All wastes are distinguished by their harmful effects: radioactivity in the case of radioactive waste. An advantage of this type of waste is that the harmful effects diminish in time. Pending this radioactive decay, the materials must be contained to avoid human exposure to radiation, which is not difficult; assurance mlust also be provided, however, that the containment is effective for very long durations, consistent with the half-lives of the radionuclides involved.

The future of the nuclear industry and its acceptance by public opinion depend to a large extent on the development of an acceptable solution for waste management and disposal. The long duration of the hazard presented by certain radionuclides found in nuclear wastes is certainly one of the reasons for public concern over definitive waste disposal.

The clay backfill that will surround a buried high level nuclear waste package in most national repository programs, could be modified to play a greater role as a barrier to radionuclide (RN) releases. The RN steady state release (Mb) rate from a clay backfill to adjacent rock is directly proportional to backfill porosity (ε), RN diffusion rate In the backfill (Ds), and RN solubility at the waste form surface (C*), and Inversely proportional to RN half-life (λ) and RN retardation coefficient (R) in the backfill [1]. We propose ways to reduce ε, Ds and C* and Increase R for important radionuclides, mostly through the addition of reactive minerals to the backfill. Silica, calcite and anhydrite may be added to precipitate and clog porosity. Increased backfill compaction similarly reducesε, Ds and Mb for all the RN's. Strongly sorbent phases can be added to selectively adsorb both cationic and anionic RN's (e.g. 1–129). However, adsorption will not Importantly reduce peak release rates of most long-lived RN's. The backfill can be poised at reducing Eh's with mineral additives to lower Ds and so immobilize radioisotopes of NI, Np, Pa, Pu, Se, Tc and U. Minerals of stable or more stable isotopes of Cs, NI, Se, Sn and U can be added to lower Ds values of the RN's and to coprecipitate them in solid solution. Phosphorite-apatites, which are known to have high selectivities for rare earths and RN's, may be added to coprecipitate Am, Np, Pu, Sr, Th and U.

The historical development of international advice on radiological protection criteria for solid waste disposal is described, and problems associated with the interpretation of existing criteria are discussed. These problems arise largely because of the long timescales over which radionuclides may be released from solid waste repositories. In an attempt to aid in the interpretation of existing international advice, the UK National Radiological Protection Board is issuing new guidance on radiological protection criteria for solid waste disposal and its application. Aspects of this new guidance are described.

Results are presented for a large number of aqueous leach tests on glass specimens doped with 237Np, 238Pu, 239pu, 241Am or 244Cm. The effects of the leaching mode (static or dynamic), temperature and environmental materials are discussed. The amount of actinides contained in the alteration film is characterized by the retention factor RF = NL(B)/NL(act). The neptunium retention factor ranged from 1 to 10 depending on the test conditions; for plutonium and americium, the retention factor varied from 6 to 1300.

In general, the retention factor increased with the temperature. The very important influence of the environmental materials shows that there is competition for actinide retention between the alteration film and the surrounding materials, but in every case the amount of ca activity released into solution is very low.

The Al2O3 rich borosilicate glass SM527 was submitted to corrosion tests with glass surface area to solution volume ratios ranging from 10 to 10000 m-1. This latter condition would correspond with a 1000 fold acceleration relative to the reference MCCI condition. Powdered glass was used to reach SANV ratios of 500 m-1 and more. The leaching solutions were either distilled water or referred to Boom clay disposal conditions. The results based on the boron concentration in solution revealed a relatively linear dependence on SA/V.t0.5 in the pure solutions, on the longer term (DW and clay water). Diffusion is suggested to be the process governing the glass dissolution, although other processes should not be excluded. In a clay / clay water mixture (slurry), long term dissolution seems to be limited by saturation. Short term data for boron are largest in the clay slurry, but with time the boron concentrations converge to similar values in the three media considered. The use of SA/V as an accelerating factor is promising, but certainly requires additional research.

This communication gives a detailed characterization of some molybdate solid solutions and cerianite-type material which formed on the French borosilicate nuclear waste glass R7T7 upon corrosion in various saturated salt solutions at 110°C, 150 °C and 190 °C. The glass contained lanthanoid elements, such as neodymium, lanthanum, praseodymium, cerium and yttrium, but did not contain actinoid elements, except uranium and thorium. Various solid solutions containing lanthanoids (Ln) were found on the glass surface after corrosion, including powellite solid solutions and cerianite-type material. The secondary solid phases are characterized based on quantitative microchemical and structural analyses. These phases are expected to incorporate actinoids such as americium and curium in acid magnesiumcontaining salt solutions. The phases then constitute an additional barrier against migration of these radionuclides, which would otherwise be in the aqueous phase.

Field leaching studies were carried out in granite at the Stripa site in Sweden and also in salt in the Materials Interface Interaction Test at the Waste Isolation Pilot Plant (MIIT/WIPP) in New Mexico. The goal of these studies is to assess the durability of various glass compositions engineered to isolate high-level nuclear waste from the biosphere. An additional goal of the MIIT study is to determine how the glasses interact with a wide array of proposed materials that may be a part of the multi-barrier waste package. These substances include metals, geological host specimens,, as well as engineered backfill and overpack materials.

Two year data on the SRLY (165/TDS) glass compositions has been extracted from both studies (Stripa and WIPP/MIIT) and five year data has recently become available from the MIIT study. Results from SEM/EDS, SIMS and FTIRRS analyses on glass/glass interfaces are presented in this paper.

Experimental studies of silicate glass/water reactions at low temperatures have previously identified the glass surface area-to-solution volume ratio (SA/V) as a significant rate determining parameter [1-4]. The value produced when SA/V is multiplied by reaction time, hereafter referred to as SVT, has been proposed as a scaling factor for comparing experimental results collected under different test conditions and for extrapolating short-term results to longer periods of time. Developing an understanding of the effect of SAN is needed for modeling experimental results where SA/V ranges in value or may vary during experiments. It is also useful to understand the effect of SA/V for modeling natural systems where this value almost certainly varies, such as during the hydrothermal diagenesis of natural glasses or projecting the long-term reaction of water and borosilicate nuclear waste glass in a geologic repository.

The dissolution behaviour of two alkali borosilicate glasses has been investigated at 90°C in the presence of potential corrosion inhibitors. The added materials were selected on the basis of surface chemistry considerations: Since the dissolution rate of silicate solid phases in alkaline solutions increases with growing negative surface charge, it should be expected that sorbing cations reduce the reaction rates. Due to the formation of insoluble hydroxides and silicates and to the dominance of neutral or negatively charged hydroxo complexes in the pH range of interest, the selection of potential inhibitors is very restricted.

The precipitated hydroxides of Al, Zn, Cd and Pb were chosen as the solid phases. Compared with the experiments in pure water, glass dissolution is enhanced in presence of Al(OH)3 and Zn(OH)2. The concentration of the aqua ions is too low for an inhibiting effect. Sorption of silicic acid on the hydroxide accelerates the glass dissolution. Addition of PbO had a strong inhibiting effect in a 28 days experiment. At longer times an insoluble lead silicate is formed and thereby the glass dissolution rate is increased. Cd on the other hand is still effective as an inhibitor after 365 days.

Based on the present results, the possibility of using glass corrosion inhibitors in a repository is considered not to be worthwhile.

Presented here are results relating to the possible causes for the very strong nonlinearity in the relationship between glass composition and glass durability, as determined by MCC 3 or Product Consistency (PCT) tests. We have previously found a rapid increase in leachate concentrations with small changes in composition (as little as 1 weight percent SiO2 or Al2O3) over some composition ranges near those of interest for nuclear waste disposal. This rapid rise occurs over a narrow transition region in composition space that provides a natural division between more durable and less durable glasses. Our new data on glasses of interest to the West Valley Demonstration Project (WVDP) suggest that this effect is not due to phase separation. Results from leach tests conducted with and without constant agitation are compared in an attempt to determine whether the physical integrity of alteration layers plays an essential role, as might be the case if formation-spallation cycles were important in this process. Microstructural data are also presented in order to compare the alteration layers on the more durable and less durable glasses leached with and without agitation.

Alteration layers developed on SRL-165 simulated waste glasses in dilute sodium silicate/bicarbonate leaching solutions have been examined by Secondary Ionization Mass Spectrometry (SIMS) using fine-scale, multiple-element depth profiling. Selected samples were examined with an imaging detector system, which demonstrated the horizontal homogeneity of the layer development at all depths within the layer. After 1 day of reaction at 90°C the reaction layer shows depletion of glass elements to a depth of 0.2 μm. The surface of the layer in contact with the solution shows enrichment of Si, Al, and alkali elements even at this short reaction time, suggesting the early stages of development of secondary aluminosilicate phases. With increased reaction time, the layer thickens to about 1.3 μm at 91 days, while the evidence for aluminosilicate development at the surface of the layer becomes more prominent. Penetration of hydrogen into the “unreacted” glass proceeds to a depth of about 0.5 μm deeper than the alkali depletion zone. This suggests the mechanism of initial reaction of the glass is by attack of the silicate structure by molecular water or hydroxide ion rather than by alkali-hydrogen ion exchange. The simple structure of the layers developed in the silicate solution is in contrast to the complexity of layer structure found when glasses are reacted in deionized water. Since the conditions for geologic disposal will be closer to those used in the silicate leaching experiments, these results hold promise for the ability to model the system to predict long-term performance after disposal in a repository.

The influence of the surface finish on nuclear glass dissolution was investigated. Seven different surface finishes were tested: a specimen as cut, three specimens polished with 220, 600 and 4000 mesh SiC paper and one with 1 μm diamond powder, a flame-polished specimen and a thermal rupture specimen.

The initial glass corrosion rates were measured after 7 and 28 days of Soxhlet leaching at 100°C. The surface finish of each glass coupon was also assessed by three-dimensional analysis before and after leaching for 28 days. Leaching solution analyses showed that the mean apparent leach rate for the first 7 days was highly dependent on the surface finish: the measured rates ranged from 3.8·g.m-2d-1 for the as-cut specimen to 1.3 g·m-2d-1 for the flamepolished specimen. The differences in the leach rates diminished considerably after 28 days of leaching: the mean rate measured between 7 and 28 days for the as-cut and polished specimens was identical (1.7 g·m-2d-1), but was still lower for the remaining two specimens. Three-dimensional surface analysis showed that leaching revealed surface defects (superficial microcracks) produced by cutting or polishing. These effects were less perceptible on the flame-polished or thermally ruptured specimens. The actual surface area of the cut and polished specimens exceeded the apparent geometrical area.

In another experiment, the leach rates measured for as-cut specimens and flamepolished specimens were measured over periods of five months or more using a thermogravity balance placed above the Soxhlet device. The experiment confirmed the existence of an initial period during which the dissolution rate for the as-cut specimen was much higher than the mean corrosion rate.

The first field tests conducted in the United States involving burial of simulated high-level waste [HLW] forms and package components, were started in July of 1986. The program, called the Materials Interface Interactions Test or MIIT, is the largest cooperative field-testing venture in the international waste management community. Included in the study are over 900 waste form samples comprising 15 different systems supplied by 7 countries. Also included are approximately 300 potential canister or overpack metal samples along with more than 500 geologic and backfill specimens. There are almost 2000 relevant interactions that characterize this effort which is being conducted in the bedded salt site at the Waste Isolation Pilot Plant (WIPP), near Carlsbad, New Mexico. The MIIT program represents a joint endeavor managed by Sandia National Laboratories in Albuquerque, N.M., and Savannah River Laboratory in Aiken, S.C. and sponsored by the U.S. Department of Energy. Also involved in MIIT are participants from various laboratories and universities in France, Germany, Belgium, Canada, Japan, Sweden, the United Kingdom, and the United States. In July of 1991, the experimental portion of the 5-yr. MIIT program was completed. Although only about 5% of all MIIT samples have been assessed thus far, there are already interesting findings that have emerged. The present paper will discuss results obtained for SRS 165/TDS waste glass after burial of 6 mo., 1 yr. and 2 yrs., along with initial analyses of 5 yr. samples.

The comparison of laboratory data from the corrosion of borosilicate nuclear waste glass (German SM513LW11 and French R7T7) with data from the Materials Interface Interactions Test (MIIT) and Repository Systems Simulation Test (RSST) illustrates the inherent limitations of in situ tests. Although in situ tests may confirm the short term behavior of waste forms and identify phenomena associated with the repository system, they do not provide the fundamental basis for the extrapolation of long-term behavior.

Previous tests have shown that glass is highly corroded in the presence of smectite 4a, a clay under consideration for use as a repository engineered barrier material. Alteration experiments with French R7T7 glass were therefore conducted in moist clay media containing 5 wt% of a siliceous additive. Five siliceous materials were tested: silica gel, pozzolana, window glass, R7T7 glass frit and R7T7 glass, all in powder form. Following an initial 28-day preconditioning period during which the clay/additive/leachant system was allowed to reach equilibrium, a glass specimen was placed for 7 or 28 days in the moist clay medium. Glass alteration was characterized by the mass loss and by surface analysis. After 28 days in contact with the silica gel and with the three glass powders the mean glass corrosion rate calculated from the mass loss values was about 0.01 g.m−2d−1, versus 1.4 g·m−2d−1 with clay alone and 0.3 g·m−2d−1 with pozzolana. The use of a suitable additive can thus considerably attenuate the corrosive effects of moist smectite in contact with R7T7 glass.

We report on the time dependence of release of glass constituents during static dissolution experiments with the COGEMA glass R7T7 in saline MgCl2 and NaCl dominated solutions at temperatures between 110 and 190°C. The experiments were performed at high S/V values to ensure silica saturation almost from the start of the tests. The results show a square root of time dependence indicating diffusion as rate-controlling with apparent diffusion coefficients similar to that of water diffusion in alkali silicate, borosilicate glasses or long-term weathered obsidians.

R717 glass alteration was investigated in the presence of four mortars prepared from four different cement grades: “CPA” Portland cement (mortar M1), CPA with pozzolana additive (M2), CPA with amorphous silica additive (M3) and “CLK” blast furnace slag cement (M4). Glass specimens were also altered in Volvic mineral water and in a cement effluent. Glass corrosion in the cement media was greater than in Volvic water, but well below what could be expected from the high pH (approx 12.5). The relatively low alteration was probably related to the protective action of the calcium-enriched gel layer that formed at the glass surface. The glass corrosion rate was 2 to 3 times lower with cement containing pozzolana or silica gel additives or with CLK cement than with CPA cement alone.

Nine samples of the “R7T7” glass composition selected to vitrify fission product solutions in France were prepared with added platinoid elements (ruthenium, rhoditun and palladium) in soluble form and as insoluble metal particles in solution, and their major properties were measured. Regardless of the initial form when added to the glass the platinoids always formed the same heterogeneous inclusions in the final glass: RuO2 precipitates which were often found as aggregates, and polymetallic (Pd, Rh & Te) inclusions.

The particles tended to settle in the molten glass. The viscosity increased by about 20% at 1100°C. The mechanical properties and short-term leach rates were not significantly affected. Crystallization increased by a factor of 2 or 3 in heat-treated glass specimens but did not exceed a few volume percent. However, as the short-term leach rate did not significantly increase, the glass properties were very satisfactory.

The effects of noble metal elements such as ruthenium, rhodium and palladium on the viscosity and electrical resistivity of simulated nuclear waste glass were studied. The glass enriched with noble metals showed the viscosity of a non-Newtonian fluid. The viscosity of the waste glass with 10 wt% Ru0 2 was 3 to 7 times higher than that of glass without noble metals. The RuO2 was mainly responsible for the increase in viscosity for the glass enriched with noble metals. Electrical resistivity of the glass with 15 wt% RuO2 was one seventh to two orders of magnitude lower than that of glass without noble metals. The three noble metals contributed to the decrease in resistivity. The quantitative effects of noble metals on these properties were obtained.