[en] The Savannah River National Laboratory (SRNL) is part of a consortium that is looking to improve the retention of aluminum, chromium, and sulfate in high level radioactive waste (HLW) glass. Such glass has been produced by the Defense Waste Processing Facility (DWPF) at the Savannah River Site (SRS) in South Carolina since it began operating in 1996 and is planned to be produced by the River Protection Project-Waste Treatment Plant (WTP) at the Hanford Site in Washington. The consortium conducting this study, which is designated as Task No.6 by the Department of Energy (DOE) Environmental Management (EM) program sponsoring this effort, is made up of personnel from SRNL, the Pacific Northwest National Laboratory (PNNL), and the V.G. Khlopin Radium Institute (KRI). Coordinated glass experimental work will be performed by each member of the consortium. The glasses that are being studied were selected to further the understanding of composition-property relationships within the glass regions of interest to both DWPF and WTP. Forty-five (45) glasses, making up the US test matrix, were batched and fabricated to support the study. The chemical compositions of these glasses were measured by SRNL's Process Science Analytical Laboratory (PSAL) under the auspices of an analytical plan. In addition, two heat treatments (quenched and centerline canister cooled, ccc) of each glass were subjected to the 7-day Product Consistency Test (PCT) to assess their durabilities. More specifically, the Method A of the PCT (ASTM C-1285-2002) was used for these tests. Measurements of the resulting leachate solutions were conducted by PSAL under the auspices of three analytical plans. A statistical review of the PSAL measurements of the chemical compositions and of the PCT results for the glasses making up the US test matrix is provided in this memorandum. Target, measured, and measured bias-corrected compositional views were determined for these glasses. The durability results for the US study glasses are compared to those of the Environmental Assessment (EA) glass. All of the US glasses yielded PCTs that are lower than those of the EA glass. The largest PCT values are those measured for the ccc versions of US-27 and US-18 whose boron normalized leachate (NL[B]) values in grams per liter (g/L) were 16.4 g/L (based on the targeted composition) and 10.7 g/L (based on the targeted composition), respectively. The 16.4 g/L is just below the value of 16.695 g/L for EA's NL[B] that was reported by Jantzen et al. in WSRC-TR-92-0346, Revision 1. For the quenched version of the glasses, the largest NL[B] value is 0.67 g/L (based on the targeted composition). Thus, some statistically significant differences were seen between the quenched and ccc versions for some of the glasses. It should be noted that the thirty (30) glasses making up the KRI test matrix were not included in these analyses

[en] Savannah River National Laboratory's frit development effort for SB4 is being driven by the most current CBU option for this sludge, referred to as Case 15C Blend 1. Candidate frits have been identified for this option via a paper study approach developed by Peeler and Edwards with the intent of down-selecting to a set of key frits whose operating windows (i.e., WL intervals that meet PCCS MAR criteria) are robust to and/or selectively optimal for this sludge option. The primary frits that appear attractive on paper (i.e., down-selected via the paper study) are now being incorporated into this experimental study. The potential for the formation of a nepheline primary crystalline phase is an important factor in frit development for SB4, due to the high Al₂O₃ content of this sludge. Based upon earlier work by Li et al., glasses that do not satisfy the constraint: (SiO₂/SiO₂ + Na₂O + Al₂O₃) > 0.62 where the oxides are expressed as mass fractions in the glass, will precipitate nepheline as their primary crystalline phase, hindering the durability of the glass. Based on the most recent compositional projection from the CBU for SB4 (Case 15C Blend 1), 16 glasses have been selected to complement the earlier work by continuing the investigation into the ability of the above constraint to predict the occurrence of a nepheline primary crystalline phase for SB4 glasses and into the impact of such phases on the durability of the SB4 glasses. Glasses were selected to cover WLs which tightly bound the nepheline discriminator value of 0.62, with the intent of refining this value to a level of confidence where it can be incorporated into offline administrative controls and/or the PCCS to support SME acceptability decisions. In addition, glass specimens at WLs of 35 and 40% will be prepared and analyzed to contribute needed data to the ComPro(trademark) database in anticipation of a variability study for SB4. The glasses in Table 4-3 are to batched and fabricated using standard procedures. Visual observations and other analytical techniques are to be used, as needed, to assess the presence of crystals with specific interest in the nepheline primary phase. The durability of these glasses (for both quenched and centerline canister cooled versions) is to be measured using the ASTM PCT Method A. The results from these efforts are to be documented in a subsequent report. The results of this study will provide valuable input for the frit development efforts and subsequent feedback to the CBU regarding the relative viability of the current SB4 option under consideration. The refined nepheline discriminator value will provide a guideline for the avoidance of nepheline crystallization in SB4 glasses and aid in down-selection of frit compositions. These data will be combined with the results of melt rate studies and a paper study of the frits robustness with regard to variability in the sludge composition to provide an optimized frit recommendation to DWPF for immobilization of SB4

[en] This study was undertaken to investigate the effect of elevated sulfate and molybdenum concentrations in nuclear waste glasses. A matrix of 24 glasses was developed and the glasses were tested for acceptability based on visual observations, canister centerline-cooled heat treatments, and chemical composition analysis. Results from the chemical analysis of the rinse water from each sample were used to confirm the presence of SO^2-₄ and MoO₃ on the surface of glasses as well as other components which might form water soluble compounds with the excess sulfur and molybdenum. A simple, linear model was developed to show acceptable concentrations of SO₄^2- and MoO₃ in an example waste glass composition. This model was constructed for scoping studies only and is not ready for implementation in support of actual waste vitrification. Several other factors must be considered in determining the limits of sulfate and molybdenum concentrations in the waste vitrification process, including but not limited to, impacts on refractory and melter component corrosion, effects on the melter off-gas system, and impacts on the chemical durability and crystallization of the glass product

[en] The objective of this study was to determine a sulfate solubility limit in glass for Sludge Batch 7b (SB7b). The SB7b composition projection provided by Savannah River Remediation (SRR) on May 25, 2011 was used as the basis for formulating glass compositions to determine the sulfate limit. Additions of Na₂O to the projected sludge composition were made by the Savannah River National Laboratory (SRNL) due to uncertainty in the final concentration of Na₂O for SB7b, which is dependent on washing effectiveness and the potential need to add NaOH to ensure an acceptable projected operating window. Additions of 4, 6, and 8 wt % Na₂O were made to the nominal May 25, 2011 composition projection. An updated SB7b composition projection was received from SRR on August 4, 2011. Due to compositional similarities, no additional experimental work using the August 4, 2011 compositions was considered to be necessary for this study. Both Frit 418 and Frit 702 were included in this study. The targeted sulfate (SO₄^2-) concentrations of the study glasses were selected within the range of 0.6 to 0.9 wt % in glass. A total of 52 glass compositions were selected based on the compositional variables of Na₂O addition, Actinide Removal Process (ARP) stream addition, waste loading, frit composition, and sulfate concentration. The glasses were batched, melted, and characterized following SRNL procedures. Visual observations were recorded for each glass after it cooled and used as in indicator of sulfur retention. Representative samples of each of the glasses fabricated were subjected to chemical analysis to determine whether the targeted compositions were met, as well as to determine the quantity of sulfate that was retained after melting. In general, the measured composition data showed that there were only minor issues in meeting the targeted compositions for the study glasses, and the measured sulfate concentrations for each study glass were within 10% of the targeted values. The results for the SB7b glasses fabricated with Frit 418 showed an apparent trend of increasing sulfate retention with increasing Na₂O additions to the 5/25/11 sludge projection. This trend appears contradictory to other recent studies of sulfate retention in Defense Waste Processing Facility (DWPF) type glasses. Additional apparent contradictions to this trend were found in the data collected in the present study. Overall, the results for the SB7b sulfate study glasses with Frit 418 and the 5/25/11 projection with Na₂O additions showed that subtle changes in this complex glass composition impacted the degree of sulfate retention. These results do however provide confidence that a 0.6 wt % sulfate limit in glass is warranted for Frit 418 with the SB7b compositions evaluated in this study. The results for the SB7b glasses fabricated with Frit 702 are consistent with those of the previous SB7a study in that Frit 702 allowed for higher sulfate retention as compared to Frit 418 for the same sludge compositions. It is recommended that the DWPF implement a sulfate concentration limit of 0.6 wt % in glass for SB7b processing with Frit 418. If a higher than projected sulfate concentration is measured when SB7b processing begins (i.e., if a sulfate concentration higher than 0.6 wt % becomes necessary to achieve targeted waste loadings), DWPF should consider a transition to Frit 702. The sulfate limit could likely be raised to 0.8 wt % by transitioning to this frit. However, if DWPF considers transitioning from Frit 418 to Frit 702, additional glasses should be fabricated to confirm this higher limit due to the issues with incorrect B₂O₃ concentrations for some of the glasses made with Frit 702 in this study. There are several factors other than sulfate retention that must also be carefully considered prior to changing frit compositions.

[en] Hydridable metal alloys are used at the Savannah River Site to process tritium. The goal of this work was to develop a mechanical alloying process as a low-cost option to produce these alloys on-site. High-speed milling at elevated temperatures has the potential to significantly reduce the time and cost of the mechanical alloying process. It was demonstrated that elemental metal powders can be alloyed in an attritor mill under argon. In order to form LaNi_4.25Al_0.75 from elemental metals it was found that lanthanum and nickel must be alloyed prior to adding aluminum. It was also demonstrated that metal powders could be alloyed in the high-speed attritor with the temperature in the mill equilibrating at ∼220 C. Optimization of the process parameters will require additional testing

[en] This Phase 3 study was undertaken to complement the previous phases of the nepheline formation studies1, 2 by continuing the investigation into the ability of the nepheline discriminator to predict the occurrence of nepheline crystallization in Sludge Batch 4 (SB4) glasses and into the impact of such phases on the durability of the SB4 glasses. The Phase 3 study had two primary objectives. The first was to continue to demonstrate the ability of the discriminator value to adequately predict the nepheline formation potential for specific glass systems of interest. The second was to generate additional data that have a high probability of supporting the SB4 variability study. To support these two objectives, sixteen glasses were selected based on the most recent SB4 compositional projection, Case 15C Blend 1.3 Four different frits were included, based on previous assessments of projected operating windows and melt rate,4, 5 with four WLs selected for each frit. Eight of these frit-sludge combinations covered WLs which tightly bound the nepheline discriminator value of 0.62, with the intent of refining this value to a level of confidence where it can be incorporated into offline administrative controls and/or the Process Composition Control System (PCCS) to support Slurry Mix Evaporator (SME) acceptability decisions. The remaining eight frit-sludge combinations targeted lower WLs (35 and 40%) and were prepared and analyzed to contribute needed data to the ComPro database6 to support a potential variability study for SB4

[en] The Savannah River National Laboratory (SRNL) Frit Development Team recommends that the Defense Waste Processing Facility (DWPF) utilize Frit 418 for initial processing of high level waste (HLW) Sludge Batch 5 (SB5). The extended SB5 preparation time and need for DWPF feed have necessitated the use of a frit that is already included on the DWPF procurement specification. Frit 418 has been used previously in vitrification of Sludge Batches 3 and 4. Paper study assessments predict that Frit 418 will form an acceptable glass when combined with SB5 over a range of waste loadings (WLs), typically 30-41% based on nominal projected SB5 compositions. Frit 418 has a relatively high degree of robustness with regard to variation in the projected SB5 composition, particularly when the Na₂O concentration is varied. The acceptability (chemical durability) and model applicability of the Frit 418-SB5 system will be verified experimentally through a variability study, to be documented separately. Frit 418 has not been designed to provide an optimal melt rate with SB5, but is recommended for initial processing of SB5 until experimental testing to optimize a frit composition for melt rate can be completed. Melt rate performance can not be predicted at this time and must be determined experimentally. Note that melt rate testing may either identify an improved frit for SB5 processing (one which produces an acceptable glass at a faster rate than Frit 418) or confirm that Frit 418 is the best option.

[en] High level radioactive wastes are being vitrified at the Savannah River Site for long term disposal. Many of the wastes contain sulfate at concentrations that can be difficult to retain in borosilicate glass. This study involves efforts to optimize the composition of a glass frit for combination with the waste to improve sulfate retention while meeting other process and product performance constraints. The fabrication and characterization of several series of simulated waste glasses are described. The experiments are detailed chronologically, to provide insight into part of the engineering studies used in developing frit compositions for an operating high level waste vitrification facility. The results lead to the recommendation of a specific frit composition and a concentration limit for sulfate in the glass for the next batch of sludge to be processed at Savannah River.

[en] processing strategy for the Hanford Tank Waste Treatment and Immobilization Plant (WTP). The basis of this alternative approach is an empirical model predicting the crystal accumulation in the WTP glass discharge riser and melter bottom as a function of glass composition, time, and temperature. When coupled with an associated operating limit (e.g., the maximum tolerable thickness of an accumulated layer of crystals), this model could then be integrated into the process control algorithms to formulate crystal tolerant high level waste (HLW) glasses targeting higher waste loadings while still meeting process related limits and melter lifetime expectancies. This report provides a review of the scaled melter testing that was completed in support of the Defense Waste Processing Facility (DWPF) melter. Testing with scaled melters provided the data to define the DWPF operating limits to avoid bulk (volume) crystallization in the un-agitated DWPF melter and provided the data to distinguish between spinels generated by K-3 refractory corrosion versus spinels that precipitated from the HLW glass melt pool. This report includes a review of the crystallization observed with the scaled melters and the full scale DWPF melters (DWPF Melter 1 and DWPF Melter 2). Examples of actual DWPF melter attainment with Melter 2 are given. The intent is to provide an overview of lessons learned, including some example data, that can be used to advance the development and implementation of an empirical model and operating limit for crystal accumulation for WTP. Operation of the first and second (current) DWPF melters has demonstrated that the strategy of using a liquidus temperature predictive model combined with a 100 °C offset from the normal melter operating temperature of 1150 °C (i.e., the predicted liquidus temperature (TL) of the glass must be 1050 °C or less) has been successful in preventing any detrimental accumulation of spinel in the DWPF melt pool, and spinel has not been observed in any of the pour stream glass samples. Spinel was observed at the bottom of DWPF Melter 1 as a result of K-3 refractory corrosion. Issues have occurred with accumulation of spinel in the pour spout during periods of operation at higher waste loadings. Given that both DWPF melters were or have been in operation for greater than 8 years, the service life of the melters has far exceeded design expectations. It is possible that the DWPF liquidus temperature approach is conservative, in that it may be possible to successfully operate the melter with a small degree of allowable crystallization in the glass. This could be a viable approach to increasing waste loading in the glass assuming that the crystals are suspended in the melt and swept out through the riser and pour spout. Additional study is needed, and development work for WTP might be leveraged to support a different operating limit for the DWPF. Several recommendations are made regarding considerations that need to be included as part of the WTP crystal tolerant strategy based on the DWPF development work and operational data reviewed here. These include: Identify and consider the impacts of potential heat sinks in the WTP melter and glass pouring system; Consider the contributions of refractory corrosion products, which may serve to nucleate additional crystals leading to further accumulation; Consider volatilization of components from the melt (e.g., boron, alkali, halides, etc.) and determine their impacts on glass crystallization behavior; Evaluate the impacts of glass REDuction/OXidation (REDOX) conditions and the distribution of temperature within the WTP melt pool and melter pour chamber on crystal accumulation rate; Consider the impact of precipitated crystals on glass viscosity; Consider the impact of an accumulated crystalline layer on thermal convection currents and bubbler effectiveness within the melt pool; Evaluate the impact of spinel accumulation on Joule heating of the WTP melt pool; and Include noble metals in glass melt experiments because of their potential to act as nucleation sites for spinel crystallization

[en] This report is the third in a series of studies of the impacts of the addition of Crystalline Silicotitanate (CST) and Monosodium Titanate (MST) from the Small Column Ion Exchange (SCIX) process on the Defense Waste Processing Facility (DWPF) glass waste form and the applicability of the DWPF process control models. MST from the Salt Waste Processing Facility is also considered in the study. The KT07-series glasses were selected to evaluate any potential impacts of noble metals on their properties and performance. The glasses characterized thus far for the SCIX study have not included noble metals since they are not typically tracked in sludge batch composition projections. However, noble metals can act as nucleation sites in glass melts, leading to enhanced crystallization. This crystallization can potentially influence the properties and performance of the glass, such as chemical durability, viscosity, and liquidus temperature. The noble metals Ag, Pd, Rh, and Ru were added to the KT07-series glasses in concentrations based on recent measurements of Sludge Batch 6, which was considered to contain a high concentration of noble metals. The KT04-series glasses were used as the baseline compositions. After fabrication, the glasses were characterized to determine their homogeneity, chemical composition, durability, and viscosity. Liquidus temperature measurements are also underway but were not complete at the time of this report. The liquidus temperature results for the KT07-series glasses, along with several of the earlier glasses in the SCIX study, will be documented separately. All of the KT07-series glasses, both quenched and slowly cooled, were found to be amorphous by X-ray diffraction. Chemical composition measurements showed that all of the glasses met their targeted compositions. The Product Consistency Test (PCT) results showed that all of the glasses had chemical durabilities that were far better than that of the Environmental Assessment benchmark glass. The measured PCT responses were well predicted by the current DWPF Product Composition Control System (PCCS) durability models. The measured viscosity values for each KT07-series glass were acceptable for DWPF processing and were well predicted by the current PCCS model. Overall, the results show that the inclusion of relatively high concentrations of noble metals (in terms of expected values for a DWPF sludge batch) had no significant impact on the properties and performance of these glass compositions. There were no significant differences in the measured properties when compared to those of the KT04-series glasses, which did not contain noble metals. Liquidus temperature measurements are still underway and there may be an impact of the noble metals on those measurements. However, no adverse effects were noted in terms of crystallization after slow cooling. At the completion of these studies, all of the data generated will be reviewed with regard to the applicability of the DWPF PCCS models and recommendations will be made as to whether the validation ranges of the current models can be extended, or whether some or all of the models need to be refit to allow for the incorporation of the SCIX streams. As changes are made to the projected sludge compositions and the volume of the SCIX material, additional evaluations should be performed.