Final report of experiments with rock blocks interacting hydraulically with smectitic pellet fills

The report describes the outcome of the work within the the project 'SU5 08.20 Impact of water inflow in deposition tunnels'. Project decision SKBdoc id 1178871 Version 3.0. Two activity plans have been used for the field work: AP TD SU50820-09-014 and AP TD SU50820-09-031. A problem in backfilling of KBS-3V tunnels with smectitic pellets surrounding highly compacted clay blocks is that water entering the fill have a very substantial effect on the manner in which water moves into or through a pellet-filled region in the period immediately following pellet placement. Channels will be formed that lead much water to the sloping front of the fill in the course of placing it. This can soften the fill and turn it into mud where the water is discharged as demonstrated by large-scale tests. The nature of such channels was investigated in the present study that comprised experiments with rock blocks equipped with nozzles for injecting water into contacting pellet fills at constant flow rates. The purpose was to identify the basic mechanisms in the first phase of hydration of pellet fills and to find out if there is a threshold flow rate for 'piping'. The question if channelling at breakthrough takes place along the contact with the confinement, as indicated by preceding tests with steel and plexiglass instead of rock, was in focus. While the mechanisms of water entering a fill from separate local spots in contacting rock are well understood, prediction of the entire wetting process of a larger pellet volume requires consideration of the interactive function of several inflow spots, representing single or networks of rock fractures. Experiments with pellet fills on a larger scale with simultaneous inflow from a number of fractures would provide further information on the wetting process. Such a test is outlined in the report