Re: Pyrite Oxidation Phreeqc Question

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

> I was hoping you might be able to provide some insight for a warning
message
I am getting running a pyrite oxidation simulation.  I am modeling
injecting
oxygenated water into a basalt aquifer containing 0.1% pyrite to simulate
what may occur during aquifer storage and recovery operations.  For my
initial simulation I have not included possible precipitation of
ferrihydrite.  There is no iron in my initial solution so the only source
of
iron is what is released due to the oxidation of pyrite.  At the point in
the simulation where iron concentrations plateau as oxygen is depleted I
get
the following warning:

WARNING: Element Fe has negative moles in solution, -2.943717e-005.
             Erroneous mole balance occurs as moles are added to produce
zero
moles.
             Caused by KINETICS or REACTION.
             May be due to large time steps in early part of KINETICS
simulation.
Using solution 2.        Richland Water (9/25/00)
Using kinetics 2.        Kinetics defined in simulation 2.

> The final iron concentration in solution however is equal to the total
amount of pyrite that has dissolved during the simulation as would be
expected.  Since I have no reactions that would remove iron, I'm not sure
how I could have negative moles in solution.  I have tried decreasing my
time step but this does not seem to help.  Any help would be greatly
appreciated.

I'm not sure exactly what happens, but I think it is possible that the
program is running correctly. Kinetics causes the program to integrate the
rate equation over a period of time.  It is possible that in the
intermediate calculations, pyrite was supersaturated and an amount of iron
was calculated to be removed that caused a negative concentration. It
depends on what rate equation you were using whether this is possible. It
is also possible that the Runge-Kutte method itself generated the negative
concentrations. The RK method uses 6 different estimates of the rate and
weights them; some of the weights are negative, which could generate a
negative iron concentration. However, if all goes well, the program should
recover and use smaller time steps to avoid the negative concentration. The
message is just a "warning" and not nececarily and error. If the program
ran to completion, it is likely it calculated the correct answer.


> I emailed you earlier today with a question which I subsequently figured
out
the answer to (message below).  I now know I should have been using the
incremental reaction step.  I tried to recall my earlier message but just
in
case it gets to you I wanted to let you know I was no longer in need of a
response.

I'm not sure what you did, but INCREMENTAL_REACTIONS shouldn't make too
much difference. It only applies to batch-reaction calculations, not
transport or advection. If you are running a batch reaction, it is probably
that the warning occurred at the earliest times of the integration, when
rates are fastest. INCREMENTAL_REACTIONS would have an effect if you
specified two or more time increments in the KINETICS keyword.

David


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