Dead volume peaks - caused by environmental contamination?

[nicole0907]

Good morning everyone!

I inherited a new (purchased March 2015) Agilent 1260 HPLC system a few months ago, with a DAD UV detector and an RID detector. The system had been previously set up with prep scale silica columns and a heptane mobile phase, running in normal phase mode. When I started using it, I did a few heptane blank injections to ensure I knew how to operate the software, and replaced the three column system with a single analytical scale silica column. After doing this a large dead volume peak started appearing on both the RID and UV detectors. The peak is interfering with early eluting target compounds on the RID (negative peak ~3-4x larger in magnitude than my target analytes), so I have been trying to troubleshoot it for over two months now with no success.

I removed the column, guard column and filter from the system, and have the pump connected directly through the autosampler to the detectors. I have tried different bottles of heptane, as well as hexane, and the peaks persist. I have tried thermostatting everything (including the ALS), and it had no effect. The peak gets larger when I inject a larger volume, and smaller if I inject a smaller volume, and moves with changing flow rate. It is sensitive to wavelength, and responds differently on different UV channels. Actuating the valve in the ALS does nothing to cause a peak on the signal trace.

I had an Agilent service tech out last week, and he checked through the whole pump, and replaced everything in the flow path in the autosampler, and there was still no effect. I put a manual injector in place of the autosampler, and am still seeing the peak when I inject a syringe of mobile phase, although when I turn the valve for subsequent injections, the peak appears to decrease as if it is something dissolved in what I am injecting. I have tried new beakers, new syringes, and even preparing vials in a different lab, and I cannot get rid of this peak. The Agilent service tech thought it was something contaminating my autosampler vials, but I am seeing it without using autosampler vials when injecting manually.

Has anyone seen anything like this before, or have any suggestions as to something I could try next? I'm getting pretty far behind on my project because of this problem, and am at the end of my rope troubleshooting.

Thank you!

[tom jupille]

First of all "t0 noise" is fairly common. Most of the time, we handle it the old-fashioned way: we ignore it. If you deal with pharmaceuticals, the USFDA suggests that any peak you quantitate should have k' > 2 (i.e,. it should elute at 3 times t0 or later). Even if you are not in pharma, that's still good advice.

If your early peaks are eluting with k' < 2, don't keep beating your head against the wall to get rid of the t0 noise, just change the conditions to pull them out to longer retention.