Shifting Standard Peaks (GPC)

[cjwaine]

Hello all,

I have spent the last few days trying to solve a number of (possibly unrelated) problems with our GPC [Agilent 1200 Series pump and autosampler, Phenomenex column, Wyatt detectors]. Our Wyatt engineer is satisfied that the detectors and analysis software are working well, and the chromatography obtained looks clean, with good peak shapes.

However, the peaks themselves are not at the expected retention time. For example, peaks for a BSA standard (MW ~70 kDa) have been seen before eluting at around 8 minutes. I am know seeing them at around 11 minutes. Similarly, the main polymer with which our research group works (MW ~10-12 kDa), which will normally elute at around 11 minutes, isn't being seen at all; the injections of this polymer consistently cause a visible disturbance to the injection (solvent) peak at around 16 minutes.

If the pores on the GPC column were becoming blocked, I would expect my peaks to shift earlier rather than later. Given that the reverse is true, the best I can hypothesise is that there is some problem with the autosampler/injection mechanism.

Any advice or suggestions would be very gratefully received - I'm fast approaching the end of my tether on this one!

Many thanks,


Chris.

[unmgvar]

have you checked that the flow is good and correct?

[cjwaine]

Yes, the pump has been recently serviced and is working well.

I'm seeing similar problems when running drug standards with a C18 RP-HPLC column, so I'm pretty convinced now that the problem lies in the autosampler.

[ivanvins]

It seems you are dealing with proteins - adsorption is the probably cause for retention time increase besides the changed flow rate. Check the elution of some well behaving low molecular weight compound (urea). Your column may deteriorated to show some active sites (exposed silanol groups on hydrophilized silica).

Autosampler may cause bad synchro with peak start - all peaks will be shifted for the same time. Or there could be a leak causing decreased flow rate - entire chromatogram will be stretched. Adsorption will cause selective shifts of some peaks only.

[remesquaddie]

have you checked that the flow is good and correct?

I have spent many years troubleshooting GPC systems, and the # 1 reason for RT changes has been FLOW RATE issues.
Trust nothing but your own flow test on a graduated cylinder, even after a pump PM. I always recommend a flow test every day first thing in the morning before running samples.
Good luck

[nschwartz]

Depending on which column from Phenomenex you have, it may be solely polymer based and contain no silica. I agree with the previous post- flow rate is the number one thing. Measure it to make sure.