Column equivalency

lmarotta

<p>Hi everybody, I have a little problem:</p><p>Last year I validated a UPLC method for a B-lactam antibiotic assay , using a ACQUITY UPLC BEH Shield RP18 1.7 um 2.1 x 50 mm column. I obtained a USP tailing close to 1.1.</p><p>Last week I purchased another BEH Shield 2.1 x 50mm, but now the tailing factor is 1.6. How can it be explained?</p><p></p><p>This is the method I'm using:</p><p>Mobile phase: phosphate buffer 0.005M pH 5.5 (69%). Acetonitrile (31%)</p><p>Flow rate: 0.8 ml/min</p><p>Column Temp. :40°C</p><p>Wavelenght:220nm</p><p>Inj. Volume:1 uL (full loop mode)</p><p>Weak wash: water (69%) Acetonitrile (31%) [800 uL]</p><p>Strong needle wash:water (25%), Methanol (75%) [400uL]</p><p>For both columns the pressure is about 9000 PSI.</p><p>Sample are dissolved with a water(80%)-acetonitrile(20%) mixture</p><p></p><p>Thanks to everybody</p><p>Lorenzo</p>

peakdalejo

Have you tried reinstalling the column or changing the collet/gold nut/ferrule assembly on the front end of the column connection?  We change columns quite frequently and often end up with poor peakshapes due to bad seating as the collet and ferrule get worn.

We hate this fitting and are eagerly awaiting an alternative that is easier to use!  We've yet to workout how to release the fitting without a lot of trouble and stabbing yourself in the finger with the little spanner thing!

Waters - please redesign something better!!

I can't see any other obvious reason for the change in tailing factor looking at your method..

Its worth a try at least...

lizh

Hi Lorenzo:

I also have a theory worth trying, becasue you are using full loop mode and a very small loop, I think what has happened is that the loop you are using is probably larger than the original loop used to validate the method.

If your current loop has a larger ID you could be injecting a large amount on column, more than you may guess. If you are close to overloading your peak shape could suffer. Try making a 0.5 uL PLNO injection and see what happens. Does peak shape improve with lower volume injected? If this helps a 1 uL out of a 2 uL loop may be the answer.

Let me know,

Liz

lmarotta

HI Liz,

Unfortunately the current loop is still the same I used for the validation. Yesterday I tried to reinstall the old column, and the tailing factor is still good (about 1.15). I’ve tested the two columns with the same mobile phase and the same sample, so it doesn’t look like an instrumental problem.

If you have other suggestion, please let me know.

P.S. I’ve validated my method with full loop mode because using PLNO I could not get a satisfactory area reproducibility (RSD% on six standard injection was >2%).

Thank you very much,

Lorenzo

dougm

Hi Lorenzo,

Could you please tell me how many columns you used during the validation of your method? Did you test multiple batches of packing material (e.g., Method Validation Kit)? Just trying to see if the peak shape differences you describe are normal in terms of column-to-column variability, or did you stumble upon a particular column / batch that gives you higher tailing.

Thanks,

--Doug

lmarotta

Hi Doug,

During the validation I used two columns (lot 0121381151 and lot 0122381861) to test method robustness, and I obtained very similar suitability parameters (tailing factor, retention times, resolution and so on) . The columns that gives bad  peak shape is Lot 0132301661.

I hope this can help,

Thanks a lot,

Lorenzo.