RE: Using Phreeqc for modeling redox reactions

["Mukhopadhyay, Biswajit" <BiswajitM@xxxxxxx>]

David:
I am sorry to bother you with the same issue.  Please help me to remove the
confusion in my mind caused by your explanation of the reactions.  It was my
understanding that when I was using pure iron electrodes, at anode ferrous
iron was being liberated into the solution.  Ferrous iron was acting as the
reducing agent to reduce Cr(VI) to Cr(III) and in this process itself was
being oxidized to ferric iron.  I have observed formation of thick yellowish
brown ferric hydroxide precipitation.  Cr(OH)3 was also precipitated.  In
the initial water Cr total [mostly as Cr(VI)] was 86 mg/l but after 20
minutes of reaction, the filtered water had no detectable amount of
dissolved Cr.

Reduction of Cr(VI) bearing solution by addition of ferrous salts (e.g.
FeSO4) is a well-known process.  However, I read a paper in which the author
casually made a statement: "removal of Cr from wastewater by chemical
reduction-precipitation is neither thermodynamically nor kinetically
efficient, especially when Cr content is low (<10 mg/l)".  I was trying to
see the justification behind his statement by performing equilibrium
calculations using Phreeqc.  Now I am totally stumped!

I agree with the above-mentioned author since I did some experiments with
groundwater from various parts of the plume and I did find that using pure
chemical method such as reduction by S(IV) containing compounds was not
producing satisfactory results when Cr concentration in the contaminated
plume was less than 10 mg/l.  But it worked fine for the samples with high
concentration of Cr (like the example I sent you).

With the electrochemical method I do not see this problem.  However one
interesting point that I find experimentally is that as Cr concentration
decreases, the administered dose of Fe(II) needs increase.  For example, the
MW-15 water (86 mg/l Cr) needed an ideal stoichiometric amount of Fe but
another water with 2.6 mg/l Cr needed almost twice the stoichiometric
amount.

Any suggestion on how a theoretical explanation can be found using Phreeqc
by equilibrium calculation?

Thanks for your time and help.

Best regards,

Biswajit


Biswajit Mukhopadhyay, M. S., M. E., Ph. D., D. Sc.
Senior Hydrologist & Geochemist
ecology and environment, Inc.
International Specialists in the Environment
biswajitm@xxxxxxx
Phone: 214-245-1051
www.ene.com

Office Address:  1200 Main Street, Suite 500
                         Dallas, TX 75202
                         USA


> -----Original Message-----
> From:	David L Parkhurst [SMTP:dlpark@xxxxxxxx]
> Sent:	Tuesday, January 14, 2003 6:24 PM
> To:	Mukhopadhyay, Biswajit
> Subject:	RE: Using Phreeqc for modeling redox reactions
> 
> 
> I think your reaction is essentially correct. You need to remove the
> EQUILIBRIUM_PHASES with O2(g). Initially I think the reaction should be as
> follows:
> 
> At the anode    Fe = Fe+3 + 3e-
> 
> At the cathode  CrO4-2 + 3e- = Cr+3, with the stoichiometry determined by
> equality of electrons.
> 
> However, you have NO3 in the system and that confuses things. If you
> simplify the system, you can see the reaction you want. Comment out the
> N(5) and reduce the amount of iron in the reaction by an order of
> magnitude. You will see that you make Cr(3) and Fe(3). If you add an
> excess
> of Fe, you eventually reduce all of the CrO4, make Fe+2 out of all the
> Fe+3, and start to reduce SO4, CO3, or water.
> 
> David
> 
> David Parkhurst (dlpark@xxxxxxxx)
> U.S. Geological Survey
> Box 25046, MS 413
> Denver Federal Center
> Denver, CO 80225
> 
> Project web page: https://wwwbrr.cr.usgs.gov/projects/GWC_coupled
> 
> 
> 
>  
> 
>                       "Mukhopadhyay,
> 
>                       Biswajit"                To:       'David L
> Parkhurst' <dlpark@xxxxxxxx>              
>                       <BiswajitM@xxxxxx        cc:
> 
In-Reply-To: <OF4DC11241.D37A7A62-ON87256CAF.00006E20@xxxxxxxx>
>                       m>                       Subject:  RE: Using Phreeqc
> for modeling redox reactions     
>  
> 
>                       01/14/03 04:36 PM
> 
>  
> 
>  
> 
> 
> 
> 
> 
> David:
> Thanks a lot for your reply.  I think the reason I see the difference in
> two
> approaches I was taking lies in your explanation that in the Reaction
> calculation Phreeqc does not differentiate between Fe(II) and Fe(0).  But
> in
> the electrochemical process I am actually introducing ferrous iron into
> the
> solution directly instead of as a salt such as FeSO4.  The reactions are:
> 
> At the anode (oxidation reaction):         Fe *  Fe+2 (aq) + 2e-
> (1)
> At the cathode (reduction reaction):      2H2O + 2e-  *  H2 * + 2OH- (aq)
> (2)
>  <<...OLE_Obj...>>
> How can I model this in Phreeqc?
> Regards,
> 
> Biswajit
> 
> > -----Original Message-----
> > From:            David L Parkhurst [SMTP:dlpark@xxxxxxxx]
> > Sent:            Tuesday, January 14, 2003 5:47 PM
> > To:        Mukhopadhyay, Biswajit
In-Reply-To: <OF4DC11241.D37A7A62-ON87256CAF.00006E20@xxxxxxxx>
> > Subject:         Re: Using Phreeqc for modeling redox reactions
> >
> >
> > > My question is, for the experiment described above, is using the key
> > word
> > EQUILIBRIUM_PHASES with O2(g) Partial Pressure (10^-0.68) the correct
> > approach.  I enclose the input file and the database [modified
> phreeqc.dat
> > as phreeqcc.dat to include all Cr(VI) and Cr(III) species] for your
> > viewing.
> >
> > If you include the O2(g), you are assuming that sufficient oxygen enters
> > the beaker to maintain atmospheric partial pressure of oxygen at all
> > times.
> > My guess is that the gas-water transfer of oxygen is relatively slow and
> > you might be better off not using the O2(g). If the transfer were really
> > fast, you would expect to see red iron hydroxide forming as well.
> >
> > Note that in the reaction calculation, PHREEQC does not consider any
> > charge
> > that you might enter ("Fe+2"). It ignores the charge and is actually
> > adding
> > elemental Fe to the solution, which is ok, that is the reaction you
> want.
> > The elemental iron must oxidize to Fe+2 (or Fe+3) and in the process
> > something must be reduced. If you include the oxygen, then oxygen will
> be
> > reduced, whereas leaving oxygen out will force the reduction the most
> > thermodynamically preferred electron acceptor available, Cr(IV) I should
> > think.
> >
> > David
> >
> > David Parkhurst (dlpark@xxxxxxxx)
> > U.S. Geological Survey
> > Box 25046, MS 413
> > Denver Federal Center
> > Denver, CO 80225
> >
> > Project web page: https://wwwbrr.cr.usgs.gov/projects/GWC_coupled
> >
> >
> 
> 
> 
>