A major issue in the cleanup of this country's nuclear weapons complex is how to dispose of the radioactive waste resulting primarily from the chemical processing operations for the recovery of plutonium and other defense strategic nuclear materials. The wastes are stored in hundreds of large underground tanks at four U.S. Department of Energy (DOE) sites throughout the United States. The tanks contain hundreds of thousands of cubic meters of radioactive and hazardous waste. Most of it is high-level waste (HLW), some of it is transuranic (TRU) or low-level waste (LLW), and essentially all containing significant amounts of chemicals deemed hazardous. Of the 278 tanks involved, about 70 are known or assumed to have leaked some of their contents to the environment. The remediation of the tanks and their contents requires the development of new technologies to enable cleanup and minimize costs while meeting various health, safety, and environmental objectives.

While DOE has a process based on stakeholder participation for screening and formulating technology needs, it lacks transparency (in terms of being apparent to all concerned decision makers and other interested parties) and a systematic basis (in terms of identifying end states for the contaminants and developing pathways to these states from the present conditions). The primary purpose of this study is to describe an approach for identifying technology development needs that is both systematic and transparent to enhance the cleanup and remediation of the tank contents and their sites.¹ The committee believes that the recommended end state based approach can be applied to DOE waste management in general, not just to waste in tanks. The approach is illustrated through an example based on the tanks at the DOE Hanford Site in southeastern Washington state, the location of some 60 percent by volume of the tank waste residues.

The approach proposed for identifying technology development needs for the remediation of high-level waste in tanks is essentially an application of systems engineering. The essence of the approach is the structuring of remediation scenarios (i.e., a reference scenario and several alternatives) to identify the technologies required to reliably achieve the goals of radioactive waste management in the face of uncertainties about the future. Identification of technology needs is based on specifying remediation goals in terms of the desired end state of

specific wastes. As used in this report, an end state is defined as the final product of a waste processing, remediation, or management scenario characterized well enough in terms of chemical, physical, and radioactive attributes to allow details of scenarios to be specified. In addition to chemical and physical properties, specifications of end states may include location, legal, regulatory, societal, and institutional factors. Owing to the emphasis on the specification of end states, the approach is referred to as the end state based approach. For the DOE Hanford waste in large underground tanks, used as an illustration of the end state based approach, three general waste end states are considered; (1) immobilized HLW, (2) immobilized low-activity waste (LAW), and (3) closed tank farms containing some amount of radioactive material.

The recommended approach consists of the following steps:

A scenario is defined for this report as a qualitative description of the transition path of waste from its initial state to a specific end state.

The key to successful application of the recommended approach is in the scoping and specification of end states and functional flowsheets. Note that there may be appropriate interim products as a result of phasing or modularizing the overall remediation process. One of the most important steps in the approach is the development of a functional flowsheet, a generalized description of processing operations (functions) linked to effect transformation of the initial radioactive waste to an end state.

The advantages of the end state based approach are many. Technology development needs are tied to specific end states (goals), and the underlying bases are easily visible. There is a traceable path from the problem to the solution through specifying the initial states or conditions, defining a reference goal and alternatives to accommodate uncertainties, identifying functional approaches to move from the initial problem to the solution, assessing the adequacy of existing technology, and implementing technology development only in those areas where technology is unsatisfactory (e.g., too costly) or inadequate to accomplish technical goals. Technical adequacy refers to the ability to reach a prescribed end state in a cost-effective manner. The existence of this traceable path is believed to be very important to the technology sponsor and users, and in gaining public support. A significant benefit of the end state approach should be a better chance of passing the scrutiny of review and oversight groups, including the U.S. Congress.

The explicit connection of the technology development program to the desired end states of the wastes is intended to provide logic and efficiency for the technology development program. The need for support of each significant technology development project may be derived from a specification of the satisfactory and plausible end state to be achieved and an assessment to determine whether additional technology development is required. If technology to

achieve the end state already exists, then justification of additional technology development would require that such development lead to increased benefits, such as reduction of implementation cost or risk, that would compensate for the projected cost of the development.

A limited application of the end state based method was developed to demonstrate the identification of technology development needs to remediate high-level waste tanks at the Hanford Site in Washington state. The results of the example are illustrative only and are based on information accessible to the committee.

The total radioactivity content of the Hanford tanks is approximately 198 million curies, with about two-thirds of it in the tank solids. The principal radioactivity of the waste comes from cesium-137 and strontium-90 and their decay products. The tanks contain solids of various types (e.g., sludge, saltcake, slurry) and supernatant liquid. The chemical constituents of the solids are mostly precipitated iron, aluminum, and other hydrated metal oxides. Saltcake is primarily crystalline sodium nitrate, and the supernatant liquid contains large amounts of dissolved sodium salts, especially nitrates, nitrites, and hydroxides.

For the Hanford tanks example, end states were postulated by the committee for three products; HLW, LAW, and closed tanks. The end state assumed for HLW is immobilization in borosilicate glass logs stored in a passively cooled, on-site temporary storage facility, and certified for transport to and acceptance in a deep geologic repository.² The end state for LAW for the committee's reference scenario in the example is an immobilized form containing most of the bulk chemicals from the tanks and a small amount of radionuclides. The LAW is assumed to be disposed of in on-site near-surface facilities. The end states for the tanks and tank farms are in situ stabilization, with or without waste in the tanks.

Three scenarios were postulated for the example. The committee's reference scenario represents the essence of DOE's currently planned approach (i.e., the Hanford baseline scenario). The committee would have chosen Hanford's baseline scenario as its reference scenario except that a well-defined baseline did not exist because of its dependence on technology yet to be provided by a private contractor. The committee defines an in situ disposal scenario as representing a budget-restricted, cost-risk balanced remediation approach. To further reduce radionuclides in the LAW, the committee developed an extensive separations scenario which reduces risk, reduces the volume of HLW, or both. The diverse nature of the waste in the Hanford tanks suggests that use of various scenarios may allow optimization of costs, schedules, and other factors, resulting in different scenarios being applied based on the initial attributes of the tank wastes.

The essence of the committee's reference scenario is to retrieve most of the waste from each of the tanks. The tanks would then be filled with gravel, capped with a multilayer barrier to prevent water ingress, and subjected to occasional maintenance and surveillance. This scenario would result in the tank farm area being perpetually unsuitable for unrestricted use. The retrieved waste would be converted into two products; (1) vitrified HLW suitable for interim on-site storage, to be followed by disposal in a deep geologic repository, and (2) immobilized LAW, containing most of the chemicals and a small amount of radionuclides, which is suitable for onsite, near-surface disposal. Functional steps in the scenario are characterization of waste in the

tanks, mobilization and retrieval of the wastes, initial solids washing and waste transfer, enhanced sludge washing, radionuclide separation and recovery functions, waste immobilization, vitrifier offgas processing, storage operations, and tank stabilization and closure. None of the committee's scenarios include transportation of waste to or disposal at an off-site repository.

In addition to the committee's reference scenario, two alternative scenarios are considered. The first is the in situ disposal scenario, which may be used if tank cleanup cost estimates exceed allocated budgets (assuming that in situ disposal, including consideration of possible increased characterization and change in regulations, is indeed less costly than other remediation options) or if it is found that some tanks can be remediated at relatively low risk using in situ techniques that do not include waste retrieval. The end state of this scenario is tanks with contents immobilized to the extent practical, completely filled with solid material to prevent tank collapse, and surrounded with protective barriers. Some functions for this scenario, such as the characterization of the tanks and their contents and secondary waste treatment, are the same as in the committee's reference scenario. Added functions include risk analysis for the selection of tanks suitable for in situ remediation, the stabilization of tanks and tank contents, and the application of enhanced barriers to reduce both the amount of water contacting the stabilized tank and radionuclide migration.

The other alternative scenario, extensive separations, could be applicable if a need exists to further reduce radionuclide contents in the LAW and/or to reduce the volume of HLW. The end states for this scenario are the same as in the committee's reference scenario except for the relative volumes of the two waste streams, and the low-activity waste that will contain significantly lower levels of radionuclides. This scenario uses all functions of the reference scenario. Additional functions include analyses required for deciding which tank waste should be subject to extensive processing, a solids dissolution step, enhanced cesium removal, separation of radionuclides, and destruction of nitrate and acids.

The three scenarios in this example involve a number of process steps that are, at this point in the description of the approach, identified only by specification of the functions to be performed. The next step in applying the end state based approach is to determine the technology requirements to implement each function in the various remediation scenarios. This step requires an examination of the functions to identify technology requirements. The final step in the application of the approach compares the requirements with available technologies to establish specific technology development needs, which is a major objective of the end state based approach.

Since the only purpose of this example is to illustrate the approach, the committee limited the detailed consideration of the functional steps in the three scenarios to a few functional process steps selected for their likely importance to the overall process and the anticipated importance of technology development to those functions. The functions selected for examination from the committee's reference and extensive separations scenarios were enhanced sludge washing and vitrifier offgas processing. The functions selected from the in situ disposal scenario were stabilization of tanks and their contents and use of enhanced barriers. The technology status for each selected function was assessed and compared to the requirements of the postulated end states to yield technology development needs. The Office of Environmental Management (EM) technology development program was then examined and evaluated in the context of these needs.

For enhanced sludge washing, the technology development needs include additional investigations on colloid formation, identification of the broader range of sludges likely to be encountered during the processing steps or waste retrieval, and reaction rates. Regarding vitrifier offgas streams, technology development needs include engineering response to information on

the specification and quantities of materials evolving from the vitrifier, including development of processes for the treatment of these streams and secondary waste. Technology development needs are also influenced by requirements for remote operations and maintenance and the effectiveness of offgas cleanup and pollution abatement.

The EM tank waste remediation programs have not defined, and the EM technology development program is not addressing, technology needs related to tank stabilization and enhanced barriers for tanks from which wastes have not been retrieved. As a result, new technology development activities would likely be required if DOE were to pursue the in situ disposal scenario.

The conclusions and recommendations, while considerable with respect to what was learned from the Hanford example, are primarily directed at the major purpose of the committee's study, that is, to describe a broadly applicable end state based method for identifying technology development needs. The example was created only for illustrative purposes, and the conclusions and recommendations based on the example should be considered in this context.

There are several major differences between the application of the approach to identification of technology needs as recommended by the committee (i.e., use of end states) and as used by the DOE. An important difference is the consideration by the committee of end states other than those for the baseline scenario as codified at a given site in various site-specific compliance agreements. The committee believes strongly that scenarios involving alternative end states may need to be considered for reasons including life cycle costs, technical failures, changing requirements, and delays in meeting schedules when originally selected end states present problems. The committee recognizes that such alternative end states are largely outside of the present plans of both the DOE remediation programs and the DOE technology development organization. However, the committee believes that DOE management and legislative decision makers should change their approach by expanding the scope of consideration of alternative end states that may be needed in the future and reflecting this in the DOE remediation and technology development programs.

At present, many public stakeholders at Hanford apparently want DOE to follow the current compliance-driven Hanford baseline approach, and they view investment of significant resources in technology development for alternative scenarios as a diversion from that effort. Some stakeholders do apparently recognize that readjustments to the Hanford baseline may become necessary if a particular approach proves to be infeasible for whatever reason (whether technical, programmatic, economic, or political). However, stakeholders generally appear to prefer DOE to limit such investments. Nevertheless, more explicit consideration of alternatives as proposed herein and greater organizational commitment to a risk-based approach could make the overall DOE program more robust with respect to unexpected developments, as well as provide a more transparent rationale for a particular approach to eventually be adopted from among the candidate approaches.

• The committee observes that technology development typically requires several years to produce deployable results, depending on the initial technology status. In contrast, regulations and stakeholder values concerning remediation issues and decisions often change much more rapidly.
• An end state based approach that includes a range of plausible scenarios can identify technology gaps in a reference flowsheet and technology development needs associated with credible alternative flowsheets.
• The committee concludes that an end state based approach for identifying technology development needs would greatly facilitate the efficiency and visibility of the DOE EM efforts to provide technologies required to remediate the HLW in tanks throughout the DOE complex, thereby providing rationale and background for appropriate funding. The committee concludes that (1) attention should be given to the development of alternative scenarios for such reasons as providing viable contingencies in case of operational problems or changes in remediation guidelines or regulations, (2) these alternatives should be pursued by application of an end state methodology, and (3) if an end state methodology is carefully applied, the overall increase in funding necessary to support technology development for alternative scenarios may be relatively modest.
• End state specifications can be grouped into those related to human health and ecological risk (e.g., safety, hazard), to cost and schedules (e.g., process reliability, performance, efficiency), and to societal externalities (e.g., policy on allowable risk, desirable end state specifications, the role of privatization). To form the basis for a technology development program, these specifications must be translated into specific requirements for each function. Some type of risk or cost assessment, together with a decision analysis, is necessary to convert broad end state requirements into quantitative end state specifications pertinent to scenario analysis.
• The committee found the EM tank waste technology development program to have a substantial but incomplete end state approach for the Hanford baseline scenario. In particular, there was no significant consideration of different alternative scenarios with associated end states, primarily because of the absence of their identification by the problem owners.
• Several health and safety risk studies have been conducted on remediation strategies under consideration. The committee was unable to find evidence that such quantitative risk studies have had an effect on the choice of remediation strategies.
• The committee is confident that the end state based approach proposed in this report and illustrated by the Hanford tanks would greatly facilitate the identification of technology development needs, thereby highlighting the technical basis of the technology development program. This can be done even though substantial uncertainties exist in the end state specifications, and the real drivers of the program may be non-technical factors. The committee believes that such an approach is critical to gaining public support and acceptance for the remediation program.
• The Hanford tank example was effective in illustrating the identification of specific technology development needs in selected functional areas and the relationship of those needs to achieving the desired end states.
• Examples of additional technology development needs, derived from the committee's scenarios and the DOE technology development program for the Hanford tanks wastes, are improved understanding of enhanced sludge processing reaction times, better information on colloid formation during sludge processing, and an improved means for preventing buildup of solids and radioactivity in the offgas system of the vitrification process.

• The privatization initiative diverts DOE attention from addressing some tank technology development needs (e.g., vitrification offgas). DOE has not explicitly stated its strategy and policy concerning its role in technology development in light of privatization of tank remediation operations. This obscures the scenarios and functions that are the targets for the technology development program and confuses the process for identifying and prioritizing needs.
• Finally, the process for making decisions related to technology needed for the remediation of radioactive waste in tanks must consider factors other than purely scientific and technical ones. These other considerations, referred to in this report as external factors, include the constraints on decision making that come from organizational changes, regulations, stakeholder agreements, congressional actions, etc. While external factors were not a focus of the committee's study, such issues as privatization and DOE's organization have a major influence on technical decisions.

• Implementation of an end state based approach is recommended for determining an appropriate technology development program to support DOE's remediation efforts for the radioactive waste in tanks. In particular, this approach should encompass a range of plausible scenarios, as opposed to only a preferred baseline approach as used at Hanford. Depending on the extent to which waste remediation has been completed, alternative scenarios may not be needed for some remediation problems. Examples of this situation could be cases where the best remediation option is obvious or where regulations mandate a particular approach with no recourse to alternatives.
• If initial conditions cannot be adequately characterized and end states cannot be completely specified for waste-containing systems to be remediated, it is recommended that, in the interim, provisional assumptions that allow scenario development and identification of technology needs to proceed should be clearly stated, preferably by problem owners, but by technology providers if necessary. A plan leading to the timely resolution of the 'open items' should be prepared and executed.
• The committee recommends that the end state based approach be applied by DOE on a much broader scale than currently apparent, consistent with a total systems approach to the remediation of contaminated sites. The success of such a strategy is dependent on the support for it from legislators and stakeholders. The need to apply the end state based approach is highlighted by the extensive uncertainty surrounding the entire tank cleanup program, including the role and responsibilities of DOE and contractors that may be part of the privatization program.
• The various steps taken in implementing the end state based approach should be documented in detail for the custodians of the waste and those persons engaged in technology development. Documentation should be the basis of reviews by oversight groups. In addition, executive-level documentation appropriate for decision makers, including DOE upper management, regulators, Congress, and the interested public, should be provided.
• The committee recommends a limited parallel pursuit of alternatives (including end states) to the current strategy of the remediation of the radioactive waste tanks developed by the privatization contractors. It is not considered prudent to rely totally on the success of the privatization process. The uncertainties and costs are great and the chances for unacceptable results are too high not to pursue viable alternatives.

The committee provides in this report a description and an example of application of an end state methodology that can lead to a rationalized technology development program. The committee believes it is highly desirable for DOE to apply such a methodology to waste remediation activities and other comparable assignments to define technology development needs clearly, conserve resources, meet schedules and cost targets, reduce risk of technology failures, and provide visibility to stakeholders of the course of action being taken to remediate sites and facilities.