Competitive sorption, solute concentration initial conditions

A discussion forum for Hydrus-1D users.
Post Reply
Hue
Posts: 6
Joined: Tue Jul 16, 2019 12:32 pm
Location: Germany

Competitive sorption, solute concentration initial conditions

Post by Hue » Tue Nov 05, 2019 1:27 pm

Hi Jirka and all,

I am using two-site sorption model to simulate a flow-through column of soil that is contaminated with multiple perfluorinated chemicals (9+) to see how they are desorbed over time. The infiltrating water is clean. The contaminants in the soil initially can be only in solid phase (as adsorbed in the soil particles) and can also include the liquid phase concentrations. I have several questions below and hope you could help.

1) Is competitive sorption possible in Hydrus?: This means, as the substances present in the transport domain are supposedly competing for the limited sorption sites, modelling a porous media with multiple solutes would give different concentration results (for respective solutes) to the results for the same solutes when they are modelled separately in single solute models, given that all other model setup conditions are the same. I haven’t yet figured out how this could be done in Hydrus-1D.

2) Up to 10 independent solutes can be defined in one model, but it is not quite so: According to the Tutorial version July 2018, page 63, up to 10 solutes can be defined. Also, in one of your earlier posts under ‚Solute mixing?‘ in 2013 on this Forum, you mentioned that the multiple solutes are treated independent, if no first-order reactions (SinkWater1', SinkSolid1', SinkGas1' in the Solute Transport - Reaction Parameters window) are specified. If so, running a multiple-solute model should give the same concentration results as if these multiple solutes are modelled separately in several single-solute models? But I got quite a different experience: I was simulating a two-site sorption model for 9 chemicals (no reacting with one another, no decay). I tried simulating all of these 9 solutes in one model, but the solute mass balance for most of the solutes was really bad (relative error could amount to thousand percentage), although the model stability was excellent (PexCourant < 2); so I tried running them in two separate models (4-solute model, and 5-solute model), and still the same type of problem occurred. Then I tried out 9 separate single-solute models, and I got very good results; water and solute mass errors were all well below 1% at every time step for all solutes. I need to simulate all these and more solutes in one model, so if I have to simulate them separately, that would be very inconvenient. Have any of you experienced the same, and have an explanation? Also related to my first question, does this treating multiple solutes as independent mean no competitive sorption is possible in Hydrus-1D?

3) Solute concentration initial conditions: On the ‚Solute Transport Boundary Conditions‘ window, initial conditions can be defined as either in Liquid Phase Concentration [Mass_solute/Volume_water, or M/L_w3] or in Total Concentrations [Mass_solute/Volume_soil, M/L_s3]. I was wondering how these initial conditions are specifically related to the Initial Conditions Concentration [Mass_solute/Volume_water] and Sorbed Concentration [Mass_solute/Mass_soil] under the Soil Profile – Graphical Editor window. In one scenario, I have only the solid phase solutes initially, which I entered as Sorbed Concentrations in the Graphical Editor, and I left the Concentrations in this window all zero. So should I select the initial conditions defined as in Total Concentrations [Mass_solute/Volume_soil] in the Solute Transport BC window? But these Total Concentrations and Sorbed Concentrations have different units. I did select this option, and got zero output liquid concentrations at all time steps for all solutes in the Obs_Node.out files. But when I selected the Liquid Phase concentration option (and still Sorbed concentrations in the Graphical Editor), which again had different units, I got non-zero concentration results. I was puzzled. In another scenario where I have both solid and liquid phase concentrations of the solutes as the initial conditions, which option should I select, the In Liquid Phase or the In Total Concentrations?

4) Initial sorbed concentration: For the two-site sorption model, the initial sorbed concentration is the solute concentration on type-2 sorption sites (kinetic sorption sites). I understand that in this model solutes would adsorb first to type-1 instantaneous sorption sites which is supposed to be always at equilibrium with the liquid phase concentration, type-2 sites would be ‚filled up‘ following this equilibrium. So if I define the initial Sorbed Concentrations without defining the initial liquid concentration (the latter is assumed equal to Zero), would Hydrus internally back-calculate the initial solid concentration sorbed on type-1 sites and hence also the initial liquid concentration in equilibirum with the type-1 sorbed concentrations, based on the solute transport parameters provided in the model? Does this mean initial mass of the solutes would be larger than what user define initially?

Sorry for a long post. And thank you for your help.
Regards,
Hue

Jirka
Posts: 4798
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Competitive sorption, solute concentration initial conditions

Post by Jirka » Tue Nov 05, 2019 5:29 pm

I will compensate your long post with a short one:
1. No. Solute are independent, except for the first-order degradation reactions.
2. If you simulate multiple same chemicals at the same time, you should get the same results for all of them and the same results as if you simulate only one (if you have parameters for all chemicals the same). I just ran Test5 for 10 chemicals and got identical results.
3. This is explained in the help and manual.
4. You can define Sorbed Concentrations on the Kinetic sites. If you keep liquid phase concentration equal to zero, it will be initially zero. An increase in liquid phase concentration would then occur due to desorption from kinetic sites.

If you can answer your own questions by simply running such an example (e.g., items 3 and 4 above), why don't you do that?

J.

Hue
Posts: 6
Joined: Tue Jul 16, 2019 12:32 pm
Location: Germany

Re: Competitive sorption, solute concentration initial conditions

Post by Hue » Tue Nov 05, 2019 9:49 pm

Thank you, Jirka, for your quick and short reply. It was helpful.

Regarding item 2, I ran again my simulation for 9 different chemicals and still got bad mass balances, while I got good results simulating them separately. I would be grateful if you could have a look at my model. I can send you the files. Just let me know.

Thanks again.
Regards,
Hue

Jirka
Posts: 4798
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Competitive sorption, solute concentration initial conditions

Post by Jirka » Tue Nov 05, 2019 11:13 pm

You can post those files right here in the forum (as others did). J.

Hue
Posts: 6
Joined: Tue Jul 16, 2019 12:32 pm
Location: Germany

Re: Competitive sorption, solute concentration initial conditions

Post by Hue » Wed Nov 06, 2019 9:33 am

The attached zip files include the 9-solute model and a single-solute model for one of those solutes that was bad in the former but good in the latter. Much appreciated for your thoughts.
Regards,
Hue
Attachments
multiple_solutes_model.zip
(1.5 MiB) Downloaded 7 times

Jirka
Posts: 4798
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Competitive sorption, solute concentration initial conditions

Post by Jirka » Wed Nov 06, 2019 5:06 pm

I think that for nonlinear and nonequilibrium solutes you can have only up to 5 solutes. Max 10 solutes are only for equilibrium conditions. J.

Post Reply