Drop in efficiency in HILIC column

[annaand]

My question - very general - what can cause a drop in efficiency?

I tried to check everything but maybe I forgot about something else.

Some details of my project are written below:

I have a TSKgel Amide-80 column by Tosoh Bioscience. It is HILIC. I try to develop chromatographic methods for separation of nucleosides. First I wanted to find optimal conditions in HPLC, using standard HPLC mobile phases. I found them, efficiency was very reasonable, resolution as well. Then I wanted to find optimal conditions for separation in EFLC (enhanced fluidity liquid chromatography). Here mobile phases is a mixture of CO2, H2O and MeOH (under pressure ~200Ba). After running it on HILIC column I wasn't able to see any peaks. I flashed the column several times but after coming back to standard HPLC mobile phase the EFFICIENCY dropped down very much. Now it looks that I cannot reproduce my previous results.

[XL]

I would repeat the column performance test recommended condition to confirm the column efficiency. Although 200 bar is not that high by the standard of modern chromatography, it is possible that somehow the column bed was demaged. It could be due to manufacturing quality or/and mis-use.

[Uwe Neue]

If my memory serves me correctly, the TSKGel Amide 80 column is a polymer-based column. Polymer-based columns can shrink and swell, depending on the solvent environment. It may very well be that one of your mobile phase conditions was not suitable for the column.

Can you confirm that this is indeed a polymer-based column?

[annaand]

As is written in the producer website there are 5 um spherical silica particles.

"TSKgel Amide-80 columns are packed with 3 (new), 5, or 10µm spherical silica particles that are covalently bonded with carbamoyl groups..."

http://www.separations.us.tosohbioscien ... ide-80.htm

Amide groups should not react with CO2, right?? Amines do but not amides, right?
Thank you for any clues,
Anna

[Uwe Neue]

You are right - a silica-based column. I do not see a primary reason why there could be a problem with carbon dioxide. What were your optimal HPLC conditions? Also, did you test the column again under the manufacturer's test conditions?

[annaand]

I haven't test the column yet with msnufscturer's condition - that is a good idea.
The HPLC condition were as follows: MeOH/aqueous acetate buffer (pH~5), 80/20 v/v, pressure p~200Ba, room temperature, injection volume = 200 nL, concentration ~2g/L, (uridine, cytidine).
Thanks again.
Anna

[Uwe Neue]

Nothing detrimental or wrong with the test procedure, at least as far as I can see.

[annaand]

Thank you, Uwe. I appreciate your comments a lot. As well as your papers

[HW Mueller]

What about the acidity of the CO2 mobile phase, maybe you got H+ reaction?
Incidentally, what sort of enhancement in fluidity did you get, any figures?

[Uwe Neue]

Hans, please clarify what you mean by "acidity of the CO2 mobile phase" and the "H+ reaction"?

[HW Mueller]

The original post states that the mobile phase consisted of CO2, H2O, and MeOH. Maybe CO2 doesn´t react directly on the column as a Lewis acid, but it certainly will form H+ with water. Maybe the H+ changed (reacted with) the column.

[annaand]

I found some data concerning pH and mobile phases with CO2
D.Wen, S.V.Olesik, Anal. Chem, 72 (2000) 475.
pH measurement of MeOH/H2O/CO2 at 20.7 MPa
MeOH/H2O/CO2 mole ratio=>pH
68.2/30.6/1.2=>4.54
65.1/29.3/5.6=>4.22
61.7/27.7/10.6=>4.38
55.7/25.1/19.2=>4.73

As you can see if the amount of CO2 increases pH does not change significantly. We have (instead of H2O) acetate buffer (pH~5, 20mM). So I assume that pH does not change.

[annaand]

HW Mueller,
A nice comparison of HPLC and EFLC you can find in a paper paper: S.T.Lee, S.V.Olesik, Anal.Chem, 66 (1994) 4498. I don't know how to insert a picture here. Sorry..
[/img]