Reaction between organic matter, limestone and AMD

[tomsanroca]

Hi

I've to model the precipitation of metals from acid mine drainage. This is the setting:

A flux (acid mine drainage) enter in a pond. This pond has a limestone layer in bottom, and overlies it an organic matter layer. Then, the flux reacts first with organic matter layer to form sulfides (so, metals can precipitate and get separated from the flux). Then, the flux reacts with the limestone to dissolve it to achieve pH gets neutralized. Finally, the flux continue the path to another pond.

My question is:

How can I model those processes? How put the organic matter layer's composition? and limestone's composition? What kind of model are those processes? mixing, tritation, reactive transport?

Please help! This is the reference: Noosai, N., Vijayan, V., & Kengskool, K. (2014). Model application for acid mine drainage treatment processes. International Journal of Energy and Environment, 5(6), 693-700.

[dlparkhurst]

Here is an example of a sequential calculation: first an initial solution is speciated, then the solution reacts with organic matter (CH2O), and finally, the reacted solution reacts with calcite.

There are a lot of choices to be made. I used a solution from the paper that you cited, but it lacks analyses for major ions Na, Ca, Mg, and Cl. I made up a charge balanced solution with Ca and Mg, but you need analytical data for these ions.  I skipped U and As just to simplify, which allowed using the phreeqc.dat database.

The extent of organic matter reaction is unknown, I have used an arbitrary 4 millimoles of CH2O. The minerals that might form are uncertain, as are the thermodynamic data for those minerals. I have included carbonates, sulfides, and some oxyhydroxides as candidate minerals. Finally, when the solution reacts with carbonate, you must consider whether the system is open or closed to CO2(g) (and O2(g)). 

There are probably many other considerations, but maybe this script will get you started. If you have data at the beginning and the end of the processes, you might consider using inverse modeling instead of this forward modeling.

Code: [Select]

PHASES
MnS(Green)          192
        MnS + H+ = Mn+2 + HS-
        log_k           3.8
        delta_h -5.790 kcal

Greenockite         332
        CdS + H+ = Cd+2 + HS-
        log_k           -15.930
        delta_h 16.360 kcal
END
SOLUTION 1
-units mg/L
pH 3.1
Fe(3) 5
Fe(2) 46.8
Mn 14.2
Al 1.14
Cd 1.1
#As 0.9
#U 0.85
S(6)  580
O(0)  5.3
Ca    1 charge
Mg    50
END
USE solution 1
REACTION 1
CH2O 1
4 mmol in 1 step
EQUILIBRIUM_PHASES 1
    Calcite   0 0
    Greenockite 0 0   #CdS
    MnS(Green) 0 0    #MnS
    Otavite   0 0     #CdCO3
    Pyrite    0 0     #FeS
    Rhodochrosite 0 0 #MnCO3
    Siderite  0 0     #FeCO3
    Goethite 0 0
    Gibbsite 0 0
SAVE solution 2
END
USE solution 2
EQUILIBRIUM_PHASES 1
    Calcite   0 10
    Greenockite 0 0   #CdS
    MnS(Green) 0 0    #MnS
    Otavite   0 0     #CdCO3
    Pyrite    0 0     #FeS
    Rhodochrosite 0 0 #MnCO3
    Siderite  0 0     #FeCO3
    Goethite 0 0
    Gibbsite 0 0
    #O2(g)    -0.7  10
    #CO2(g)   -3.4  10
END