Arsenite sorption on Goethite

[Gunther_Schletter]

Hey after listening to some of your feedback on my other post, started to have the feeling the trend in our observed effluent data for arsenic does resemble a sorption trend. I?ve recently been trying to incorporate that into the model but have been having some trouble. Using example 19 from the PHREEQC manual and example 10.3 from Geochemistry, Groundwater and Pollution (Appelo, Postma).
We have some solid phase data from our experiments and in the initial phase (prior to any treatment) we had 0.6 grams of iron which we chose to enter in as Goethite. At the time Goethite was chosen because we had seen it referenced frequently in literature and had surface area and site density.

When I check the output file under the surface composition section Goet_sOHH3AsO3 and Goet_wOHH3AsO3 are present however the number of moles are extremely low ~e-125. Not sure what is causing it.

I?ve tried changing the surface type from equilibrium to general but still had the same result. My thought was maybe to change to another iron mineral and use the data specifications from: [https://www.hzdr.de/db/RES3T.queryData] but, from what I can find from literature arsenite is fairly adsorbable to Goethite so I think the problem lies elsewhere. Below is the batch reaction but, have also tried with a transport step and it still results in the same issue.

Any input would be greatly appreciated!

Goethite features found via:
Dixit, S., & Hering, J. G. (2003). Comparison of arsenic (V) and arsenic (III) sorption onto iron oxide minerals: implications for arsenic mobility. Environmental science & technology, 37(18), 4182-4189.
Mathur, S. S., & Dzombak, D. A. (2006). Surface complexation modeling: goethite. In Interface Science and Technology (Vol. 11, pp. 443-468). Elsevier.

SURFACE_MASTER_SPECIES
   Goet_s   Goet_sOH
   Goet_w   Goet_wOH
SURFACE_SPECIES
# strong binding site--Goet_s,
   Goet_sOH = Goet_sOH
   log_k   0.0
   Goet_sOH + H+ = Goet_sOH2+
   log_k   7.47    
   Goet_sOH = Goet_sO- + H+
   log_k   -9.51    
Goet_sOH + H3AsO3 = Goet_sOHH3AsO3
log_k -105
       mole_balance Goet_sOHH3AsO3

# weak binding site--Goet_w,
   Goet_wOH = Goet_wOH
   log_k   0.0
   Goet_wOH + H+ = Goet_wOH2+
   log_k   3.71435700000       
   Goet_wOH = Goet_wO- + H+
   log_k   -10.508450000    

       Goet_wOH + H3AsO3 = Goet_wOHH3AsO3
       log_k -105
       mole_balance Goet_wOHH3AsO3

SURFACE 1
    #-equilibrate with solution 1
    Goet_sOH Goethite        equilibrium_phase 0.0004 4806
    Goet_wOH Goethite        equilibrium_phase 0.016

EQUILIBRIUM_PHASES 1
    Goethite  0 0.000291 precipitate only

SOLUTION 1
    temp      15
    pH        7.87
    pe        -4
    units     mmol/kgw
    density   1
    Alkalinity 5.81
    #As_five   0.2 umol/kgw
    As(3)  1.388 umol/kgw
    Ca        1.921
    Cl        6.798
    Fe     0.895 umol/kgw
    K         0.169
    Acetate       0.18 mmol/kgw
    Mg        2.304
    Mn        282 ug/kgw
    N    1.357
    Na        9.482
    S(6)      2.719
    -water    1 # kg
END

#---------------------------------------#
# First Introduction of Groundwater YCB
#---------------------------------------#
SOLUTION 0 injection
    temp      15
    pH        7.77
    pe        -1.87
    units     mmol/kgw
    density   1
    Alkalinity 9.51
    Acetate       0.5 mmol/kgw
    Ca        1.799
    Cl        3.695
    Fe     9.45538 umol/kgw
    K         0.041
    Mg        1.074
    Mn        316 ug/kgw
    N     0.149
    Na        8.264
    S(6)      0.807
    #As_five   0.267 umol/kgw # Only including As(III) for right now
    As(3)  1.188 umol/kgw
    -water    1 # kg
END

[dlparkhurst]

I did not check the reference, but the log Ks of -105 are the reason that you are getting no sorption of As(3). The following plots the sorbed concentrations of As on weak and strong sites as a function of log K. You need log Ks above about 4 to have any effect on the As concentration.

BTW, there appears to be a problem when using precipitate_only with surfaces if the mineral does not precipitate. I suggest you remove the precipitate_only constrant.

Code: [Select]

SURFACE_MASTER_SPECIES
   Goet_s   Goet_sOH
   Goet_w   Goet_wOH
SURFACE_SPECIES
# strong binding site--Goet_s,
   Goet_sOH = Goet_sOH
   log_k   0.0
   Goet_sOH + H+ = Goet_sOH2+
   log_k   7.47   
   Goet_sOH = Goet_sO- + H+
   log_k   -9.51   
Goet_sOH + H3AsO3 = Goet_sOHH3AsO3
log_k -105
#       mole_balance Goet_sOHH3AsO3

# weak binding site--Goet_w,
   Goet_wOH = Goet_wOH
   log_k   0.0
   Goet_wOH + H+ = Goet_wOH2+
   log_k   3.71435700000       
   Goet_wOH = Goet_wO- + H+
   log_k   -10.508450000   

       Goet_wOH + H3AsO3 = Goet_wOHH3AsO3
       log_k -105
#      mole_balance Goet_wOHH3AsO3

SURFACE 1
    Goet_sOH Goethite        equilibrium_phase 0.0004 4806
    Goet_wOH Goethite        equilibrium_phase 0.016

EQUILIBRIUM_PHASES 1
    Goethite  0 0.000291 #precipitate only
END
SOLUTION 1
    temp      15
    pH        7.87
    pe        -4
    units     mmol/kgw
    density   1
    Alkalinity 5.81
    #As_five   0.2 umol/kgw
    As(3)  1.388 umol/kgw
    Ca        1.921
    Cl        6.798
    Fe     0.895 umol/kgw
    K         0.169
    Acetate       0.18 mmol/kgw
    Mg        2.304
    Mn        282 ug/kgw
    N    1.357
    Na        9.482
    S(6)      2.719
    -water    1 # kg
END
USER_GRAPH 1
    -headings               logk Dissolved weak strong
    -axis_titles            "Log K of As surface species" "Molality" ""
    -initial_solutions      false
    -connect_simulations    true
    -plot_concentration_vs  x
  -start
10 GRAPH_X LK_SPECIES("Goet_wOHH3AsO3")
20 GRAPH_Y LOG10(TOT("As")), LM("Goet_wOHH3AsO3"), LM("Goet_sOHH3AsO3")
  -end
    -active                 true
END
SURFACE_SPECIES
Goet_wOH + H3AsO3 = Goet_wOHH3AsO3
    log_k 0
Goet_sOH + H3AsO3 = Goet_sOHH3AsO3
    log_k 0
USE solution 1
USE equilibrium_phases 1
USE surface 1
END
SURFACE_SPECIES
Goet_wOH + H3AsO3 = Goet_wOHH3AsO3
    log_k 2
Goet_sOH + H3AsO3 = Goet_sOHH3AsO3
    log_k 2
USE solution 1
USE equilibrium_phases 1
USE surface 1
END
SURFACE_SPECIES
Goet_wOH + H3AsO3 = Goet_wOHH3AsO3
    log_k 4
Goet_sOH + H3AsO3 = Goet_sOHH3AsO3
    log_k 4
USE solution 1
USE equilibrium_phases 1
USE surface 1
END
SURFACE_SPECIES
Goet_wOH + H3AsO3 = Goet_wOHH3AsO3
    log_k 6
Goet_sOH + H3AsO3 = Goet_sOHH3AsO3
    log_k 6
USE solution 1
USE equilibrium_phases 1
USE surface 1
END
SURFACE_SPECIES
Goet_wOH + H3AsO3 = Goet_wOHH3AsO3
    log_k 8
Goet_sOH + H3AsO3 = Goet_sOHH3AsO3
    log_k 8
USE solution 1
USE equilibrium_phases 1
USE surface 1
END
SURFACE_SPECIES
Goet_wOH + H3AsO3 = Goet_wOHH3AsO3
    log_k 10
Goet_sOH + H3AsO3 = Goet_sOHH3AsO3
    log_k 10
USE solution 1
USE equilibrium_phases 1
USE surface 1
END


[Gunther_Schletter]

Well? that?s embarrassing. Thank you so much that makes total sense. I had initially tried varying the log_k but, I was running into a different issue initially where the adsorbed mineral wasn?t forming at all in the surface composition section and then I didn?t try varying it again after it had been resolved. -105 log K was just the last one I had tried.

I had used the precipitate only function because after I had put in the acetate reaction you had mentioned in my other post, I was seeing a dissolution of the initial Goethite (evident from a sharp increase in effluent iron) specified in the surface block. So then I used precipitate only in the hopes it would stop the dissolution of Goethite.

After varying the log_k it did resolve the issue in the batch experiment but when I put it in with the transport step Goethite was basically nonexistent after transport began. I figured because the saturation index for Goethite was below 0. I then checked for another iron mineral that had an SI above 0 and opted for Mackinawite. So I won?t need the precipitate function anymore and will be more cautious using it in the future.

Sorption reaction is working well with Mackinawite. Thanks again!