I am analysing a mixture of ether derivatives of Fructose using DRI detector. Chromatographic conditions are
Column: Intersil ODS, 5um, 250 X 4.6 mm
Mobile phase: ACN : Water (60:40)
Flow rate: 1mL/min
Column temp.: 30C
DRI temp: 50C
HPLC: HP1100

Sample was prepared in mobile phase.A negative peak is appearing in about 6 minutes. The size of the negative peak increases with the injection volume. When no injection injection is made the negative peak does not appear. Retention time of the negative peak varies with change in mobile phase composition.

I could not figure out the reason for this negative peak. How can I eliminate this negative peak?

What is your sample dissolved in? Don't forget. DRI will see everything, even small differences in the composition of the sample solvent and the mobile face.

Sample was dissolved in the mobile phase.It was taken from the mobile phase reservoir, in order to avoid the difference in the composition of the sample solvent and the mobile phase.

Does the negative peak appears when you are doing a blanck injection (just by switching the valve from load to inject)?

No, the negative peak did not appear.

Do you see this negative peak in a standard, prepared from highly purified material, or only in "real" sample? Where does that sample come from, what can be in there besides your derivatives?

I assume that you see your fructose peak(s), and that this second negative peak is simply unusual, because it is negative. Don't forget, that the differential refractometer measures differences in the refractive index. Therefore a negative peak is just an unknown peak, which could be your reagent, part of your sample, the solvent in which the sample is dissolved - anything.

Check the valve that isolates the reference chamber on your detector. If it is not closing or only partially closing, you can get a negative peak. This is your analyte going through the detector for a second time.

I was also under the impression with DRI that the column and detetor temperatures should be similar. Could the 20 degree difference be the cause?

Differences in temperature can cause drift, but not a peak.

This negative peak is appearing not only when standard or sample is injected but even when only the mobile phase (sample solvent) is injected. As far as same mobile phase is used the negative peak is very reproducible.

We tried with same column and detector temperature as well. Still it is appearing.

It is behaving like a peak in reverse phase chromatography. When the proportion of water is increased in the mobile phase the retention time of this negative peak increases.

Do you get the peak when you go through the mechanics of injection without injecting anything?

Do you have an on-line degassing unit in your LC? If so, you are seeing an air peak. Inject air onto the LC (empty vial) and see if the peak becomes large. If so, prepare new mobile phase and turn off the degasser.

Yes, it is air peak. Once the on-line degasser was off the negative peak vanished.

Thank you for all the suggesions.

How can a degasser produce an air peak? Itīs easy to understand a shift in baseline, but a peak? I am seriously considering to get a degasser, are there some negative aspects of these which are not well known?

When the degasser is removing air from the mobile phase, dissolved gas in the sample will show up as a peak. Turn degasser off (and vent to atmosheric, of course)and and now the mobile phase and sample are at same gas saturation- No peak. I have seen this on many many methods at lowish UV wavelengths (205-220 nm). I've seen a many-millivolt drop in baseline response when a degasser was switched on. Of course, the dissolved gas peak invariably elutes with your active!

Allright, so there is nothing unusual about this type of degasser.