A review of porosity and Diffusion in Bentonite (pdf) (2.4 MB) - Posiva

25ion, the structure expands further into the 3-layer hydrate when the RH increases above85 % (Bérend et al. 1995).With increasing spacing, the interlayer water can be expected to gain the density of freewater, as observed in the XRD studies. Thus, the decreasing density calculated byKaraborni et al. and others (e.g. Marry and Turq, 2002) may be an artifact that is associatedwith approximations in the MD simulations (Young and Smith 2000).In the model for calculating the distribution of the porosities, it was assumed that thedensity is 1 kg/L (f dens = 1), and that may look erroneous given the results in Table 4.However, the XRD measurements are done at less than 90 % relative humidity, correspondingwith activity of water less than 0.9. On the other hand, in free porewater inwhich the ionic strength is below 1, the activity of water is close to 1 and hence, the assumeddensity may be realistic.4.1 Experimental verification is necessary; consistency checks arepossibleThe example calculation in Section 4.3 of the fluxes of 3 differently charged tracersthrough the porosity domains hinges on the parameters discussed before, but offers atemplate for testing by experiment. The proposed setup can be reproduced in a laboratoryexperiment by assembling 2 pieces of bentonite with the properties of the boundaryconditions and measuring concentration profiles after given times. The bentonite needsto be cleaned from reactive accessory minerals and washed. The resulting diffusion profilescan be modeled with PHREEQC and used for verifying model-assumptions and -parameters. In particular, the large flux of Sr 2+ through interlayer water, which is partlya result of the tortuosity factor found in MD calculations, needs experimental validation.The MD calculations consider a single, rectangular interlayer unit and do not accountfor irregular packing of the units. Thus, the real tortuosity factor for cations in interlayerwater is probably larger. Also, the same tortuosity factor was adopted for cations andanions in free porewater and DDL water. However, as illustrated in Figure 5, the tortuositymay be larger for anions than for cations and neutral species, and the viscosity ofDDL water may be higher than of free porewater.Consistency checks are possible of the parameters used for DDL water. The relativeamounts of free porewater and DDL water, and the average potential of DDL watershould agree with the anion-accessible porosity, the potential that is used for calculatingthe depletion or enrichment of anions or cations, the external surface area ofmontmorillonite in contact with DDL water, the thickness of DDL water, the averagepore-size, and the charge of montmorillonite that is counterbalanced in DDL water. Forexample, if the anion-accessible porosity is measured to be 0.034, the following equationresults for I - :an, Ifree FDDL DDL 0. 034 exp(41) RT tottot