Surface Complexation and Arsenic Adsorption

[Pesquerj]

Hey, new here, I am trying to recreate some of the modeling done in this paper "Dynamics of arsenic adsorption in the targeted arsenic-safe aquifers in Matlab, south-eastern Bangladesh: Insight from experimental studies" Robinson et al, 2011, Applied Geochemistry

The base for the model is generating a surface complexation using minteq.v4.dat and DDL, I am having several issues, first and foremost I get this when trying to input Alkalinity as HCO3

ERROR: Concentration data error for alkalinity in solution input.

Then, several surfaces reactions that are needed are not included in the database so I input them manually under Surface Species like so

SURFACE_SPECIES

   Hfo_wOH + CO3-2 + H+ = Hfo_wCO3- + H2O
   log_k 12.56

   Hfo_wOH + CO3-2 + 2H+ = Hfo_wHCO3 + H20
   log_k  20.62

   Hfo_sOH + H4SiO4 = Hfo_sH3SiO4 + H20
   log_k 4.28

   Hfo_wOH + H4Si04 = Hfo_wH3SiO4 + H20
   log_k 4.28

   Hfo_sOH + H4SiO4 = Hfo_sH2SiO4- + H20 + H+
   log_k -3.22

   Hfo_wOH + H4SiO4 = Hfo_wH2SiO4- + H20 + H+
   log_k -3.22

   Hfo_sOH + H4SiO4 = Hfo_sH2SiO4-2 + H20 + 2H+
   log_k -11.69

   Hfo_wOH + H4SiO4 = Hfo_wH2SiO4-2 + H20 + 2H+
   log_k -11.69

These are the ones missing from minteq.v4 that they did also use in the article.

But when I try to run it with them, it does not work I get these errors and I am not sure what's wrong really (it does function properly when I take them out and only include the Surface data I want to run for Hfo)

SURFACE 1

   Hfo_sOH        5e-5    600.    0.09 
        Hfo_wOH        2e-3
   
Surface 2
   Hfo_sOH        4.5e-5    600.    0.79 
        Hfo_wOH        1.8e-4
   
Surface 3
   Hfo_sOH        3.5e-5    600.    0.66 
        Hfo_wOH        1.4e-3

And these are the errors;

ERROR: Elements in species have not been tabulated, H20.
ERROR: Reaction for species has not been defined, H20.
ERROR: Elements in species have not been tabulated, H4Si04.
ERROR: Reaction for species has not been defined, H4Si04.
ERROR: Calculations terminating due to input errors.

Then I am trying to understand part of what they did, seems to me like they modeled the surface complexation in equilibrium with the mean water chemistry for one date (I have included that as solution 1) and then applied the command USE as in USE Solution 1 USE Surface 1 end. The thing is they also then equilibrated it with each batch experiment solution, and I am not sure how to compute that in the model to get a user graph that will show the isotherms for Arsenic adsorption on the each 3 surfaces.

I know it is a lot to ask, but will really appreciate some help, thanks in advance.

[Pesquerj]

I realized I had some 0 instead for Os in there so that was a quick fix, I am still struggling with the last question, in the article they mention that "The sustainability of targeting the oxidized sediments as a
source of As-safe drinking water is influenced by the capacity of
these sediments to attenuate As in groundwater that may flow
in from As-rich reduced aquifer sediments. Adsorption isotherms
were generated for the three oxidized sediment samples by combining
the sediments with solutions with low (0.2–1 mg L1) and
high (5–100 mg L1) As(V) concentrations. The quantity of adsorbed As(V) was calculated as the difference between the
initial and final dissolved As(V). For all experiments with the
low As(V) solutions more than 95% of the initial As(V) was adsorbed.
Experiments conducted with high As(V) solutions produced
similar isotherms for the three sediment samples"

And I am not sure how to add that to my script after I have modeled the 3 surface complexation that they give away as examples of what they used.

This is what they said about it "The PHREEQC model was then applied to simulate the measured
experimental adsorption isotherms and test the extent to
which the model, which considers adsorption to amorphous Fe(III)
oxides only, could predict the adsorption behaviour of the sediment.
The initial surface composition was simulated to be that in
equilibrium with the mean field water chemistry in the oxidized
sediments in November 2006. This surface composition was then
equilibrated with each batch experiment solution"

[dlparkhurst]

I think I can address the last part of your post. Assuming you have defined all the sorption reactions and log Ks correctly, PHREEQC has the capability to calculate the composition of a surface that is in equilibrium with a solution. The definition, say, SURFACE 1; -eq 1, indicates surface 1 will be calculated to in equilibrium with solution 1.

So, the article has taken a water from the oxidized zone and calculated the surface composition that would be in equilibrium with it. They then take that surface and react it with another water of a different composition (say, USE solution 2; USE surface 1) and look at the calculated change in arsenic concentration.