measurment of MEOH/aqueous mobile phase pH

[pipettemonkey]

If one wishes to measure the actual pH of the mobile phase as it passes through the detector (organic + buffer), will a pH meter calibrated in standard aqueous buffer be accurate? If a (digital) pH meter could be re-calibrated in MEOH/buffer? Would this work, in principle? (this has been suggested, )

With the standard lab pH meter, is it possible to calibrate it in meoh, H2O without ultimately having to take apart the pH meter?

This business of non-aqueous pH measurement has been the focus of several papers. I think data exists for some commonly used buffers in varied MEOH fractions. I wish to know the pH values as a function of methanol fraction, (across the pH range) in multi-component buffers (acetate, phosphate, citric acid, for instance).

[Uwe Neue]

You can measure the pH in the presence of the organic solvent after calibrating your pH meter in water. This is fine and has been advocated in several publications by Bosch and Roses. This is simpler than the other approach.

[Mark Tracy]

There is only one small problem. The reference electrode has a different response to MeOH/H2O than to plain water, and different types of references have different responses to MeOH/H2O. This difference is a constant for any given electrode chemistry and MeOH/H2O ratio. As long as you use a pH combination electrode of the same type, you can transfer your data between systems. It need not be the same manufacturer, just the same chemistry, for example Ag|AgCl|4M KCl. Fortunately, for most purposes these offsets are minor.

[Bill Tindall]

Mark, I am bewildered by your response. The reference electrode of a "pH electrode" never sees the solvent being measured for it is isolated from this solvent by the fill solution and frit/plug/fiber or whatever the junction is made from. The junction certainly sees the solvent being measured and there can be some junction potential issues to deal with, but the sample solution can have no effect on the reference potential.

[Bill Tindall]

1. The glass electrode responds nearly ideally in many solvents-acetic acid, alcohols and glycols, THF, pyridine, etc- as well as mixtures of these solvents and water. Hence, comparitive measurements of pH can reliably be made in these solvents and solvent mixtures, so long as the solvent composition remains constant among the measurments compared. In these cases 60 mV (about) of potential difference between solutions can be interpreted as a 10X (about)difference in hydrogen ion activity. The "abouts" are because the electrode response may not be perfectly ideal, but it will be close.

2. Comparing hydrogen ion activities between solvents of quite different composition is not easy for a host of reasons that are too complex to be described here (but I do consult).

3. If you want to do relaible thermodynamic calculations using pH values measured in these partially aqueous solutions (such as calculate fraction of acids dissociated, or calculate a pKa) then you must either calibrate with a pH standard developed for the solvent (the most reliable approach) or rely on tabulated fudge factors for comparing pH standards between solvents and water.

4. If you are making comparitive measurements for whatever reason, calibrate in water and measure away. Just use a reference electrode designed to be used in nonaqueous solutions (it will have a sleave type junction that minimizes junction potential problems) If you want to do more with the measurements then you may need to do something different.

[Mark Tracy]

Bill,
I was trying not to say too much. I was trying to address the issue of transferring results from one instrument to another. Most all the other things, thermodynamics, etc. transfer easily. Your advise of using a special reference electrode is good because not only does it eliminate clogging, it eliminates the one variable to data transfer. That is junction potential. The bridging solution in many reference electrodes is KCl for the good reason that K+ and Cl- have very similar diffusion and transport rates in aqueous media. That minimizes junction potential and its thermal coefficient, so you can ignore the issue in most applications. In non-aqueous systems, the junction potential is no longer minimized. There are for instance double-junction electrodes where the bridging solution is KNO3, that will behave a bit differently than single-junction electrodes bridged with KCl. As long as you use a reference electrode with a consistent bridging solution, the offset will transfer from one instrument to another.

Fortunately, the effect is small, and the cure is easy.