Liquidus temperatures (TL) were measured for high-level waste (HLW) borosilicate glasses covering a Savannah River composition region. The primary crystallization phase for most glasses was spinel, a solid solution of trevorite (NiFe2O4) with other oxides (FeO, MnO, and Cr2O3). The TL values ranged from 859 to 1310°C. Component additions increased the TL (per mass%) as Cr2O3 261°C, NiO 85°C, TiO2 42°C, MgO 33°C, A12O3 18°C, and Fe2O3 18°C and decreased the TL (per mass%) as Na2O −29°C, Li2O −28°C, K2O −20°C, and B2O3 −8°C. Other oxides (CaO, MnO, SiO2, and U3O8) had little effect. The effect of RuO2 is not clear.

The liquidus temperatures, TL, of rare earth-alumino-borosilicate glasses were measured as a function of glass composition. The TL values ranged from 1153°C to 1405°C. Three primary crystalline phases were identified in the study. The most frequently encountered was a rare earth silicate phase in which the TL values ranged from 1153°C to 1405°C. A12O3 was encountered in glasses with an A12O3:SiO2 mass ratio greater than 1, with TL values ranging from 1242°C to 1305°C. Alumino-silicate crystals were encountered as the primary phase in glasses with less than 43 mass% of mixed rare earth oxides (Ln2O3) with TL values between 1164°C and 1255°C. A linear relationship between total mixed rare earth oxide concentration and TL was found within all three primary phase fields. In the rare earth silicate primary phase field, normalizing the Ln2O3 concentration with its mean ionic radius enhanced this linear relationship.

The solubilities of Hf and Gd in sodium alumino-borosilicate glasses based on the target compositions were examined and confirmed by electron microprobe analysis. The measured compositions of essentially crystal-free glasses are generally homogeneous and close to the target compositions. Therefore, the solubilities of Gd and Hf in sodium alumino-borosilicate glasses based on the target glass compositions are valid. However, for glasses containing precipitates (crystals grown from the melt) and undissolved HfO2 with overgrowths, the chemical compositions are often heterogeneous and may be significantly different from the target compositions. Precipitated crystalline phases include a rare earth silicate with the apatite sructure (NaGd9(Si5.25B)O26) in a gadolinium sodium alumino-borosilicate glass and a HfO2 phase in hafnium sodium alumino-borosilicate glasses.

Mixed wastes involve radioactive constituents and hazardous components and must be conditioned to be disposed. Joint vitrification of low and intermediate level radioactive wastes (LILW) from nuclear power plants, together with spent catalysts, sorbents, and cathode ray tube (CRT) glass is proposed as a suitable means to condition these wastes. Preliminary experiments on vitrification in crucibles in a laboratory-scale resistive furnace were carried out that demonstrate possibility of 40 wt.% LILW salt loading in a batch. Homogeneous silicate- and borosilicatebased glasses have been obtained and characterized. A phase separation problem at high sulfate and chloride content caused a “yellow phase” to be formed. To prevent this, sulfates and chlorides must be separated from the LILW. The most promising method of mixed waste vitrification is inductive melting in a cold crucible (IMCC) because of its high throughput rate, high achievable temperatures, and long lifetime of the apparatus. Bench-scale tests have been carried out. Preliminary testing of IMCC of spent CRT glass and a combined batch of this glass and LILW salt surrogate was also conducted. LILW salt loading in the batch reached 30 wt.%. The starting melt was formed by heating electrically conductive silicon carbide rods in a high frequency electromagnetic field. The CRT glass melt has low electric conductivity and high viscosity at temperatures of about 1200°C, which does not allow the required IMCC starting melt to be formed. Addition of up to 30 wt.% of surrogate RW salts to the batch increases the electric conductivity and reduces the viscosity of the melt to values that support IMCC. The batch is preferably fed as a calcine.

Technetium-99 and cesium-137 are two radionuclides in high level waste (HLW) that can volatilize from high temperature melters during the immobilization of the HLW into a borosilicate glass. At Savannah River Site (SRS) we have obtained data that indicate that this volatilization is small from the full scale production melter in the Defense Waste Processing Facility (DWPF). These data were obtained during the vitrification of the first HLW macrobatch at SRS. This campaign lasted ∼2.5 years and produced ∼9 hundred metric tons of glass from ∼1.6 million liters of HLW. Losses of Tc-99 and Cs-137 were determined by comparing their measured concentrations in the glass with their respective predicted concentrations based on the composition of the HLW being vitrified. For three glass samples taken during the campaign, the measured and predicted concentrations agreed within 7% or better indicating a small loss of either radionuclide. The DWPF melter operates with a cold cap on the surface of the melt. This cold cap could enhance the radionuclide retention, especially Tc-99.

At the Savannah River Site (SRS) we are currently finalizing the design for a multi-system vitrification process that will be installed in the F-Canyon Multi-Purpose Process Facility (MPPF), an existing highly shielded, remotely operated facility. Authorization to proceed beyond the preliminary design based on the recommendation of a Formal Design Review Board was requested in May of 1999.

The Savannah River Technology Center (SRTC) Process Development Group has been conducting research and developing a process to identify equipment design bases and process operating parameters since 1996. The goal of the project is to stabilize a tank of ∼11,000 liters of nitric acid solution containing valuable isotopes of americium (Am) and curium (Cm). Vitrification has been selected as the most attractive alternative for stabilization and provides the opportunity for recovery and eventual reuse of the actinides. The final glass form will be placed in interim storage awaiting a disposition by the Department of Energy. This paper presents a brief history of the stabilization program and an overview of the entire Am/Cm stabilization process. This paper also provides details of a specific processing issue related to drain tube pluggage (devitrification) that was encountered during the development of the baseline batch vitrification process, and the remedy employed to reduce the potential for further drain tube pluggage.

Liquidus temperature (TL) studies of high-Zr high-level waste (HLW) borosilicate glasses have identified three primary phases: baddelyite (ZrO2), zircon (ZrSiO4), and alkali-zirconium silicates, such as parakeldyshite (Na2ZrSi2O7). Using published TL data for HLW glasses with these primary phases, we have computed partial specific TLS for major glass components. On the Na2O-SiO2-ZrO2 submixture, we have determined approximate positions of the boundaries between the baddelyite, zircon, and parakeldyshite primary phase fields. The maximum that can dissolve at 1150‘C in a borosilicate HLW glass subjected to common processability and acceptability constraints appears to be 16.5 mass% ZrO2.

A thermochemical representation of the Na-Al-Si-B-O system relevant for nuclear waste glass has been developed based on the associate species approach for the glass solution phase. Thermochemical data were assessed and associate species data determined for binary and ternary subsystems in the Na2O-AI2O3-B2O3-SiO2 system. Computed binary and ternary phase diagrams were compared to published diagrams during this process, with adjustments in data made as necessary to obtain consistent thermodynamic values. The resulting representation for the four oxide system was used to help understand the problem of nepheline precipitation in certain waste glass formulations.

Glasses related to those used for the vitrification of high-level waste (HLW) have been produced from the sodium-lithium borosilicate system. Their thermal and structural characteristics have been measured for a wide range of boron oxide contents (NBS series) and for a range of alkali oxide to boron oxide ratios, R, for several fixed silica to boron oxide ratios, K (ABS series). The NBS series of glasses was seen to exhibit a maximum in the glass transition temperature, Tg, and a minimum in the fraction of 4-coordinated borons, N4, at 22 mol% and 29 mol% boron oxide, respectively. With variation of K and R, initial results indicate an increase in Tg to a maximum before a gradual decrease. Density measurements show a general increase before remaining within experimental error at larger R. Initial α-particle irradiation tests have been carried out on several base glass samples and show evidence of B(α,n) reactions, the occurrence of which could have important consequences for the future viability of wasteforms designed for reprocessing of higher burn-up fuel.

Using electron energy-loss filtered transmission electron microscopy (EFTEM), we have observed the formation of silicon-rich zones on the corroded surface of a West Valley (WV6) glass. This layer is approximately 100-200 nm thick and is directly underneath a precipitated smectite clay layer. Under conventional (C)TEM illumination, this layer is invisible; indeed, more commonly used analytical techniques, such as x-ray energy dispersive spectroscopy (EDS), have failed to describe fully the localized changes in the boron and silicon contents across this region. Similar silicon-rich and boron-depleted zones were not found on corroded Savannah River Laboratory (SRL) borosilicate glasses, including SRL-EA and SRL-51, although they possessed similar-looking clay layers. This study demonstrates a new tool for examining the corroded surfaces of materials.

This paper reports the results of corrosion tests with glass-bonded sodalite, a ceramic waste form (CWF) that is being developed to immobilize radioactive electrorefiner salt used to condition spent sodium-bonded nuclear fuel, and with sodalite and binder glass, the two major components of the CWF. These tests were performed with dilute pH-buffered solutions in the pH range of 5-10 at temperatures of 70 and 90°C to determine the pH dependences of the forward dissolution rates of the CWF and its components. The tests show that the pH dependences of the dissolution rates of sodalite, binder glass, and glass-bonded sodalite are similar to the pH dependence of dissolution rate of borosilicate nuclear waste glasses, with a negative pH dependence in the acidic region and a positive pH dependence in the basic region. The dissolution rates are higher at 90°C than at 70°C. Our results on the forward dissolution rates and their temperature and pH dependences will be used as components of a waste form degradation model to predict the long-term behavior of the CWF in a nuclear waste repository.

Borosilicate glasses loaded with ∼10 wt % plutonium were found to produce plutonium-silicate alteration phases upon aqueous corrosion under a range of conditions. The phases observed were generally rich in lanthanide (Ln) elements and were related to the lanthanide orthosilicate phases of the monoclinic Ln2SiO5 type. The composition of the phases was variable regarding [Ln]/[Pu] ratio, depending upon type of corrosion test and on the location within the alteration layer. The formation of these phases likely has implications for the incorporation of plutonium into silicate alteration phases during corrosion of titanate ceramics, high-level waste glasses, and spent nuclear fuel.

The potential effect of alteration phases on the kinetics of glass corrosion has been examined in a preliminary series of Product Consistency Tests (0.5, 1, 3, 7, 35, and 91 days). Crushed samples of a relatively simple Li-Na-Ca-K-Al-B-Si glass were reacted in the presence of a relatively high ionic strength fluid, to which various alteration phases (analcime, adularia, chabazite, or Na-montmorillonite) were added as “seed-crystals”. The release of boron and lithium were used to monitor the corrosion rate of the glass. In general, corrosion rates varied only slightly between the tests with different seed-crystals types. Boron and lithium contents in tests with analcime or adularia were slightly higher than tests with Na-montmorillonite or chabazite present. Silicon concentrations did not display any consistent variation over the testing interval, remaining relatively similar to the starting leachant value of 3.5 × 10−2 M. The concentration of aluminum, however, decreased significantly during the first 35 days of testing and could be inversely correlated to boron and lithium concentrations. The concentration of aluminum then increased between 35 and 91 days, whereas boron and lithium concentrations remained relatively static. The noted correlation between aluminum and boron (or lithium) suggests a coupling of the rate of glass corrosion with aluminum concentration.

The intrinsic dissolution rates of nine borosilicate waste glasses were extracted from the results of MCC- I tests conducted for durations long enough that the solution pH reached a nearly constant value but short enough that the buildup of dissolved species did not affect the dissolution rate. The effects of the pH and temperature on the measured rates were deconvoluted to determine the sensitivity of the rate to the glass composition. The intrinsic dissolution rates were similar for all of these glasses and were not correlated with the glass composition. The mean and standard deviation of the intrinsic dissolution rates of these glasses are log {ko/[g/(m2•d)]} = 8.2 ± 0.2.