Initial Mols in RM

[bkazamel]

Hello,

I am simulating water draindown and mineral reactions in a soil column using Reactive Transport Modelling, and have a question about how PHREEQC handles the initial mols of minerals in the KINETICS block. My model has 100 cells, which we are assuming to be 1 m3. Is the initial mols in the KINETICS block referring to the initial mols of each individual 1m3 cell, or the initial mols spread evenly across all 100 cells? As well, are we correct in assuming that each cell has dimensions of 1 m3, although the RM is a 1D model? Or, does the volume of each cell relate to the volume of water defined in the SOLUTIONS block (default 1 kg, or 1 L)?

Cheers,
B

[dlparkhurst]

The default representative volume is 1 L. It is best to keep the RV such that there is 1 L of water--within an order of magnitude or two. 1 m3 might work if porosity is low, but it is on the high side for RV.

If you are using InitialPhreeqc2Module to populate the cells of your model, the method SetUnitsKinetics determines how the moles defined in the KINETICS definition (InitialPhreeqc instance) is converted for the module. The options are moles per RV, moles per L of water, and moles per L of solids.

[bkazamel]

Thank you for your quick reply.

I am using the GUI for reactive transport modelling, and populating the inputs of our cells manually (each cell is identical anyway). Could you clarify what you mean by keeping the RV such that there is 1 L of water? Not sure what is meant by RV. For reference, I are modelling an unsaturated soil with porosity of 0.475, and degree of saturation of 65%.

Cheers,
B

[dlparkhurst]

Sorry, I thought you were using PhreeqcRM, which is a library for coupling geochemistry to multicomponent transport models. It allows definition of a representative volume (RV).

With PHREEQC and the GUI PhreeqcI, there is not an explicit representative volume. A cell is created by defining SOLUTIONs, EQUILIBRIUM_PHASES, KINETICS, etc having the same user number. A SOLUTION normally has 1 kg water, or approximately 1 liter (although it could be different). EQUILIBRIUM_PHASES, KINETICS, etc are defined by a number of moles (no reactant data are defined as moles per liter or moles per kilogram water, just moles). So, when the solution and reactants are combined in a cell, the moles in the KINETICS definition is normally in contact with about  one liter of water.

I think it best that you use a SOLUTION with about 1 kg (~1 L) water and scale the moles of your other reactants (KINETICS) to that volume of water.

It is possible to adjust the volume (mass) of solution with -water in SOLUTION or MIX and then scale your other reactant moles accordingly.  Note that in TRANSPORT, the the volume/mass of water of SOLUTION 0 will propagate through the column so that after one pore volume, any original masses or volumes in a cell will be replaced by that of SOLUTION 0.

[bkazamel]

Hello, thanks for that clarification. That logic makes sense. So, would it be fair to say that the mass and volume of soil reacting in each cell can be estimated by determining how much soil volume would contain 1 L of water at any given porosity?

Cheers,
B

[dlparkhurst]

In your case, in spite of what I said, it may be easier to scale the amount of water to that in a liter of porous medium; at least that is easier for me to conceptualize. You can use -water in SOLUTION to adjust the mass of water (and solutes) in the solution. Also MIX 100; 1 0.5; SAVE solution 1; END would produce SOLUTION 1 with half the mass of water and half the volume.

However, the PHREEQC transport capabilities are fairly limited, and do not explicitly consider unsaturated flow. It is not possible to preserve variations in water content in the column. TRANSPORT will eventually replace any initial variations in water volume with the volume of SOLUTION 0. So, your only real option is to choose a single value for the volume of water in each cell of your column and ensure that SOLUTION 0 has that same volume.