Abstract:Thermal gradients during the early,non-isothermal period of near-field evolution in a KBS–3 repository for nuclear wastes could alter the mineralogy of the bentonite buffer and cause the constituent clay particles to become cemented together by mineral precipitates. The present study evaluates a reactive-transport model of buffer cementation using published experimental results on the hydrothermal alteration of clay buffers embedded in crystalline rocks. Model predictions are qualitatively consistent with experimental observations of regions of the buffer exposed to the highest temperatures,which indicate precipitation of secondary phases(anhydrite ± amorphous silica(opal-CT) ± calcite) and alteration of Na-smectite to Ca-smectite ± saponite. The model does not predict extensive dissolution of quartz,kaolinite and feldspars in this region,however,which conflicts with observations from one of the field tests. This disagreement may be due to the fact that the hydrothermal transformation of high-charge smectites to low-charge smectites in kaolinite-smectite mixed layers cannot presently be accommodated in the model due to a lack of basic thermodynamic data. The otherwise good qualitative agreement overall between model predictions and experimental observations suggests that the model can presently be used to carry out sensitivity analyses of the potential for buffer cementation to occur in the KBS–3 near field using expected time-temperature and resaturation histories extending over time scales of hundreds for this system of years.