The folloivlng documents in the NUREG serles are avallable for purchase from the Government Printing Offlce:formal NRC staff and contractor reports, NRC-sponsored conference proceedings, International agreement reports, grantee reports, and NRC booklets and brochures. Also avallable are regulatory guldes, NRC regulatlons In the Code of Federal Regulatlons. and Nuclear Regulatory Commission Issuances.
ABSTRACTThe contact between an obsidian flow and a steep-walled tuff canyon was examined as an analogue for a high-level waste repository. The analogue site is located on the southwest rim of the Valles Caldera in New Mexico, where the massive Banco Bonito obsidian flow filled a paleocanyon in the Battleship Rock tuff. The obsidian flow provided a heat source, analogous to waste panels or an igneous intrusion in a repository, and caused evaporation and migration of water. The tuff and obsidian samples were analyzed for major and trace elements and mineralogy by INAA, X R F , X-ray diffraction, and scanning electron microscopy and electron microprobe. Samples were also analyzed for D/H and 39Ar/40Ar isotopic composition.Overall, the effects of the heating event seem to have been slight and limited to the tuff nearest the contact. There is some evidence of devitrification and migration of volatiles in the tuff within 10 meters of the contact, but variations in major and trace element chemistry are small and difficult to distinguish from the natural @re-heating) variability of the rocks. Apart from devitrification, the p~ncipal mineralogic change in tuff near the contact is the development of feldspar-sifica linings on voids in the pumiceous tuff matrix; we found no significant development of zeolites in the samples examined for this study. Age determinations by 394r/40Ar are ambiguous, and do not provide resolution necessary to map paleoisotherms from the heating event.A simple model is developed to predict the temperatures in the obsidian and tuff as a function of time. The model predicts the movement of a boiling front several tens of meters into the tuff over several hundred years, and is relatively insensitive to the assumed saturation. Flowing-steam experiments show that F and C1 are readily lost by the glassy tuff at temperatures as low as 200 "C. Loss of F is consistent with diffusionantrolled release, and could produce condensates with tens to hundreds of ppm F in the region behind the boiling zone.
ISeveral mechanisms for producing compositional variations in the tuff are examined. A simple evaporation-capillarity model can explain the concentration of F above background levels. It is shown that the remobilization of silica, predicted by several workers, would probably not be detected in bulk analyses, and may be masked in microscopic analyses by the effects of alkali metasomatism. Significant gas phase transport of metals at the analogue site is shown to be improbable.iii