pKa and Buffering capacity

[ajey]

how to start RP-hplc method development for a drug knowing pKa and which buffer (acetate, phosphate etc) should be given preference? and What is buffering capacity?
Regards,
Ajay

[Uwe Neue]

First, I don't think that the pK of your analyte is terribly relevant for setting up a method, especially since both the pKa of the analyte and the pKa of your buffer will change as you add the organic modifier. However, you can avoid changes in retention with buffer pH by choosing a buffer with a pKa that is +/- 2 pKa units away from the pKa of your analyte. As you avoid changes in retention with the buffer pH, you are also depriving yourself from the most powerful method for changing the selectivity of a separation. This is why I am of the opinion that to worry about the relation between buffer pKa and analyte pKa is misguided.

However, what I do recommend is to use a buffer near its pKa, such as phosphate around pH 7 and acetate around pH 4.5 or ammoniumbicarbonate around pH 10. The reason is that the buffer capacity is maximum at the pKa. Buffer capacity means that if you add a small amount of an acid or base to the buffer, the pH will not change. (Buffer capacity is the inverse of the first derivative of the titration curve: change in concentration / change in pH).

[mohan_2008]

I agree with Uwe.

However, there are certain situations wherein the molecule has multiple pKas' - say 3 and 7.

In situations like this, there is no point in debating about the optimal working pH, trying to keep away from the analyte pKa's etc.

I would go for a intermediate pH say 4.0, and use a HIGH Buffer concentration say about 40-50 mM. (Make sure, % organic doesn't precipitate the Buffer). This way, your selectivity is not compromised and you have adequate buffering to hold the RT.

[syx]

I agree with Uwe.
However, there are certain situations wherein the molecule has multiple pKas' - say 3 and 7.

it will be the problem if the analyte is amphoteric.

[Uwe Neue]

The analyte is less important than a good buffer with good pH control.

Also, for amphoteric analytes, there is often a pH region of pH independent retention. For example for an aliphatic carboxylic acid and an aliphatic amine, there is a pH range around 7 where the analyte is zwitterionic, and the retention is stable with a phosphate buffer.

[syx]

Dr Neue, do you mean that we could use the pH in the mid region between the pKAs for amphoteric substances?

[Uwe Neue]

Yes. Why not?

The retention of the analyte is stable. It may be shorter than under other conditions, but it is stable...

[syx]

thanks for the enlightenment... it should be the solution for our problem when handling amphoteric substances.
could it be applied to a mixture of acidic and basic analytes too?

[Uwe Neue]

Fully ionized acids and fully ionized bases at neutral pH will give stable retention.

[Rafael Chust]

I use an expression that is "a variable that remains constant is no longer a variable", which complies with Uwe previous post: totally ionized compounds give reproducible retention time.

[DM909]

Well, I think it is important to have an idea of your analyte pKa and to select your buffer pH accordingly when developing a method. Ionized species will come out like a shot on RP columns and this can be readily fixed by changing the pH. If we consider an acidic species of pKa 5, working at +2 pKa units will give the ionized species with very low retention time while working at -2 pKa units will give the neutral species with longer retention times. Select a buffer that happens to have its natural pKa close to your target pH. You can find the suggested buffer ranges of many common buffers in many places. I think the instruction manuals that come along with Waters columns actually contain a reference table of this sort.

For someone without much experience who has to develop a method, I think this is an important starting point.

[HW Mueller]

One can not generalize the behavior of ions on RP like that. Two personal examples: Baclofen, a gamma-amino acid retained beautifully on a C-18 at its pI. Pertechnetate (TcO4-) can be chromatographed nicely on a RP column.

[DM909]

Yes, definitely. You can have ionized species chromatographing beautifully on a C18. I doubt TcO4- is a very good illustrative example though, since most people are dealing with organic molecules. On the other hand, an amino acid at its pI has a net charge of zero and is retentive, which is great for RP. Still, you're comparing two very different species.

I agree that generalizations are not laws. However, for someone with limited experience who has to develop a method from scratch, the pKa guideline is useful.

[HW Mueller]

Hardly anybody said that considering the pKa, or fiddling with the pH is useless. For instance, I understand Uwe´s statement such that you don´t necesarily have to shun the ionic form. He has never, to my knowledge, stated that pH adjustments are useless. On the contrary, he has in my memory been pointing out relentlessly that pH is probably the most powerful tool if the molecules are responsive (in equilibrium with H+), etc.

Again: The zwitterionic form of the amino acid is usually its most polar, thus least retentive form (in RP), since it has at least two ions on it. What can be more polar than a + and - on the same molecule?!

Also, why would one want to misinform greenhorns?

Furthermore, I did not compare anything, I mentioned two widely differing ionic species on purpose.