C. L. Crawford, D. M. Ferrara, R. F. Schumacher, and N. E. Bibler
Westinghouse Savannah River Company
Aiken, SC 29808

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

This report has been reproduced directly from the best available copy.

Available for sale to the public, in paper, from:  U.S. Department of Commerce, National Technical Information Service, 5285 Port Royal Road, Springfield, VA 22161,  phone: (800) 553-6847,  fax: (703) 605-6900
email:  orders@ntis.fedworld.gov
online ordering:  http://www.ntis.gov/help/index.asp

Available electronically at  http://www.osti.gov/bridge

Available for a processing fee to U.S. Department of Energy and its contractors, in paper, from: U.S. Department of Energy, Office of Scientific and Technical Information, P.O. Box 62, Oak Ridge, TN 37831-0062,  phone: (865 ) 576-8401,  fax: (865) 576-5728,  email:  reports@adonis.osti.gov

Hanford tank waste consists of about 190 million curies in 54 million gallons of highly radioactive and mixed hazardous waste stored in underground storage tanks at the Hanford Site in Washington State. The tank waste includes solids (sludge), liquids (supernatant), and salt cake (dried salts that dissolve in water to form supernatant). The tank waste will be remediated through treatment and immobilization to protect the environment and meet regulatory requirements. The U.S. Department of Energy’s (DOE’s) preferred alternative to remediate the Hanford tank waste is to pretreat the waste by separating it into low-activity waste (LAW) and high-level waste (HLW), followed by immobilization of the LAW for on-site disposal and immobilization of the HLW for ultimate disposal in a national repository.

The Savannah River Technology Center (SRTC) at the Savannah River Site (SRS) in Aiken, South Carolina, has been involved in research and testing with actual radioactive Hanford tank samples to demonstrate and verify the conceptual design of the waste treatment and immobilization plant (WTP) processes. This paper describes the crucible-scale vitrification and associated wasteform product tests in support of the WTP at Hanford. The two different LAW glasses produced in this study were from pretreated Envelope A (Tank 241-AN-103) and Envelope C (Tank 241-AN-102) waste. The HLW glass was produced from Tank C-106 HLW sludge and the HLW radionuclide products separated from Hanford Site tank samples AN-103, AN-102 and AZ-102. Pretreatment of these three supernates consisted of characterization, strontium and transuranics removal by precipitation and filtration, and final Cs-137 and Tc-99 removal by ion exchange (IX). The glasses were produced from formulations supplied by Vitreous State Laboratory of the Catholic University of America (CUA). Formulations were based on previous surrogate testing and the actual characterization data from the radioactive feed streams. Crucible-scale vitrifications were performed in platinum/gold crucibles in a custom-designed furnace fit with an offgas containment system. Both LAW and HLW melter feed slurries were evaporated, calcined, and then melted at 1150°C. The LAW and HLW glasses were heat-treated per a modeled centerline cooling curve for the LAW canister and HLW canister, respectively.

All glasses were analyzed for chemical and radionuclide content and examined for crystalline phase identification. The glasses were also durability tested using the ASTM Product Consistency Test (PCT) procedure and were tested by the U.S. Environmental Protection Agency’s Toxicity Characteristic Leaching Procedure. The LAW glasses were also analyzed for regulatory analyses by Babcock and Wilcox Technology (BWXT) Services, Inc. per U.S. EPA SW-846 protocol. These analyses included cyanide, ignitability, RCRA metals to the Universal Treatment Standard (UTS), volatile and semivolatile organic compounds, dioxins and furans, pesticides, and polychlorinated biphenyls (PCBs).

Remediation of Hanford tank waste involves pretreatment of the waste to separate it into fractions of LAW and HLW, followed by immobilization of the LAW for on-site disposal and immobilization of the HLW for ultimate disposal in a national repository.

LAW waste is composed of the tank waste liquids (and dissolved salt cake) and contains the bulk of the tank waste chemicals and certain radionuclides (e.g., cesium, technetium, strontium, and transuranics) that must be mitigated prior to immobilizing the waste. LAW is a mixed, characteristic, and listed waste regulated under the Resource Conservation and Recovery Act of 1976 (RCRA), and must meet certain treatment standards and performance standards for on-site disposal of the final waste form. The HLW waste is composed of the long half-life radioactive tank waste solids and the radionuclides separated from the LAW fraction. HLW is a mixed, characteristic, and listed waste regulated under RCRA, and must meet specific treatment and performance standards for storage and repository disposal of the final waste form.

The U.S. DOE has established the Office of River Protection (ORP) in Richland, Washington to manage and oversee the design, construction, and commissioning of a new Waste Treatment and Immobilization Plant (WTP) that will treat and immobilize the waste for ultimate disposal. To accomplish the ORP mission, DOE established the River Protection Project (RPP). The RPP contractor is responsible for designing, constructing, commissioning, and supporting the transition of the WTP. The DOE has developed the contract specifications for this effort.¹ Bechtel National is the responsible contractor, teamed with Washington Group International (WGI), to develop the WTP. A WTP Research and Technology Testing Program Plan was developed as part of the WTP Conceptual Design. The R&T Testing Program Plan describes the research and testing work that will be conducted to support process and facility design, as well as qualification testing of the immobilized waste forms (IHLW and ILAW). WGI has contracted several organizations to assist in the research and technology testing program. GTS Duratek and Vitreous State Laboratory (VSL) have performed surrogate glass and melter development work² and Battelle/Pacific Northwest National Laboratory (PNNL) has demonstrated crucible-scale vitrifications producing LAW and HLW glasses.³

SRTC has produced and characterized two LAW glasses and a single HLW glass from actual radioactive Hanford Site waste samples. The two waste samples used for the LAW glasses are referred to as Envelopes A (AN-103) and C (AN-102). The HLW glass was produced from a mixture of the following waste streams: (1) caustic-leached HLW sludge slurry, referred to as Envelope D from Hanford Tank 241-C-106, (2) eluate waste streams containing high levels of Cs-137 derived from ion exchange (IX) pretreatment of Envelopes A, B, and C (these three tank wastes were also treated for Tc-99 removal by IX but the Tc-eluates were not used in this HLW glass study), (3) surrogate solids representing a Sr-90 and transuranics (TRU) precipitate slurry derived from precipitation treatment of Envelope C, and (4) glass-forming minerals. This paper describes production, characterization, and associated wasteform testing of these crucible-scale glasses.

Vitrification and leach testing for this demonstration was performed in a radiochemical hood for LAW and remotely in the SRTC Shielded Cells Facility (SCO) for the HLW. The demonstration was performed in four phases, consisting of feed stream preparation, vitrification, glass analysis, and leach testing.

The feed stream preparation for the LAW consisted of mixing a prescribed amount of pretreated LAW supernate with about 10 different glass forming minerals directly into the 600-mL platinum/gold vitrification crucibles. The feed stream preparation phase for HLW involved mixing the waste forms (HLW filtered and dried Envelope D sludge, Cs-137 eluate, and surrogate Sr-90/TRU solids) and glass-forming minerals in a 600-mL platinum/gold crucible. Liquid Cs-eluate waste streams produced from ion exchange treatments of Envelopes A, B, and C were combined and concentrated by evaporation by a factor of ~ 10x before they were mixed with the chemicals. Tc-99 IX eluates are also targeted for inclusion in WTP HLW vitrification flowsheet, but were not used in this study. Feed stream samples were analyzed using inductively coupled plasma-emission spectroscopy and mass spectrometry (ICP-ES and ICP-MS), atomic absorption spectroscopy, gamma spectroscopy and scintillation counting prior to batching. Results from the analyses were transmitted to VSL at Catholic University, where developmental work had been performed with waste simulants.² Results from this development work were the basis for glass formulations used at SRTC. Glass-forming mineral batches were dissolved by both Na₂O₂ fusion and acid dissolution and analyzed by ICP-ES prior to being mixed with the waste streams.

Vitrification of the mixtures was performed in three steps inside custom-designed DelTech, Inc. furnaces (DelTech, Inc., Denver, CO). LAW glasses were produced in a top-loading electrically heated furnace inside a radiochemical hood. The HLW glass was produced using a remotely operated, front-loading electrically heated furnace. Water was initially evaporated from the crucible slurry feed mixtures and collected in one of the off-gas system traps. The off-gas system included a condenser, a dry-ice trap, charcoal filters, and a vacuum pump. The off-gas system was developed to keep hazardous species from the exhaust to the radiochemical hood and to the shielded cells environment. The evaporation stage involved heating the mixture from 50 to 200°C at a rate of 10°C per 30 minutes. The mixture was then calcined by heating it to 900°C at a rate of about 100°C per hour. After the temperature had been held at 900°C for 30 minutes, the off-gas system was disconnected. Finally, in the vitrification stage the material was brought to the melt temperature of 1150°C and held there for typically 4 hours before the resulting glass waste forms were cooled according to prescribed centerline canister cooling programs representative of the different LAW and HLW design canisters.

The product glasses were dissolved by Na₂O₂ fusion and acid dissolution and analyzed using ICP-ES and ICP-MS for elemental and individual radionuclide isotope concentrations. Reference analytical standards were also dissolved and analyzed for quality assurance purposes. Liquid scintillation counting and gamma spectroscopy were also performed to obtain concentrations of radionuclides, including Cs-137/Ba-137m and Sr-90/Y-90. Microstructure analysis of the LAW and HLW glasses used scanning electron microscopy (SEM), energy dispersive x-ray (EDAX) and x-ray diffraction (XRD) analysis.

Two different leach tests were performed on both the LAW and HLW glasses. Durability was measured using the ASTM C-1285 standard nuclear waste glass durability test, commonly referred to as the product consistency test.⁴ This is a crushed glass leach test at 90°C for 7 days using deionized water as leachate. Triplicate tests were performed in sealed stainless steel vessels. Final leachate pH’s were measured, and final elemental concentrations of the filtered, acidified leachates were measured by ICP-ES.

Release rate measurements for toxic substances were determined using the US EPA Toxicity Characteristic Leaching Procedure (TCLP, Method 1311). The TCLP is a standard room temperature crushed glass static leach procedure for 18 hours using acetic acid extraction fluid as leachate. Duplicate tests were performed in sealed polyethylene bottles that were constantly rotated at 30 rpm. A blank and a standard (TCLP Metals in Soil, Environmental Resource Associates, Arvada, CO) were also included with the duplicate glass TCLP tests. Final leachate pH’s and final elemental concentrations of the acid-digested leachates were measured by ICP-ES and AA for total elemental concentration.

LAW. Characterization of the feed streams was performed to determine the concentration of species important to the vitrification process. The major characteristics of the various radioactive feeds used in the vitrification of the LAW and HLW glass are described below. Table I shows characterization of the pretreated AN-103 and AN-102 LAW fraction supernates, the respective amounts on oxide basis of both the waste and glass formers, along with the target glass compositions (final column) supplied by VSL. Contract specifications require at least 14 wt% and 10 wt% waste sodium as oxide for the Env. A and C glasses, respectively. Glass forming minerals used in the radioactive crucible studies were similar to those used in developmental surrogate work at VSL. Sugar was added as reductant to both LAW feeds at targeted levels of 12 moles carbon per 16 moles NO_x in the feed. Nitrate and Nitrite anions were abundant in both feed supernates at levels of about 40% of the total sodium molarity.

HLW. Table II shows that HLW Tank C-106 sludge solids contained total elemental compositions of, primarily, Al, Fe, Na, and Si, with minor components of Ag, Ca, K, Mn, Ni, P, Pb, and Ti also present. The solids also contained about 580 m Ci/g of Cs-137 and 1,200 m Ci/g of Sr-90, as well as significant amounts (Total Alpha = 10 m Ci/g) of transuranics of the Am, Cm, and Pu isotopes. Table III shows the analyzed major species present in the concentrated Cs-eluate waste stream. These results were obtained from detailed characterization of the final 100-mL Cs-eluate concentrate. This eluate concentrate solution contained, primarily, nitrate anion from nitric acid used to elute the Cs and sodium with other cations, Cr, Cu, Ni, and Pb. The eluate also contained about 31,500 m Ci/mL of Cs-137. Table III also shows the characterization of the 150-mL-washed Sr/TRU slurry used in the HLW glass formulation. The actual radioactive Sr/Mn-solids contained a total activity of about 15,900 m Ci of Sr-90, 1,950 m Ci of Cs-137 and 73 m Ci of TRU isotopes. However, inadvertent recombination of this stream with remnants of the original Sr/TRU testing necessitated use of a Sr/TRU surrogate in the actual HLW glass formulation. The dry powder surrogate was formulated to simulate all of the major elemental species in the radioactive Sr/TRU precipitate slurry.

The waste streams detailed in Tables II and III were combined with a nonradioactive glass former batch consisting of the minerals boric acid, lithium carbonate and silica. The ratio of chemical compounds used in preparing the final 80 grams of HLW glass was determined from analyzing the HLW streams shown in Tables II and III, and considering the corresponding target glass composition. Table IV lists the amounts of each stream used in the HLW glass formulation. The total targeted waste loading on an oxide basis in the HLW glass was 56 wt%; the total HLW sludge loading on an oxide basis in the HLW glass was 36.7 wt%; and the total HLW sludge iron content in the glass was targeted to meet the contract minimum limit of 12.5 wt% as Fe₂O₃.

LAW. Table V shows the measured composition of significant species on an oxide basis of the two LAW glasses. Summation of the various components on oxide basis gives about 97 to 100% total for AN-103 and about 102 to 107% for AN-102. Boron is indeterminate by acid dissolution due to use of boric acid and both sodium and nickel are not available in the peroxide fusion method. Both glasses

contained targeted amounts of waste sodium. The radionuclide contents of the glasses are shown to be below the contract maximums for Cs-137 (< 3 Ci/m³), Tc-99 (< 0.1 Ci/m³), Sr-90 (< 20 Ci/m³) and TRU (< 100 h Ci/g) indicating that each original tank waste was adequately pretreated to achieve the necessary removal of these three radionuclides and the transuranics. Current WTP engineering plans are to require 10X less Cs-137 in the LAW glasses, i.e., <0.3 Ci/m³, which will require removal of Cs-137 to lower levels than were targeted for pretreated AN-103 LAW fraction in this study. The AN-102 LAW fraction was pretreated to remove enough Cs-137 to meet the 0.3 Ci/m³ level for the glass produced in this work.

The LAW glasses were also analyzed using EPA SW-846 protocol, which included analyses for cyanide, ignitability, RCRA metals, volatile and semivolatile organic compounds, dioxins and furans and pesticides, and PCBs. Detailed results from these analyses are not presented in this summary paper, as the data are currently being documented for access in the public literature through WSRC technical reports. However, initial conclusions from these regulatory analyses data shown that both AN-103 and AN-102 LAW glasses do not exhibit any of the tested characteristics of hazardous waste as defined by the Washington State Department of Ecology.

HLW: The HLW glass elemental analysis is shown in Table VI. The glass was analyzed to contain on an oxide basis, primarily, Si, Fe, Na, B, Al, Li, Mn, and Sr. The results shown in Table VI are averages of two separate glass samples from the -200 mesh fraction not used in the PCT durability test. Duplicate results agreed to within 10% or better. Table VI also shows the HLW radionuclide results for the glass. Total alpha and beta were measured by liquid scintillation counting, Cs-137 by gamma spectroscopy, Tc-99 by ICPMS, and Sr-90 by separation and beta-counting.

The actinides Am-241, Cm-244, and Pu isotopes were detected by separation and counting methods. Sr-90 and Y-90 are in secular equilibrium, and their activities are equal. Ba-137m is in secular equilibrium with Cs-137, and its activity is equal to 95% of that for Cs-137 (5% of the Cs-137 decays directly to the stable Ba-137).

XRD and SEM analyses were performed on several pieces of the 100–200 mesh LAW and HLW glass granules prepared for the PCT. No crystalline phases were measurable in the LAW glasses, and microscopy data from these glasses confirmed their amorphous structure. Two phases were identified in the HLW glass: one that appeared featureless and apparently amorphous and one identified as a crystalline phase indicated by similar square-shaped spots on the glass surface. These phases were examined by energy dispersive x-ray analysis (EDAX). The amorphous phase was the glass itself, and the EDAX showed peaks resulting from the major glass components such as Si, Sr, Mn, and Fe. The EDAX spectrum of the crystalline phase indicates that it was primarily Fe with minor peaks attributed to Ni and Cr. The x-ray diffraction pattern from analysis of milligram quantities of the powdered HLW glass indicates a peak that is best matched with trevorite, NiFe₂O₄, a crystalline phase in the spinel family of crystals with generalized formula ((Fe,Ni,Mn)(Fe,Cr)₂O₄). Quantification of the extent of crystallization by comparing the crystal surface area versus amorphous surface area on a single glass piece indicates an estimated 1.5% crystallinity; however, based on the PCT results discussed below, it did not significantly affect the glass’s durability compared to the EA glass. VSL and PNNL researchers have also reported similar results for observation of spinel phases in heat-treated HLW glasses in recent studies with surrogates² and actual Hanford Site tank waste.³

Durability of the LAW and HLW glasses was measured by the Product Consistency Test, or PCT (ASTM Test C 1285-97).⁴ Triplicate samples of the ground glass were subjected to the 90° C PCT, along with the appropriates blanks and reference low-activity reference material glass (LRM for the LAW glasses⁵) and environmental assessment (EA)⁶ glass for the HLW glass. Normalized releases were calculated based on the average measured composition from glass dissolution and analysis given previously in Tables V and VI. The average normalized releases for B, Si, and Na, and the standard deviations from triplicate tests are reported in Table VII in units of grams glass leached per square meter of glass exposed in the PCT. Average measured values for the LRM and EA glasses tested concurrently with the LAW (LRM comparison) and the HLW (EA comparison) glasses are also presented. Published values for LRM and EA are also presented for comparison.^5,6 Normalized mass losses for both the LAW glasses are well below the contract specification maximum for LAW glasses of <2 g/m². Also, normalized mass losses for the HLW glass are clearly below the average of the EA glass, indicating the HLW glass is more durable than the EA glass.

The toxic leach characteristics of the HLW glass waste form were measured by the TCLP on duplicate LAW and HLW glass samples along with a blank and a standard. The testing was performed by BWXT, Inc. for the LAW glasses at 23 ± 2ºC, performed remotely in the SRTC shielded cells at 18 to 21ºC because the remote environment could not be controlled to exactly 23 ± 2ºC. The average concentrations in leachates resulting from tests of the LAW and HLW glasses are reported in Table VIII. All glasses were shown to pass the TCLP for the Universal Treatment Standards limits.

Laboratory-scale tests conducted at SRTC have successfully demonstrated (1) the production of two immobilized LAW borosilicate glass products incorporating actual Hanford Site decontaminated LAW fractions at the targeted waste loadings based on waste sodium, and (2) the production of a HLW glass incorporating Envelope D HLW sludge from Hanford Site Tank C-106 and other secondary waste from pretreatment of Tanks AN-103, AN-102, and AZ-102. Glass analyses indicate that both the LAW and HLW glass waste form compositions were close to target formulations. Waste sodium was oxide-loaded above the contract minimums of 14 and 10 wt%, respectively, for AN-103 and AN-102. A total waste loading of 36.7 wt% Envelope D sludge solids on an oxide basis was demonstrated with attainment of the contract minimum of 12.5 wt% Fe₂O₃ from the HLW sludge. Microstructure analysis of the LAW glasses indicated no detectable crystalline phases, whereas similar analysis of HLW glass showed the presence of nickel spinel trevorite (NiFe₂O₄) crystalline phase. These crystals did not degrade the leaching characteristics of the glass; since the glass waste form passed standard durability PCT leach tests and TCLP leach tests. Both of the AN-103 and AN-102 LAW glasses do not exhibit any of the EPA SW-846-tested characteristics of hazardous waste as defined by the Washington State Department of Ecology (WAC 173-303, May 2000, Dangerous Waste Regulations, Washington Administrative Code, as amended).

1. WTP Contract DE-AC27-01RV14136, Mod. M016 (see http://www.handord.gov/orp/contracts/deac27-01rv14136/toc.html).
2. (a) I. S. Muller et al., "Glass Formulation & Testing with RPP-WTP LAW Simulants", VSL-01R3560-2 (Feb. 23, 2001); (b) W. K. Kot and I. L. Pegg, "Glass Formulation & Testing with RPP-WTP HLW Simulants", VSL-01R2540-2 (Feb.16, 2001).
3. (a) G. L. Smith et al., "Vitrification and Product Testing of AW-101 and AN-107 Pretreated Waste", PNNL-13372 (Oct. 2000);
   (b) G. L. Smith et al., "Vitrification and Product Testing of C-104 and AZ-102 Pretreated Sludge Mixed with Flowsheet Quantities of Secondary Wastes", PNNL-13452 (Feb. 2001).
4. "Standard Test Methods for Determining Chemical Durability of Nuclear Waste Glasses: The Product Consistency Test (PCT)", ASTM C1285-97, Annual Book of ASTM Standards, Vol. 12.01, Philadelphia, Pa, pp. 774-791 (1995).
5. (a) W. L. Ebert and S. F.Wolf, "Round-Robin Testing of a Reference Glass for Low-Activity Waste Forms", ANL-99/22 (Oct. 1999); (b) D. Peeler et al., "Characterization of the Low Level Waste Reference Glass (LRM)", WSRC-TR-99-00095, Rev. 0, (Mar. 30, 1999).
6. C. M. Jantzen, N. E. Bibler, D. C. Beam, C. L. Crawford, and M. A. Pickett, "Characterization of the Defense Waste Processing Facility (DWPF) Environmental Assessment (EA) Glass Standard Reference Material (U)", WSRC-TR-92-346, Rev. 1 (1993).