Surface complexation modeling of silica particles

[rrahimi]

Hi everybody. I am working on the surface complexation modeling of silica particles in simple NaCl
solutions in Different pH values. However, the model is only capable to reproduce experimentalzeta potential
values at neutral pH (pH=7). The further the pH value gets from 7, the higher the error value becomes. Can
anybody tell me what is wrong in my code? it is worth mentioning that I have changed the K values and the
surface site densities too, but the same occurs again. Thanks a lot in advance.

TITLE Franks G.v

SURFACE_MASTER_SPECIES
Q_s     Q_sOH 


SURFACE_SPECIES

Q_sOH = Q_sOH
log_k  0



Q_sOH + H+ = Q_sOH2+
log_k  -1.1 



Q_sOH = Q_sO- +H+
log_k  -8.1







SOLUTION 1
    temp      25
    pH     9
    pe        4
   -pressure 1
    redox     pe
    units     mol/kgs
   
   # Ca        147.49
    Cl        0.1
    #K       0.1
    #Mg        17.46
    Na        0.1 charge
    #S(6)      0.89
    #Sr        8.44
    -water    1 # kg
   

   
SURFACE 1
    -sites DENSITY
    Q_sOH      10   3.033      4.5
    Donnan
   

    
PHASES
     Fix_H+
     H+ = H+
     log_k  0.0   
EQUILIBRIUM_PHASES
FIX_H+ -3 HCl 1
END

[dlparkhurst]

I can't help you much, other than to say that the surface potential is not the same as the zeta potential. Your system, with just NaCl, is pretty simple; it only depends on two log Ks and how you estimate the zeta potential.

There was another post recently that you may want to look at https://phreeqcusers.org/index.php/topic,1571.msg5101.html#msg5101

[rrahimi]

Thank you so much for your help. As you mentioned, zeta potential is different from the surface potential. However, there are several research papers reporting the use of surface complexation model to predict the zeta potential ( they assumed that the surface potential is the same as the experimental zeta potential and tuned their model parameters (log_k and surface site densities). In my case, no matter how the parameters are tuned, the model only predicts the data in neutral pH.

[dlparkhurst]

You don't show the data you are trying to match. The basic shape of the surface potential versus pH is pretty well set with the complexation reactions that are defined--a transition from positive to negative as pH increaces. You can adjust the PZC and to some extent the slopes by adjusting the parameters. Here is the result for the parameters that you provided.

Code: [Select]

TITLE Franks G.v
SURFACE_MASTER_SPECIES
    Q_s           Q_sOH       
SURFACE_SPECIES
Q_sOH = Q_sOH
    log_k     0
H+ + Q_sOH = Q_sOH2+
    log_k     -1.1
Q_sOH = Q_sO- + H+
    log_k     -8.1
PHASES
     Fix_H+
     H+ = H+
     log_k  0.0
END
SOLUTION 1
    temp      25
    pH     9
    pe        4
   -pressure 1
    redox     pe
    units     mol/kgs
   
   # Ca        147.49
    Cl        0.1
    #K       0.1
    #Mg        17.46
    Na        0.1 charge
    #S(6)      0.89
    #Sr        8.44
    -water    1 # kg
SURFACE 1
    -sites DENSITY
    Q_sOH      10   3.033      4.5
    Donnan
END
USER_GRAPH 1
    -headings               pH phi
    -axis_titles            "pH" "Surface potential, volts" ""
    -initial_solutions      false
    -connect_simulations    true
    -plot_concentration_vs  x
  -start
10 graph_x -la("H+")
20 graph_y EDL("psi", "Q")
  -end
    -active                 true
END
USE solution 1
USE surface 1
EQUILIBRIUM_PHASES
FIX_H+ -9 HCl 10
END
USE solution 1
USE surface 1
EQUILIBRIUM_PHASES
FIX_H+ -8 HCl 10
END
USE solution 1
USE surface 1
EQUILIBRIUM_PHASES
FIX_H+ -7 HCl 10
END
USE solution 1
USE surface 1
EQUILIBRIUM_PHASES
FIX_H+ -6 HCl 10
END
USE solution 1
USE surface 1
EQUILIBRIUM_PHASES
FIX_H+ -5 HCl 10
END
USE solution 1
USE surface 1
EQUILIBRIUM_PHASES
FIX_H+ -4 HCl 10
END
USE solution 1
USE surface 1
EQUILIBRIUM_PHASES
FIX_H+ -3 HCl 10
END
USE solution 1
USE surface 1
EQUILIBRIUM_PHASES
FIX_H+ -2 HCl 10
END
USE solution 1
USE surface 1
EQUILIBRIUM_PHASES
FIX_H+ -1 HCl 10
END
USE solution 1
USE surface 1
EQUILIBRIUM_PHASES
FIX_H+ 0 HCl 10
END


[rrahimi]

Thank you so much. I very new to Phreeqc and did not know I can plot the results of my simulations in a single graph. Is there any way to include my experimental data in the model too so that I can see the experimental data and the simulations results in a graph side by side ( to visually assess the precision of my model as a whole).  once again, thanks for the great help. :)

[dlparkhurst]

Look at -plot_tsv_file in the description of USER_GRAPH. Here is a made-up example that you can plot by adding the following to USER_GRAPH. There are options to change the symbols, colors, etc.

Code: [Select]

-plot_tsv_file psi.txt