The sorption of radionuclides onto mineral surfaces is described by K_d-values based on surface complexation models (SCMs). To establish a respective data base, the following strategy is obeyed:

• In a first step, a literature survey  (cf. here, Tab. 1) mainly based on the RES³T database (Rossendorf Expert System for Surface and Sorption Thermodynamics) helped to define the chemical system, i. e. the mineral properties and the set of surface species. RES³T is a digitized thermodynamic sorption data base and is implemented as a relational data base. It is mineral-specific and can therefore also be used for additive models of more complex solid phases such as rocks, sediments or soils.
• The value for the specific surface area is strongly dependent on the sample history and grain size fraction. Here, experimentally determined values are used.
• All reaction constants are converted to infinite dilution (when necessary) by assigning activity coefficients based on the Davies Equation (Davies (1962)) to all dissolved species.
• Since reported reaction constants are related to varying site densities, it is necessary to convert the values to a reference state to be able to compare results. Here, the procedure according to Kulik (2002) is followed, based on a reference surface site density of 2.31 sites nm^-2. This value, of course, is then also used for subsequent predictive modeling.
• The values of pK₁ and pK₂ for the two successive protolysis steps always refer to the following deprotonation reactions, with =X-OH denoting a generic neutral surface binding site:
  pK₁:           =X-OH₂⁺  =  =X-OH + H⁺
  pK₂:           =X-OH  =  =X-O^- + H⁺ 
• When the data situation was too sparse, various approximations are used to derive a sensible parameter, namely the estimations based on crystallography and thermodynamics or the extrapolation from chemically similar systems (with regard to both mineral and sorbent) by applying the Linear Free Energy Relation­ships (LFER) as, e. g., described by Dzombak and Morel (1990).
• After normalization, the records with the same reaction (mineral surface protolysis and surface complex formation) are compared and assessed to identify and exclude outliers and doubtful data points. The remaining sets are then averaged to obtain respective surfacec complexation para­meters and also an estimation of their uncertainty.

References:

Davies, C. W., Ion Association, Butterworths, Washington (1962).

Dzombak, D.A., and Morel, F.M.M., Surface complexation modeling, Hydrous ferric oxide, Wiley, New York (1990).

Kulik, D.A., Sorption modelling by Gibbs energy minimisation: Towards a uniform thermodynamic data base for surface complexes of radionuclides, Radiochim. Acta, 90, 815 p., (2002) .

Sverjensky, D. A., Sahai, N., Theoretical prediction of single-site surface protonation equilibrium constants for oxides and silicates in water, Geochim. Cosmochim. Acta, 60, pg. 3773-3797 (1996) .