Is it possible there was a partial clog in the in the rotor slots? The could cause the sample to come out of the injector as a stream rather than a slug of sample. Did the backpressure decrease with the new rotor installation?

Another possibility is a scratch in the rotor that didn't cause a leak, but did provide a place for the sample to reside until it was eventually washed out.

Most rotor seals are inert, but your molecular target is particularly sticky. Some LC's constantly wash the needle and the rotor. In that case always use a strong solvent that is guaranteed to solubilized the chemical.

Dear Micheal, Chromatographic peak shapes can be distorted due to fluidic as well chemical reasons_outside the column_. The initial shape of an injection band would also determine the final peak shape. As Jack stated above, a scratch in the rotor can potentially provide an additional volume which takes some time to completely sweep the whole analyte, this results in tailing. This effect should not affect the retention time per se. A significant increase in retention must be originating from chemical reasons (perhaps residue build up in the rotor system).

You can see my article on peak shapes pg 10 onward. You might find the table useful.

Thanks lot for the answers and for the paper!

I forgot to mention that the method covers some 700 compounds - only those 3 particularly tricky compounds were affected, so I guess there must be some chemical reason. My guess was that the inert seal material has been abraded thus exposing chemically active sites, but I was not able to see any scratches or particles. The injector is washed with a strong solvent.

A general question deriving from this issue: Do you replace spare parts such as the rotor valve in regular intervals or do you use them until the end of their lifetime?

The scratches are often microscopic. Some latest instruments, keep track of injection counts and show the remaining life of rotor seal. Agilent can do an "Autosampler Leak Test" in the diagnostics. It can tell you how much liquid is leaking from the rotor, metering valve etc.

"I wonder if it was really the condition of the old rotor seal that caused the problems - has anyone experienced something similar?" - It probably was. This is an extremely common problem found in all models of HPLC system. Those pesky sticky samples can sometime help point out weaknesses or defects which may have been overlooked over time. Regular maintenance and inspection of the parts can help to greatly reduce their occurrence.

Your Agilent 1290 UHPLC system has many seals of this type, not just the injector rotor seal, and all require regular inspection and replacement when worn. Have you had formal training in performing the routine maintenance services for your HPLC yet? You should. They will help you troubleshoot and solve issues faster, plus discover what types of replacement intervals are normal for your system. Each one is different, but in all cases the parts require regular inspections and replacement when worn or damaged.

HPLC injector rotor seals are NOT inert and are in fact high wear items subject to chemical attack and surface wear under normal conditions. Often made from machined materials such as Vespel, Tefzel or PEEK, each has specific limits and applications which must be followed for long life (note that 'PEEK' is becoming increasingly popular for UHPLC applications, but we must remember that it is highly susceptible to damage from strong acids, THF, dichloromethane, chloroform and many other solvents). No matter the type of seal, all require regular replacement (I teach in my classes that replacement should occur every 6, 12 or 24 months, depending on applications and use). Scratches are the worst and contribute directly to carry-over contamination. Sometimes you can easily observe the scratches directly on the rotor surface connecting the ports together, sometimes not. They should all be visually inspected, but the lack of any scratches or surface damage is not an indication that they are OK. Their surface wears over time (or from chemical attack) resulting in seal leakage which requires replacement. Just as with piston seals, needle seats, multipurpose valve seal and replaceable filter frits, everything wears out and needs regular service. BTW: I wrote an article on Injector Seals which will be out in March of this year (stay tuned).

In the meantime, here are some related articles which may help you.

• "Troubleshooting HPLC Injectors (Manual and Automated) "; https://hplctips.blogspot.com/2013/06/troubleshooting-hplc-injectors-manual.html
• "Carry-Over (Carryover) Contamination in HPLC and LC-MS Systems"; https://hplctips.blogspot.com/2015/02/carry-over-carryover-contamination-in.html
• "HPLC Maintenance & Repair Parts To Have on Hand for HPLC Systems"; https://hplctips.blogspot.com/2013/11/maintenance-parts-to-have-on-hand-for.html