unknown injection peak

[zhensr]

Hi, I need some help here.

We have Waters Prep HPLC. Each time when the injector is switched, there will be a peak showing up. When I started a run, the software showed waiting for injection, after the injection, there is a peak detected right after the switch is made so the diode array UV VIS will register a wrong baseline. During the run, each time I swtiched the injector back and forth, a peak will show up. Most likely it is not contamination peak, I have clean the injector, and UV VIS detector did not show any distinct feature.

Any suggestions and solution?

Thanks for your help.

[tom jupille]

Turning the valve generates a slight pressure pulse, which changes the refractive index a bit. UV detectors all respond somewhat to RI, so what you're seeing is not at all unusual. If possible, delay a bit before zeroing the baseline.

[Chromatographer2010]

Also partially clogged sample loops and/or injection valve ports can increase the pressure fluctuation but typically the pressure fluctuation takes time to propagate through the sytem to the detector. It also typically has a distinctive up/down or down/up characteristic to it. Also if it was a contamination then it would also be delayed a bit from the turning of the valve.

I think the waters PDA detectors actually report zeros for the first few data points when it is performing the auto zero. If there is any drift during this time then the baseline will jump right after it finishes the autozero.