Chromatography Forum

how to start RP-HPLC method development for a drug knowing only its pKa value: means selection of mobile phase and column etc.

Generally speaking, you want your MP's pH to be a couple of units away from an analyte's pKa to prevent peak broadening or splitting.

Other good reason is the retention time; an acid molecule retains longer with a low pH mobil phase; and a bacis drug retains longer in a high pH mobil phase.
the reason is this.
For example; if you have Benzoic Acid (BzCOH); in a low pH mobil phase this molecule do not turns to ionize; so remains neutral and is strongly retained in Stationary phase; as higher the pH as more ionition you get and as lees retention there is.

BzCOH + OH- ↔ BzCO- + H2O

So you can start trying to find the best pH; and testing alterning MeOH, and MeCN as organic modifier.

We have been through this splitting of peaks before. Can´t think of any acid or base which equilibrates slow enough with its conjugate to cause splitting on a chromatographic time scale. A pH mismatch of mobile phase and sample solvent could do this, possibly, or just broadening. Bad peak shapes can also result if your pH allows interaction between SiO- and a cation.
Good luck if you have no idea about the polarity and other aspects of the compounds. You may have to play with RP, HILIC, IC, even NP.

Honestly, knowing only the pKa value will let you take a guess at the identity of any significant ionisable functionality (i.e. it falls within the typical pH range for chromatography) should one exist. In many cases, one might not be able to say with confidence whether that function is actually acidic or basic as there are examples of each type running the other side of 7.

To that end I submit that knowing only the pKa value leaves you in the exact same spot as not knowing the pKa value. Square one.

What might I do at square one? Follow a four-column approach in reversed phase: four complimentary stationary phase packings with a combination of low and high pH mobile phases with differing organic modifiers, as appropriate. And go from there...

As already stated, it is preferable to stay away from the drug's pKa to get a good peak shape and good retention.

But in real life, you often end up with a mobile phase that is close to the pKa. pH is often the most powerful parameter to separate closely related compunds. Closely related compounds with slight differences in pKa (e.g. related peptides) can often only be separated with slight adjustments of the pH around the pKa of the main peak.

Someone mentioned retention time and the desire to increase retention by avoiding the analyte being ionised. Absolutely true, but even more: if the pH is close to the pKa, then the degree to which the analyte is ionised is very pH dependent, so the slightest error in preparation of buffer will cause a big change in retention time. Methods are more stable if pH <> pKa.

Hi Hans,

Would you please explain in more technical terms the following statement?
“A pH mismatch of mobile phase and sample solvent could do this, possibly, or just broadening.â€

When you know pka value for this drugs....this information has great information for your way.....the great choice is a reverse phase for column (C18 more shipper)....for mobile phase whith litle bite below value pH...more details some friends...explain better.... another choice because the value the pKa below value is a HILIC column....

I only want to reply to the pKa/pH question. There is a nice example in Uwe's "Trouble Shooting Guide" (and he himself has pointed it out here frequently): in general separations at pH=pKa are possible without any peak distortion. Only if the buffer capacity of the mobile phase is too low (or not existent at all) will you see peak tailing/broadening.

Example in the guide: ibuprofen at pH 4.4, once with 5mmol/L acetate buffer (no peak tailing) and once with 5mmol/L phosphate (no buffer capacity, peak tailing) . The pKa of ibuprofen is 4.5

Thankyou Sir

Somehow I missed Danko´s questions. Robert already answered one.
On the pH mismatch: In my experience this has given more severe effects than organic modifyer mismatch. For instance, if the pH of the sample has the analyte in a low retentive state it might "wash in" the analyte until the pH of the mobile phase takes over. I have obtained peaks of enormous width if the pH of sample converted part of the stat. phase to the SiO- version and the analyte was cationic.

So, it’s a kind of separation after all (not necessarily in 2 distinctive peaks) – whether it’s the actual sample/analyte in its self, or the sample solvent is the cause. Wouldn’t you agree that it corresponds to some sort of gradient – even though unintended?
Never mind the silanols - because if they weren’t there the ionized part of the analyte would just be in a less retentive state (i.e. will move faster than the protonized ditto.

Best Regards

You are shifting equilibrium position when shifting pH. One has different species (different distribution, hence different retention characteristics) in the column at different times. In principle this is not different than organic modifyer incompatibility.
Regarding ionized silanos, one can severely interfere (shift) an equilibrium, even the equilibration time. But, by the time you exit, lets say two peaks, from the column they will be the same equilibrium mixture. Thus no separation.

Hi Hans,

You are shifting equilibrium position when shifting pH. One has different species (different distribution, hence different retention characteristics) in the column at different times. In principle this is not different than organic modifyer incompatibility.

Exactly!
So, it’s not a question of acid/base equilibration rate (whether it’s fast or slow), as you mentioned in an earlier post (current thread and elsewhere), but local conditions at a particular location in the column.

Otherwise; silanols etc. is a completely different matter and has nothing to do with the current subject.

Best Regards