Post-preventive maintenance visit peak shape problems

jbarr

<p><span>Hi,</span></p><p></p><p><span>I was wondering if somebody could help us out with this rather tricky problem we are currently faced with.</span></p><p></p><p><span>We received around a month ago 2 yearly PM visits for our aquity UPLC systems. All seemed to have gone well; BSM pressures and stability were normal, whilst our test mix ( a mixture of small peptides - we now recognise that a more diverse test mix would be better) was spot on compared to what we saw before the PMs. There then followed a long series of experiments profiling metabolites in culture medium samples using HSS T3 columns - again, the experiments went well; we had no reason to believe that we'd have problems when switching to one of our prefered configurations for serum extract (MeOH protein precipitated) profiling using BEH C8 1.0*100 mm 1.7 um columns. We were then quite surprised to see such poor chromatography for the serum extracts - see attched file showing the approprate retention time region; the upper chromatogram shows what we had before the problem occured, the lower chromatogram shows what we have now. In general, peak shape is poor, with the more-retained lipidic region being severely affected. Also, subsequent injections show a huge, broad carry over peak bang in the middle of the chromatogram - this peak simply wasn't present before. We've begun to attempt to troubleshoot the problem; we've cleaned the whole system with an IPA / MEOH / ACN / FA mix, and prepared new mobile phases, to no avail so far. We've confirmed that the samples are not to blame, being able to obtain good chromatography on a third instrument. We're trying to elimiminate different effects in a logical manner, but will soon be running out of ideas.</span></p><p></p><p><span>It's almost as if something is sticking to certain compounds before they reach the column. </span></p><p></p><p><span>I was wondering if anybody has seen a similar problem on their system?</span></p><p></p><p><span>The serum extract chromatographic conditions are:</span></p><p></p><p><span>Column: 1.0*100 mm AQUITY UPLC BEH C8 1.7um</span></p><p><span>Mobile phase: A = Water + 0.05%FA, B = ACN + 0.05%FA, weak wash = H2O, stromg wash = ACN</span></p><p><span>Gradient: Hold 0% B 1 min, increase to 50% over further min, reach 100% B at 7 min and then hold for 5 min</span></p><p><span>Loop: 2uL</span></p><p><span>Injection: 1 uL partial loop with needle overfill</span></p><p><span>Weak wash = 600 uL</span></p><p><span>Strong Wash = 200 uL</span></p><p><span>Detection = ESI +ve ion mode</span></p><p></p><p><span>Thanks in advance for your help.</span></p><p></p><p><span>Jonathan</span></p>

lizh

Hello:

THis needs your field Service guys back. It feels like there is a bad connection - as I am an instrment person, so this could be a mistake, I will float around the team here, but any chance we could see the same print out with counts rather than % on the y-axis. That would be very helpful.

Liz

jbarr

Hi:

Thanks once again for your quick reply. This problem has now been traced to the columns that were being used (they were brand new); we either receieved a dodgy batch or the columns were inadvertently exposed to a pre-system clean up solution that we tend to use between runs.

Regards,

Jonathan

dougm

Hi Jonathan,

Could you please expand on your 'pre-system clean up solution?' The reason I ask is that I am not aware of any 'dodgy' batches of UPLC columns - at least based upon current customer returns/complaints.

Thanks!!

Regards,

--Doug

jbarr

Hi Doug.....sorry, shouldn't have blamed the columns.....I'm sure we were at fault! The cleaning solution is (or should have been) IPA / ACN / MEOH / H2O + 2% Formic acid, except seal wash which is left in H2O; 50 full-loop injections are performed with no column present. The experiment mobile phase is then replaced and the column, and column pre-filters, are fitted - presumig that the columns were ok, then my only explanation for our observations is that the experiment mobile phase was not left sufficient time to purge before the new columns were fitted.....or, maybe some other kind of mobile phase prep error.....we're always trying to improve our protocols, but it's impossible to cover all mistakes / eventualities.