What do you think of this baseline?

[Noser222]

Been having some problems and I'm wondering if it's the water supply or the acetonitrile. I eliminated the buffer salts and still get this. I know the isocratic hold at the beginning makes it worse, but I need that to be able to do a 100 uL injection volume on the method.

The water is 18 megaohm. I don't know the TOC, but it does pass through an activated carbon filter. The Acetonitrile is Acros HPLC Grade.

I should mention I'm at a new workplace and my previous employer had a water system and we monitorled TOC, which was typically 20 ppb. I frequently did long isocratic holds at the beginning of my methods and recall better looking baselines than this.


Column: Restek Ultra Phenyl 4.6 x 100 3 micron
UV @ 254 nm

Solvent D = 2.5% Acetonitrile in Water
Solvent C = Acetonitrile
Flow 2.0 mL/min
Gradient program:

Time Flow %C %D Curve
1 0.01 2.00 0.0 100.0 6
2 3.00 2.00 0.0 100.0 6
3 6.50 2.00 100.0 0.0 6
4 9.00 2.00 100.0 0.0 6
5 9.50 2.00 0.0 100.0 6

[Noser222]

not sure why the image isn't showing, but here's a link to it

http://farm3.static.flickr.com/2470/361 ... c9.jpg?v=0

[Uwe Neue]

Could be the charcoal - err... activated carbon filter. Could also be the phenyl of the phenyl column. What do you see when you run the same experiment with a C18 column?

[Bruce Hamilton]

Initially, just perform the simple test of a set of increasing equilibrium times between injections, to ascertain that the the peaks are from your mobile phase system.

Then you can perform the same increasing equilibrium times with a high water mobile phase, and also a high acetonitrile mobile phase. If they show a difference in unwanted peak intensity, you know the mobile phase component to focus on ( solvent, filters, degasser ).

If they don't, you have to look at cleaning your column and system from the mixing valve forward. Also ensure that you fully flush the disused solvent lines ( if a quaternary pump ) - it's amazing how they can introduce spurious peaks.

Please keep having fun,

Bruce Hamilton

[Noser222]

Could be the charcoal - err... activated carbon filter. Could also be the phenyl of the phenyl column. What do you see when you run the same experiment with a C18 column?

They just installed the filter on the water system and it doesn't seem to make a difference, ie. the baseline looked the same when I used water that did not pass through it. I have not yet done the C18 experiment myself, but baseline issues seem to be plaguing the lab. I've been here a few weeks and that's what I'm told.

Initially, just perform the simple test of a set of increasing equilibrium times between injections, to ascertain that the the peaks are from your mobile phase system.

I did this and they did increase with a longer time.

If they don't, you have to look at cleaning your column and system from the mixing valve forward. Also ensure that you fully flush the disused solvent lines ( if a quaternary pump ) - it's amazing how they can introduce spurious peaks.

Tonight, all of the HPLCs, they are Waters 2695 units, some older than others, are being flushed with 0.1N Nitric Acid through all solvent lines 25% each A,B,C,D.

Is bleeding of phenyl groups something that is a typical problem? I haven't had too much experience with phenyls.

[Uwe Neue]

At 254, you can see bleeding of phenyl columns. How much you see depends on the recent history (first gradient run or second gradient run) quality of the packing, storage conditions, design of the packing etc.

I would do the experiments with a C18, because they will be more conclusive with respect to mobile phase issues.

[DR]

-adding to Uwe's comments-

If it does the same thing w/ a C18 column, try filtering your A-phase destined water with an Empore Extraction Disk.

[Noser222]

I wish I could do the experiment, but I'll be moving to another company soon and in a process development position with no HPLC work.

[tom jupille]

Good luck and great success in your new position. Per my comments "Around the water cooler", please keep coming back!

The present problem looks like a classic case of contamination in the "weak" solvent reservoir. The problem is that it could come from *anywhere*: water system, dirty glassware, the ACN, etc., etc.

The Empore trick is probably the best workaround if your successors can't eliminate the cause.