Help! - Negative Peaks

[Chris]

I am currently running a method that uses Calcium Chloride to produce a complex of the active and then detection by Fluorescence is performed, however over the past few months we have noticed problems with poor sample repeatability and intermittent low standard injections.

The method details are as follows:

Mobile phase: 1% Calcium chloride in 80/20 Methanol/Water
Extraction Solvent: 1% Calcium Chloride in Methanol
Detection: Ex318, Em390nm
1ml/min, 20µl, C18 column
RT of active = 5.0min

So far during the investigation blank solutions have generated positive and negative peaks at 2.7 and 5.0 minutes (i.e. interfering with the component of interest).

Methanol gave a positive peak at 2.7mn and negative at 5.0 (approx 2.5% of nominal)
Mobile phase gave negative at 2.7 positive at 5.0 (approx 0.2%)
Water (2 sources) gave negative at 2.7 and positive at 5.0 (approx 2.5%)

More confusing was that extraction solvent generated positive peaks at 2.7 and 5.0min (approx 8% of nominal) whilst a standard stock diluted into both mobile phase and extraction solvent gave similar responses.

Peaks in the extraction solvent (at approx 2.5%) have been observed on using alternative mobile phase and extraction solvent on another system (but with the same detector).

I am currently looking at different batches of Calcium Chloride but can anyone offer any other suggestions/possible causes.

[Chris]

Last night I inject more extraction solvent

The old extraction solvent repeatably gave positive areas at 2.7 and 5.0 min (over100% of peak height at 5.0min) howevere from another vial this decreased to approximately 10% of peak height.

New extraction solvent using the same calcium chloride gave a postive at 2.5 and a negative at 5.0 (approx 3%), this was consistent for 3 vials.

New extraction solvent with Calcium chloride from another supplier gave a postive peak at 2.7 and 5.0 (again at over 100% of peak height at 5.0) again consisitent over 3 vials.

Not quite sure what is going on now.

[HW Mueller]

What are your column dimensions? Could it be that all this takes place at or near the dead volume? It´s nothing unusual to get some strange peaks in fluorescence there. Though I don´t quite follow all of what you did I suggest to do this: Check whether the peaks are due to fluorescence or scattering. If fluorescence is involved (for instance in the mobile phase) you have to clean up your act. If it is scattering you can´t do much, if anything, to lower it. You can lower the MeOH content and move the analyte back. (It´s possible that lowering the MeOH also lowers possible scattering).
Also, I suspect that there may be some mobile phase unequilibria involved, since it seems things are not repeatable?

[Chris]

Dimensions are 4.6mm x 25cm.

I also tried this on dual detection and at 265nm (Abs max of analyte) this confirmed that the peak at 5.0 is the analyte in the solutions which are supposed to be blank. What confuses me is that one of these blank solutions was prepared from fresh reagents and gave peak areas (both UV and flouresence) at 200% of the normal peak height.

I have also tried a new column to ensure I wasn,t just flushing the active off and this also gave peaks in the balnk solutions.

[HW Mueller]

If I understand you correctly you got carryover from somewhwere. Why don´t you excite your fluorescence at the max wavelength?? You get fluorescence that far away?? Move that analyte to a higher retention time. Check whether you have fluorescence rather than scattering.

[Chris]

Could you explain the light scattering effect further, but would the peak by UV detection indicate that the peak is actually a component (either interfering or active).

I have had the fluoresence detector checked and it seems to be ok, I also tried a new column and the extraction solvent gave the positive peak at the RT of the active again but the chances of solvent contamination with the acive at a concentration of over 100% of the normal peak height seems mininimal.

[HW Mueller]

Since I am not entirely sure of your terminology and of what you did, exactly, I am just giving some general info which might pertain to your problem. The first thing to do when you get a "blank" peak is to check whether it is analyte and, if so, where you carryover source is. In your case I would also move the analyte to higher RT. Also if you get negative peaks in Fluorescence you have a raised baseline, which could be due to scattering (usually Rayleigh) or fluorescence. You distinguish between scattering and fluorescence by running spectra at different excitation wavelength: If the emission spectrum max shifts as does the excitation you have scattering, if the max stays the same (the intensity can change) you have fluorescence. A negative peak can mean that you inject something of lower fluorescence or scattering ability than your mobile phase. If you do fluorescence you better know whether you are dealing with fluorescence or scattering.

[HW Mueller]

I forgot to mention, above, that if you have a raised baseline you can also get a negative peak when you inject a very strongly absorbing (UV) non-fluorescing substance. It reduces the light available for either scattering or fluorescence.
Also, when you run fluorescence spectra, be aware of scattering overtones (at 2x the excitation wavelength, grating monochrometer).

[HW Mueller]

I forgot to mention, above, that if you have a raised baseline you can also get a negative peak when you inject a very strongly absorbing (UV) non-fluorescing substance. It reduces the light available for either scattering or fluorescence.
Also, when you run fluorescence spectra, be aware of scattering overtones (at 2x the excitation wavelength, grating monochrometer).

[Chris Pohl]

Chris,

To elaborate a bit further on the comments above, I would say is that the chances are very high that the calcium chloride you are using contains a fluorescent impurity which elutes at five minutes. I would switch to a new source of calcium chloride (purchase highest-quality reagent available).