Those areas that the committee believes are most important for reducing the costs of the D&D of the GDPs are identified in this chapter. These major recommendations address a variety of approaches for reducing D&D costs relative to previous cost estimates and specific technical demonstrations required, as well as providing broader suggestions for planning, management, and regulatory coordination that should lead to a cost-effective D&D process. Management of the DUF6 (depleted uranium hexafluoride) inventory is also addressed.
The committee's review of the two D&D cost estimates by Ebasco and TLG of $16.1 and $13.9 billion, respectively, identified significant opportunities for cost reduction. The case for lower cost is supported by the actual cost data reported for the BNFL D&D of the Capenhurst GDP in the United Kingdom. That effort cost about $160 million, a hundred times lower than the Ebasco cost estimate (see Chapter 4). Compared to the U.S. GDPs, the Capenhurst plant was much smaller, of somewhat different design, and incorporated some different materials. However, even if the Capenhurst D&D costs are scaled-up to the situation for the U.S. GDPs, these scaled-up costs are a factor of 8 to 15 times less than the cost estimate of $16.1 billion.
There have also been a number of D&D projects conducted in the United States for nuclear power reactors. The D&D of the Shippingport Atomic Power Station, for example, indicated the importance of a streamlined management approach and thorough planning, with a decommissioning plan incorporating a detailed technical baseline, as part of developing a cost estimate and engaging in the actual D&D.
Based on reported D&D costs for the Capenhurst GDP, experience from other D&D projects, and a review of the cost estimates, the committee concluded that there are significant opportunities to reduce the D&D costs of the GDPs. While the magnitude of the potential cost savings is uncertain, it could equal or exceed 50 percent of the current Ebasco estimate of $16.1 billion.
Recommendation 1. The technical and management approaches used successfully for the D&D of the Capenhurst gaseous diffusion plant and for recently completed D&D projects with U.S. power reactors should be carefully considered by the DOE to reduce costs for the D&D of the U.S. GDPs.
Coordinated planning at the DOE headquarters level, as well as across the complex of the three GDPs and at each site, will all be required to ensure that D&D is integrated effectively with other operating or cleanup activities at the sites and that resources are used effectively. DOE headquarters-level planning would outline decisions on D&D financing, on integration of D&D with other DOE programs; and on the broad contracting, regulatory, and stakeholder involvement approaches for D&D. A complex-level master plan for the three-plant enrichment complex as a whole would coordinate such decisions as the sequence of plant cleanup, the priority actions to be taken, allocation of funds among the sites, and cleanup strategies, including approaches to waste management and recycling. Priority setting would be based on analysis of relative risks, costs, and social values. Site-specific plans are also needed to coordinate D&D, environmental remediation efforts, and management of the DUF 6 inventory at each of the three sites. For example, it would be costly if previously cleaned areas of soil and the groundwater were recontaminated during D&D operations. For the Oak Ridge GDP, the site plan should be coordinated with the plans for the whole Oak Ridge Reservation. It was not clear to the committee that the idea of cleaning the areas that the buildings occupy to greenfield status makes much sense if other parts of the sites remain contaminated or if DUF6 continues to be stored on site.
Coordinated planning efforts are also called for by budgetary uncertainties. If planned contributions to the D&D Fund continue, a total of about $7.2 billion will accrue. The fiscal year contributions to the fund limit the rate at which D&D could be accomplished. Since the inception of the fund, most of the projects funded have been for environmental remediation. This funding allocation may be driven by the priority given to such important problems as remediation of waste ponds or contaminated soil or groundwater. If such a spending profile persists, however, and better controls on the fund are not established, funding may not be available for D&D of the GDPs.
A detailed D&D plan, which the committee believes should not take more than 18 to 24 months to prepare, needs to be developed delineating the sequence of activities necessary to incrementally achieve the D&D of the facilities. The sequence of tasks should be based on considerations of cost and risk. Uncertainty regarding the final end state should not delay the development of an initial D&D plan, which would be refined as circumstances change. The decommissioning plan should incorporate all major assumptions (technical, cost, and institutional), a proposed management organizational structure for both DOE and the decommissioning operations contractor, tradeoff studies for determining an optimized decommissioning sequence, a detailed work breakdown structure, and a detailed cost estimate and schedule. The detailed work breakdown structure would specify, for example, the sequence of steps required to remove the equipment, decontaminate the equipment and buildings, demolish the buildings, recycle material, and dispose of wastes. A new cost estimate would be derived from the detailed D&D plan and used as a basis for soliciting competitive bids from the private sector for execution of the work; a competitive bidding process would be the most likely process to identify cost-effective approaches.
Recommendation 2. DOE should develop three plans, namely, headquarters-level, GDP complex-level, and GDP site-level, that address and integrate the
D&D of the facilities, environmental remediation activities, and management of the DUF6.
Contracting And Management
Although the committee's analysis indicates that D&D could be accomplished much more cost effectively, the cost savings may not be achieved without a significant change in the management and contracting approach through which DOE oversees and conducts project planning and execution. DOE has traditionally managed its major projects at operating (or formerly operating) sites using an M&O (management and operating) contractor. This management approach, which was assumed in the Ebasco estimate, includes multiple layers of management and results in an unnecessarily high ratio of the cost of management and professional services to the cost of execution of the physical decommissioning. This approach is inherently more expensive than that used in the Capenhurst D&D and in other private sector D&D projects. Large reductions in D&D project costs are unlikely to be achieved under the currently proposed project management approach using multiple prime contractors. Experience on other DOE projects demonstrates conclusively that this concept results in much higher costs relative to projects managed by other government agencies and the private sector.
A more cost-effective approach would be to use a management structure employing an independent decommissioning contractor, who would assume total responsibility and accountability for all aspects of the D&D. For example, a DOC (decommissioning operations contractor) approach was used for the Shippingport D&D effort. Such a contractor would be selected through an open competitive bidding process based on demonstrated experience in successful management of D&D projects. Improvements in the cost effectiveness of projects could be achieved by incorporating financial incentives in contracts to the DOC and to all subcontractors. As part of the process of reducing costs, every aspect of the D&D effort needs to be examined closely to identify the most cost-effective alternatives for accomplishing each task and to eliminate redundant and excessive management oversight, while complying with health, safety, and environmental protection requirements.
Recommendation 3. An independent contractor should be selected through open competition and should be assigned total responsibility and accountability for all aspects of the assigned D&D work.
Prioritized Cost And Risk Reduction
Currently, there is no quantitative risk analysis of the nonoperating plant at Oak Ridge. The committee believes that, because people living near the enrichment plant sites are not exposed to the contaminants within the buildings, the near-term risk to the public is quite low. Although uranium is radioactive, its primary risk to human health is its chemical hazard upon ingestion or inhalation. Uranium in the buildings is contained, for the most part, inside process equipment and does not present a hazard. However, the deposits of highly enriched uranium require a safeguards and security regime to be maintained. The vast majority of the uranium at the GDP sites is in the DUF6: a major release of DUF6 is a very unlikely event. For such a large
cylinder rupture, the HF (hydrogen fluoride) concentrations at the site boundary would depend on atmospheric conditions. Very low concentrations of HF below hazardous levels are odorous, presenting a warning to on-site employees; however, a rapidly moving cloud of HF from a short-term, large release could potentially engulf on-site employees under certain conditions. The primary risk during D&D will be worker risk, not only from possible chemical or radioactive contamination but from industrial accidents. Minimizing such risks will require strict adherence to adequate worker health and safety protection. Efforts to reduce D&D costs cannot come at the expense of protecting workers, public health, or safety.
Moving forward expeditiously with the planning and execution of D&D is important for a number of reasons. In the event of delays in D&D, annual surveillance and maintenance costs will lead to substantial expenditures. With time, deterioration of the buildings can exacerbate these costs, and risks to individuals will increase. The risk posed to the workers, the public, and the environment by the present situation may not be large, but it will probably increase with time. The continued presence of highly enriched uranium entails an expensive security and safeguards regime. Furthermore, as noted above, there are uncertainties about the future availability of funds for D&D.
Planning for D&D does not imply that no cleanup is achieved until a plan is finalized. Developing an adequate and robust D&D strategy, detailed plan, and cost estimate will take time. Uncertainties over final decisions, such as site release criteria, end uses of the sites, free-release criteria for cleaned materials, and location of low-level radioactive waste sites, must be considered in developing plans and estimated costs. The D&D of the three sites could very well occur over a period of several decades, and political priorities and budget commitments may also change over that time.
Because of these uncertainties, the committee proposes that a prioritized cost- and risk-reduction approach be taken. Such an approach identifies conditions at the sites that, if not quickly remediated, could lead to increased risks or costs as a result of delay. For example, it appears to the committee that the removal of the deposits of highly enriched uranium from the Oak Ridge process equipment should be a first priority. This approach would reduce the costs of enforcing safeguards and security and reduce the risk of nuclear criticality accidents. The optimal sequence of D&D actions will be decided during the detailed decommissioning planning and cost estimation process. A prioritized cost- and risk-reduction approach allows initiating D&D operations during the planning process by identifying priorities for immediate cleanup. It also allows scheduling of projects within the detailed D&D work plan that will minimize risks and costs to workers and the public and increase the likelihood that D&D efforts will be accomplished given the uncertainty about future available funds. A prioritized cost- and risk-reduction approach should be flexible and should not preclude alternative end states. Such an approach could embody a stepwise, detailed decommissioning plan as a means of sequentially and incrementally achieving the D&D.
Recommendation 4. A prioritized cost- and risk-reduction approach should be used as the basis for developing the D&D plan. This approach should be used to accomplish D&D activities prior to the completion of the entire plan.
There are numerous laws and regulations, as well as regulatory bodies at federal, state, and local levels, that will affect D&D. The health and safety of the workforce, as well as the potential impacts on the local community, need to be addressed. Guidelines for decommissioning have been published by the Nuclear Regulatory Commission and DOE. Cooperative efforts to revise them on the basis of current radiation protection concepts are under way by the EPA, DOE, and the Nuclear Regulatory Commission. Available draft proposals have radionuclide concentrations equivalent to annual dose equivalents of 15 mrem per year, based on generic exposure scenarios, for the release of sites and materials. The agencies recommend that each exposure scenario be evaluated for a specific site. The large number of regulators with jurisdiction over the enrichment plants and their decontamination, as well as the large number of applicable laws and regulations, virtually ensure an overlapping, conflicting, and potentially costly regulatory regime under which to conduct D&D.
Recommendation 5. The committee recommends that DOE seek coordination of all regulatory aspects of D&D with the appropriate state and federal agencies early in planning to provide consistency during D&D planning and execution.
Site planning and the associated planning for D&D of the buildings should be undertaken in consultation with the stakeholders. This process of communication among various interested parties—public groups, regulators, workers, and DOE—and DOE decision makers should start at the very beginning of planning. For example, a consensus-building process would elicit public advice on the incremental cost- and risk-reduction approach and on the desired end states of the sites, taking into account costs, risks, and social values. Given the contentious issues that have emerged relating to many sites in the DOE weapons complex as a result of past practices, it is essential that a credible and meaningful stakeholder and public involvement process be implemented that ensures smooth planning and implementation of D&D. Increased attention to stakeholders and the public concerning D&D at the three GDP sites can further result in decisions that enjoy wide public acceptance, reduce conflicts among stakeholders and citizens, and minimize costly delays. Effective efforts to integrate the multiplicity of citizen and stakeholder interests are needed to provide meaningful inputs to decision making on such issues as health and safety, budgets, employment, and end states.
Recommendation 6. A stakeholder involvement program should be pursued to obtain timely and substantive public participation and input to ensure that social values are reflected in policy decisions.
Waste management is a major cost factor. Large quantities of low-level radioactive waste, hazardous waste, and mixed waste will result from the D&D and will require processing, packaging, transportation, and disposal. Although low-level radioactive waste burial costs are
uncertain, recent experience indicates that storage capacity is very limited, costs are increasing rapidly, and some sites are closing. Even if disposal costs are low, the long-lived radioactivity of uranium requires monitoring over millennia or permanent, safe disposal. Opposition to the proposed transportation and disposal of the very large volumes of low-level radioactive waste is likely. These concerns suggest that a strategy of waste minimization should be pursued. The location of the waste disposal sites and the likely cost of waste transportation and disposal need to be resolved.
Generation of mixed waste during D&D should be avoided because its processing and disposal entail a complex and costly regulatory regime. Improperly planned D&D operations could also contaminate otherwise clean material, for example, during asbestos removal operations, or areas around the buildings during D&D operations. Creation of secondary waste streams, such as contaminated water from decontamination processes, should be minimized. Improper management of discarded clothing and protective gear can also generate significant quantities of low-level radioactive waste.
One approach to waste minimization is to decontaminate materials for reuse. For example, concrete might be used for road fill (as was done at Capenhurst), and the large quantities of copper, nickel, and steel could be cleaned to surface release standards and recycled. Metal difficult to decontaminate might be compacted for low-level radioactive waste disposal. The geometric shape of some of the components, such as the diffusion barriers, would make it difficult to ascertain whether adequate decontamination has been achieved. These materials might be melt refined, but release of the subsequent metal would require volumetric standards for free-release of melted metal. Such standards do not currently exist in the United States, although the EPA and Nuclear Regulation Commission have begun preliminary work on surface and volume contamination standards. The cost of low-level radioactive waste burial and the market price of particular materials could economically justify the recycling of those materials to the commercial sector and reduce D&D costs significantly.
In the event that cleaned metals cannot be released to the commercial sector, they might be cleaned to standards adequate for controlled use within the DOE complex. For example, steel might be used for shield blocks or for radioactive waste canisters. The avoided environmental and economic cost of producing such materials from virgin ore would have to be balanced against the costs of recycling and the avoided costs of burial. If material could not be cleaned sufficiently for recycling to the commercial sector, the DOE complex, or the Department of Defense, it would have to be packaged, transported, and placed in low-level radioactive waste disposal facilities. Considering the large costs of low-level radioactive waste management, there are significant opportunities to optimize waste packaging and transportation to achieve cost savings.
The committee notes that radioactive, hazardous, and mixed wastes from other activities on the Oak Ridge Reservation are being temporarily stored within the GDP process buildings. This practice complicates, and could delay subsequent D&D efforts and could also engender costs during D&D that should not be ascribed to the D&D program.
Recommendation 7. An integrated, optimized waste management plan must be developed that encompasses material reuse, recycling, packaging, transport, and
waste disposal. Consistent with cost reduction and public health and environmental protection, materials should be cleaned to free-release standards and released to the commercial sector for recycling. Material that cannot be cleaned to free-release standards should be considered for recycling within the DOE or Department of Defense complexes in applications where slightly contaminated materials are acceptable, such as for shield blocks or waste containers.
Need For New Facilities
Previous cost estimates assume that large, expensive facilities will be required to decontaminate the process equipment, systems, and structures. A high-assay decontamination facility was proposed to decontaminate equipment containing highly enriched uranium deposits, and a low-assay decontamination facility was proposed to decontaminate equipment containing low-enriched uranium deposits. Decontamination techniques to be used in these facilities included mechanical removal, ClF3 (gaseous chlorine trifluoride) treatment, and high-pressure water jets. Metal parts that were difficult to decontaminate would be melted. The variety of technologies and capabilities, as well as the large size of the decontamination facilities, led to very high estimated capital and operating costs.
The committee believes that the decontamination processes in these facilities could be simplified. The need for a new, expensive high-assay decontamination facility could be eliminated by constructing small, limited-purpose shops for removal of highly enriched uranium and housing equipment in existing buildings. Low-assay decontamination could also be housed in existing buildings. Removal of uranium deposits could be accomplished in a number of ways, such as by gaseous decontamination or spray booths, using aqueous solutions and incorporating criticality safe piping, or mechanically. The approach taken would depend on criticality considerations and on a number of factors, such as enrichment level, size of deposits, and cost. After deposit removal, aqueous decontamination with criticality control could be used, as it was successfully for the D&D of the Capenhurst GDP. Selection of the best uranium removal and decontamination technologies and sequence will require systems engineering and cost tradeoff studies along with data from focused technology demonstrations, as discussed below. Simplification of the process and the use of existing buildings would reduce costs.
Recommendation 8. The high-assay decontamination facility should be eliminated; the low-assay decontamination facility should be simplified to focus primarily on aqueous decontamination and should be housed in existing buildings.
The cost estimates assumed the construction of a new administration building at the Oak Ridge GDP site, with space for several thousand people. Such large staffs are a result of the management and contracting approach assumed in the cost estimates. The committee believes that existing space could be modified to provide appropriate office space for administrative functions, especially if the committee's recommended changes in management and contracting are implemented, which should reduce the size of the management and professional staff.
Recommendation 9. Rather than constructing a new administration building, existing facilities should be used to house the management and professional D&D staff.
D&D Technology Issues
Proven technologies are available for the D&D of the GDPs. These technologies include technologies for characterization, disassembly, removal of uranium deposits from the process equipment, decontamination of the process equipment and buildings, melt refining and recycling of metals, and treatment of wastes. However, there are some uncertainties about technical effectiveness, such as the degree to which certain technologies can remove 99Tc (technetium-99) or decontaminate to required levels, and about what degree of cost savings can be achieved. Determining answers to such questions would require focused demonstrations, not major research and development programs. Such focused demonstration efforts on currently available technologies would help D&D planners select the most appropriate technologies based on considerations of cost, environmental protection, performance, and safety.
The GDP plants are major structures with extensive floor and wall areas, containing large pieces of equipment in a repetitive design arrangement. The committee believes that there are opportunities for automation and robotics in many of the repetitive operations involved in removal and transfer of equipment, size reduction of equipment components, and decontamination process operations. Such opportunities could reduce costs and reduce worker exposure to radiation and hazardous substances. The large areas within the buildings encourage automated characterization.
After treatment of the process equipment with gaseous ClF3, BNFL used aqueous processes at Capenhurst to decontaminate the cascade equipment. There are some uncertainties about the aqueous process, such as the degree to which 99Tc contamination can be removed. Furthermore, it is not clear that the physical characteristics of the diffusion barrier material, which contains much of the valuable nickel, will allow decontamination to free-release levels by aqueous decontamination. The most effective approach for this material may be melt refining, with the uranium removed in the slag. The best approach for 99Tc removal is uncertain.
A few, highly focused demonstration programs are needed. By demonstration program, the committee does not mean very large programs involving costly expenditures over many years. Technical and cost effectiveness should be demonstrated at a sufficient scale to allow scaleup of data to actual D&D operations. Much of the decontamination work can probably take place in the laboratory; demonstration of robotic systems may need to take place at a larger scale, such as in the plants.
The most appropriate approach to resolve the technical and cost uncertainties may be to conduct small demonstrations of different decontamination technologies for treatment of a stage or cell from the cascade to obtain required data on operating and decontamination factors. Based on these data, a decontamination process should be chosen and applied to the high-enrichment sections of the cascade. During the early stages of these D&D operations, careful record-keeping, data collection, and analysis should be conducted to serve as a basis for future D&D activities.
Recommendation 10. A few highly focused D&D demonstrations should be undertaken to verify the cost and effectiveness of specific technologies, including the following two:
- optimization of aqueous decontamination to remove radioactive surface contamination from materials and process equipment, with special attention to 99Tc; and
- support of current DOE robotics programs, with highly focused demonstrations to verify potential cost savings and safety benefits.
Recommendation 11. A modest research program should be established to develop methods to decontaminate the diffusion barrier material effectively.
Safeguards and Security
The assumption was made in the Ebasco cost estimate that the gaseous diffusion technology would be declassified prior to D&D. Costs will be larger if the diffusion barrier remains classified and the D&D has to be carried out in a "secure" environment. Furthermore, the D&D of the GDPs will require the handling of special nuclear materials. The regulatory requirements to safeguard these materials entail significant costs. These costs could be reduced significantly if less-stringent requirements could be applied. For example, special nuclear material should be removed from the high-enrichment sections of the cascade prior to the start of large-scale D&D operations so that safeguards and security requirements can be relaxed.
Recommendation 12. To reduce costs without compromising information security for the gaseous diffusion technology, DOE should try to define physical security requirements that allow uncleared workers under adequate supervision to conduct D&D operations. In addition, DOE should conduct an in-depth evaluation of the safeguards and security requirements during D&D to determine how their impact on D&D cost could be reduced.
A DOE study has found that past practices for storage of DUF6 have been inadequate in several respects. There have been no serious consequences, however, and there is a vigorous program to correct past deficiencies.
There is general agreement, however, that DUF6 is an unsuitable chemical form for long-term storage; it is too reactive and too volatile. Eventually DUF6 needs to be converted to the more suitable form uranium oxide (U3O8). No large-scale uses for the DUF6 have been identified, and the most promising potential uses do not preclude conversion to oxide.
Estimates prepared for DOE indicate that costs for the conversion of DUF6 to U3O8 will be high, over $2 billion. It should be possible to realize cost savings by optimizing a number
of factors, such as plant size and the conversion schedule. Conversion processes are conventionally used in the nuclear fuel industry, and several alternatives are known. The processes are rather simple, so that large cost reductions through new technology do not appear likely. Considering cost, risk and social values, the most attractive of the known processes can be chosen.
Significant savings in the costs of long-term storage should be possible by improving the physical properties of the U3O8; in particular, increased particle size and much higher packing density should be possible, which would reduce storage costs based on volumetric fees. This area promises benefits from a limited research and development program.
Recommendation 13. The committee recommends that, if consistent with the prioritized cost- and risk-reduction process, the DUF6 should be converted to the more stable chemical form, U3O8, for storage or disposal.