Samples of borosilicate glass, high-silica glass, tailored ceramic, and SYNROC, incorporating simulated Savannah River high-level defense waste sludges, were leached by the MCC-1 procedure for times up to 28 days. Cesium, uranium, and cerium leach rates are reported for waste forms containing a composite sludge, at 40°C in deionized water, and at 90°C in deionized water, silicate water, and brine. The ordering of the waste forms from best to worst differs for each element leached, and none of the forms show a clear advantage for all the key radwaste elements. Some cesium leach rates for forms containing high-aluminum or high-iron sludges also are presented. So far, only small effects of sludge type have been observed, with one exception. This study is one of several inputs for selection of an alternative waste form for Savannah River waste.

This work is one of the first attempts to work out a proper technique for the determination of the diffusion of the phosphate glass components into various rocks by using X-ray microanalysis. Under study was thermal and radiationenhanced diffusion of phosphorus, chromium from phosphate glasses into the samples of basalt, metagabbro, metadunite and quartz at high temperatures (to 600°) during gamma irradiation. Radiation enhanced diffusion of ions into rocks.

Titanate processing parameters including temperature, waste loading, redox conditions, and additives used to promote specific phases were evaluated with respect to their effect on leaching of Cs, Mo, Ca, Sr, Ba, U, Gd, and Ti in the range of 22°-150°C in leachates which included deionized water, acidic (pH 2), and basic (pH 12) solutions. Surface analyses and microstructural characterization were used to relate observed leaching behavior to processing parameters. Redox conditions were found to be very important with respect to Cs (and Mo) retention. Two Cs-bearing phases were found in titanate prepared under oxidizing conditions. One had a hollandite structure and the other contained Cs, Mo, Ca, and Fe as major constituents. The latter phase, which was more susceptible to leaching, was not observed in titanates prepared using adequate reducing conditions. Where applicable, a reference glass (PNL 76–68) was included in the leaching tests. Comparative leach rates for elements common to both waste forms were generally one to four orders of magnitude lower for the titanates within the range of conditions used.

Iron-enriched basalt (IEB) has been investigated as a containment medium for defense transuranic (TRU) and defense high-level wastes (HLW). IEB samples have been fabricated incorporating simulated and actual defense TRU wastes generated at the Rocky Flats Plant (RFP), and simulated defense high-level wastes. These samples were tested for leaching resistance and fracture toughness. Cesium and strontium volatility from an IEB melt was also assessed, using radioactively spiked samples.

Pollucite (CsAlSi206) has been suggested [1,2] as a suitable phase for the incorporation and immobilization of 137Cs, on the basis of its refractory nature and leach resistance. The phase also occurs in tailored ceramic formulations designed for the disposal of commercial reprocessing waste [3,4] and zeolites which are heavily loaded with Cs [5,6].

Various ceramic materials as well as Formula 127 glass have been developed to immobilize ICPP zirconia calcine. Tailored, titanate and matrix encapsulated ceramics were prepared by cold pressing and sintering reactants, while glass-ceramics were prepared by melting in a manner similar to that of preparing the glass. X-ray diffraction techniques were used to determine the presence of radionuclide and calcine matrix element host phases in each ceramic material. The aqueous leachability, practical loading capacity and the incorporation of waste elements into host phases was investigated and compared to those of Formula 127 glass through application of the Soxhlet and MCC-1 static leach tests. Potential process requirements for each material were evaluated based on the number of steps needed for laboratory-scale preparation.

Large samples (up to 15cm diameter) of SYNROC D containing simulated Savannah River (SRL) waste sludge have been prepared and performance tested. Waste loadings of 60-65 wt% for the SRL composite sludge have been achieved; this corresponds to a waste concentration (volumetric) loading of approximately 2.3-2.5 g/cm3. A typical SYNROC D sample has a density of about 4.0 g/cm3 with less than 0.2% porosity. The compressive and flexural strengths of SYNROC D are 51,200 and 9,400 psi, respectively and Young's Modulus is 20.1 × 106 psi by ultrasonic measurement. The quantity of respirable fines (less than 10 μm) generated during a constant energy density impact (10J/cm3) was less than 0.16 wt%. Values for the thermal conductivity (22°C) and the thermal expansion coefficient (22–950°C) were measured to be 1.7 W/m·K and 11 × 10−6 K−1, respectively. Normalized elemental leach rates determined by 28-day MCC-1 tests (90°C) for Cs, Sr, U and Nd are 0.7, 0.3, 5×10−5 and < 8×10−3 g/m2·d respectively.

The electronic charge states and site symmetries of the radioactive ions incorporated in nuclear waste forms are of considerable importance in determining the physical and chemical properties of these materials. An in situ characterization of these ions is, unfortunately, often difficult – especially when a mixture of charge states and local crystal symmetries exist. The application of Mbssbauer spectroscopy represents a powerful technique for obtaining solid state chemical information.

Current plans for high-level radioactive waste disposal call for long term emplacement of canisters of a suitably leach resistant waste form in deep underground repositories. Proposed regulations [11] call for canisters with a designed lifetime of 1000 years. After this initial period, most of the remaining activity in the waste will be associated with the actinide elements. Thus, the leach behavior of the actinides takes on special importance. In particular, the leaching mechanisms must be understood if we are to attach any confidence to long term predictive extrapolation of short term leach tests. The actinides, unlike the rare earths, are not particularly similar in chemical behavior nor is their behavior reliably modeled by that of other elements. Thus, more studies directed toward the behavior of the actinides in candidate waste forms are important.

Results are presented for leach testing at 95°C and 200°C of SYNROC containing 9% and 20% simulated high level radioactive waste, synthetic hollandite and pervoskite samples, and natural zirconolite and pervoskite samples. Single phase synthetic minerals show much higher leach rates than natural mineral samples and polyphase SYNROC samples. Natural zirconolite samples with low radiation damage have leach rates at 200°C based on U which are identical to those measured on SYNROC samples. Natural zirconolites with very large accumulated α dose and radiation damage have leach rates at 200°C which are only 5 times higher than those of low dose samples.

One of the major reservations regarding the use of glass as a primary nuclear waste form is its ability to retain hazardous radioactive nuclei when exposed to ground water at tenmeratures above about 75C. It is now generally accepted that the removal of radioactive nuclei from glass in an aqueous environment proceeds via two main mechanisms, i.e., either by a diffusion process (which may be driven or undriven) or by direct matrix dissolution - which, in its initial phase, also results from a diffusion related process [1]. The term "leaching" is generally employed in describing the overall process resulting in the removal of nuclei (whether they are radioactive or not) from glass that is exposed to an aqueous environment.

Glass ceramics and ceramics based on minerals of proven thermodynamic stability in Canadian Shield groundwaters are being developed to host the wastes arising from future CANDU* fuel recycling. Sphene-based glass ceramics, produced by melti0n g Na2 O-CaO-Al2O3-TiO2-SiO2 compositions at 1400° C, cooling, ang then reheating the phaseseparated glasses to approximately 1000° C, show strong partitioning of waste ions into the sphene (CaTiSiO5 ) lattice. Leaching results suggest that these materials are durable in deionized water. In simulated groundwaters, the sphene phase appears to suffer no net leaching. Leach testing and manufacturing methods are discussed.

The current status of SYNROC C research and development by the Australian Atomic Energy Commission is reviewed. A non-radioactive fabrication demonstration line designed to produce 10 cm o.d., 90 cm long, cylinders of SYNROC canned in stainless steel by the method of in-can hot pressing is being commissioned. Leach tests are proving the excellent leach resistance of SYNROC. Accelerated radiation damage testing using fast neutrons has simulated storage times of up to 6.7×105 years. Thermophysical properties of SYNROC have been measured over the temperature range 20–650°C.

The planned Swedish KBS glass corrosion investigation program comprises experiments with inactive glasses containing simulated waste, prolonged in-situ tests, the characterization of corrosion products, immiscibility studies, and corrosion experiments with “hot” glass.

This presentation gives a short description of the entire program. It focuses thereafter on some recent leaching results with the inactive KBS glass qualities ABS 39 and ABS 41, which were leached in a manner similar to the PNL MCC–1 test procedure.

Standardized comparative drop weight impact tests were conducted on solid alternative waste forms under consideration for immobilizing Savannah River Laboratory (SRL) defense wastes. The fragment size distributions were measured in the size ranges of ֮5 to 8000 μm. All waste form fragment size distributions could be described accurately by lognormal plots. Respirable sizes (≤10 μm) were measured. Borosilicate glass and SYNROC specimens yielded the same mass fractions of respirable sizes; FUETAP concrete, high silica and alkoxide glass specimens yielded ֮2–3 times more mass fractions of particles of respirable sizes, whereas tailored (Spinel) ceramic specimens yielded the smallest mass fractions of respirable sizes, ֮2–1/2 times less.

A general method for characterizing the major practical effects of accidental impacts on waste packages and for evaluation of scale-model tests has been partially developed. Impact fracture of brittle waste forms has been shown elsewhere to produce particulates whose size distributions are described by the lognormal probability distribution. The model proposed for fragment generation involves the transformation of impact (kinetic) energy into elastic strain energy which is followed by fracture and energy dissipation into heat by the fracture particulates. The peak stresses developing during compressions are approximated as a function of time using elastic theory for a wide range of practical impact conditions for typical (glass) waste forms. The proposed methodology requires experimental validation in terms of correlation of stress and energy parameters with particulate parameters describing the results of fracture. Two kinds of preliminary correlations are presented: (1) calculations of stress parameters for a range of impact conditions; and (2) averages of particulate parameters obtained in standard impact tests.

Nitric acid waste can be immobilized in an amorphous aluminosilicate matrix that is produced by polymerization of metal alkoxides [1]. Unlike vitreous aluminosilicate host matrices obtained by melting, neither alkalies, alkaline earths, nor borates are required since no melting is involved. Instead, the alkoxide glass formers and radionuclides react directly to form a randomly cross-linked structure of silica and alumina polyhedra. Structurally, the amorphous product formed through chemical polymerization is similar to a true glass except that it has not experienced melting. As a result, neither a glass transition temperature nor a specific volume temperature dependence would be expected in such an amorphous material. As a chemically durable waste form, the "gel-glass" product should be substantially better than ordinary waste glass because it can be made without the addition of either alkalies or boron.

In order to provide technical support to the NRC, Brookhaven National Laboratory has performed leach tests and immersion tests using organic liquids and organic saturated water of concentrated decontamination waste solidified using a vinyl ester-styrene binder.

The leach tests measured the release of Fe, Ni, and Co from these forms in deionized water, groundwater, and seawater. After 64 days, the mean fraction released normalized by V/S for iron was 5.1 ± 1.9 × 10−3 cm in deionized water, 7.1 ± 1.1 × 10−3 cm in groundwater, and 9.0 ± 3.2 × 10−3 cm in seawater. For nickel, 64 day release rates observed were 4.6 ± 1.6 × 10−3 cm, 4.6 ± 1.0 × 10−3 cm, and 5.9 ± 0.6 × 10−3 cm in deionized water, groundwater, and seawater, respectively. After 50 days, the 59 Fe release rates were 5.9 ± 0.7 × 10−3, 4.8 ± 2.4 × 10−3, and 2.8 ± 1.0 × 10−3 cm in deionized water, groundwater, and seawater, respectively. For 60Co, the 50 day releases in the three respective leaching media were 6.0 ± 1.7 × 10−3 cm, 6.8 ± 1.0 × 10−3 cm, and 2.3 ± 0.2 × 10−3 cm.

Immersion tests of waste forms prepared at a solidification demonstration held at the Dresden Nuclear Power Station were conducted in toluene, xylene, and water saturated with toluene and xylene. During immersion of samples in the pure organics, large changes in sample volume and weight were observed. Total weight changes of 9.6 ± 0.3% and 21.6 ± 0.7% were observed after 839 hours of immersion in xylene and toluene, respectively. Air drying of the samples led to an overall weight loss of 23.5 ± 0.7% for xylene and 35.6 ± 0.6% for toluene. Qualitatively, similar changes were observed for immersion tests using organic saturated water. Severe sample deterioration was observed in this case, however. The cause of this deterioration is not known.

The following study is a portion of a comprehensive research program that is examining the stability of a variety of nuclear waste forms. In addition to the bulk waste forms, important individual radiophases are being studied to obtain a more complete understanding of the behavior of the components of complex multiphase radwaste systems. The stability of the strontium radiophase in supercalcine and an iodine-radiophase will be discussed.

The strontium radiophase in this study was Sr-powellite and the iodine radiophase was I-sodalite. Each radiophase was incorporated into bulk compositionally adjusted portland and aluminate cements. Two processing variables were studied: curing at 60°C and warm pressing at 150°C and 345 MPa.

Sr-powellite in portland cement leaches incongruently; a combination of dissolution and diffusion-controlled exchange of Ca for Sr is demonstrated. In the warm-pressed aluminate cement these reactions are masked by reactions with the curing cement. I-sodalite leaching data indicate dissolution, dominant at long times, combined with diffusion, dominant at short times.

The stability of a tailored ceramic waste form under hydrothermal conditions was the subject of several previous papers [1,7–10] from this laboratory. One of the results reported in these studies was the apparent dissolution and reprecipitation of the alkaline-earth molybdate phases (scheelite-structure phase). The composition of the scheelite-structure phase after hydrothermal treatment was different from that before treatment (approximately Ca95 Sr5 after and Ca35 Sr40 Ba25 before). A schematic phase diagram was presented at that time to explain the results obtained.

This paper is a report of our attempt to experimentally determine the phase relations in the alkaline-earth molybdate (Ca, Sr, and Ba) and verify our previous interpretation. For this work, we prepared compositions throughout the ternary system and heated them at 1200°C for periods of 2 to 24 hours. The experimental products were characterized by x-ray diffraction and SEM/EDX examination.

At 1200°C, the two-phase solvus is nearly symmetric and extends from near the pure calcium and barium end-members to about 35 mole percent strontium at the critical point of the solvus. The increase in grain size, uniformity of composition within and between grains of the same phase and the approach to “textural equilibrium” in the 1200°C experiments all strongly suggest that equilibrium was attained.

New, calcium-rich compositions are suggested for the scheelite-structure phases in the supercalcine-ceramics based upon the interpretation of data from this study. This observation forces a re-examination of the assumption concerning the partitioning of Sr and Ba amongst the phases in the supercalcine-ceramic.