Re: Phreeqci 2.6 charge balance question

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

> No, I don't think the program is calculating incorrectly.  I think I
didn't
fully understand how the Advection keyword worked.  Realistically the
phenomena I was concerned about (solid phases resulting from the initial
condition dissolving) was very small after one pore volume and probably not
worth worrying about.  If I ran a large number of steps I think I'd get
what
I'm looking for, but I think just running a few pore volumes through is
close enough.

I'd have to see the input file to understand exactly what the problem is.
It seems that the calculations are adequate, so maybe it will keep until
the effect is significant.

> I did try running ten separate simulations representating ten steps
through
the reactor, where at each step I saved the solution and used it on the
next
one, each time using the same set of equilibrium phases and reaction.

Advection will automatically save the results after a "shift"; reaction
does not change; so you would need only to redefine the equilibrium_phases.
I think this outline would work.

SOLUTION
EQUILIBRIUM_PHASES
REACTION
END
SELECTED_OUTPUT
ADVECTION  # 1 shift only
END
EQUILIBRIUM_PHASES
ADVECTION
END
EQUILIBRIUM_PHASES
ADVECTION

> I got pretty much the same answer.  The bad thing about this way is, I
guess if I
use the selected output keyword to write to a spreadsheet, I would have to
write to ten separate spreadsheet files to save the results for plotting.
I
don't suppose there's a way for results from different simulations to be
written to the same spreadsheet file is there?

Once a file is defined in selected_output, that file is used for each
subsequent simulation. If you want a second file, then you need another
SELECTED_OUTPUT data block with a new -file. In general you only need to
define one file. Writing to the file can be suspended by using PRINT;
-selected_output false. Writing to the selected_output file will be stopped
until a simulation is read with PRINT; -selected_output true.

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



                                                                                                           
                      "Brian Park"                                                                         
                      <bpark@xxxxxxxxxx        To:       "David L Parkhurst" <dlpark@xxxxxxxx>             
                      >                        cc:                                                         
                                               Subject:  Re: Phreeqci 2.6 charge balance question          
                      03/14/03 06:38 AM                                                                    
                                                                                                           
                                                                                                           




Hello,

No, I don't think the program is calculating incorrectly.  I think I didn't
fully understand how the Advection keyword worked.  Realistically the
phenomena I was concerned about (solid phases resulting from the initial
condition dissolving) was very small after one pore volume and probably not
worth worrying about.  If I ran a large number of steps I think I'd get
what
I'm looking for, but I think just running a few pore volumes through is
close enough.

I did try running ten separate simulations representating ten steps through
the reactor, where at each step I saved the solution and used it on the
next
one, each time using the same set of equilibrium phases and reaction.  I
got
pretty much the same answer.  The bad thing about this way is, I guess if I
use the selected output keyword to write to a spreadsheet, I would have to
write to ten separate spreadsheet files to save the results for plotting.
I
don't suppose there's a way for results from different simulations to be
written to the same spreadsheet file is there?

Thanks,
Brian Park

----- Original Message -----
From: "David L Parkhurst" <dlpark@xxxxxxxx>
To: "Brian Park" <bpark@xxxxxxxxxx>
Sent: Thursday, March 13, 2003 1:07 PM
Subject: Re: Phreeqci 2.6 charge balance question


>
> >What happens is, as the water flows through, various metals are
> precipitated
> as they should (Cu, Zn, Cd as sulfides, Al as hydroxide).  But eventually
> towards the end of the reactor the Cu/Zn/Cd concentrations increase
> slightly
> which should not happen - they should either decrease or stay the same,
not
> increase.  I have the equlibrium phase amounts set to zero, so that they
> should not dissolve but can precipitate if supersaturated.  It seems like
> during the advection sequence, the phase amounts for those metals get set
> to
> nonzero amounts, so that if conditions dictate they dissolve slightly.
>
> Right. If a phase precipitates during the simulation, it will redissolve
if
> conditions become undersaturated. I'm not sure why you think that the
> concentrations should not increase. Do you think the program is making an
> incorrect calculation? or that the program is calculating correctly, but
> you want to eliminate this effect in some way? My guess is that the
> calculations are done correctly, but perhaps there is an effect that you
> have not expected.
>
> >Is there a way to get around that with the advection keyword, to have
each
> shift contact equilibrium phases with zero amounts?  What I'm trying to
do
> is predict water quality change as it flows through the reactor.
>
> Unfortunately, there is a dissolve_only option, but not a
precipitate_only
> option. There is no easy way to do what you want. You could introduce
> kinetic reactions that precipitate the phases, but do not dissolve. This
> should work, but there can be problems if you try to make the rates fast
to
> simulate equilibrium. You can also redefine the EQUILIBRIUM_PHASES in the
> column, but at the extreme you would only run ADVECTION for one step at a
> time, redefine EQUILIBRIUM_PHASES, and then run another ADVECTION for
> another step.
>
> 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
>
>
>