Columns experiments

[adrianosbz]

Hi everyone,

I am a Masters degree student in Geoscience and I had done a leaching columns experiment with Ni-slag, soil, and mixture of both in which I would like to use a Phreeqc modeling. I was trying to apply a Tom's example involving pyrite oxidation and AMD (Acid mine drainage). I have tried to follow colleague's suggests, but the program gives warnings over input data. Can anyone say me what the script I should use? Or could anyone go through me a model script that I can apply to leaching columns experiment?

Best regards,

Adriano

[dlparkhurst]

First you need to decide what the important processes are. You mention pyrite oxidation, so do you have low pH, high iron effluent from your column? Surface complexation might be important for certain elements: metals, arsenic, selenium.

I think inverse modeling is a good approach to try to understand the major reactions. Inverse modeling is good when you have reactants with known stoiciometries (pyrite, Fe(OH)3, gypsum, calcited, etc). Exchange is fairly well defined, but surface complexation is tougher to know the stoichiometry of the reactions. Trace elements are also difficult to treat in inverse modeling. Still, I think it is a good process to look at the major ions in solution and try to deduce the sources and sinks for each. Even if you do not achieve a quantitative model for the reactions that account for the effluent composition, it will be a useful exercise. 

[adrianosbz]

Hi David,

The columns simulate Ni-slag heaps disposed over and wrapped with oxisol. As the nickel ore is lateritic originated in ultramafic rocks in the Middle-West of Brazil, we do not have problem with pyrite, therefore there is not AMD. I take the Tom's example with pyrite oxidation as base model to create my simulation, though without pyrite. I am going to add thermodynamic Ca-Mg-Al-Si-glass data,  'cause this is the main composition of slags. So we mixed oxisol and slag, in various proportions, adding distillate and deionized water in equilibrium with air to simulate one hydrologic cycle with 2,400 mm of rain water. What we want to simulate would be the leaching conditions of this material after that distillate and deionized water pass through columns, and try to compare the results with those that had been gotten with leaching column experiment. We know that soil has several kaolinite, besides rutile, anatase, and gibbsite, which favor oxi-hydroxi-Fe accumulation. In this context, the Cr(VI) content is the main object of this investigation. This Cr is present as trace element, which enrich the Ni-slag and its release is cause of great environmental concern. For the moment, DPC tests confirm that there is no Cr(VI) present into leachate solution.

[dlparkhurst]

I think you just need to start putting together an input file. You know best what you want to do. It is much easier to respond to specific questions.

I still think it is useful to use inverse modeling to consider the sources and sinks for the major elements.

[adrianosbz]

Hi, Mr. Parkhurst,

Considering yours last message, I tried to do a simulation inverse, introducing new parameters. However, much errors has arisen which interrupted the running simulation:

ERROR: Equation has no equal sign.
   Mg2SiO4
ERROR: Parsing equation.
ERROR:         Mg2SiO4
           Mg2SiO4 + 4CO2 + 2H2O = 2Mg2+ + 4HCO3- + SiO2
           Log_k  0.0
ERROR: Equation has no equal sign.
   Log_k0.0
ERROR: Parsing equation.
ERROR:         Log_k  0.0
           delta_h   16.11
   Lizardite
           Mg3(Si2O5)(OH)4
ERROR: Equation has no equal sign.
   Mg3(Si2O5)(OH)4
ERROR: Parsing equation.
ERROR:         Mg3(Si2O5)(OH)4
           Mg3Si2O5(OH)4 + 6CO2 + H2O = 3Mg2+ + 6HCO3- + 2SiO2
           Log_k  0.0
ERROR: Equation has no equal sign.
   Log_k0.0
ERROR: Parsing equation.
ERROR:         Log_k  0.0
   Chromite
              FeCr2O4
ERROR: Equation has no equal sign.
   FeCr2O4
ERROR: Parsing equation.
ERROR:            FeCr2O4
         FeCr2O4 + 4 OH- + 1.75O2 = 2CrO4-2 + 0.5Fe2O3 + 2H2O
              Log_k_ 0.0
ERROR: Equation has no equal sign.
   Log_k_0.0
ERROR: Parsing equation.
ERROR:            Log_k_ 0.0
      Glass
         Ca0.015Fe0.095Mg0.066Na0.025K0.01Al0.105S0.003Si0.5O1.35
ERROR: Equation has no equal sign.
   Ca0.015Fe0.095Mg0.066Na0.025K0.01Al0.105S0.003Si0.5O1.35
ERROR: Parsing equation.
ERROR:       Ca0.015Fe0.095Mg0.066Na0.025K0.01Al0.105S0.003Si0.5O1.35
         Log_k   -99
   END
ERROR: Could not find phase, Forsterite.
ERROR: Could not find phase, Lizardite.
ERROR: Could not find phase, Chromite.
ERROR: Could not find phase, Glass.
ERROR: Calculations terminating due to input errors.

Follow the script used:

TITLE Leaching Ni-Slag Column Experiment (KINETICS with oxisol mineralogy)

######INFILTRATING SOLUTION######

SOLUTION 0   #Distillate and deionized water (assuming it is rainwater)
Units   ppm
pH   6.45
temp   25
Na   2.05
K   0.35
Ca   1.42
Mg   0.39
Cl   3.47
S(6)   2.19
N(5)   0.27
N(-3)   0.41
EQUILIBRIUM_PHASES 1   #equilibrates with the atmosphere
CO2(g)   -3.5
O2(g)      -0.699
SAVE SOLUTION 0
END
######INITIAL COLUMN CONDITIONS######
SOLUTION 1   #Initial solution in column (assuming it pure water, fixed pO2)
    Temp   25
    pH      6.45
    pe      4
    redox   pe
    units   mmol/kgw
    density 1
    -water  1 # kg
PHASES
INVERSE_MODELING 1
        -solutions   1   2
        -uncertainty   0.025
        -range
        -phases
                Forsterite
                CO2(g)
                Lizardite
                Chromite              precip
                Glass                 dissolve
                -balance
                Ca   0.05   0.025
PHASES
Forsterite
        Mg2SiO4
        Mg2SiO4 + 4CO2 + 2H2O = 2Mg2+ + 4HCO3- + SiO2
        Log_k  0.0
        delta_h   16.11
Lizardite
        Mg3(Si2O5)(OH)4
        Mg3Si2O5(OH)4 + 6CO2 + H2O = 3Mg2+ + 6HCO3- + 2SiO2
        Log_k  0.0
Chromite
           FeCr2O4
      FeCr2O4 + 4 OH- + 1.75O2 = 2CrO4-2 + 0.5Fe2O3 + 2H2O
           Log_k_ 0.0
   Glass
      Ca0.015Fe0.095Mg0.066Na0.025K0.01Al0.105S0.003Si0.5O1.35
      Log_k   -99
END

SOLUTION 1
        -units  mmol/l
        pH      8.45
        Si      0.324
        Ca      0.130
        Mg      0.463
        Na      0.011
        K      0.04
        Al      0.009            
        Fe      0.005
        B      0.051
        Cr      1.154e-4
        Cu      2.047e-4
        Zn      0.0012        
        Alkalinity   1.22e-3
        Cl      0.013
        N(5)   0.044
EQUILIBRIUM_PHASES 2
Quartz   0   17.92
Kaolinite   0   0.90
Chromite   0   0
Glass      0   0
Forsterite   0   0
Lizardite   0   0
GAS_PHASE 1   #Fix the partial-pressure of O2 in the top cell to simulate UZ
-fixed_pressure
-pressure 1.001
-temperature 25
O2(g   0.2
SAVE SOLUTION 1
SOLUTION 2-50   #Initial solution in column (assuming it pure water, fixed pO2)
    Temp   25
    pH      7
    pe      4
    redox   pe
    units   mmol/kgw
    density 1
    -water  1 # kg
EQUILIBRIUM_PHASES 2-50
Quartz   0   17.92
Kaolinite   0   0.90
Chromite   0   0
Glass      0   0
Forsterite   0   0
Lizardite   0   0

SAVE SOLUTION 2-50
END

######TRANSPORT######
TRANSPORT
-cells   1
-length   0.45 # 1 cell @ 0.45m = 0.45 m total
-dispersivity   0.005 #m
-flow_direction forward
-boundary_conditions fluxed   fluxed #at column ends
-diffusion_coefficient   1e-9 #m2/s
-punch_frequency   100 #punch out every 52 shifts
-time_step   144288 s #based on silty clay k of 0.25 cm/hr or 7e-7 m/s
-shifts 52

######OUTPUT FILE CONTENT######
_OUTPUT SELECTED
-pH true
-pe false
-step false
-state false
-simulation true
-solution true
-distance true
-saturation_indices Gypsum
-totals Ca

END
KNOBS
    -iterations            100
    -convergence_tolerance 1e-008
    -tolerance             1e-015
    -step_size             100
    -pe_step_size          10

USER_GRAPH 1
    -initial_solutions      false
    -connect_simulations    true
    -plot_concentration_vs  x
    -active                 true

As you can observe, for some values I have done an estimate. But I think that this has produced some inconsistencies which I do not get to solve nto can simulate my leaching column experiments. Could you help me, please? I have pH, electrical conductivity, cations, and anions data which I used at the simulation. So I would like to know if I can simulate after how many water volume there would be a stabilization in the leching process.

[dlparkhurst]

I have responded to several of your inquiries, but I am not going to respond to more input files that have input errors. You must revise your input file until it works, either by reading the manual keyword descriptions and examples, or using PhreeqcI to guide the input. PHASES data blocks are not that difficult. You can find examples in every database file, and you can probably find the minerals that you want among the database files.

Use scientific experimentation with your input file to deduce the cause of errors, and the effects of varying input on the calculated results. Modeling is only intended to help you think.

If you have a long input file, please attach it for the post to make it simpler for the users of the forum to read your question. Please try to make your questions as specific as possible.

If you want to do INVERSE_MODELING, you only need two SOLUTIONS (initial and column effluent), potentially PHASES or EXCHANGE_SPECIES to define reactants not already in the database, and INVERSE_MODELING. I suggest you focus on these data blocks. Consider sources and sinks of the major elements in these solutions, and elements in the phases that you think are responsible for the reactions in the column.