SURFACE input: number vs concentration of sites

[kbelli3]

I'm having trouble understanding why the SURFACE input is the number of surface sites (in moles) rather than the concentration of surface sites (in moles per liter).   

As an example, I'm trying to model uranyl adsorption onto ferrihydrite.  If the total number of surface sites is 1e-5 moles, it seems the equilibrium concentration of dissolved uranium will be different if that number of surface sites equilibrates with 1 mL of a solution of a given uranium concentration versus 1 L of the same solution.  Since the number of surface sites is finite rather than a concentration, that number of sites will more significantly decrease the concentration of dissolved uranium in the 1 mL solution compared to the 1 L solution where the concentration of dissolved uranium will not change substantially.  As the volume of solution is not provided in PHREEQC (at least not in the surface complexation example in the user guide), I don't understand how PHREEQC takes this into account.

I'm likely missing a small detail somewhere, but I haven't been able to rectify this in my mind.  Any help would be great.  Thanks!

Keaton

[dlparkhurst]

The amount of water in a solution is defined in the SOLUTION (or SOLUTION_SPREAD) definition. By default it is 1 kg of water, but can be changed by the -. By default it is the
-water identifier. The solution volume will be calculated from the speciation of the solution, but will be similar to the mass of water for dilute solutions.

SURFACEs and all other reactants (EQUILIBRIUM_PHASES, EXCHANGE, KINETICS, etc) are define in terms of moles. The main reason is to make their definitions independent of the solution that they are added to. The same number of moles are defined for a surface  whether the solution it is reacted with has a volume of 1 L or 0.5 L.

You may disagree, but I think it is easier to define extensively in moles than intensively as mol/L. Note that volume is temperature and pressure dependent. Thus, if SURFACE sites were defined as mol/L, the reaction system would have a different number of moles of surface sites if the same solution were used, but the temperature differed.

[kbelli3]

Ah, I see - I wasn't aware I could adjust the mass of water.  I do agree that it's more practical to define surfaces in terms of moles rather than concentrations for most problems.  Thanks for the clarification!  I really appreciate your quick response. 

[dlparkhurst]

Just for completeness, I should say that reactions also change the mass of water and solution volume. The most obvious is removing H2O to simulate evaporation. Precipitation of gypsum, for example, removes 2 moles of water per mole of gypsum, and almost all reactions make a subtle difference through hydrolysis reactions.