Chromatography Forum

DR, I see what you're saying, but I would argue that the information is consistent with contamination in the A solvent (not necessarily in the aqueous part of the A solvent). I can easily imagine a non-polar contaminant in the ACN which would be totally retained at 30% and elute at, say, 50%; in other words, I don't think we can exonerate the ACN.

Either way, though, the Empore filtration is woth a shot.

agreed - though experience forces me to glower preferentially at the water.

No argument there; it *is* the most likely suspect.

longer run times at 30% produce even larger mystery peaks.

This result was not confirmed by ejharrels. In the initial description the the peak is absent with a clean column then the peak size/area increases with increasing number of injections and eventually stabilizes.

I asked if different re-equilibration times had been run and, if so, what the results were but no details were given.

It's straight forward to determine if/which organic modifier is giving the peak by running one or the other (not a mixture). If the peak intensity does not change when using methanol instead of acetonitrile or vice versa then the water is the source. The peak elution time will be different with different organic modifier but it can still be tracked, i.e. peak/no peak, same area, same UV spectrum, etc.

However, the increase in peak size with increasing number of injections indicates perhaps something else is going on.

I guess I rushed my reading a bit...
Yes - this change over sequential injections on a given column could be pointing to the build up of something found in the samples. Better sample cleanup (going from a 0.45µ to a 0.22µ syringe filter) or regular reverse flushing of the column may help (though reverse flushing carries a modest risk as a function of the different porosities found in inlet vs. outlet frits in some column lines these days).

Yes - this change over sequential injections on a given column could be pointing to the build up of something found in the samples. Better sample cleanup (going from a 0.45µ to a 0.22µ syringe filter) or regular reverse flushing of the column may help (though reverse flushing carries a modest risk as a function of the different porosities found in inlet vs. outlet frits in some column lines these days).

But the peak is still present and growing even when running a sequence of non-injections (I do this by leaving the sample location blank when programming the sequence). I should receive the empore sample pack today...I'll update as soon as I have results

215nm is really a tough wavelength, especially when running gradient elution - almost everything can show up. At 215nm, your method can hardly be robust and reliable evevn if you can identify the source of the issue this time. Instead of pin point to the exact culprit, maybe try a longer UV wavelength? Have you collected whole UV spectrum of the "ghost"? ...

The results are in!!!

After receiving my empore disks I prepared two batches of mobile phase. The first batch was prepared using only the buffer and then filtered using the C18 empore disk. The second batch was a mixture of buffer and 22.5% ACN (the starting conditions of my gradient). Each batch was then run on the same instrument using the same column. However the gradient was adjusted for the second batch to compensate for the ACN in Mobile Phase A to keep the total amount of ACN throughout the run identical to the original gradient.

The run using the first batch (not premixed) behaved the same as outlined in my previous posts. However, the run using the premixed batch looked much better. By the end of my sequence the contaminant peak had grown slightly but was still too small to integrate accurately and was smaller than my method LOQ (thus it will pass typical non-interference criteria for method validation)! My next course of action will be to run a much longer sequence to identify when the peak does grow large enough to not satisfy non-interference criteria...I can then provide an instruction in the final method that sequences should be limited to that number of injections. I anticipate it will be over 30 injections which is well over the usual sequence length utilized by our QC department.

Thanks everyone for your guidance...I can't think of a better way to start my weekend

Glad to hear it.

And thanks for the update. It's the way we all learn from one anothers' experiences!