Compacted bentonite is a candidate backfill material for the disposal of ra
dioactive waste. A correct understanding of the processes affecting diffusi
on in compacted bentonite is needed to predict radionuclide migration.
Using distribution coefficients (K-d) obtained in batch experiments and geo
metric factors relevant for compacted bentonite, it is not possible to pred
ict or explain the apparent diffusivities (D-a) typically observed. Here, a
n approach is presented that integrates mechanistic sorption and diffusion
models in combination with thermodynamic data and bentonite characteristics
in order to predict D-a for cesium. Key features of this approach are (i)
to calculate the correct K-d value at high dry densities of the bentonite,
and (ii) to calculate the correct constrictivity based on an electric doubl
e layer model.
The integrated sorption-diffusion model present in this paper assigns excha
ngeable, immobile ions to the Stem layer, whereas ions located in the diffu
se layer are considered mobile. Constrictivity is interpreted as the ratio
of the average concentration of an ion in the diffuse layer to its concentr
ation in the bulk solution. We show that in this way, the diffusion of Ca i
n Kunigel-V1 as well as Kunipia-F bentonite can be modeled successfully for
a wide range of dry densities. Stern plane potentials calculated for these
conditions also agree with measured zeta-potentials.