Trouble with ZIC HILIC

[ed_eddo]

For my diploma thesis I try to set up a method to analyse melamine and cyanuric acid simultaneously in various food samples.

Liquid chromatography was carried out with a Agilent HPLC-System (Series 1200). The LC column is a silica-bonded ZIC-HILIC, 2,1 x 150 mm, 5 micro-meter.

Solvent A is 250 ml H2O, 250 ml ACN and 0.38 g NH4Ac
Solvent B is 25 ml H2O, 475 ml ACN, 25 micro-liter HCOOH

mobile phase gradient
time A B Flow
0.0 0 100 0,4 ml
3.5 0 100 -"-
6.0 35 65 -"-
6.5 35 65 -"-
7.0 75 25 -"-
9.0 75 25 -"-
9.2 0 100 0.6 ml
12.0 0 100 0.6 ml
12.5 0 100 0.4 ml
18.0 0 100 0.4 ml

Mass spectrometry was carried out with a Applied Biosystems (API 4000).

My problem is that I haven't any experience in the work with a ZIC-HILIC column. The retention time for melamine changes from run to run. From 8,35 min at the beginning till 5,49 min after serveral runs (may be 30 runs). After approximately 30 runs, the retention time is stable. The retention time for cyanuric acid is stable from the beginning.

I already tried an other gradient from the US-FDA. But every day the same problems.

Any suggestion? What could be the problem?

[Bryan Evans]

This assay works very well on Unison UK-Amino:
http://www.imtakt.com/TecInfo/TI438E.pdf

The advantage to UK-Amino over HILIC phase is better
retention of cyanuric acid.

In fairness to ZIC HILIC, I would make sure the pumps
on your LC are working properly.

(you should probably use AcOH with ammonium acetate)

[Bruce Hamilton]

Search this site for Melamine and you will find an ardent proponent of the ZIC-HILIC column for melamine analysis ( Einar Pont? ) from Merck Sequant AB. You should be able to email him for advice.

Alternatively, Vlad Orlovsky from SIELC Technologies, also often present on this site, can probably help provide a solution using their columns.

Bruce Hamilton.

[ed_eddo]

Thank you very much! I will email him!

[wan]

Hi Ed,

Is the retention time problem solved for you?

I am experiencing something similar, except that for mine, the retention time shifts more frequently. It shifts from 2.30 to 1.83 from run #1 to #5, then stabalise at ~1.83 (1.81, 1.83, 1.82) from run #5-7, then shifts again to ~1.90 plus and more for run #9 onwards.

I had only started using the LCMSMS system for a month, but the RT is usually stable and does not shift like it did recently.

Do let me know if you have any solutions.

I'll post my conditions tomorrow when I have access to the instrument desktop.

Thanks!

Regards,
Wan

[HW Mueller]

This method has a gradient for org. modifyer, pH, and flow rate? Maybe temperature fluctuations on top?

[danko]

Guys,
The issue is inadequate equilibration. HILIC needs longer equilibration time than RP for instance and especially in gadient mode.
In this particular case I would add at least 3 - 5 min at the end of the run. In addition I'd run at leat 3 - 4 times the gradient (condition column) when new column is taken in use.

Best Regards

[wan]

Hi,

We equlibrate the column for a longer time (60 mins, instead of 5-10 mins used previously) and the retention time is more or less stable at 1.57. So it seems the shifting RT was due to what Danko said.

The weird thing was we had always equlibrate the column for 5-10 mins before and the RT was fine, thus I am not sure why we need a longer time now.

Also we are having the problem of jagged peaks. Below are the conditions we used:

Mobile phase: (A) 10 mM ammonium acetate in water (B) 10mM ammonium acetate in 95/5 ACN/H2O; 75% (B); isocratic
Flow Rate: 0.3mL/min
Column: HILIC 2,1x100 3micron
Ionization mode: ESI

Sample is dissolved in 95:5 ACN:5mM ammonium formate. The reason why we used ammonium formate here is becase we were using it as part of the mobile phase earlier. Will this cause any problem?

We had added 1% formic acid into the buffer before when we were using ammonium formate, but it does not help the peak shape. We had also increase the salt conc from 5 to 10 mM.

The jagged peaks occurs for both samples and standards, so we doubt there is any problem with them.

I hope someone can help as I am very new to LCMSMS. Thanks!

Regards,
Wan

[Bintang]

Dear Wan
If you are experiencing problems with applications using a ZIC-HILIC column please contact us at

support@mercksequant.com

Our support staff will do their best to help you.

[wan]

Hi Bintang,

I'm not using a ZIC-HILIC, but thanks anyway.

Regards,
Wan

[danko]

Hi Wan,

Try replacing the column, keeping the rest constant (unchanged). I assume the problem you described has occurred underway and thus not inherent.
What's the column's chemistry btw?

Best Regards

[wan]

Hi Danko,

I'm using Atlantis HILIC column and sample is an amine.

Today we ran gradient and obtained serious tailing. The peaks also remained jagged. The tailing is not as serious for isocratic, but again, the peaks remained jagged.

We tried changing the ionspray voltage, but that does not help.

Besides changing the column, is there other parameters I can tweak to obtain better peak shape?

Thank you!

Regards,
Wan

[danko]

Did the problem you describe occur underway, or is that the way things have always been?
A chromatogram illustrating the issue would really be helpful.

Best Regards

[wan]

Hi Danko,

I was not the one who run the earlier samples, so I'll check the earlier data and post again. At least from the time I started joining into the project, the peaks had been jagged. However the tailing wasn't that serious before.

I'll post the chromatograms later. Thanks!

Regards,
Wan

[wan]

Sorry, I forgot to add. We're using API5500. Another scientist, who saw raw data obtained using that API instruments, realised that the chromatograms all have jagged peaks for the same analyte.

I am beginning to wonder if it is the nature of the analyte when used with API instruments. Is that even possible?

Regards,
Wan