Mean ion activity coefficients of NaCl solutions

[ammonge]

Hello,

I am new to PHREEQC. I am having difficulty trying to get the mean ion activity coefficient (gamma_plusminus) as a function of NaCl molal concentration or ionic strength for NaCl solutions. I am basically trying to reproduce Fig. 1A of Appelo (App. Geochem., 55, 2015: 62?71) but including additional databases in the comparison. My problem is double: on the one hand, I am not being successful in implementing reactions in which increasing concentrations of NaCl are added to a starting pure water "solution". On the other hand, if I create a solution of, say, 2 mol/kgw of NaCl, the mean ion-activity coefficient values I compute from the individual log(gamma_i) for Cl- and Na+ that PHREEQC reports in the "Distribution of species" section of the output are significantly lower than those plotted in the aforementioned reference for the same thermodynamic database and conditions. For this, I am using gamma_plusminus = gamma_Cl-*gamma_Na+. There is also a slight difference between the reported gamma_i and the ones I can calculate from the reported solution activities and molalities.

Any help would be welcome. Many thanks in advance.

[dlparkhurst]

Mean activity coefficient is sqrt(gamma(Na+)*gamma(Cl-)) and in general for a salt CnAm

Code: [Select]

mean_gamma = [gamma(C)^n*gamma(A)^m]^(1/(n+m))

Here is a script you can use:

Code: [Select]

SOLUTION 1
REACTION 1
NaCl 1
6 mol in 20 steps
USER_GRAPH 1
10 GRAPH_X RXN
20 GRAPH_Y [GAMMA("Na+")*GAMMA("Cl-")]^(1/2)
END

Spoiler alert. Version 3.8.0 of PHREEQC and PhreeqcI at https://github.com/usgs-coupled has a new mean-activity coefficient capability. There is a new keyword MEAN_GAMMA that defines the stoichiometry of each salt and a function MEANG that calculates the mean activity coefficient. Phreeqc.dat, Amm.dat, and pitzer.dat have MEAN_GAMMA data blocks with definitions for most salts suitable to the databases.

Code: [Select]

SOLUTION 1
REACTION 1
NaCl 1
6 mol in 20 steps
USER_GRAPH 1
10 GRAPH_X RXN
20 GRAPH_Y MEANG("NaCl")
END


[ammonge]

Many thanks, Dr. Parkhurst.

Before your response I was getting the activity coefficients from succesive runs of PHREEQC each for a certain NaCl concentration, so the script is most helpfult. I had tried to implement a reaction to do the job but to no avail. It seem pretty simple in your response, but as a newbie I was a bit lost. What's more important is that I was calculating the mean ion activity coefficients without the square root, just as the product of the gamma_i values, so no wonder why my values were so different.

I will check version 3.8.0 straightaway. Many thanks for the hint.