|
|
|
Quantification of Water-Content of Simulated Nuclear Waste Glasses Using Nuclear Reaction Measurements Supported by EMSL Operations and PNNL Laboratory Directed Research and Development.
Hydrogen and water are commonly found in several stages of generation, storage, and disposal of nuclear waste (Murray 1994). The presence of hydrogen or water can have significant effects on the electrical, mechanical, and chemical properties of some materials. Hydrogen is chemically versatile. As it has a valence of both +1 and –1, it can react with most elements. Additionally, hydrogen is many orders of magnitude more mobile than other common contaminants. High-energy (MeV) ion beam techniques have been employed to measure hydrogen concentration profiles in a variety of materials, including glasses (Khabibullaev and Skorodumov 1989). Recently, Lanford has presented the general considerations in hydrogen analysis and practical information on hydrogen profiling and data analysis in his excellent review (Lanford 1992, 1994). All commercial glasses contain a certain amount of water in the form of hydroxyls (or hydrogen oxide). Hydroxyls (Brown and Kobayashi 1996; Barthlow 1983) enter a glass structure and change the properties of a glass through the reaction: H2O (g) + º Si-O-Siº « 2º Si-OH. Effects of water content on processing (Brown and Kobayashi 1998; Jewell et al. 1990), electrical conductivity (Brown and Kobayashi 1998; Jewell et al. 1990), mechanical strength (Shin and Tomozawa 1996), phase separation (Tomozawa 1998), structure (McGahay and Tomozawa 1994), hygroscopicity (Tomozawa et al. 1994), radiation effect (Jewell et al. 1993), and thermal properties of silica and silicate glasses have been extensively studied. We have applied nuclear reaction analysis (NRA) to investigate the hydrogen in simulated nuclear waste glasses. The 1H (19F, ag)16O reaction used in this investigation requires a resonance energy that was available by use of a 6.471 MeV 19F beam. In this technique, the sample to be analyzed is bombarded with 19F beam with energy at or above the resonant energy, and the number of characteristic gamma rays produced in the target is measured with a scintillation detector. When the sample is bombarded with 19F at the resonance energy, the gamma-ray yield is proportional to hydrogen on the surface of the sample. When the sample is bombarded with 19F above the resonance energy, there are negligible reactions with the surface hydrogen because the energy is above the resonance energy; however, as the 19F ions penetrate the sample, they lose energy and reach the resonance energy at some depth. Now, the gamma-ray yield is proportional to hydrogen at this depth. Thus, the hydrogen concentration as a function of depth (profile) is determined by measuring the gamma-ray yield as a function of beam energy. In the present investigation, hydrogen quantification was calibrated using a 1.0 x 1017 H+/cm2 irradiated SiC standard. Surface characteristics are represented by prepared sample surfaces. The assumptions are 1) hydrogen-content is directly proportional to the water-content of the glass; that is, no dissolved hydrogen present in the glass (except in the case of hydrogen-treated samples) and 2) the hydroxyls species are homogeneously distributed throughout the sample. This summary outlines the results on hydrogen or water content and their profiles in waste glasses. A detailed description of results are published elsewhere (Sundaram et al. in press). A glass sample of about 1.5 cm ´ 1.0 cm ´ 0.2 cm was sliced off the bulk glass piece. After metallographic polishing, the samples were ultrasonically cleaned by placing them in a beaker of ethyl alcohol and acetone for 5 minutes. The glass sample was removed from acetone using a forceps, air-dried, and wiped with ethyl alcohol. Finally, the sample was dried in an oven at 150°C for one hour. The target melter glass composition was also prepared in the laboratory, using conventional chemicals (carbonates, oxides, and hydroxides). This glass was melted at 1150°C for 2 hours. These glass surfaces were used as cast and not prepared for NRA measurements. All the glasses were annealed at 500°C overnight and furnace-cooled before measurements were taken. NRA measurements were first taken from the surface into bulk and then vice versa to establish the applicability of the NRA technique for measuring bulk hydroxyl species in glasses. Figure 9.8 shows clearly no significant difference in these spectra in the bulk of the sample from the melter test. High hydrogen content at the surface followed by a sharp decrease into the bulk clearly showed that a surface layer formed during surface preparation of the samples. A melter glass was compared with the corresponding laboratory-prepared glass counterpart, as shown in Figure 9.9. Comparing the bulk concentrations, the melter glass had approximately double the water content of the laboratory glass. This is attributed to the water vapor present in the off-gas stream during melter operation. The marked difference at the surface compared to the bulk is attributed to the presence of a surface-modified layer due to preparation procedure. This was later confirmed by x-ray photoelectron spectroscopy (XPS).
The NRA technique has been shown to be a useful tool in quantifying the hydrogen/water content of laboratory as well as melter waste glasses. The technique can differentiate the surface region from the bulk glass region. The melter glass has approximately double the water content of the laboratory glass. This has been attributed to the water vapor generated during slurry-fed melter operation. The preliminary hydrogen concentration—redox of feed as well as glass redox data—suggest that the hydrogen content is a good indicator of optimum processing conditions in a melter. References Barthlow, R. F., J. Non-Cryst. Solids, 56, 331 (1983). Brown, J. T., and H. Kobayashi, Am. Ceram. Soc. Bull., p. 81 (1998). Brown, J. T., and H. Kobayashi, Glass Ind., p. 16 (1996). Jewell, J. M., C. H. Shaw, and J. E. Shelby, J. Non-Cryst. Solids, 152, 32 (1993). Jewell, J. M., M. S. Spess, and J. E. Shelby, J. Am. Ceram. Soc., 73(1), 132 (1990). Khabibullaev, P. K., and G. G. Skorodumov in Determination of Hydrogen in Materials, Nuclear Physics Methods (Springer-Verlag, Berlin, Germany) 1989. Lanford, W. A., in Handbook of Modern Ion Beam Materials Analysis, Editors: J. R. Tesmer and M. A. Nastasi, (Materials Research Society, Pittsburgh, Pennsylvania) pp. 193-204, 1994. Lanford, W. A., Nuclear Instrum. Methods in Phys. Res. B 66, 65 (1992). McGahay, V., and M. Tomozawa, J. Non-Cryst. Solids, 167, 127 (1994). Murray, R. L., in Understanding Radioactive Waste, Battelle Press, Fourth Edition, 1994, pp. 71-158. Shin, D. W., and M. Tomozawa, J. Non-Cryst. Solids, 203, 262 (1996). Sundaram, S. K., S. Thevuthasan, and D. E. McCready, MRS Proceedings (1999), in press. Tomozawa, M., H. Li, and K. M. Davis, J. Non-Cryst. Solids, 179, 162 (1994). Tomozawa, M., Phys. Chem. of Glasses, 39(2), 65 (1998).
William R. Wiley Environmental Molecular Sciences Laboratory Feedback: webmaster@emsl.pnl.gov Revised: June 12, 2001 Security & Privacy PNNL-13147 |
