. For these cases, the PNA concluded that the SRS canyons may be needed to provide a backup stabilization capability.
To determine the programmatic risks associated with baseline and backup stabilization strategies, the PNA used a methodology adopted from the aerospace industry known as "technical maturity" (TM) scoring.  This procedure is an attempt to quantify the uncertainties associated with a particular stabilization approach and combine them into a single number. The risk of failure of the approach can then be calculated from the TM score and the "need date." While the TM approach can be useful, it can also be extremely misleading unless the results are taken in context. In particular, the TM approach, coupled with the extremely optimistic 94-1 IP need dates, results in an inherent bias against new and innovative technologies and in favor of established ones. In most cases, technologies that are not fully developed have poor TM scores and compare unfavorably to canyon processing.
However, the TM methodology does not take into account a number of other considerations which may favor the development of new technologies, including the non-proliferation benefits of early canyon shutdown, reduced environmental impacts and avoidance of intersite transport. The methodology also does not account for the possibility that the TM scores of new technologies can be improved rapidly by controllable factors, such as funding levels and administrative support. Seen in this way, the conclusions of the PNA can be interpreted not as a "shopping list" for the canyons, but as a blueprint for placement of resources to avoid further use of the canyons.
The programmatic risk associated with a stabilization strategy is also strongly dependent on the "need date" for availability of the technology. The 94-1 IP "need date" of 2002 is only approximate and was not derived from a detailed technical assessment of the health and safety risks of all materials in their current storage configurations. Relatively stable materials, such as those in categories (b) and (c) above, can be prepared for safe interim storage using processes with low technical risk. The opportunity would then be available to make a less hurried decision about the best way to prepare these materials for final disposition.
A Strategy for Accelerated Canyon Shutdown
If DOE decides that delaying shutdown of the SRS canyons to 2010 or beyond is undesirable, it will have to implement a unified strategy to resolve outstanding uncertainties in the viability of existing stabilization baseline and backup technologies. Some of these questions may be resolved by modifying administrative requirements, such as DOE-3013 or safeguards termination limits (STLs). Others may require a renewed effort to develop innovative processing and packaging technologies. In addition, a sensible policy needs to be devised on how to coordinate development of geologic repository waste acceptance criteria (WAC) and disposal strategies for wastes with relatively large repository performance uncertainties, such as Al-clad, Hanford N-Reactor and EBR-2 spent fuels. A greater focus on enhancement of engineered barriers to compensate for potential inadequacies in waste form performance (the so-called "supercanister" concept) may eliminate the possibility that these fuels will have to be eventually reprocessed.
An accelerated canyon shutdown strategy should have the following elements:
- Commitment by DOE to restrict reprocessing at SRS to those materials that pose immediate safety risks;
- Identification of candidate materials for canyon processing which could result in a significant (>6 month) delay in canyon shutdown;
- Classification of materials according to hazard and revision of "need dates" to more accurately reflect risk;
- Safety- and security-based revision of administrative limits (3013 standard, STLs, WACs) to avoid the unnecessary reprocessing of stable materials and those of low proliferation concern;
- Aggressive development of one or more "multipurpose processing technologies" that can accommodate a wide range of residue types and do not require plutonium separation;
- Pursuit of innovative packaging and backfilling solutions to reduce the risks of direct disposal of DOE spent fuels to a level comparable to that of commercial spent fuel.
To its credit, DOE is beginning to examine some of these issues, both individually and in the context of the complex-wide Environmental Management Integration (EMI) effort. However, the goal of EMI is to minimize the cost and time required to accomplish the overall EM mission by considering the DOE complex as a whole, and is neutral on the question of whether extended use of the canyons should be part of this effort. The objective of accelerated canyon shutdown can only be realized if it is imposed on EMI as an explicit constraint.
A few examples of how this strategy might be applied to different classes of materials are given below.
1) Spent fuel
(a) Al-clad fuel:
The list of materials considered "potential" canyon candidates by the PNA includes the inventory of aluminum (Al)-clad spent fuel assemblies from foreign and domestic research reactors (FRRs and DRRs) which is scheduled to be received at SRS, of which most contain uranium enriched to greater than 20% U-235 (highly-enriched uranium, or HEU). The bulk of this material is in good condition and DOE believes that it can be safely stored in carefully controlled wet storage until alternative treatment and/or packaging technologies become available.  However, reprocessing remains an option because it has been the traditional disposition route for Al-clad HEU spent fuel at SRS. Also, some observers have raised questions with regard to the acceptability of direct geologic disposal of this material because of the risks of accelerated corrosion of the metallic fuel cladding and matrix, occurrence of repository criticality events and theft of the HEU for weapons use.
The PNAs designation of Al-clad RR spent fuel as a potential canyon candidate is consistent with DOE's path forward for the material, as set out in the May 1996 Record of Decision on a Nuclear Weapons Nonproliferation Policy Concerning Foreign Research Reactor Spent Nuclear Fuel (FRR ROD).  In the FRR ROD, a three-point management strategy was outlined. First, DOE would immediately begin a program to demonstrate the viability of non-reprocessing treatment and/or packaging technologies to prepare Al-clad spent fuel for geologic disposal. If none of these alternatives had been demonstrated by the year 2000, DOE would then consider reprocessing the spent fuel as long as the F-Canyon were still operating to stabilize at-risk materials. In addition, DOE would carefully monitor the condition of FRR spent fuel in wet storage, and would permit reprocessing of fuel elements in the F-Canyon should safety problems arise.
The three-point strategy in the FRR ROD was constructed to balance the concerns of a number of stakeholder groups. The state of South Carolina, which supports additional reprocessing missions at SRS, opposed the return of FRR spent fuel to SRS unless it were subsequently reprocessed. In contrast, some national arms-control organizations supported the takeback of FRR HEU spent fuel on non-proliferation grounds, but opposed the reprocessing of any material that did not pose acute safety risks. Although it was an acceptable compromise at the time, the FRR ROD was flawed in several respects and set the stage for future complications.
For example, SRS will be accepting FRR and DRR spent fuel until around 2010 and 2035 respectively, yet the policy is unclear regarding the fate of spent fuel accepted after the canyons finally cease operating to stabilize at-risk materials. If an alternative packaging or treatment technology has been demonstrated by 2000, then it can be used for the entire spent fuel inventory. However, if the alternative technology program is not successful by that date, DOE will have three options. It will either have to continue the search for alternatives, locate enough at-risk material throughout the complex to keep the canyons operating through 2035, or amend the FRR ROD to permit operation of the canyons solely for reprocessing stable RR spent fuel. 
Another problem with the formulation of the FRR ROD is that it does not provide objective criteria for determining whether an alternative treatment and/or packaging technology is "ready for implementation." This raises the possibility that the results of the program can be manipulated to support a political decision on the future of the SRS canyons. This is a concern because there is a considerable lack of enthusiasm for the alternatives development program at SRS, where it is based, as well as strong support for continued reprocessing at SRS among powerful members of Congress who play key roles in authorization and appropriation of DOE funding. Moreover, the arbitrary imposition of the year 2000 as a need date for demonstration of alternatives may not permit sufficient time for resolution of outstanding technical issues, such as whether Al-clad spent fuel has adequate durability under typical repository conditions. Since the FRR and DRR spent fuel inventory may cause the biggest delay in the current schedule for canyon shutdown, DOE should be making development of alternatives a high priority.
As a result of the inherent bias in the TM scoring system, the PNA gives a pessimistic appraisal of both direct disposal of Al-clad spent fuel and the backup technology under development, "melt-and-dilute." Melt-and-dilute is a non-aqueous process which converts Al-clad HEU spent fuel into stable, low-enrichment ingots without separation of plutonium and HEU from fission products. Even though the technology is straightforward, and will eliminate the concerns regarding repository criticality and durability of Al-clad spent fuel, it receives a low TM score because it is currently in development and the necessary facility remains to be built. However, with adequate funding and support, the TM can be expected to improve rapidly in a short time.
Moreover, because the bulk of this fuel can be safely stored for a long time, the true need date for availability of a disposal technology is decades away. The need date of 2000 imposed by the ROD was not based on safety and health considerations but was a policy decision (dictated largely by the anticipated availability of the canyons). Thus the rapid timetable mandated for development of a disposal technology is not technically based and should be extended.
(b) N-Reactor fuel:
The PNA includes the Hanford N-Reactor spent fuel on its list of "potential" canyon missions because of concerns regarding the performance of uranium metal fuel in a repository. However, common sense dictates that 2,100 MT of spent fuel, much of it in degraded condition, is not a suitable candidate for transport clear across the continental United States. In addition, use of the canyons for this purpose would require construction of a new head-end at an estimated cost of $200-400 million, followed by a four-to-five-year campaign that will increase the separated plutonium stockpile by 4 tonnes. The N-Reactor fuel inventory is clearly a problem that should be resolved at Hanford.
The designation of the current baseline as "high-risk" also does not appear warranted. While there is insufficient information at present about the performance of the fuel in a repository, there is little risk in the path to conditioning the fuel for interim dry storage. If it is determined that the fuel ultimately will require stabilization for final disposal, a dry, controlled oxidation process can be used, and the resulting oxide can be immobilized together with Hanford high-level wastes (HLW). Since design of the HLW immobilization plant at Hanford is still in the early stages, there is ample opportunity for integrating such a process into the plant.
(a) RFETS classified metal parts:
The PNA recommends that this category of material be sent to SRS for a 2-year HB-line campaign to convert it to an unclassified form. DOE should consider a variant of the non-aqueous ARIES process now under development at LANL as an alternative. The ARIES plant is being designed to handle a variety of different feeds, both in pit and non-pit form.
(b) RFETS impure plutonium oxides (<50% Pu):
Here is another example of how strict application of the TM methodology can result in a questionable decision. The baseline strategy for this material is to stabilize it using the Plutonium Storage and Packaging System (PuSPS) developed by British Nuclear Fuels Limited (BNFL), for storage in 3013 cans.  However, concerns have arisen that the PuSPS process, which was designed for pure metals and oxides, may not be able to stabilize impure oxides to the same level of safety as pure oxides that meet DOE-STD-3013. There are questions concerning the accuracy of the "loss-on-ignition" (LOI) test specified in DOE-STD-3013 as a measure of water content, especially for impure oxides (<88% Pu).  At the present time the process parameters required to stabilize impure oxides to a degree of safety comparable to pure oxides are undefined. There are also uncertainties related to the performance of the PuSPS system for impure feeds. The low TM score resulting from these "immaturities" in storage requirements and process development led the PNA to recommend that impure oxides be considered a "potential" candidate for a 3.5-year campaign in H-Canyon.
However, with appropriate surveillance, the risks of storing this material are low, and even if it could not be brought into strict 3013 compliance the risks would most likely still be low. The time and money required to send this material through the canyons would be better spent carrying out the laboratory work necessary to validate or, if necessary, to revise DOE-STD-3013 and/or to adjust PuSPS accordingly so that it will be capable of stabilizing impure oxides for safe storage.
c) RFETS pyrochemical salts:
The preferred alternative in the 1997 draft Environmental Impact Statement (EIS) on RFETS residues for stabilization of pyrochemical chloride salts is pyro-oxidation, followed by a salt washing process (water leach at LANL for calcium chlorides and salt distillation at RFETS for other types).  The PNA ranks this baseline as "high risk" and consequently considers this material to be a "potential" candidate for a 3-year campaign in F-Canyon. However, the risk is not associated with the pyro-oxidation stage, which is a well-established technique using existing equipment to convert plutonium and other reactive metals to oxide embedded in a stable salt matrix.  The low technical maturity of the baseline is associated with the subsequent salt distillation stage at RFETS, since the final process design has not yet been optimized and the equipment is yet to be built.
Pyro-oxidation is itself sufficient to stabilize the salts for safe interim storage. However, additional processing may have to take place prior to final disposal because the salts have, on average, very high plutonium concentrations and are regarded as attractive materials for theft. Nearly the entire salt inventory exceeds the current DOE safeguards termination limit (STL) of 0.2 weight-percent plutonium appropriate for the material, and therefore cannot be directly disposed of in the Waste Isolation Pilot Plant (WIPP), where no safeguards controls are planned. DOEs preferred alternative is to use salt washing for salts that exceed 10 weight-percent Pu to obtain a lean fraction which will meet the STL and a rich fraction which can be stored according to 3013 and transferred to the Office of Fissile Material Disposition (MD).  For salts with Pu concentrations below 10 weight-percent, RFETS is seeking a variance from the STL to permit direct disposal in WIPP in small-capacity "pipe overpack" containers. However, some parties within DOE have raised concerns about granting such a high STL variance, and the viability of this alternative is still in question.
It is paradoxical and wasteful to spend time and effort extracting one metric ton of plutonium from the salts for non-proliferation reasons when the goal of MD is to eventually contaminate and dilute it once again. Use of a direct disposal approach would make much more sense. This would entail either dilution to below the STL and disposal in WIPP or "immobilization" in a glass or ceramic and "can-in-canister" disposal in SRS vitrified high-level waste canisters. Because the former approach could result in a large increase in waste volume unless a high STL variance were approved, the latter approach, which would meet the "spent fuel standard," may be preferable.
Time and resources would be needed to develop an immobilization approach capable of handling the residual chlorides in the pyro-oxidized salts. One concept that shows promise is the Glass Materials Oxidation and Dissolution System (GMODS), which could separate volatile chlorides from actinide oxides and produce a waste form suitable for can-in-canister disposal in a single step.  However, development of GMODS for this application was proposed but not pursued by DOE.
d) "Rich" materials (> 50% Pu):
The PNA refuses to close the door on canyon processing of "rich" materials such as pure metals and oxides, despite the fact that these materials can be easily stabilized for long-term storage using PuSPS, and MD has said that it will accept all materials in this category. The concern is that these materials may contain impurities that will cause problems with the two disposition approaches that have been selected, fabrication and irradiation of mixed-oxide (MOX) fuel and ceramic immobilization. MD is in the process of developing "acceptance criteria" for the plutonium it accepts for disposition, and it may ultimately determine that some "rich" plutonium may contain unacceptable levels of impurities.
However, the disposition program is scheduled to continue through the 2020's, and it would be foolish to preserve the canyons for this length of time as a backup for "cleaning up" MD materials. If impurity issues eventually render MOX infeasible, then immobilization will always be available, which has an inherently greater flexibility than MOX to accommodate variations in feed quality. Blending strategies are also being developed to produce homogeneous and acceptably pure feeds. DOE should be able to rule out potential use of the canyons for this purpose without concern of undue risk to the disposition program.
DOE is also considering using existing facilities both within or adjacent to the SRS canyons for other activities related to disposition, such as can-in-canister immobilization. This could be an acceptable option if it were possible to segregate and decommission the dissolution and separations equipment, and to allow full access by the IAEA to the disposition facilities. However, for the reasons discussed above, it is not clear whether this will be possible. If not, then the additional capital expenditure for a new disposition facility would clearly be justified from an arms control perspective.
Conclusions: The Benefits of New Technology Development
DOEs strategy for nuclear material stabilization puts little emphasis on the development of innovative technologies, and relies instead on utilization of historic processes, where possible. This strategy appears to be advantageous in most cases from the standpoints of cost and timeliness, and it is clearly appropriate for unstable materials for which no viable alternatives are immediately available.
However, historical processing technologies were designed to recover fissile materials from waste streams, whereas the goal today is to "unseparate" excess fissile material stockpiles and stabilize residues for storage and disposal. Separation of plutonium from waste is an intermediate and unnecessary step that is likely to increase overall non-proliferation risks and environmental impacts relative to direct immobilization and disposal. For materials which do not pose acute risks, these considerations may warrant development of new non-separation technologies, even if this strategy requires greater near-term expenditures and a relaxation of aggressive cleanup schedules.
Construction of new stabilization facilities will provide the opportunity for incorporating features into the design that will facilitate application of international safeguards, as is being done at the new Actinide Packaging and Storage Facility (APSF) at SRS.
The benefits of new non-separation technology development to U.S. industrial leadership overseas should also be considered. Aqueous separation is a 1940s-era technology which is widely disseminated. However, there is a niche in the market for more proliferation-resistant technologies for nuclear waste management.
For example, there is a growing need, both domestically and overseas, for innovative ways to manage Al-clad spent fuels. Many research reactor operators currently deal with their spent fuels by reprocessing them or sending them to the U.S. under the takeback program. However, both of these options are of limited duration. As reactors convert to LEU fuels, the incentive to reprocess the fuels will be reduced, and ultimately nations will have to develop direct disposal strategies. If the results of ongoing research indicate that direct geologic disposal of these fuels is not feasible, a U.S.-designed "melt-and-dilute" (or "melt-and-recast" for low-enrichment fuels) system could be a valuable export to these nations.
In Russia, the three production reactors will continue to use Al-clad fuel even after their cores are converted, and the fuel may be HEU-based as well (although progress is being made toward a cooperative U.S.-Russian program to develop an LEU-based fuel). Minatoms intention is to continue reprocessing the spent fuel. A joint program to develop a "melt-and-dilute" system could provide a path forward for the near-term shutdown of the Russian separations plants.
The U.S. and Russia have an historic opportunity to shut down their Cold War-era reprocessing plants and lock in an unofficial weapons plutonium production moratorium which has been in effect for several years. Unfortunately, inertia, lack of vision and a refusal to appreciate the value of such an act may delay this event for decades and jeopardize its ultimate accomplishments.
Update: September 1998
There have been several developments since the attached paper was completed in April 1998 which have a bearing on the discussion. These include:
- The timetable for completion of F-canyon processing of Rocky Flats Pu fluorides, SS&C and scrub alloy has slipped from mid-FY 2000 to the end of FY 2001.
- Progress has been made at SRS on development of the "melt-and-dilute" process for aluminum-clad research reactor fuel, including an analysis demonstrating the feasibility of using existing 105L-Reactor facilities for the process.
- DOEs Office of Safeguards and Security approved the requested STL variances for residues at Rocky Flats with Pu contents below 10 weight-percent. This enabled DOE to change the Preferred Alternative in the RFETS Residues EIS for most electrochemical salts from salt distillation to repackaging with blend-down and direct disposal at WIPP. While this action greatly reduces the possibility that these salts will have to be reprocessed at SRS in the future, it is unclear whether the variance is truly consistent with DOE policy for management of other plutonium-containing materials.
- In the Draft Surplus Plutonium Disposition EIS, DOE has proposed that if further research determines that aqueous processing will be necessary to convert pure plutonium forms into a feed suitable for MOX fabrication, a new "plutonium polishing module" would be constructed as an adjunct to the pit conversion and/or MOX fabrication plant. The Draft EIS does not appear to consider the possibility that existing F- or H-Area facilities will be used. However, the "Pu polishing" plan does not have the status of a "proposed action" in the Draft EIS, so other possibilities may eventually be proposed in future NEPA actions.
- A National Research Council report was issued which stated that DOE "should have given more careful consideration to the conventional reprocessing option" for treating Al-based RR fuel. The report claimed that this action "does not appear to be in conflict with the DOE decision to phase out reprocessing at Savannah River." However, the report takes an extremely narrow view of the security issues involved and does not consider many of the broader issues discussed in the paper.
1. Fetter and von Hippel (1995).
2. See, e.g., DOE (1996).
3. Gore (1997).
4. NAS (1994).
5. DOE (1998).
6. DOE (1998).
7. Gore (1997).
8. DOE (1997a).
9. DOE (1997b).
10. Guevara and Klipa (1998).
11. DNFSB (1994).
12. DOE (1995).
13. DOE (1997c).
14. DOE (1996).
15. DOE (1996).
16. Since the FRR ROD was issued, DOE has determined that should the decision be made to reprocess RR spent fuel, H-Canyon, and not F-Canyon, would be used. Thus the ROD may hae to be amended before this action could proceed. One consequence of this change is that H-Canyon, unlike F-Canyon, will produce purified HEU solution prior to isotopic dilution. Such a process would present even greater challenges to international monitoring under a fissile material cutoff regime.
17. SRS does not currently plan to operate the canyons beyond 2010 for the relatively small number of RR spent fuel elements it expects to receive annually after that date. However, this underscores the fact that a direct disposal solution for at least a portion of this fuel will have to be developed in any case.
18. Strictly speaking, this material cannot meet DOE-STD-3013 because it contains a plutonium concentration below the 3013 threshold of 50 weight-percent. However, there are efforts underway to change this adminstrative limit, which has no safety justification.
19. A limitation on the water content within a package is essential for the safe long-term storage of plutoium. This limitation is specified in DOE-3013 in terms of the weight loss experienced by a plutonium sample when heated, otherwise known as the loss-on-ignition (LOI).
20. Since issuance of the draft EIS, there are indications that DOE has decided not to pursue distillation at RFETS, but only at LANL, perhaps because of its low technical maturity.
21. DOE (1997a).
22. DOE (1997a).
23. Forsberg et al. (1995).
[DNFSB (1994)] U.S. Defense Nuclear Facilities Safety Board, Recommendation 94-1, May 26, 1994.
[DOE (1998)] U.S. Department of Energy, Office of Environmental Management, Accelerating Cleanup: Paths to Closure, Draft, DOE/EM-0342, February 1998.
[DOE (1997a)] U.S. Department of Energy, Draft Environmental Impact Statement on Management of Certain Plutonium Residues and Scrub Alloy Stored at the Rocky Flats Environmental Technology Site, DOE/EIS-0277D, November 1997.
[DOE (1997b)] U.S. Department of Energy, Savannah River Site Chemical Separations Facilities Multi-Year Plan, September 1997.
[DOE (1997c)] U.S. Department of Energy, Plutonium Focus Area, Research and Development Plan, DOE/ID-10561, Revision 2, November 1997.
[DOE (1996)] U.S. Department of Energy, Record of Decision on a Nuclear Weapons Nonproliferation Policy Concerning Foreign Research Reactor Spent Nuclear Fuel, May 13, 1996.
[DOE (1995)] U.S. Department of Energy, Defense Nuclear Facilities Safety Board Recommendation 94-1 Implementation Plan, February 28, 1995.
[Fetter and von Hippel (1995)] Steve Fetter and Frank von Hippel, "A Step-By-Step Approach to a Global Fissile Materials Cutoff," Arms Control Today, October 1995, 3.
[Forsberg et al. (1995)] Charles Forsberg, Edward Beahm, George Parker, Jeff Rudolph, Karla Elam and Juan Ferrada, "Conversion of Plutonium Scrap and Residue to Borosilicate Glass Using the GMODS Process," Proceedings of the U.S. Department of Energy Plutonium Stabilization and Immobilization Workshop, Washington, DC, December 12-14, 1995, p. 177.
[Gore (1997)] The White House, Office of the Vice President, Gore-Chernomyrdin Commission, Fact Sheet on the U.S.-Russian Plutonium Production Reactor Agreement, September 23, 1997.
[Guevara and Klipa (1998)] Arnold Guevara and George Klipa, U.S. Department of Energy, Nuclear Materials Processing Needs Assessment Presentation to EM-60, March 16, 1998.
[NAS (1994)] National Academy of Sciences, Management and Disposition of Excess Weapons Plutonium, National Academy Press, Washington, DC, 1994.
What's New Pu Disposition Home Page