Treatment of multi-dentate surface complexes and diffuse layer implementation in various speciation codes

[Display omitted] •Pitfalls associated with multidentate surface complexation modelling exist.•Different codes may yield different results with the same model parameters.•Simple corrections of intrinsic equilibrium constants eliminate the discrepancies.•Need for conversions of equilibrium constants between concentration scales. Spectroscopic studies and atomistic simulations of (hydr)oxide surfaces show that ionic aqueous adsorbates can bind to one, two, three, or four surface oxygen atoms (sites), forming multi-dentate species in surface complexation reactions. The law of mass action (LMA) for such reactions can be expressed in several alternative scales of surface concentration (activity). Unlike for mono-dentate surface complexes, the numerical value of the equilibrium constant is not independent of the choice of the surface concentration scale. Here, we show in a number of examples that the different formalisms implemented in popular speciation codes (MINEQL, MINTEQ, PHREEQC, and ECOSAT) yield different results for the same systems when the same parameters are used. We conclude that it is very important to generate general equations to easily transfer stability constants between the different concentration scales. It is of utmost importance for application of these models to reactive transport that the implementation in both the model fitting and speciation codes, and in the transport codes, is transparent to users. We also point to the problem that the implementation of the diffuse layer formalism in the various codes is not necessarily generally applicable. Thus, codes like VisualMinteq or MINEQL involve the Gouy–Chapman equation, which is limited to symmetrical (z:z) electrolytes, while PHREEQC and ECOSAT use general equations. Application of the former two to environmental problems with mixed electrolytes will therefore involve an inconsistency.