Re: Phreeqc modeling of kinetic Fe(OH)3 reductive dissolution

["David L Parkhurst" <dlpark@xxxxxxxx>]

There are other ways to set this up, but it is simplest if you consider the
consumption of organic carbon as the kinetic reaction. Then all you must do
is add the correct amount of CH2O and let equilibrium do the rest. Signs
may be confusing, but positive "moles" times positive coefficient of CH2O
is positive, which means CH2O is added to the solution. Fe(OH)3(a)
dissolves to equilibrium in response to the addition of CH2O. Carbon from
CH2O must end up as C(IV)(carbonate species) and/or C(-IV) (methane),
because those are the only aqueous species for carbon that are defined to
the program. With Fe(OH)3(a) present, the carbon ends up in carbonate
(oxidized) and Fe(OH)3 is reduced because of thermodynamics.

Another reason to set it up this way is that the number of surface
complexation sites can be related (proportional) to the number of moles of
an EQUILIBRIUM_PHASE, so as you develop this model, you can allow for
surface sites that diminish as Fe(OH)3 dissolves.




SOLUTION 0
      pH    7
      Na    10
      C     10 charge          # note total C ends up at 1.2 mmol for
charge balance at pH 7
EQUILIBRIUM_PHASES
      Fe(OH)3(a)  0.0   .01
KINETICS
Fe(OH)3_reduction
      -formula    CH2O 1
      -steps 86400 172800 345600 691200 1382400 # 1, 2, 4, 8, 16 days
RATES
Fe(OH)3_reduction
      -start
10 m_feoh3 = EQUI("Fe(OH)3(a)")
20 k = .05/(24*3600)    # constant /sec
30 rate = k*m_feoh3
40 moles = rate*TIME    # TIME in sec
50 SAVE moles
      -end
END



David Parkhurst (dlpark@xxxxxxxx)
U.S. Geological Survey
Box 25046, MS 413
Denver Federal Center
Denver, CO 80225


                                                                                                                             
                      "Roden, Eric E"                                                                                        
                      <Eric.Roden@xxxx         To:      "'dlpark@xxxxxxxx'" <dlpark@xxxxxxxx>                                
                      gov>                     cc:                                                                           
In-Reply-To: <1B1080D3ADDF8D4AB59832DE0A02C52A6BAE59@xxxxxxxxxxxxxxxx>
                                               Subject: Phreeqc modeling of kinetic Fe(OH)3 reductive dissolution            
                      02/13/02 11:20                                                                                         
                      AM                                                                                                     
                                                                                                                             
                                                                                                                             



Dear Dr. Parkhurst:

I wonder if I might trouble you with a question regarding the use of
Phreeqc
(version 1.5.08 for Windows) for a paper I'm working on to be submitted to
Applied Geochemistry. This is actually the same problem, more-or-less,
which
I had written to you about sometime ago, as I was just starting to try my
hand at mixed equilibrium-kinetic geochemical modeling. I wish to simulate
changes in pH and mineral precipitation accompanying kinetic reductive
dissolution of hydrous ferric oxide [HFO, assumed to be represented by
Fe(OH)3] by dissmilatory Fe(III)-reducing bacteria in HCO3-buffered medium.
An approximate set of starting conditions for such a simulation is as
follows:

pH 7
10 mM NaHCO3
10 mmol/L Fe(OH)3

We have good evidence that bacterial HFO reduction follows first-order
kinetics, with a rate constant of ca. 0.05/d in our normal culture systems.
Hence, the initial idea would be to predict changes in pH and dissolved
inorganic carbon speciation assuming that HFO reductive dissolution took
place according the following reaction:

Fe(OH)3(s) + 0.25 CH2O + 1.75H+  =>  Fe2+(aq) + 0.25HCO3- + 2.5H2O

in which the labile organic carbon (CH2O) is present in excess and
therefore
does not need to be tracked during the simulation, and in which Fe(OH)3
consumption is depicted as a first-order reaction

RFe(OH)3(s) = -k[Fe(OH)3(s)]

Alternatively, the reaction could be modeled as a surface-area controlled
process according to a more standard formulation such as

RFe(OH)3 = -k(m/m0)^n

Ultimately I would like to include other heterogeneous reactions and their
impact on pH changes during HFO reduction, namely: (i) H+ complexation by
Fe(OH)3 surfaces, whose abundance would be declining as reductive
dissolution takes place, e.g. as a first approximation in direct proportion
to molar/mass concentration and an assumed surface area of 600 m2/g; (ii)
precipitation of FeCO3(s) (siderite) and/or Fe(OH)2.

I have made some initial attempts to set-up a Phreeqc simulation of the
simple case where only reductive dissolution of Fe(OH)3 takes place, but I
am uncertain about how to depict the parallel consumption of the mineral
phase and H+, together with production of Fe2+(aq) and HCO3-.

I would be very grateful if you could provide assistance in setting-up this
problem. If I can get this to work, I can think of a great many uses for
Phreeqc in understanding and interpreting experimental studies of bacterial
Fe(III) oxide reduction and associated aqueous/solid-phase geochemical
reactions.

Best regards, Eric

Eric E. Roden
Current address (01/07/02 to 05/10/02):
Analytical Microbiology
Pacific Northwest National Laboratory
900 Battelle Blvd, Mail Stop P7-50
Richland, WA   99352
(509) 373-1043 (office)
(509) 376-5154 (lab)
(509) 376-1321 (fax)
(509) 627-0118 (home)

Permanent address:
The University of Alabama
Department of Biological Sciences
A122 Bevill Bldg 7th Ave
Tuscaloosa, AL 35487-0206
(205) 348-0556 (office)
(205) 348-1813 (lab)
(205) 348-1403 (fax)
(205) 349-1134 (home)