The treatment of sorption and retardation in the assessment of geological barriers to contaminant transport

The approaches used in long-term performance assessments to determine the impact of sorption in the geosphere on retarding the movement of radioactive contaminants have been reviewed. A comparison is made between the approaches used in the 1994 assessment of the Whiteshell Research Area in Canada and in similar assessments conducted in other countries. The long-term performance assessments reviewed were the AECL-EIS, the TVO-92 assessment conducted in Finland, the PNC-H3 project from Japan , the SKI Project-90 and SKB-91 assessments performed in Sweden, the Kristallin-1 assessment done in Switzerland, and the NIREX-95 assessment performed in the United Kingdom. In each of these assessments, the effect of sorption in the geosphere on contaminant transport was assessed using the 'Kd concept'. The sorption data were selected to reflect the anticipated effects of the reference geochemical conditions of the site being assessed. A number of approaches were taken to account for the uncertainty in the values selected for the sorption coefficients. Contaminant transport was modelled using a continuous porous medium approach or a dual-porosity approach, with the latter being more widely used. In both approaches, the retardation of contaminant transport due to sorption on the host solid is described by a retardation factor. In the present report, sorption values used in the assessments were compared and differences between the values selected for similar materials and geochemical conditions were noted, particularly for the redoxsensitive elements. A comparison of retardation factors required the introduction of a 'utilization factor' into the retardation equation to account for the effects of matrix diffusion on retardation. The comparison illustrated the impact of time dependence on retardation in the matrix diffusion approach vis-a-vis the time-independent retardation approach used in the AECL-EIS. (author)