Agilent ChemStation Integrator

[gpronger]

We use Agilent's Environmental ChemStation for GC work, but find that integrator does not wrok well for more complex chromatograms (particularly when negative peaks occur as is common with ECD detectors). These are fairly typical environmental samples where not much can be done to improve the matrix.

Is there an after market integrator that can be loaded replacing "genie" that works better?

Thanks in Advance.

Sincerely,
Greg Pronger

[AICMM]

gpronger,

Have you considered that it may be a dirty detector? This phenomenon is pretty common with HP ECD's as they get dirty (in my experience.) Could have them sent off to be cleaned. Also, if I understand correctly, you don't really want to integrate the negative peak, you just don't want it messing up your baseline? If you negative peaks are early, you could just put a baseline event in or you could use a baseline back event near the end of the run.

Sorry I could not be more helpful.

Best regards.

[mbicking]

Are you using the GC or GC-MS version of software?

In the GC version, you can add an Integration Event to turn on testing for negative peaks. This might help.

But as the other post mentioned, consider cleaning/baking your detector first.

[gpronger]

Thanks for the responses.

We will periodically send the detectors in for cleaning which helps, but particularly when pushing the sensitivity of the detector, you will experience negative peaks with dirtier samples.

Regarding the baseline events, there is a lot you can do with them (we are using "genie" which is the integrator developed for GC data versus the "RTE" integrator, traditionally used for mass-spec data).

The problem often appears that once the algorithm "decides" where the baseline is, it locks into it pretty strongly. For instance, if you have a typical baseline rise towards the end of the run, the system will often hold the baseline horizontal, dropping down peak areas to this horizontal line versus following the baseline rise. You can then force the baseline up, using a "Baseline Now" or "Baseline Next Valley" timed event, but rather then smoothly following the baseline rise, it is likely to draw a straight line from the start of the rise to the point of the baseline event.

I was hoping that one of the third party vendors had written a "smarter" routine that could be plugged in replacing "genie".

Greg

[mbicking]

I guess I'm not familiar with the specifics of this particular integration algorigthm, but I'll do some checking at my end. If I find anything I will post it here.

With the usual ChemStation integrator, you can select "Advanced" baseline options, which should take care of the baseline wander. In my experience, that option works better about half the time, and it is never clear why.

There are no third-party add-ins for the integrator. The program is not designed in a way that would easily allow that.