Conceptual Models > Database selection and modification

pH scales in PHREEQC


I am using PHREEQC to try and model boron in seawater (with the different cation pairs) using the Pitzer database.
Because I work in seawater, there are several ways to define pH.

I read on the reply #1 on this post (,1117.msg3354/topicseen.html#msg3354) that NBS convention is not implemented.  It was also said that the Macinnes convention (which is the default, as said in the PITZER datablock documentation) scales the activity coefficients and I assume that it applies to H+. But how to relate it to other pH scales? I have some measurements in both the total and NBS scale and I wanted to use that as an input to PHREEQC.

I found an article by Nir et al. 2015 (10.1016/j.epsl.2015.01.006) that use PHREEQC to do pretty much what I want to do. They mention in the caption of their figure 2 that the "Macinnes activity scale" is about 0.19 higher than the Total pH scale. I wanted to confirm this by asking the community and make sure that I can interpret my data correctly.

Thank you all for your time!

Because individual activities are unmeasurable, there is not a correct answer. I certainly don't have an answer for you. I can tell you what happens in PHREEQC with pitzer.dat. The activity of H+ is fixed at the the value assigned by the specified pH. Raw individual-ion activity coefficients are calculated by the Guggenheim/Pitzer equations, which are only meaningful in neutral combinations for mean-activity coefficients. Finally, when using the McInnes convention, a KCl solution of the same ionic strength as the target solution is used to calculate the the mean activity coefficient of KCl. K+ and Cl- are assigned individual-ion activity coefficients equal to the square root of the mean activity coefficient (calculated from the KCl solution). The K+ and Cl- activity coefficients are used to correct the raw activity coefficients to individual ion coefficients.

In detail, the activity of H+ is used unchanged in the calculations, and the activity coefficient of H+ only affects the calculated molality of H+. So, whatever pH is entered, the corresponding activity of H+ is used in all of the mass action equations of the model. But, ultimately, I do not know what value you should use for pH.

If you are really persistent, you could look at the solutions used for the NBS scale and calculate the pH from pitzer.dat for comparison.


[0] Message Index

Go to full version