Re: O2(g) Minteq

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

Thank you for bringing the discrepancy about O2(g) to my attention. Your
analysis is correct, although mixing values from different databases is
probably the type of thing that caused the problem in the minteq data base.
The delta H for O2(g) in the minteq data base appears to be a poor
selection or an error. The phreeqc.dat delta H gives the correct tendency
for solubility. However, to confuse things more, I think the value in
llnl.dat (delta H = -2.9 kcal; -12.1336 kJ) gives a better fit to
solubility measurements.

I am reluctant to change minteq.dat in the distribution of PHREEQC. I
checked and the data in minteq.dat are consistent with the database for the
MINTEQ program, so PHREEQC results with minteq.dat should be consistent
with results from the program MINTEQ. I am not going to attempt to analyze
all the thermodynamic data, it is just too big a task. I am reasonably
confident of the basic data in phreeqc.dat and wateq4f.dat. Also, the data
in llnl.dat has probably been through more rigorous analysis than the other
databases. In general, I am somewhat suspicious of minteq.dat, just because
I found many errors in stoichiometry when I was translating the data to
make minteq.dat, but I made no attempt to evaluate the log K's and delta
H's.

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



                                                                                                                           
                      "Kaoru Masuda"                                                                                       
                      <k-masuda@xxxxxxxxxxx         To:      "David L Parkhurst" <dlpark@xxxxxxxx>                         
                      elco.co.jp>                   cc:                                                                    
In-Reply-To: <002701c253ca$4f6f7940$033a310a@xxxxxxxxxxxxxxxxxxxxx>
                                                    Subject: O2(g) Minteq                                                  
                      09/03/02 10:19 PM                                                                                    
                                                                                                                           
                                                                                                                           



Dear Dr. Parkhurst,

I am one of the PHREEQCI users.
I think I found some discrepancy on deltaH of O2(g) dissolution.
When I tried to calculate O2 solubility in water at different temperatures
using Minteq.dat that is bundled with PHREEQCI.
(I usually use Minteq.dat because it contains a lot of data for solutility
 products.)
For example, when I set the partial pressure logP=-2,
calculated concentrations of O2 in the solution are
 2.19E-5 mol/L at 25C
 2.37E-5 mol/L at 40C

On the other hand, using PHREEQC.dat,
 2.19E-5 mol/L at 25C
 1.89E-5 mol/L at 40C
My understanding is that gas solubility decrease at higher temperature.

O2 gas data in Minteq.dat and Phreeqc.dat are followings.
Minteq.dat
   O2 + 4H+ + 4e- = 2H2O
   logk1 = 83.12  delta_H1=-133.83
Phreeqc.dat
   O2 = O2
   logk2 = -2.96  delta_H2=-1.844

I think the data in Minteq is the combination of
Henry's constant and dissociation constant of H2O;
2H2O = O2 + 4H+ + 4e-
  logk3 = -86.08 delta_H3=134.79

and relationship between  logk1,k2,k3 is
 logk1=logk2 - logk3.

I think this is the same for delta_H and
 delta_H1=delta_H2 - delta_H3.
If it's true, deta_H1 should be -136.634

If I replaced O2(g) data in Minteq.dat by
O2 = O2 logk= -2.96  delta_H2=-1.844
or I replaced delta_H by -136.634 ,
calculation results were same as those by Phreeqc.dat.

I will appologize if I am wrong or this problem is already known.

Best regards,

Kaoru Masuda,
Kobe Steel, Ltd.