Chromatography Forum

http://www.chromforum.org/viewtopic.php?t=2617

Best Regards

Thanks Danko fo the tip, here's the profile of an injection of USP grade water for injection: The 1st signal at the end of the column volume is normal, it is is the 2nd bump (marked ???) that confuses me, pity that I have to extend the run to almost 50 min instead of 30 min. Any comments?

This is very much a longshot and still requires that the column has reversed-phase properties:

It looks exactly the same as when did SEC on a system that someone had used for m-cresol and phenol. Those substances has huge UV absorbance, and even the slightest carry-over from seals and such resulted in extra peaks. But again, it has probably nothing to do with your situation.

Thanks Miro for the chromatogram. It has definetively nothing to do with RP interaction on your SEC column. It might be possible to have some IEX interaction on a silica based SEC column, but not RP. It is in your case a system peak, most likly linked to the time constant, or better to say data aquisition rate of your detector. What you see is a RI effect coming from your UV detector at 215 nm and, I estimate, a very low data aquisition rate of your detector. Please increase that. In addition please check if your flow rate is constant and if you have constant back pressure.

If you have DAD detector you should be able to see if there is something eluting at 40 minutes, or if it is some other strange phenomenon.

Gerhard, do you think it is a pressure shock that causes a RI effect (some switching going on at the time)? If not then something is retained, why not via RP?
If I remember correctly then miro mentions a phosphate buffer. Is there something else in there, because for a pure phosphate buffer the system at ~22min seems strange to me. Firstly it is very strongly negative, secondly jt has a positive peak on the negative one (incidentally I don´t see any time consant or collection rate influence on that peak, but who knows?).
Also the obvious thing to do in a situation like this is to go through all the motions of a usual injection without injecting anything. Also a spectrum of the peak is imperative. Injection of air could be helpful.

Dear all, in reply to some of your comments:
Gerhard: Yes I have a very stable pressure throughout the whole run (20 bars) as I usually display its scale in my real time window, as for the data acquisition rate, I have to check that and get back to you as I usually leave this setting to the default value, but do you really think this would help?? and would it affect the rest of my profile in any way? Please elaborate more on this specific setting as I honestly don't know about it.. I only know about the detector response which I know could be increased if i have a noise problem, not the case here.

Mattias: No I only have a dual UV detector

HW Mueller: I use a 0.1M sod. phosphate + 0.1M NaCl buffer, I have also tried this buffer without the NaCl for method development purpose but the bump also existed.
As for the signal at 22min, please correct me if I'm wrong, but I believe it is usually sample related where its salts and all the completely retained components (in SEC case passing through all the small pores) and takes a whole column volume to elute, and what appears is usually not a peak but as you see in the profile is like a signal disturbance/deflection... am I correct?
i would like to try the air injection but wouldn't that insert air bubbles in the autosampler (also the system and column in general)? Unfortunately the software I use (LC solution) has no option to make a blank injection that involves the actual movement of the injector... I only have the option to insert a vial position of (-1) but this just runs the method with no injector movement, otherwise, the only option is to inject water as I had did and attached the above profile...

Awaiting your comments

I would follow Mattias's suggestion and add 5% ACN to the MP and see what happens to this peak. I bet it is going to move.

First of all it is unbelievable for me that anybody would be so short sighted as to put apparatus on the market that doesn´t allow to check what the switching does.
Air injection is not a problem if you don´t inject a liter. It is a standard procedure here if there is some strange peak in the chromatogram to inject maybe 5 to 50µL air.
I don´t recall ever seeing a positive peak superimposed on negative peak of H2O unless there was something in the water. usually I used acetone to mark what I call the "smalls" peak, always got a normal positive peak.

What is a dual wavelength detector? You can not take a spectrum?

Miro2009,

You can silulate the injection procedure without injecting anything by performing a 0 (zero) ul injection.

Best Regards