Chromatography Forum

hi everybody;
I received a sample, run LC on it ( grad 5-95 ACN/H2O HCOOH 0.1%) and got 2 peaks (no base line separation). Problem is that it was supposed to be only one. I tried different methods: grad, isoc, formic acid, ammonium ac, methanol, acn, diff columns (all RP); and I couldn't get base line separation.
The UV spectrum of both compounds are almost the same and they have the same MS spectra.
The NMR guy got to the conclusion that they are rotamers (don't ask me how).
I find kind of difficult that a C18 column can distinguish between rotamers (it is not a peak split).
I will appreciate any comments on this

Thanks, and have a great weekend

Sergio

We have seen one example (ok, a few, but they were analogues with a common set of bulky groups) in our labs where rotamers were separated by RP HPLC. The presence of rotamers was confirmed by variable-temperature NMR and sure enough there was a very good correlation at ambient temperatures (we did not perform any high temperature HPLC).

It's not rare, one well-studied system is rapamycin, which has 3:1 ratio of trans/cis amide bond rotamers that are usually well separated by standard reverse phase conditions.

There's heaps of published data on the effect of temperature on the equilibrium. If you know the analyte's structure, you can predict the problem - that may have been what your NMR guy did.

I found that the rapamycin equilibrium was fairly fast, and you could see product rearranging on long HPLC runs. We had a client whose CRO claimed our product wasn't pure ( because of the second rotamer ), but they were using a sixty minute elution, and our 6 minute elution gave a different result.

We had to educate them about rotamers before they would accept the data, despite Wyeth's Rapamune FDA application, and published research papers discussing the issue.

Please keep having fun,

Bruce Hamilton

It is not that uncommon. Peptides with a proline in the structure or drugs like enalapril or captopril also with a prolein structure wil do similar things. You may be able to make one peak out of two by changing the temperature or the pH.

Sugars give two peaks too, but here it is a ring-opening and -closing phenomenon.

Thanks everybody for the reply. I increased the temp and got one peak, so i would say: problem solved.

I didn't think that the stereochemistry of this molecule will affect the speed or the inter-conversion. It's a small, not very complicated molecule; but I can't argue with the experimental data.

Thanks again

have a nice weekend.
Sergio

PS: I'm so keen looking for this weekend that I forgot that yesterday was Thursday

Sergio,
Is it possible to get a look at the structure, or an "anonymised" version of it? We deal with a lot of small molecule stuff and hence don't come across it as often as Uwe and Bruce do; like I mentioned, until a project containing analogues of the following compound we hadn't seen (or identified it, I guess) any separation of rotamers by HPLC.


We found that if you replace the two ethyl groups on the aromatic ring to methyl the effect is not seen in ambient-temperature HPLC or NMR. If you have one ethyl, one methyl then rotamers are observed in NMR but not HPLC.

Looking at the proline-containing Captopril example, I'd be interested to know if the effect goes away with removal of the methyl group on the thiol sidechain.

JA,

Unfortunately I can't show you the structure. I tried to trim it, but in order to show the stereochemistry of interest I have to leave too much.

Thanks again for your input

Sergio