Stability analysis on pH

[#1]

Hello,

I am running a stability study on several lots of a buffer solution based on the measured pH over time.

I noticed that the stability analysis in JMP fits the data to a linear degradation model.

Is this appropriate to use for a pH analysis since the pH scale follows a log base 10 relationship?

Would it be more appropriate to first convert the values to [H3O+] concentration? (Or, is there an alternate model to use in JMP?)

Thank you,

Cam

[#2]

It is desirable to to model the response in the natural scale. You measure pH because it is more meaningful. Because the p function linearizes relationships with hydronium activity! So go ahead and enter pH directly as the response for stability analysis.

[#3]

Welcome to the community!

I'm not sure what it is you want to do?  Do you want to understand how consistent the pH is from lot-to-lot, within lot, over time?  What about measurement error?  Do you expect the pH to change over time or are you expecting it to be relatively constant? How are the "lots" made?

As Mark suggests, there is no reason to not just plot the pH values over time and perhaps a nested study would be appropriate.

There are, of course, multiple ways to perform the analysis in JMP.  The graphical approach (traditional control chart method) is, IMHO, the best, but regardless the graphical technique, you'll still need to interpret the charts from the eyes of the scientist.