Chloride accessible porosity fractions across the Jurassic sedimentary rocks of northern Switzerland

The Jurassic clay-rich sedimentary sequence in northern Switzerland was investigated in the context of nuclear waste disposal. Tracer profiles across this sequence play an important role when assessing transport processes and long-term evolution of the environment. Negatively charged clay-mineral surfaces cause a depletion of anions in the porewater close to the surfaces compared to free porewater further away. The chloride accessible porosity fraction (f(Cl)) is an important parameter to scale Cl bulk porewater concentrations from aqueous extraction to free porewater concentrations, which are required to establish anion profiles used for transport modelling of anionic tracers. A vast set of f(Cl) data from advective displacement (AD), squeezing (SQ) and through-diffusion (DI) experiments allowed for method comparison and yielded insight into controlling factors of f(Cl) across the sedimentary sequence. Two observations were made: (1) f(Cl) varies strongly depending on lithology and partly on ionic strength; (2) at high ionic strength, the three methods yield consistent results, but at low ionic strength, f(Cl) from DI are lower than those from AD and SQ. The observations are discussed in terms of influencing factors, including structural charge, ionic strength, rock texture, pore sizes and the fact that the different methods sample different parts of the pore space. The stronger variation of fCl from DI experiments with the ionic strength of the experimental porewater is in agreement with theoretical concepts. The volume fraction of small pores (r < 2 nm) inversely correlates with fCl; it accounts for similar to 50% of the anion depleted pore volume. Two empirical approaches were tested to derive functions for f(Cl) valid for the lithologically diverse sequence: the clay-content model uses a constant f(Cl) at a clay-mineral content >30% and extrapolates to a f(Cl) of 1 at zero clay content. The formation model applies formation- or member-specific average values of experimental f(Cl) for each unit. Both approaches work reasonably well, with the formation model producing less apparent outliers of free porewater concentrations than the clay-content model. Still some scatter remains, particularly in heterogeneous and low-porosity formations. This work results in a comprehensive and coherent fCl dataset for all studied Jurassic formations and serves as a basis to interpret the profiles of anions with regard to transport across the entire low-permeability sequence.