Measuring pH in 95% DMSO solutions?

[David F. Spencer]

In article <20021030163433.3406.qmail at ww02.hostica.com>, jsp at bio.auc.dk 
wrote:

> Hi there,
> 
> 
> 
> I'm trying to make a 95% DMSO, 4.5% H2O, 0.5% CH2ClCOOH solution at pH 5, 
> by adjusting pH with HCl.
> 
> 
> 
> Im using a standard pH-meter (Hamilton minitrode), but I can't seem to 
> get stable readings due to the low water activity. If I clean the 
> electrode and re-introduce it into the DMSO solution I get a completly 
> different reading. Does anyone have a good suggestion how to accurately 
> measure the pH?
> 

There are many layers to the problem presented here.

First, chloroacetic acid is a fairly strong acid (pKa 2.85) so the fact 
that you have been adding HCl to lower the "pH" to 5 should be the first 
sign that this is not a simple problem. The normal strategy here would 
be to make up a 20X stock of chloroacetic acid (because you have room 
for only 5ml per 100ml) with the pH adjusted to 5 (with say NaOH) and 
add that to the DMSO. However, that would not give a pH of 5 when added 
to the DMSO, assuming that you could measure the pH of this solution 
with a conventional pH meter, and also assuming that the whole concept 
of pH is applicable to what is essentially a non aqueous solution. Note 
that even in aqueous solution adjusting chloroacetic acid to a pH of 5 
would be tricky because that is more than 2 pH units above the pKa so 
less than 1% of the chloroacetic would be free acid.

The pK of acids (and bases) is a function of several factors the simple 
Henderson-Hasselbach equation does not account for, such as ionic 
strength and the dielectric constant of the solution. The dielectric 
constant for DMSO (45) is considerably less than that of water (78.5 at 
25C) and that factor shifts the observed pKa of a carboxylic acid up 
(i.e., towards 7). There are equations to calculate this more properly 
but they are not trivial to work with. The ionic strength factor is 
another can of worms.

You would predict that in 95% DMSO choroacetic acid would be poorly 
ionized and so trying to adjust the pH down to 5 with acid would not be 
useful. The analogy would be trying to adjust the pH of say a NaCl 
solution to 5 or 7, whatever. You can only attain a stable pH if one of 
the solutes in a solution has a pK in the general vicinity of the 
desired end point. With the chloroacetic acid in DMSO example, if the 
observed pKa for the acid has risen to say 6 to 7, there is little 
buffering capacity at pH 5 (if that even means anything in this 
situation). If the chloroacetic acid is essentially unionized in DMSO 
the problem is even more fundamental and so buffering is not possible.

You did not explain what you were trying to do or why but I assume this 
can't be an established protocol. You might get better feedback here if 
you elaborated more on the specific goals. I assume that you know that 
it is advisable (before dispensing) to purge the DMSO stock bottle (pure 
liquid) with a gas such as nitrogen to clear out the dimethyl sulphide, 
a breakdown product of DMSO.

Dave

-- 
David F. Spencer, PhD
Dept. of Biochemistry and Molecular Biology
Dalhousie University
Halifax, Nova Scotia, Canada

DSpencer at Dal.CA