All,
I am currently using an HPLC method which employs an ACN/Water gradient (%A= 90:10:0.1; Water:ACN:HoAc;%B= 90:10:0.1 ACN:Water:HoAc); 40%B to 100%B from 0 to 30 min. with re-equilibration, wavelength = 260nm, at 1ml/min., and Col Temp = 40C.
The problem is that I have detected an unid "hump" (@ 137 mVs area response) in my chromatogram at about 20 min into my 40 min run. I've have been using the same HPLC system (HP 1100 w/in-line degasser), HPLC column (YMC Pro-C18;150x4.6mm; 3um),and acetic acid, but different lots of ACN from the same vendor. I thought I solved the problem when I changed the lot# of ACN, but the problem still exists. My question is could it be contamination from the glassware, or something else. Your input would be greatly appreciated.

It would be nice to see an actual chromatogram to tell, but at least two possibilities come to mind:

1. A late-eluting peak from a previous injection. Check out John Dolan's "LC Troubleshooting" column in the March issue of LC/GC for a good example of what this can look like. A good diagnostic for this is to do a couple of "blank" or "dummy" (no injection) gradients to see if the problem goes away.

2. Refractive index mismatch between the A and B solvents. This is usually a fairly broad hump and is more pronounced at *longer* wavelengths. If you have another detector available, you might try swapping detectors and see what happens. Some models are inherently better than others, but the problem can be exacerbated on *any* detector if the cell is slightly out of alignment with the optical bench.


-- Tom Jupille / LC Resources Inc.

Here's my 2 cents worth:
1) If you change your gradient profile, does the peak shift?
2) What about injections (blank/dummy) running 100% A or 100% B.

you can change the lines of your solvents may be a pump is not pulling and mixing properly from those lines.Try this some times it helped me out.

We've been having similar problems with a method which is CH3CN:0.05M NH4H2PO4, initial conditions 35:65 increasing to 70:30 over 15min gradient then held for 15mins, 239nm, 1.5ml/min, 25cm x 0.46cm spherisorb ODS2 column. it varies from a 'peak' to humps around the 15-20 min mark!! We thought it's probably a combination of 'contaminants' from the buffer salt/glassware AND water !! I've used a prep guard column to flush the aqueous part to gain some success. However, there is still interference around the 15-19min mark ( max 1mV)even starting a run without injection. This may still be due to contaminants somewhere in the glassware/tubing, flush solution (Gilson injector used). Any further ideas would be much appreciated !!

I concur with Mr J. above, but if not, here's my laundry list:

1. Mobile phase: Could it be a contaminant in your water or HOAc? If the peak size varies with re-equilibration time (larger peak w/longer re-equilibration time), is independent of sample type, and independent of the column used, a component in your mobile phase is a likely culprit.. Is your water stored for any length of time in contact with plastic, like in a carbuoy? In my own work, I've found this to be the most common cause of oddball peaks - and in extreme cases, irreproducible chromatography and non-linear detector response.

2. System hardware: Have you thoroughly washed your system (say, neat MeCN/50C/60 min @ 1mL/min in this case)? Are your columns and instruments dedicated to one method and/or user or are there multiple users and multiple methods per instrument? Incompatabilities between past sample preparations and mobile phases may be problematic if the instrument isn't rinsed out with a mutually compatible solvent between each different method set-up. How old are the frits in your autosampler if you've got one? How clean is your needle/syringe/loop?

3. Column: Is the hump independent of your column - does it occur EXACTLY the same way (RT/size/shape) even on a brand new column of the same type?

4. Sample: Is the hump independent of sample type? Does it occur in standards as well as samples and blanks? What solvent are these made up in? (I'm assuming phase A, but if not, I've seen that have a HUGE impact on chromatography - often dependent on injection volume).

5. Wierd stuff (we're getting into Act of God territory): How's the quality of your sparge gas? What kind of environment is your machine in? Dumb question: Are you filtering your MP? If so, are you sure the filters are both compatible with your MP and uncontaminated? Is your filtration apparatus clean (I assume so...)? Is your vacuum pump belching noxious fumes? I've seen 'em do that and contaminate MP, albeit rarely.

I've only very rarely traced spurious peaks to any solvent fresh from the bottle - assuming, of course a freshly opened bottle of HPLC grade or better solvent.

Good luck and let me know what you find - I'm curious.


Chris

It seems that the unidentified “hump” is from the particular lot of Glacial Acetic acid that I have been using. I conducted several experiments in which I extended my long organic hold with the following HPLC conditions - at 260nm; C18 column; 100 X 4.6mm id;3um; 1mL/min flow; %A (90:10:0.1; Water:ACN:acetic acid), %B (90:10:0.1; ACN:water:acetic acid), with initial %B at 40%, to 100%B at 30 min with a 7 min %B hold (instead of 2 min. %B hold), and then back to initial conditions. There seems to be an unid pk within my chromatogram , in the blk injections at @34 min (which was the “hump” which was originally carrying over to my next injection).
The question I still have is, can it be a by-product from the manufacture of acetic acid, or decomposition product of acetic acid? Note: Acetic Acid (with unid impurity; originally opened 8/18/97) has been stored at RT in enclosed cabinet.
Conducted the same exact experiment using rival manufacturer (also note: brand new bottle; opened 8/16/99) and unid peak is not present. Your help would be greatly appreciated.