Interim Report -- Committee on Cesium Processing Alternatives for High-Level Waste at the Savannah River Site (1999)

Advisers to the Nation on Science, Engineering, and Medicine

National Academy of Sciences

National Academy of Engineering

Institute of Medicine

National Research Council

The National Research Council empaneled a committee ¹ at your request ² to provide an independent technical review of alternatives for processing the high-level radioactive waste salt solutions at the Savannah River Site ³ . You requested that the Research Council provide you with an interim report that identifies significant issues or problems with the processing alternatives before the Department issues a draft environmental impact statement (EIS), which is planned for release in October 1999. The committee’s interim report is provided in this letter. This report has been reviewed in accordance with the procedures of the National Research Council ⁴ and reflects a consensus of the committee.

The information used to develop this interim report was obtained from several sources. The committee reviewed published documents that describe the salt processing program at Savannah River and the screening process used to select alternative processing options ⁵ . The committee also held an information-gathering meeting on September 13-14, 1999 in Augusta, Georgia to receive briefings from Department of Energy (DOE) staff, Westinghouse Savannah River Company (WSRC) staff, and National Laboratory scientists ⁶ on the alternative processing options and current and planned R&D activities.

The committee does not yet have enough information to fully address its statement of task ⁷ . However, based on the information gathered to date, the committee has reached several conclusions that it believes will be helpful to DOE in finalizing the draft EIS. These conclusions are described in the following paragraphs and are organized around the four bullets of the

committee’s statement of task. The committee also offers several recommendations in the concluding paragraphs of this report.

• Task 1: Was the process used to screen the alternatives technically sound and did its application result in the selection of appropriate preferred alternatives?

The screening process (see Attachment C) to identify cesium removal alternatives was undertaken by the Salt Disposition Systems Engineering Team under the sponsorship of WSRC. This team was comprised of 10 members with expertise in science and engineering, operations, waste processing, and safety and regulations. The team interacted with experts throughout the DOE complex and undertook a historical review and literature survey to identify about 140 possible processes that could potentially be used to process the high-level waste salt solutions at Savannah River. These processes were grouped into an “initial list” of 18 alternative processing options, which were subsequently screened using a multi-attribute analysis to obtain a “short list” of four alternative processing options: small tank tetraphenylborate (TPB) precipitation, caustic side solvent extraction, direct disposal in grout, and crystalline silicotitanate (CST) ion exchange. This screening process has been reviewed by numerous groups, including two expert teams assembled by DOE, and has received generally favorable marks.

The committee has not yet had an opportunity to perform a detailed examination of this screening process. Therefore, a full response to this part of the task statement must be deferred to the committee’s final report. However, the committee does have one comment at this time relative to this question: Given the ambitious schedule that the Department has defined for selecting and implementing a process for treating the cesium-bearing salt solutions at Savannah River—a draft EIS is to be issued in October 1999, a Record of Decision (ROD) is to be made in spring 2000, and the selected option is planned to be implemented no later than 2008 ⁸ —a negative answer by the committee to this statement-of-task question could delay Savannah River’s plans to process this waste and could markedly increase the total cost of the processing operations ⁹ . This question could have been asked earlier to permit more meaningful input into the screening process.

The storage of high-level liquid wastes in underground tanks, some of which are several decades old, represents a potential hazard to workers and the environment at the site and a continuing burden on U.S. taxpayers. The committee shares the Department’s (and WSRC’s) sense of urgency to address this hazard by removing and treating the waste as soon as safe and practical. Consequently, in addressing this part of its statement of task, the committee will be asking the question “Did the screening process lead to the identification of technically sound options for processing the waste?” The committee’s initial impression is that the screening process did result in the identification of several potentially viable alternative processing options. The committee will perform a more detailed review of the overall screening process during the remainder of this study.

• Task 2: Was an appropriately comprehensive set of cesium partitioning alternatives identified, and are there other alternatives that should be explored?

Given the compressed schedule for producing this interim report, the committee, has focused most of its attention on the four options that were included in the “short list” of alternatives discussed in Attachment C. The committee has not yet had the opportunity to perform a detailed review of the full list of alternatives for processing the salt solutions that were identified by WSRC through its alternatives screening process. The committee did, however, perform a cursory examination of the list of 18 alternative processing options developed by the Salt Disposition Systems Engineering Team. These options included the approaches that were known to committee members to be useful for processing cesium-bearing alkaline salt solutions. Thus, the committee’s initial impression is that no major processing options were overlooked in the screening process. However, the committee will perform a more thorough review of alternative processing options for the final report.

• Task 3: Are there significant barriers to the implementation of any of the preferred alternatives, taking into account their state of development and their ability to be integrated into the existing SRS HLW system?

The committee examined the four alternatives that passed the multi-attribute screening process (i.e., small tank TPB precipitation, caustic side solvent extraction, direct disposal in grout, and CST ion exchange) to assess whether there were significant barriers to implementation. The committee concluded that any of these four alternatives could probably be made to work if enough time and funding were devoted to overcoming the remaining scientific, technical, and regulatory hurdles. However, the time, cost, and technical risk of implementation of each option could vary widely because all are at different states of development.

A preliminary summary of the scientific, technical, and regulatory hurdles for each option is summarized below. For the small tank TPB precipitation, caustic side solvent extraction, and CST ion exchange processing options, the remaining hurdles are both scientific and technical in nature and include the need for obtaining a better understanding of basic chemical processes. The direct disposal in grout option appears to be technically mature but faces significant regulatory hurdles.

Small tank TPB precipitation. The small tank TPB precipitation option was developed by WSRC staff to “engineer around” the benzene production problem discovered during large tank in-tank precipitation (ITP) operations (see Attachment C). The development of this option was based on the belief by WSRC staff that they adequately understood the basic chemistry and process phenomena that led to the earlier difficulties with the ITP process.

This new option was designed both to reduce the production of benzene during processing and storage of the waste and to reduce benzene explosion hazards. As currently designed, the process will employ 57,000 liter (15,000 gallon) stainless steel reaction vessels and short processing and storage times that, taken together, allow less TPB to be used in processing and reduces the time available for radiolytic and catalytic decomposition of the TPB to form benzene. Additionally, the reaction vessels are designed to maintain a positive pressure so that the head (open) space can be blanketed with a nitrogen atmosphere to reduce explosion hazards. This design also allows for the capture and treatment, if desired, of benzene evolved during processing operations.

Although a positive pressure would allow the reaction vessels to be blanketed with inert gas, this design could promote the release of benzene and possibly of radionuclides to the environment should a leak occur. The standard practice in biological and radiological facilities is to maintain a negative pressure relative to the atmosphere so that leaks result in inflows rather than outflows. The use of positive pressure reaction vessels for this process may require additional containment and safety procedures to minimize hazards should leaks occur.

Small tank TPB precipitation appears to be the alternative preferred by WSRC for processing the cesium-bearing salt solutions at Savannah River. WSRC has over 16 years of experience with TPB, and has developed a rudimentary understanding of the cesium precipitation process through R&D work done during ITP development and operations. This option appears to have the most advanced R&D and engineering development of all of the options except direct grouting.

Nevertheless, the committee believes that additional R&D is required to demonstrate that this option could be successfully implemented to treat the cesium-bearing salt solutions at Savannah River. WSRC lacks an adequate understanding of the chemistry underlying the TPB decomposition process and the catalysts and catalytic reactions responsible for benzene generation. In place of such an understanding, WSRC appears to be focusing on an engineering-design solution based on untested assumptions about maximum likely benzene production and catalytic pathways. The WSRC staff who briefed the committee indicated that WSRC has a limited understanding of the mechanisms of catalysis responsible for benzene production and also that WSRC has collected little experimental data on the sources or roles of likely catalysts such as soluble transition metal complexes and dispersed palladium metal particles.

The committee believes that WSRC must obtain a better understanding of the chemistry of the TPB decomposition process before this option could be selected and deployed to treat the cesium-bearing salt solutions at Savannah River. The extreme complexity of the chemical system in the alkaline tank waste at Savannah River—which consists of more than 35 elements in a variety of phases and chemical compounds, including solid and liquid complexes— increases the likelihood that significant and unanticipated technical problems will be encountered unless benzene generation and release processes are better understood ¹⁰ . The committee believes that it would be advantageous from both time-efficiency and cost- efficiency standpoints to undertake this R&D work before the process is selected and deployed. The alternative—namely, to proceed with deployment immediately and engineer around the gaps in chemistry knowledge—carries a high technical risk and could result in a repeat of the ITP failure.

CST Ion Exchange. Crystalline silicotitanate is an inorganic material that has a high selectivity for cesium over other alkali metals and is thus a potentially useful ion-exchange material for cesium removal from alkaline tank waste. Although ion exchange for cesium removal is a well known technique, CST ion exchange has never been used in a large-scale nuclear waste application, and CST has never been manufactured in commercial-scale quantities. Consequently, additional R&D will be required to demonstrate this technology for cesium removal from tank waste at Savannah River.

The committee learned during its information-gathering meeting that WSRC has discovered two significant and potentially insurmountable problems with the CST ion exchange process. First, WSRC staff discovered that cesium is desorbed (i.e., released) from CST at elevated temperatures—this phenomenon was observed to occur at temperatures of 50 °C and probably operates (albeit at reduced rates) at lower temperatures, including at the planned 25 °C to 30 °C processing temperatures. This desorption process appears to be irreversible—that is, once cesium is released, it is not reabsorbed once temperatures are lowered. Second, CST appears to react with constituents in the alkaline tank waste to produce new solid phases that may be capable of plugging the ion exchange columns. Because either of these problems could lead to extended and costly shutdowns of tank waste processing operations at the site if this processing option were to be implemented, they must be resolved before this process can be deployed.

WSRC staff do not yet understand the chemical processes responsible for either of these phenomena, although they speculated to the committee that the tank waste reactions with CST may be due to manufacturing impurities. Clearly, additional R&D work is required to address these problems.

Caustic side solvent extraction. Solvent extraction is a mature and widely implemented technology for separating uranium and plutonium from acidic solutions (e.g., in the PUREX process), but it has never been used to treat highly alkaline wastes like the cesium-bearing salt solutions at Savannah River. There is a great deal of process experience with solvent extraction across the DOE complex, induding at Savannah River. Furthermore, this “all liquid” process is highly compatible with the existing high-level waste system at the site. The process would produce a liquid cesium-bearing solution that could be sent directly to the Defense Waste Processing Plant without additional processing steps.

The caustic side solvent extraction process has been developed by scientists at the Oak Ridge National Laboratory and appears to the committee to be a potential “break-through” technology. However, caustic side solvent extraction is at an immature stage of development relative to the other three options and, although there do not appear to be any insurmountable problems with this process, additional R&D will be required to demonstrate that it could be successfully implemented to treat the cesium-bearing salt solutions at Savannah River. In particular, additional R&D needs to be done to obtain performance data under real-life conditions—for example, R&D to determine the radiation stability of the solvent system (including the stability of the cesium chelator, diluent, and modifier; see Attachment C), the ability to scrub and recycle the expensive solvents, and the ability to mitigate contaminant formation during processing. It may be possible to answer these questions relatively quickly in a small-scale pilot study under process conditions. Additionally, the cesium chelator for this system has never been manufactured in commercial quantities, so a significant scale-up of manufacturing capabilities would have to be demonstrated to ensure that reagent-grade material could be produced in quantities required for processing the cesium-bearing salt solutions at Savannah River.

Direct grouting. This process is very similar to the so-called “saltstone process” that was to have been used to dispose of the salt solutions from the ITP process. As noted previously, this is a very mature technology and has already been demonstrated at the site for less radioactive salt solutions. Some additional R&D work may be needed to develop grout that will

retain cesium to satisfy regulatory requirements. In general, cesium sorption on cementitious material like Portland cement is lower than virtually all other radionuclides. Given the large inventory of cesium that would be disposed of in grout under this option, the demonstration of compliance under applicable regulations (see the next paragraph) may require the development of grout formulations with a higher cesium sorption coefficient. Additionally, engineering design would be required to develop a facility that could be remotely operated and maintained to protect workers from high radiation fields. However, the required R&D and engineering work appear to be relatively straightforward, and the committee knows of no insurmountable problems with this option.

The major hurdles with the direct grouting option are regulatory in nature ¹² . The process would be intended to produce Class C low-level waste that would be disposed of at Savannah River in a land-disposal facility (see Attachment C). To be eligible for such onsite disposal, the cesium-bearing salt solutions would have to be declared to be “incidental waste” under DOE Order 435.1 ¹³ . To be declared as incidental waste, DOE would have to demonstrate that the wastes “Have been processed, or will be processed, to remove key radionuclides to the maximum extent that is technically and economically practical” and that the waste “Will be managed to meet the safety requirements comparable to the performance objectives set out in 10 CFR Part 61, Subpart C ....” The latter criterion would require DOE to undertake a detailed performance assessment analysis to demonstrate that disposal of the grout onsite would meet the U.S. Nuclear Regulatory Commission’s radionuclide release criteria for land disposal facilities as well as intruder barrier requirements for Class C waste ¹⁴ . However, DOE would not be required to seek formal NRC approval for this assessment. Additionally, DOE would likely be required to seek permits from South Carolina and/or EPA to operate this facility because it contains other regulated wastes ¹⁵ . It is not clear to the committee whether resolution of these regulatory issues would be possible under the current schedule constraints.

“Front-End” Actinide and Strontium Removal. As noted in Attachment C, the four cesium removal options discussed above are designed to process waste streams that have been treated to remove actinides and strontium. Savannah River plans to remove these radionuclides at the “front end” of processing operations by treating the waste with monosodium titanate (MST), which absorbs actinides and ion exchanges with strontium. To the committee’s knowledge, this process has not been used elsewhere to process tank waste and therefore would be a first-of-its-kind implementation, subject to the usual problems inherent with such applications.

In fact, there do appear to be some remaining technical questions that will need to be resolved before this process could be implemented successfully at Savannah River, and the committee learned at its information-gathering meeting that WSRC appears to be pursuing these questions vigorously. In particular, MST reaction kinetics are not well understood—work by WSRC staff suggests that actinide and strontium reactions with the MST may proceed more

slowly than anticipated, thus requiring higher MST concentrations to achieve the required throughputs. However, titanium (a significant component of the MST) is incompatible with borosilicate glass, so there are limits on the amount of MST that could be used in processing operations. There are some indications that these limits could be approached or exceeded to meet actinide removal requirements for the planned disposal of the residual salt solutions in the saltstone facility. Recent work by WSRC suggests that waste dilution and waste blending (i.e., combining waste from different tanks) may be required to meet the performance requirements for this process. This additional processing will add time, cost, and technical risk to waste processing operations.

To the committee’s knowledge, WSRC is not considering any alternative processes for removal of actinides and strontium from the tank waste. Thus, this process must work as intended for the high-level waste processing program to succeed. Additional R&D work, especially on reaction kinetics, will be required to demonstrate that this process can meet both performance and regulatory requirements. If this work identifies any insurmountable problems, then WSRC will have to find alternative processes for removing these radionuclides. The committee’s initial impression is that this process can be made to work, but WSRC must get on with the pilot-scale testing to demonstrate that this process will achieve the needed throughput.

• Task 4: Are the planned R&D activities, including pilot-scale testing, adequate to support implementation of a single preferred alternative?

A consideration of planned R&D activities will be a major component of the committee’s future work, and at this point in the study the committee only has enough information to make two general observations about ongoing and planned R&D activities. The committee’s first observation is that R&D resource allocations for the four alternative processing options have been markedly inequitable. In FY99, R&D funding for the four alternative processing options totaled about $11 million—about $4.4 million for small tank TPB, $6.0 million for CST ion exchange, $0.3 million for caustic side solvent extraction, and $0.3 million for direct grouting. Funding for the R&D work on solvent extraction was provided not by WSRC, but through DOE’s Office of Science and Technology. Part of this funding inequity can be traced to the late 1998 decision by WSRC to pursue only one primary (small tank TPB) and one backup (CST ion exchange) option for processing the cesium-bearing salt solutions. However, this funding disparity appears to be primarily responsible for the different levels of technical maturities of the four processing options, independent of their likelihoods of success. The committee got the sense from its discussions with WSRC and DOE staff at the information-gathering meeting that WSRC did not appear to be serious about pursuing R&D on any option but small tank TPB precipitation.

Second, WSRC does not appear to have a well thought out R&D plan for the small tank TPB precipitation and CST ion exchange options ¹⁶ . There are no written R&D plans for either of these options. Moreover, in response to committee questions, the WSRC and DOE participants at the committee’s information-gathering meeting were unable to describe an R&D scope for these options that would resolve the outstanding issues. Rather, the committee was presented with lists of research needs, but these needs were not prioritized. The committee was puzzled by the lack of program planning and pursuit of important uncertainties. Even for the WSRC-

• Conclusions and Recommendations. As noted previously, DOE plans to release a draft EIS in October 1999 that will be used as a basis for a spring 2000 ROD that selects a single processing option. WSRC is now in the process of preparing the draft EIS, and the committee was told by WSRC staff that the draft EIS would likely recommend the selection of small tank TPB precipitation as the preferred processing option. Based on the committee’s initial review of the processing options, it is not clear that the small tank TPB precipitation option favored by WSRC will necessarily be the committee’s preferred choice after the committee’s detailed review is completed.

Although the committee recognizes the need for selecting and implementing an option as soon as possible—both because of the high ($400 million per year) operating costs for the high-level waste system at Savannah River and because of the potential hazard posed by the aging waste tanks—the committee concludes from the preceding discussion that there are significant technical or regulatory risks in selecting any of the four options, including the small tank TPB precipitation option that is apparently preferred by WSRC. Therefore, the committee recommends that WSRC pursue vigorously one primary and several backup options for processing the cesium-bearing salt solutions at Savannah River until the remaining technical and regulatory issues are resolved. This may require that DOE delay the issuance of the planned EIS and ROD, or that DOE adopt a phased-decision approach in the EIS and ROD that would allow several processing options to be pursued in parallel until a clearly superior option is identified. DOE and WSRC should enlist the help of specialists in industry, National Laboratories, universities, and other federal agencies to resolve these scientific, technical, and regulatory issues.

To meet the ambitious time schedule for selecting and implementing any processing option, the committee concludes that DOE and WSRC will have to develop and implement a sharply focused R&D program to resolve the open issues. To this end, the committee offers the following recommendations:

1. Actinide and Strontium Removal. WSRC should continue its efforts to address the remaining technical questions concerning reaction kinetics of the MST process for removal of actinides and strontium from the tank wastes and get on to pilot-scale testing as soon as possible.

2. Small Tank TPB Precipitation. WSRC should develop and implement a vigorous, well-planned, and adequately funded R&D program to address the remaining scientific hurdles with the small tank TPB precipitation. R&D should, at a minimum, address TPB decomposition process(es) and the envelope of catalysts and catalytic reactions responsible for benzene generation.

3. CST Ion Exchange. A vigorous, well planned, and adequately funded R&D effort should be undertaken to address the remaining scientific hurdles with the CST ion exchange option. This R&D should address, at a minimum, the cesium desorption process and reactions between CST and the alkaline waste. This effort also should be pursued independently of WSRC, which does not have the needed R&D expertise on site.

4. Direct Grouting. WSRC and DOE should undertake a vigorous program to determine the regulatory acceptability of the direct grout option through discussions with relevant staff at DOE, the U.S. Nuclear Regulatory Commission, the U.S. Environmental Protection Agency, and South Carolina Department of Health and Environmental Control. These discussions should focus on the likely feasibility of demonstrating compliance with regulations and the strategy needed to achieve regulatory approvals.

5. Caustic Side Solvent Extraction. A vigorous, well planned, and adequately funded R&D effort should be undertaken to address the remaining scientific and technical hurdles with the caustic side solvent extraction option. This R&D should address, at a minimum, the stability of the solvent system in radiation fields, the ability to scrub and recycle the solvents, the ability to mitigate contaminant formation during processing, and the ability to produce the chelating agent in quantities necessary for this application. If started immediately, it may be possible to complete this work by next spring, in time for the final EIS. This effort should be pursued independently of WSRC, which does not have the needed R&D expertise on site for this particular solvent extraction system.

The committee’s next information-gathering meeting will be held in Augusta, Georgia on November 21-22, 1999, and a major part of that meeting will be devoted to further discussions of the R&D needed to resolve the issues identified in this report. The committee plans to ask Department and WSRC staff to report on their future R&D plans to resolve the open issues. The committee will provide a critique of these plans in its final report, which it hopes to issue by April 2000.