Sample trapped in loop???

[Baffled Chromatographer]

While troubleshooting some bizarre peak shapes, I recently resorted to a simple "loop injection" or "flow injection" by removing the column from the system.

What I observed I find to be inexplicable: at low % organic, different peptides wash out of the loop at different rates, some with sharp peaks and others very broadly, and one not at all. When I raise the % organic, the peaks sharpen and the non-eluting peptide also elutes.

Now to the particulars:

I'm using a 5 microliter PEEK loop on a 0.25 mm bore injector valve
The flow rate is 50 microliters per minute
Mobile phases: A = 0.1% formic acid, B = ACN/water/formic (90/10/0.1%)

The peptides are leucine-enkephalin, bradykinin, angiotensin I, LHRH, and somatostatin-14.

My detector is +ESI-TOF-MS and I obtain the peaks from extracted ion chromatograms of each peptide.

Performing a flow injection at 5% B, I see the following:

Leucine-enkephalin and bradykinin give sharp peaks with a FWHM of approx 0.2 min. This seems reasonable to me for flow injection of a 5 microliter loop at 50 microliters per minute.

Angiotensin I and LHRH yield broad peaks are over 2 minutes wide with a long tailing.

Somatostatin-14 does not elute.

Upon increasing to 15%B, all peaks become sharper and somatostatin-14 elutes. At 50%B all peptides are uniformly sharp.

I think we can rule out unswept volumes in the system since the leucine-enkephalin and bradykinin both yield sharp peaks.

I have already attempted the following:

ESI spray efficiency effects masquerading as peak shape effects. No change when peptides run together or separately. No change when peptide concentration is reduced 10-fold (1-5 pmol down to 100-500 fmol).

Flow rate. I manually checked the flow rate at TOF inlet. It's the same at all %B.

Gradient issues. I decreased the percentage of ACN in B to 20% and repeatedusing 22.5%B and 67.5%B (which match the actual percentages of ACN used above), I observed the same results.

Changing the loop. I tried a 1 microliter PEEK and a 20 microliter steel loop. Although the peak profiles are different, the same general trend is preserved.

Impurities: the peptide spectra are clean all the way across the band.

The only interpretation that I'm left with is that these peptides bind to either PEEK or stainless steel, but I've never heard of such behavior before.

Can anyone shed some light on this??

[Baffled Chromatographer]

Here are the extracted ion chromatograms that I described in my previous posting.

Again, these are flow injections with no column.

Angiotensin I, LHRH, and Leu5-enkephalin were 20 nM in 0.1% formic acid and somatostatin-14 and bradykinin were 100 nM in 0.1% formic acid. The 5 microliter loop was over-filled with 50 microliters of sample.

[tom jupille]

Off the top of my head, I'd say you're getting hydrophobic binding on the PEEK or possibly to the polymer used in the valve innards. Try changing to a stainless steel loop and see if that changes things.