RT Issues with low pH sample and high pH mobile phase?

[indium]

For HPLC, I would imagine that there would be retention time reproducibility issues if the sample is dissolved in a low pH (say pH 2) diluent and the mobile phase is alkaline, say pH 9. Is this true? As a rule of thumb should the sample solvent necessarily be at the same pH as the mobile phase?

[tom jupille]

In our Troubleshooting course, we tell people that the rule of thumb is: "to do the best chromatography, disturb the equilibrium of the system as little as possible". This has to be taken within reason (it implies that you will get the best results if you never inject a sample ), but it does provide a good guideline. Certainly dissolving the sample in the mobile phase is the best general practice.

In the specific case of an acidic sample and a basic mobile phase, you will not necessarily have problems if:

• - the the sample volume is sufficiently low
  - the mobile phase buffer capaciity is sufficiently high.
  - the method is sufficiently robust with respect to pH changes.

The further you move from those requirements, and the greater the difference between the sample and the mobile phase, the more likely you are to have problems.

[Uwe Neue]

I agree 100% with Tom on how to minimize the problem.

However, the expected observation when injecting a sample made up at one pH into a mobile phase at another pH is not so much a problem with the reproducibility of the retention time, but a problem with peak distortion. This actually can go both ways - tailing and fronting, depending if the analyte has more or less retention at the pH of the sample solvent.

[indium]

Actually, I have a compound that is known to degrade under basic conditions, say, a half-life on the order of about a day at high (14) pH. The compound is most stable at low pH. Therefore, for RP-HPLC purity method development purposes I have kept the sample solution at pH ~ 2. The compound has three pKa's which cover the pH range up to about 8 such that no pH in this region is more than 1.5 pH units away from a pKa. I started out column scouting using a simple gradient of dilute acid and acetonitrile and got a decent separation (or so I think since the sample contains unknowns and I only use UV diode array detection). I was wondering if it was worth checking mobile phase's at higher pH or if the risk of on-column sample degradation is worth the investment in time.

[Bill Tindall]

If the pH 9 mobile phase has sufficient buffering capacity and the sample size is small enough, there is no problem with this analysis. acid base reactions are diffusion controlled so the sample will be instantly neutralized when it hits the base. I will bet that a 0.1M buffer at pH 9 and a 5 uL sample will work just fine.

[indium]

Not sure if you answered my last question. When the sample hits the column head the solvent will be neutralized, but as the sample itself passes through the column in a high pH mobile phase will it degrade if it is known the sample is unstable at high pH?

[tom jupille]

If your sample is unstable at high pH, then it certainly will degrade in a high pH mobile phase. The question is "how fast".

• If the degradation is slow compared to the residence time on the column, then the effect will be minor.
  
  If the degradation is very fast compared to the residence time on the column, then you will see either:
  • a. only the degradant(s) or
    b. a single peak representing the average equilibrium mixture of the original compound and the degradant(s).
  Which one depends, of course, on whether the degradation goes to completion or results in an equilibrium mixture.
  
  If the degradation rate is comparable to the residence time on the column, then you can see a variety of problems ranging from split peaks or shoulders through tailing to "blobs".

Note that these problems are different from the more general "insufficient buffer" pH-related peak shape problems referred to by Uwe in his post.

[JNF]

Just inject a freshly prepared sample and a degraded sample a see the difference.

[Einar Ponten]

Sometimes it is possible to add preservatives to the mobile phase.

For example, the problem you describe could be valid for a thiol that undergo metal catalyzed oxididation to a disulfide (approx within a day) and EDTA might inhibit the degradation.