Weird Tailing?

[syx]

Dear members,
We found this weird phenomena (I do not know what it is):

The rear-end of the first peak never meet the baseline until it reach the second peak.

The first peak is the analyte:

(pKa1 (25°) 2.5; pKa2 4.0; pKa3 6.7; pKa4 10.1)

The second is unknown.
I have tried mobile phase with different pH (2.0 - 6.5), buffer concentration (0.025M - 0.050M), or adding TEA... but could not found better shape.
Sample concentration is about 10 - 5 ppm in 0.1 N HCl (dissolution medium).
Please give me any advise. Thanks in advance.

Regards,
Siswanto

[AdrianF]

This could be due to a disturbance in equilibrium caused by the 0.1M HCl Try diluting the sample with mobile phase. You could also try dissolution using 0.01M HCl which I have read is just as effective,(Sorry I don't have the reference)

[tom jupille]

This is a "long-shot" (and I'm not enough of an organic chemist to do more than speculate), but is there a chance that your analyte can exist in two forms which equilibrate/interconvert fairly slowly in solution (keto-enol tautomerization? or maybe a sterically hindered rotation around one of the single bonds?)? That can give the sort of thing you're seeing.

A quick test would be to try increasing the temperature and see if that hump get's smaller (or, if you can go sub-ambient, try decreasing the temperature and see if the hump resolves into a clear peak).

[Uwe Neue]

On the right hand side of your molecule is a proline. There is a slow structural interconversion around the proline nitrogen.

This interconversion will speed up, if you increase the temperature. It could also be a function of pH.

[vickig]

Your profile looks like in addition to the main compound you have one (or more) diastereomers. Since your molecule has more than one chiral center your second peak and the hump between may be from diastereomers.

[syx]

This could be due to a disturbance in equilibrium caused by the 0.1M HCl Try diluting the sample with mobile phase. You could also try dissolution using 0.01M HCl which I have read is just as effective,(Sorry I don't have the reference)

It is a compendial dissolution procedure, we could not change the dissolution media.
I had think the same idea that it may come from the solvent. We injected analyte in mobile phase as the solvent and the problem was still appeared.
I will try to increase the temperature and report the result here.

[mardexis]

Hi Siswanto,

It looks to me like a bunch of stuff just sluffing off the column after your peak of interest. If you can I would suggest monitoring at a couple of different wavelengths. If it is tailing or diastereomers then the chromatogram will look pretty much the same though though at a different overall intensity. If you have some interferants coming out behind your peak of interest you will either see more or less of a hump there.

Is this aqueous only? No organic? What wavelength.

[syx]

Hi Siswanto,
Is this aqueous only? No organic? What wavelength.

The mobile phase contains less than 10 part of acetonitrile. The solvent (dissolution medium) for samples is 0.1 N HCl.
Detection wavelength: 215 nm.

[syx]

will be updated...

40 C

The second peak became smaller...

45 C

The second peak was gone, but the main peak was still tailing.

50 C

Tailing...
Is it ok if the temperature is increased higher?
Maybe I need to adjust the mobile phase composition right now...

[Uwe Neue]

Looks exactly like the proline anomer conversion.

[syx]

What should we do to control this conversion?

[syx]

We change the pH from 6.5 to 2.0. We get the 'best' shape when the temperate is set to 60 C:


Tf = 1.67
tR = 1.057'
k' = 6.05
Wb = 0.322
N = 172.254!

We do not have any idea how to increase the plate number...
Any suggestion?

[Uwe Neue]

Looks OK to me. Alternatively, if the packing is suffisicently stable, you can consider raising the temperature even further.

[syx]

Ok, thanks for all... I think I do not want to take the risk by increasing the temperature.

[Uwe Neue]

Another contribution to the low plate number could be extra-column effects, or the way you heat the solvent. Can you give us the column dimensions, flow rate, and the details how you bring the solvent to temperature?

On the other hand, for the purpose of this assay, the plate count is not terribly important.