why HPLC baseline spike a lot

[AUlee]

Hi, I met a trouble in my HPLC analysis: the baseline spike a lot, and in the running time it often spikes as high as 1000. When the baseline is up and down, the detector reads "alarm". I ran 50 samples yesterday. At first, the baseline was good enough as before. But it began to spike from the 4th sample, and it lasts through the following 7 samples' running. Then turn back to good and straight shape for next 10 samples. Then spike till all of the rest samples.

Column: Aminex HPX-87P (half year using)
Condition: 0.6ml/min, 85 degree C, P=850 psi (stable)

what could be the problem? I restarted it also today. It still spike from the very beginning.
Thanks for your help and time. Appreciate it very much.
Li

[tom jupille]

Two most probable causes:
- air bubbles
- electronic noise

Try shutting off the pump while still monitoring the baseline. If the spiking continues, it is electronic (perhaps a dirty or corroded connection or a bad ground).

If the spiking goes away, then it is air bubbles. Check to make sure your degasser is functioning properly. Also check for any loose connections or fittings on the intake side of the pump (these can sometimes aspirate a bit of air).

If all else fails, consider putting a capillary restrictor on the outlet of the detector. That will add some back pressure and tend to keep air in solution. Be careful not to use too much or you can cause a leaky cell!

[agilentchina]

It is possible the problem of your column, the solid phase of this type of column is easy to flush away. Please try to change the cloumn.

[DSP007]

Electronic nouse is has a very distinctive look.
This is a sharp narrow emission up and down.

Tom Jupille has postulated the first tests to be carried out.

[tom jupille]

per agilentchina's suggestion: that's another possibility that I have encountered (but not recently!) and forgotten about . A good test for that is to tap on the column with the handle of a screwdriver and see if that generates spikes. If it does, that suggests that you have a bad outlet fitting/frit/screen on the column.

[Alex Buske]

What detector?
How is it connected - analog or digital?
Are the spikes just 1point wide?

[AUlee]

Thanks soooo much for all your reply.

Sorry for my delayed feedback. I was still working on it these days. It seems to me that when I turned off all the system overnight. Then restarted it the second day. It has no spike at all. But when it runs 20 hours or so, probably spike appears.

As to Alex's questions,
Detector: RID-10A, Refractive Index Detector
Connection: digital
spike descriptions: firstly is the baseline swaying at the same time "alarm" reads on the detector. If let it still run with the "alarm" blinkering on detector, the spike will happens unexpectedly. Some of them are as high as 100. The highest is more than 1000. The width usually more than 1 point wide. It has no certain width. Basically, it looks like chromatography of GC/MS (not the mass graph). Do you know what I'm saying?

[AUlee]

Again, thanks a bunch for all the replies.

It seems that we've found the causing problem for this. A parameter in detector, "total energy", goes down to a incorrect scale. This probably caused by sample stuck or anything blocking.

So, we washed the detector following the manual. It works well. NO "alarm" blinkering at all today.

Considering the post might be referred by future members. I simply put the causing problem here. Hopefully, it works for anybody one day.

Regards,
Lee

[tom jupille]

Glad you got it solved!
Thanks for the update!