Appendix C

Meeting Summary, January 28-29, 1999

JANUARY 28, 1999
Open Session

Location: National Academies Beckman Center, Irvine, CA.

Attendance: G. Choppin (chair), M. Apted, P. Baisden, E. Flanigen, C. Hussey, F. Mansfeld, L. E. McNeese, R. Osteryoung, P. Shewmon, R. White.

Augmented by guests S. Lesica (DOE), R. Benedict (ANL), W. Ebert (ANL), T. Fanning (ANL), D. Abraham (ANL), S. Johnson (ANL), J. Ackerman (ANL).

Staff: D. Raber and C. Murphy.

Gregory R. Choppin presented opening remarks discussing the schedule for the remainder of the year. This time frame includes release of two reports—a status report on the demonstration project and a report on waste-form issues. The committee will work on its final report. Release of this report is dependent on the completion of ANL's demonstration project.

Sue Lessica, DOE-EM, also gave opening remarks concerning DOE's plans related to repository waste issues and goals for the meeting.

Robert W. Benedict spoke regarding the Spent Fuel Treatment Demonstration Project Status. Following presentation of sodium-bonded fuel treatment plans and the demonstration project milestones was the FCF Mk-IV electrorefiner experience. For the Mk-IV, 90 driver assemblies have been chopped and 84 driver fuel assemblies have been electrorefined. A three-month repeatability run with 12 assemblies was completed in two months. The last 16 fuel assemblies will be used to test possible improvements.

The Mk-V electrorefiner uses a modified anode design. This is meant to meet the concerns of uranium hold-up that have been observed previously. With Mod A-version 1, rotation is counter-clockwise, the scrapers lead the baskets, and clearance in front of the scraper was increased. Mod A-version 2 has a double gap in front of the scraper. Mod A-version 3 also has a double gap in front of the scraper and, in addition, has a double gap between the basket and the cathodes.

Tests were performed under various conditions to determine the highest voltage possible before cutoff. Using the Mk-V Anode version 1, a 0.6 V cutoff was observed with one empty basket in each channel (test designation Mk-V-7A). A 0.6 V cutoff was also seen going from 200 A to 100 A with one empty basket in each channel (Mk-V-7B). A 0.75 V cutoff was observed under these conditions with empty baskets in the inner channel (Mk-V-7F). In these examples all baskets in the outer channel were loaded with partially dissolved fuel segments.

With the Mk-V anode version 3, a 0.6 V cutoff was seen when going from 200 A to 100 A with empty basket in each channel (Test designation Mk-V-7C). A 0.6 V cutoff was also observed with rotation increasing from 40 to 60 rpm and one empty basket in each channel (Mk-V-7D).



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization Appendix C Meeting Summary, January 28-29, 1999 JANUARY 28, 1999 Open Session Location: National Academies Beckman Center, Irvine, CA. Attendance: G. Choppin (chair), M. Apted, P. Baisden, E. Flanigen, C. Hussey, F. Mansfeld, L. E. McNeese, R. Osteryoung, P. Shewmon, R. White. Augmented by guests S. Lesica (DOE), R. Benedict (ANL), W. Ebert (ANL), T. Fanning (ANL), D. Abraham (ANL), S. Johnson (ANL), J. Ackerman (ANL). Staff: D. Raber and C. Murphy. Gregory R. Choppin presented opening remarks discussing the schedule for the remainder of the year. This time frame includes release of two reports—a status report on the demonstration project and a report on waste-form issues. The committee will work on its final report. Release of this report is dependent on the completion of ANL's demonstration project. Sue Lessica, DOE-EM, also gave opening remarks concerning DOE's plans related to repository waste issues and goals for the meeting. Robert W. Benedict spoke regarding the Spent Fuel Treatment Demonstration Project Status. Following presentation of sodium-bonded fuel treatment plans and the demonstration project milestones was the FCF Mk-IV electrorefiner experience. For the Mk-IV, 90 driver assemblies have been chopped and 84 driver fuel assemblies have been electrorefined. A three-month repeatability run with 12 assemblies was completed in two months. The last 16 fuel assemblies will be used to test possible improvements. The Mk-V electrorefiner uses a modified anode design. This is meant to meet the concerns of uranium hold-up that have been observed previously. With Mod A-version 1, rotation is counter-clockwise, the scrapers lead the baskets, and clearance in front of the scraper was increased. Mod A-version 2 has a double gap in front of the scraper. Mod A-version 3 also has a double gap in front of the scraper and, in addition, has a double gap between the basket and the cathodes. Tests were performed under various conditions to determine the highest voltage possible before cutoff. Using the Mk-V Anode version 1, a 0.6 V cutoff was observed with one empty basket in each channel (test designation Mk-V-7A). A 0.6 V cutoff was also seen going from 200 A to 100 A with one empty basket in each channel (Mk-V-7B). A 0.75 V cutoff was observed under these conditions with empty baskets in the inner channel (Mk-V-7F). In these examples all baskets in the outer channel were loaded with partially dissolved fuel segments. With the Mk-V anode version 3, a 0.6 V cutoff was seen when going from 200 A to 100 A with empty basket in each channel (Test designation Mk-V-7C). A 0.6 V cutoff was also observed with rotation increasing from 40 to 60 rpm and one empty basket in each channel (Mk-V-7D).

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization When going from 200 Ah to 100 Ah, a 0.6 V cutoff was seen with two empty baskets in the outer channel (Mk-V-7E). Planned tests for the Mk-V ACM with simulated EBR-II blanket fuel will be performed at ANL-E with a 10-inch module. Tests include a run with the Mk-V-8 repeating operating conditions used in run Mk-V-7F with full loading of undissolved fuel segments in the anode baskets in the outer channel and empty baskets in the inner channel. Also, the operating conditions used in run Mk-V-8, except using full loading of undissolved fuel segments in the anode baskets in both channels will also be performed (Mk-V-9). Finally, run Mk-V-8 or -9 will be performed with materials compatibility coupons. Proposed conditions for sustained operation of the ACM with simulated EBR-II blanket fuel include using an anode drive rotation speed of 60 rpm, a voltage cutoff of 0.75 V, a 7 weight % uranium concentration in the salt, an electrodeposition current of 600 to 100 A, and a stripping current of 600 A. Laboratory test conclusions were as follows. A procedure has been developed for sustained operation of the Mk-V ACM. A throughput rate of 7 kg of uranium product per 24-hour per module was demonstrated. Reversing the direction of rotation of the anode drive can eliminate stalls. A higher throughput rate may be achieved if sustained operation can be demonstrated with fuel loaded in the inner anode baskets. The following run conditions were used for the second irradiated blanket fuel loading. The first segment/product collector no. 1 yielded (including salt) 12.6 kg. The deposition was 200 A for 200 Ah with a 0.45 V cutoff. Reverse rotation was used for the 2-minute washing cycle. For stripping, 600 A was maintained until 0.28 cutoff voltage was reached (approximately 150 Ah). This was followed by reverse rotation for the 2-minute washing cycle. Under these conditions, no significant anode rotation stalls occurred. For the second segment/product collector no. 2, the yield (including salt) was 7.5 kg. The deposition was 600 A to 200 A for 200 Ah with a 0.45 V cutoff. Reverse rotation was maintained for the 2-minute washing cycle. For stripping, 600 A was maintained until the cutoff voltage (0.28 to 0.45 V) was reached (approximately 150 Ah). Reverse rotation was then used for the 2-minute washing cycle. Under these conditions, one significant anode rotation stall occurred. Significant accomplishments in the treatment process include processing 84 of the 100 driver assemblies by driver treatment. Repeatability of 16 kg of uranium per month for three months has been completed. Eight assemblies have been treated in one month. In addition, 850 kg of low-enriched uranium (LEU) have been cast, and 570 kg of LEU have been shipped. Blanket treatment started in August 1998. The Mk-V electrorefiner has run two batches of irradiated blankets. Results from the Mk-V anode-cathode module tests at ANL-E are being used to guide Mk-V operating parameters and improvements. The blanket chopper is operational. For HIP scale-up, three 8-inch diameter HIP cans have been processed, and plans for processing an 18-inch can are ongoing. The ANSO HIP process was used in these cases. The HIP process

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization was chosen because it takes a powder and gives a ceramic, and the repository will not accept powders. Future plans for the spent fuel treatment demonstration include the following. Driver treatment should be completed by May 1999. Blanket treatment of 150 kg of uranium per month is planned for summer 1999. Twenty-five blankets may not be done by September 1999. Three irradiated metal ingots should be completed by April 1999. Five radioactive demonstration ceramic waste cans will be produced by summer 1999. Robert W. Benedict, ANL-W, spoke about Plans for Waste Form Testing and Qualification. Waste form testing discussion areas include the waste-qualification process, data requirements, specific tests, demonstration items, and future plans. Waste-qualification activities involve several organizations and layers of interaction between the waste-form producer, in this case DOE-NE and -EM, and the repository operator, DOE-RW. Documents required of the waste-form producer include waste-acceptance product specifications, a waste compliance plan, and a waste qualification report. These are sent to the repository operator for the license application, which also requires performance assessment of the waste. The geologic repository project requires data to support its analysis. An Environmental Impact Analysis includes performance assessments, which consist of a viability assessment and the license application. Waste Acceptance Criteria contain the Waste Acceptance System Requirements Document (WASRD) and the Mined Geologic Disposal System Waste Acceptance Criteria. The Geologic Repository Project is responsible for total system performance assessment. The basic objective of the waste disposal system is to contain and isolate the radioactive wastes so that the dose impact to humans is attenuated to a relatively benign level. Requirements include limited water contacting the waste package, a long waste package lifetime, low rates of release of radionuclide from breached waste packages, and radionuclide concentration reduction during transport from the waste packages. The WASRD lists some specific waste form data requirements. For chemical composition, “The Producer shall report to RW the chemical composition and crystalline phase projections for vitrified HLW.” In addition, “The Producer shall report to RW the oxide concentration of elements present in concentrations greater than 0.5% by weight and the estimate of the error of these concentrations for vitrified HLW.” For the radionuclide inventory WASRD states, “The Producer shall report to RW the estimated total and individual canister inventory of radionuclides (in curies) that have half-lives longer than 10 years and that are or will be present in concentrations greater than 0.05% of the total radioactive inventory. The estimates shall be indexed to the years 2010 and 3110.” For phase stability and integrity WASRD states, “The Producer shall ensure the phase structure and composition of the vitrified HLW are not degraded after initial cool-down by maintaining the waste form below 400 °C to ensure the glass transition temperature is not exceeded.

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization For product consistency WASRD states, “The Producer shall demonstrate control of waste form production by comparing production samples or process control information, separately or in combination to the Environmental Assessment benchmark glass using the Product Consistency Test or equivalent.” And, “For acceptance, the concentrations of lithium, sodium, and boron in the leachate, after normalization for the concentrations in the glass, shall be less than those of the benchmark glass.” For reporting requirements, the WASRD specifies, “The Waste Compliance Plan (WCP) shall describe the Producer plan for demonstrating compliance with RW acceptance criteria, including a description of tests, analyses, and process controls to be performed by the Producer. The WCP also identifies records that will be provided as evidence of compliance.” And, “The Waste Qualification Report shall compile the results from waste form testing and analysis to demonstrate the ability of the Producer to comply with RWS acceptance criteria.” Data supplied for the Yucca Mountain Environmental Impact Statement include waste form descriptions, including volume, mass, density, chemical composition, and metric tons of heavy metal. The packaging description includes the number of cans, and kilograms of material per canister. The source term description includes a description of the total radionuclides, average concentration per canister, and the canister with the highest concentration. Additional source term information includes the degradation/release route, where the assumption is made that resistance is at least as high as EA glass. The dose rate and estimate of criticality are also assessed preliminarily. These data were submitted through DOE-EM. A first submittal was made in July 1997, and a final submittal was made in May 1998. ANL's approach to data for the repository performance assessment model makes certain that inputs for biosphere and geosphere models are identical to those used in evaluating commercial spent fuel. Waste packaging degradation assumes the DOE standard canister will be used. Inputs specific to the performance assessment model for the ceramic and metal waste forms are as follows: (1) radionuclide inventory per waste package for both waste forms, (2) degradation/release rate for each waste form as a function of pH and temperature, (3) surface area estimates for both waste forms, and (4) physical properties. Waste test matrices supply experimental data for various needs. Repository performance assessments need uncertainties for the degradation/release rate model, physical properties, the source term, and phase stability. Waste specifications need ranges for phase distribution, chemical composition, the product consistency method, phase stability, and hazardous properties. Process qualification needs acceptable ranges for raw material specifications and process operating parameters. The test matrices group samples by types and sizes. Several ceramic waste sample types are used for different data needs. Consolidated samples are used for preliminary screening in all three areas and are a principal data source for repository performance assessment and phase distributions. Cold component and radiation effects samples support repository performance

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization assessment data needs exclusively. Demonstration-scale and process materials principally support process qualification and product consistency data needs. Different ceramic waste sample sizes are selected to optimize resource requirements. Lab-scale uniaxial press samples are used to meet special handling requirements. Lab-scale hot isostatic pressing (HIP) samples are the majority of samples. Demonstration-scale HIP samples are used to study process qualification. Witness tube samples are the proposed sampling method for product consistence. Scale-up samples show the feasibility to qualify processing of larger batch sizes. Production size samples will be evaluated during the actual process qualification of production equipment. Analyses support data needs for the performance assessment. For the repository performance assessment and waste specification, attribute analyses primarily support physical properties, phase distribution, phase stability and chemical composition. Characterization, accelerated, service conditions, and natural analog support degradation and release modeling. The Product Consistency Test-A (PCT-A), density and x-ray diffraction are primary analyses for the product consistency method and process qualification. The Material Characterization Center Test (MCC-1) and scanning electron microscopy are supporting analyses for the product consistency method and process qualification. Demonstration data will support the initial draft of the preproduction documents for the DOE-RW Review. The Waste Acceptance Product Specifications (WAPS) describe the generic waste form and waste packages. These include the isotopic inventory, waste physical characteristics, phase stability, and package design. The Waste Compliance Plan (WCP) describes the experiments, tests, and measurements that will be performed to demonstrate that the waste processes will produce a product that complies with the requirements in the WAPS. These include product consistency measures and important physical property measurements. Waste qualification activities to be completed by August 1999 include demonstration waste form testing matrices, which will be complete except for radioactive samples and test lasting greater than one year. Methods will be established for applying both waste form models to the repository time frame. Sensitivity performance studies will be preformed using bounding waste degradation models. These should help identify the major parameters that drive repository performance for the ceramic and metal wastes. For both waste forms, degradation modeling refinements based on experimental data will be incorporated into the simplified TSPA model. The Preliminary Waste Acceptance Product Specification and Waste Compliance Plan will be discussed with DOE-RW. The product consistency method will be identified and tested. Major process parameters will be identified for full-scale operations qualifications. Future waste test matrices for license applications will be established. Postdemonstration waste qualification activities will include complete waste test matrices that support repository license application and the waste qualification report. The waste qualification report will be written. The product consistency method will be finalized; the waste process qualification tests will be performed with full-scale equipment.

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization Bill Ebert spoke about Testing to Support Qualification of the Ceramic Waste Form (CWF). Information needs that are addressed by the testing program include repository performance assessment needs, waste specification needs, and process qualification needs. Repository performance needs entail measuring the matrix corrosion behavior and radionuclide release, providing for a mechanistic model with model parameters, and providing for model confirmation. Waste specification needs cover identification of phases containing radionuclides, and providing for methodology to monitor the CWF product consistency. For process qualification, a database must be provided to establish process-operating parameters. In a performance test of reference sodalite CWF in a seven-day PCT (a test required by the WASRD), it was found to perform similarly to HLW glass. The American Society for Testing of Materials (ASTM) method for developing a mechanistic model being followed for CWF corrosion consists of identifying important phases and corrosion modes, conceptualization of the model, conducting tests to provide parameter values and to develop the model, conducting tests to verify the model, and developing material behavior predictions. Scoping tests provided many insights. A suite of scoping tests was designed with predisposition toward any particular corrosion mode or model. Test results provide insight into information needs. The data to be discussed include measured corrosion behavior of sodalite, glass binder, and the CWF, the possible rate expression for CWF dissolution and estimation of values of model parameters, preliminary modeling results under Yucca Mountain scenarios, and a summary of the current testing matrix. A number of scoping tests are used to identify key corrosion modes. The Solution Exchange Test distinguishes between release by dissolution of free salt, release of occluded material, and framework dissolution. The product consistency tests study solution effects on dissolution behavior of the CWF under concentrated solution conditions. The MCC-1 static leach tests characterize dissolution behavior of the CWF in the absence of solution effects under dilute solution conditions. The pH stat tests characterize pH-dependence of the CWF dissolution rate. Accessible free salt measurements measure the soluble salt in the CWF. The vapor hydration tests characterize advanced stages of corrosion, identify alteration phases, and give insight into long-term stability. Solution exchange tests were developed to measure diffusion-controlled release of concrete waste forms. They involve periodically removing solution and replacing it with fresh leachant. The tests were used to distinguish between leaching of occluded salt and release as zeolite and sodalite matrices dissolved. Tests were conducted in demineralized water with 4-µm salt-loaded zeolite grains and thermally converted 4-µm salt-loaded sodalite grains. Key results were as follows: Li, Na, and Cl released faster than the zeolite dissolves; Li, Na, and Cl released slower than the sodalite dissolves; and zeolite and sodalite dissolve at similar rates. The MCC-1 static leach tests with the CWF were tests developed in the early 1980s to “distinguish differences in the leaching behavior of candidate waste forms.” It is a water-dominated system in which solution effects are minor (except for pH). Tests are conducted in

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization demineralized water with monolithic samples at 90 °C. A low solution/volume (S/V) ratio is not representative of Yucca Mountain disposal site conditions. Key results of the MCC-1 static leach tests with the CWF include high NL(Cl) and NL(Na) due to dissolution of the free salt. The NL(Si) is greater than the NL(B), which indicates sodalite dissolves faster than the glass binder initially, but appears to approach saturation after about 28 days. (The normalized elemental mass loss, NL, allows direct comparison of the release of different elements from a material and releases in tests at different S/V ratios, the surface area of material to solution/volume ratio. One- and three-day tests are used to estimate the forward rate. Product consistency tests (PCTs) were developed in the early 1990s to “evaluate whether the durability and elemental release characteristics of nuclear, hazardous, and mixed waste glasses have been consistently controlled during production.” It is the glass-dominated system in which solution effects quickly becomes significant. Tests are conducted in demineralized water with crushed samples at 90 °C. A high S/V ratio is representative of Yucca Mountain disposal site conditions. A number of key results were obtained from PCTs with the CWF. Preferential initial release of Cl and Cs due to dissolution of the free salt was observed. The NL(Cs) decreases with time, and may be incorporated into the alteration phase. Release of B is greater than Si and indicates that the glass binder dissolves faster than sodalite. The solution approaches saturation with respect to sodalite within 91 days. Some miscellaneous issues that must be resolved include the question of how addition of radionuclides affects corrosion behavior of the CWF. Tests must be run where UCl3 (which transforms to UO2) and Pu/CdCl2 (which transforms to PuO2) are added to the CWF. When responses in three-day MCC-1 tests were compared, which is sensitive to material properties, the result was that it was not possible to distinguish between responses of the CWF, the CWF + U, and the CWF + Pu. Another question is how pH affects the dissolution rate. Glass in buffered flow-through tests and natural sodalite in MCC-1 tests show a “V-shaped” pH dependence. Glass, sodalite, zeolite, and the CWF were all evaluated in pH stat tests (PCT-type: crushed glass, 80 °C, two hours). All showed a “V-shaped” pH dependence. The corrosion model for aluminosilicate minerals and glass was originally developed for aluminosilicate minerals and modified for application to HLW glasses. The rate-determining step is hydrolysis of the Si-O bond at the surface. The rate expression includes terms for the forward rate and chemical affinity for dissolution. The rate depends on dissolved orthosilicic acid, but not on time. Tom Fanning spoke regarding Ceramic Waste Form Modeling, Preliminary Results. The objectives are to develop a waste form degradation and radionuclide release model for the CWF. The purpose of the model is to predict long-term behavior based on short-term tests of important processes. It will also provide conservative release estimates based on observed waste form behavior. A second objective is to demonstrate waste form performance in a repository

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization environment. The radionuclide release model will be incorporated into the repository performance assessment model. In addition, improvement of modeling of the process is important under repository conditions. For the preliminary calculations, the goal is to determine the importance of sodalite dissolution parameters under repository relevant conditions. Sodalite is also compared with HLW glass under steady-state flow modes. The mineral dissolution equation is used for both HLW glass and sodalite. The global parameters fix temperature at 90 °C, and pH at 9. The assumption is made that incoming groundwater is undersaturated with respect to the host rock: Qin = 56 g/m3. The rate equation used is R = kf(T, pH)(1−Q/Ksp) + klong. Optimistic values for HLW glass include kfglass = 0.5 g/m2/d (low end of measured values), and kfVA = 0.17 g/m2/d (based on a 5-component analogue). For “typical” HLW glass, klong was ignored. Kspglass is typical, but HLW glass results are not strongly sensitive to small variations in the solubility product. Conservative values are used for sodalite. The value for kfsod is based on early tests, but sodalite results are not very sensitive to the forward rate. The value for Kspsod = 107 g/m3, but more recent results indicate a value of 60 to 70 g/m3. The benefits of the glass binder were ignored. Dilute conditions have a stronger effect on sodalite, so that Qin < Kspsod < (QinVA = Kspcrist) < Kspglass < Kspsilica. Two different modes for the contact of groundwater with the waste form are a pool mode and a drip flow mode. The pool mode assumes all water enters a large breach and forms a pool around the waste forms. It does not imply that the entire package is flooded. The flow rate is 0.25 m3/y (the long-term average in VA). The surface area is 0.25 to 25 m2 (the maximum for the CWF is ~25 m2). Sodalite dissolves faster than TSPA-VA results if the Log10(q/s) > −0.58, the exposed area < 1 m2 (q = 0.25 m3/y), the flow exceeds 6.5 m3/y (s = 25 m2), and the maximum flow in VA is < 1 m3/y. Sodalite dissolves more slowly than the typical HLW glass, and sodalite dissolution (grams/day) is nearly independent of the exposed surface area. The drip flow model assumes flowing water accumulates silica until it reaches the bottom of the waste form. The drip rate is calculated as 2.5 × 10−3 m3/y, and the drip area is 0.25 to 25 m2. Low flow rates lead to silica saturation. The affinity term, 1 – Q/Ksp, dominates the calculation. The term klong prevails under low flow (high surface area) conditions. Sodalite dissolves slower than HLW glass under these conditions. A lower concentration of silica is needed for saturation, and sodalite dissolution is 3.5 times lower than typical HLW glass. The sodalite dissolution (in grams/day) is independent of surface area. The conclusions from these models are that sodalite dissolves slower than HLW under the considered flow conditions, that the comparison to HLW is conservative, and that future calculations will include temperature and pH effects. Bill Ebert addressed Current Work and Test Matrix. Ongoing work and future plans involve measuring parameter values needed to model the waste form under controlled temperature and pH conditions. Also, data must be provided to demonstrate the use of the mechanistic model. A database must also be generated to evaluate potential methods for monitoring product consistency. The applicability of the model for behavior of demonstration-scale materials under conditions relevant to the disposal system must be confirmed. Finally, the identity and

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization distribution of salts, PuO2, etc., formed during processing for demonstration-scale samples must be determined. The test matrix for qualification of the CWF contains a number of tests to provide model parameter values. The MCC-1 with glass and sodalite measures parameters (k0, ƞ, and Ea), and with the reference CWF for model confirmation. Another test is the pH dependence on acid and base sides at 90 °C before the demonstration, and other temperatures after the demonstration. Long-term PCTs with glass, sodalite, and reference CWF measure model parameters [H4SiO4]sat and klong. Other tests include SEM, TEM, x-ray diffraction (XRD), the density of starting materials and reacted solids for phase identification, radionuclide distribution, and physical characterization. Other tests in the matrix confirm the applicability of the model for long-term disposal. These include VHTs and long-term PCTs with glass, sodalite, and reference CWF for advanced corrosion (e.g., long-term stability, and ID alteration phases). Also required are thermal degradation and low-temperature VHTs. Drip tests are needed to confirm the applicability of the behavior model under repository-relevant conditions with the reference CWF, reference CWF and U, and demonstration-scale samples. Finally, SEM, TEM, and XRD are required for phase identification, radionuclide distribution, and physical characterization. Tests are also required to monitor product consistency. Evaluation of MCC-1, PCT, and soluble salt are needed to measure consistency of the waste forms during production. Daniel Abraham presented the metal waste form qualification update. The focus of the work is on qualification of MWF alloys, specifically those waste forms suitable for repository disposal, and to generate input needed for TSPA analysis. Waste form qualification involves experimental testing, with data feeds into the model development, and model development. The MWF model will be incorporated into the Repository Integration Program (RIP) performance assessment software. Experimental testing encompasses study of alloy microstructures, mechanical and thermophysical properties measurement, and corrosion testing. The modeling approach uses known stainless steel degradation mechanisms as a basis for MWF modeling. It also uses functional dependencies developed for stainless steels. The baseline MWF is a SS-15Zr alloy. The bounding composition is SS-15Zr-11U-0.6Ru-0.3Tc-0.1Pd. The experimental ranges have a Zr content of 0 to 20 weight %, noble metal content of 0 to 4 weight %, Tc content of 0 to 2 weight %, and U content of 0 to 11 weight %. The number of alloy compositions is 39 and there are 111 ingots. A review of the alloy microstrucutre reveals an intermetallic phase [Zr(Fe,Cr,Ni)2+x], and a stainless steel phase containing ferrite and austenite. Noble metal-rich precipitates are not observed and actinides are present only in the intermetallic. Type MCC-1 immersion testing of the MWF takes place in Teflon™ vessels at 90 °C in simulated J-13 and deionized water. Current tests examine specimen corrosion behavior over a

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization range of Zr and noble metal contents. The test duration ranges from 90 days to 365 days to longer terms. New data from the 90-day tests suggest that specimens are either unaffected or show minor tarnish, the weight change of the specimens is small, fission product content in test solutions is small, and normalized corrosion losses are small. These results are consistent with previous tests on the SS-15Zr alloys. Immersion testing at 200 °C is designed to accelerate the corrosion rate. Test data presented previously show the weight change of the alloy specimens is small, normalized loss of fission products is < 0.1 g/m2, and silver loss from SS-1Ag-2Nb-1Pd-1Ru (no Zr) alloy is ~144 g/m 2. Examination of specimen surfaces by optical microscopy has revealed that 15 and 20 weight % Zr specimens show uniform corrosion, while a low-Zr alloy (SS-5Zr-2Nb-1Pd-1Ru) showed pitting. Pulsed-flow immersion tests were performed at 90 °C in a simulated J-13 solution/deionized water, for up to 629 days. Tests were interrupted periodically to obtain leachate solutions for elemental analysis. Thirty-six samples contained Tc and/or U. Current results extend previously presented dated for test periods up to 275 days. SEM examination of an SS-15Zr-2Tc sample that was tested in deionized water for 449 days showed mild surface tarnish and some corrosion in the casting pores. The immersion tests indicate that dissolution is not a dominant release mechanism. Also, corrosion appears to be retarded by passivation. Under these conditions, fission products and actinides may be trapped in the passivated oxide layer, hence limiting release. In vapor hydration tests, the corrosion rate of the MWF is accelerated by exposure to steam. The tests are performed in stainless steel vessels at 200 °C for 56 and 182 days. Previous experiments have shown that corrosion layers are typically 1 µm thick (56 test SS-15Zr). Current experiments examine corrosion behavior for seven alloy compositions. Fifty-six day tests were terminated recently. Results show that weight gains were small, and localized attacks (pitting) were observed on samples containing ≤ 5 weight% Zr. The corrosion layer thickness and the nature of corrosion products on the specimen surfaces will be determined in February, and the 182-day tests will be terminated. The purpose of electrochemical corrosion testing is to obtain a relative measure of corrosion rates and to screen out alloy compositions that may not be suitable for repository disposal. The corrosion rate measurement by the linear polarization method is based on ASTM G59. Tests on “cold” specimens are complete and tests on “spiked” specimens (containing Tc and/or U) have begun. Test data show that corrosion rates for alloys that contain from 5 to 20 weight % Zr are similar. A relatively high corrosion rate was observed for SS-1Ag-2Nb-1Pd-1-Ru (no Zr) alloy in a pH 2 solution. Radiation can affect corrosion in three ways. Changes in local water chemistry can have an effect due to formation of radiolysis products (e.g., H2O2). Structural damage may occur to the protective oxide layer. Radiation may also have an effect due to changes in the electronic properties of the oxide. Calculations performed at ANL-E suggest that radiation levels in the MWF will be too low to affect corrosion.

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization In summary, corrosion problems, if any, may exist at low-Zr contents for the samples exposed in the electrochemical, vapor hydration, and immersion tests. Also, MWF corrosion appears to be limited by passivation behavior of the SS base alloy. Steve Johnson spoke about A Long-Term Study of the Effects of Alpha Damage to the Ceramic Waste Form. The effect of radiation damage, specifically alpha decay damage, on an ordered (crystalline) waste form has been studied by several groups but it varied depending on the crystalline host. In an alpha decay event there are two particles that may cause damage: the alpha particle and the recoil nucleus. Ionization damage may result in the following effects: covalent and ionic bond rupture, valence changes or localized electronic excitations, enhanced diffusion processes, and decomposition. Ballistic damage may result in atom displacement. The present study will evaluate the ceramic waste form and the long-term effect of alpha radiation on that waste form. The anticipated plutonium (239Pu) loading in the actual waste form to be produced is 0.2 to 1.0 weight %. Samples made with a higher loading (up to 2.6 weight %) of 239Pu confirmed the similarity of the final product to that made with a lower loading such as 0.6 weight %. This observation allows an increase in the Pu loading without changing the fundamental properties of the sample. The samples produced as a part of this study contain 2.5 weight % 238Pu, and fission products present equivalent to ~100 drivers processed. The loading of Pu utilized is 3 to 12 times that anticipated in the actual waste form to facilitate the experimental methods employed and to accrue the most data in the shortest time frame. The purpose of the test matrix is to develop methods and techniques to evaluate the effects on the CWF of long-term exposure to alpha radiation. These techniques must be sensitive to macroscopic and microscopic changes during the study. The time frame to the test is 4 years. Contained in the test matrix are measurements of density to determine macroscopic swelling. Product consistency tests (PCT) determine the release rates of all elements, in particular Pu, Cs, and I. X-ray diffraction (XRD) analyzes phase-specific swelling or change. Scanning electron microscopy (SEM) checks for microstructural changes, as does transmission electron microscopy (TEM). Thermal properties tested include specific heat and expansion behavior at different temperatures. These tests or methods when taken together will yield a complete picture of the effect of alpha radiation to the CWF. Preliminary results show that 238Pu and 239Pu samples are equivalent and valid for this study. Analysis by XRD shows that in the 239Pu-sodalite CWF, a plutonium oxide and a (minor) halite phase are present. The same result was seen with the 238Pu-sodalite CWF. Likewise, for SEM/TEM analysis both 238Pu and 239Pu samples show agreement with XRD and phases present with small grain size (less than 30 µm in diameter). In the leach test, 238Pu and 239Pu release rate results for Pu and Cl are comparable. Density is comparable for both products. SEM reveals three primary phases present in the 238Pu-loaded CWF: sodalite, glass, and plutonium oxide. TEM bright field imaging of 238Pu-loaded CWF also shows three primary phases: sodalite, glass, and plutonium oxide.

OCR for page 37
ELECTROMETALLURGICAL TECHNIQUES FOR DOE SPENT FUEL TREATMENT: An Assessment of Waste Form Development and Characterization Product consistency test/density results indicate that the normalized release rate is ~1E−4 g/m2 day for Pu. A full elemental analysis is still pending. The release rates will be monitored with time using “aged” material to probe for enhanced release of material. Typical increases in the release rates vary from 10 to 100 times those for other CWFs. The density is 2.42 g/cm 3. This will be monitored with time to probe for macroscopic swelling. An accelerated alpha damage study is under way with sample production initiated in October 1998. The majority of the test matrix has been accomplished for the early time period samples. JANUARY 29, 1999 Closed Session Attendance: G. Choppin (chair), M. Apted, P. Baisden, E. Flanigen, C. Hussey, F. Mansfeld, L. E. McNeese, R. Osteryoung, P. Shewmon, R. White, C. Murphy. The entire meeting was conducted in closed session. Following a preliminary discussion of committee balance and composition, a detailed review of the previous day's presentations took place. The committee then discussed writing assignments and generated findings and conclusions for this report. Reviewers' comments for the committee's report 8 were also discussed.1 1   National Research Council, Electrometallurgical Techniques for DOE Spent Fuel Treatnent: Status Report on Argonne National Laboratory's R&D Activity as of Fall 1998, National Academy Press, Washington, D.C., 1999.