HPLC unknown contamination

[saioa]

Hello everyone!
This week I am having a terrible headache with the HPLC since I have a contamination. When I run HPLC grade water as a sample I can observe different peaks in the chromatogram. I use a gradient which starts with 90% HPLC water and 10% MeOH (or ACN, it does not matter) and I raise to 100% MeOH in 45 min, leave it for 3 min and then I have 22 min equilibration time with starting conditions.
Since I know that water could be the cause of this contamination I have changed couple of time the reservoir containing water and I use HPLC grade water.The glassware where I keep the water was carefully washed in a dishwasher, with acetone and HPLC gradient water. I have change the sample (vial) couple of times. I have washed the column over the weekend with ACN another night with MeOH. Nothing has worked. I still get peaks in DAD and one peak in FD detectors. I have also taken the column (C 18) to another HPLC and I still have the same problem.
So it seems that the column is contaminated somehow. Previously I injected food extracts (cooked food components like e.g. casein-linoleic acid were extracted with 1% acetic acid, filtered and injected. Usually mobile phase is a mixture of 0.5% formic acid-MeOH). After the injections the column was always washed. I also would like to mention that I changed the guard cartridge few days before the problems started.
Has anyone any idea of what can be the problem and how to solve it?
Thank you very much in advance!
Saioa

[tom jupille]

Before doing anything else, run the "three blank gradients" test as described on our web site here:
http://www.lcresources.com/resources/TSWiz/hs400.htm

If the size of the contaminant peaks increases with increasing equilibration time, then you have contamination in the "A" reservoir, and the problem is not in your column or guard cartridge.

[saioa]

Thanks Tom for you answer. I really apreciate it.
I have performed the "three blank gradients" and as you suspected peaks in the run after longer equilibration time have larger area (I do not see how to upload the chromatogram, but for example one peak area in FD increased from 648619 (18 min equilibration-the usual) to 939070( 54 min equilibration)).
If this is a contamination from "aqueous" reservoir, how could I wash it? I have already washed it with new HPLC grade water with no success. Would I need to wash it with organic solvent like ACN or MeOH? Any other tips?
Thanks again
Saioa

[HW Mueller]

Probably you have top wash it with a better H2O. You can wash forever if your water is contaminated.

[saioa]

Well, maybe then I should open a new HPLC grade water bottle see if that helps. We have not had problems before with this water...
Thanks,
saioa

[tom jupille]

Possible sources of contamination (some of which you have already considered):
- water
- residual detergent on glassware
- contact with pH electrode
- additives, buffers, salts, etc.

If you can't find (and eliminate) the source, there are a couple of cleanup options:

- If you have a "high-pressure mixing" (two-pump) gradient system, a guard cartridge installed between the A pump and the mixer can strip out the contamination.

- If you have a "low-pressure mixing" (one-pump) gradient system, you can use a solid-phase extraction disk as suggested here:
viewtopic.php?t=130

[saioa]

Based on the hypothesis that the HPLC grade water I was using was somehow contaminated (even if I open a new bottle), I have bought another HPLC grade water:
CHROMASOLV® Plus, for HPLC (Sigma-Aldrich)
and I have performed the test described by Tom. Once more I find different peaks in the chromatogram. It is true that the amount of peaks have decrease a little, but still I find quite many. It is true that when longer equilibration time is used some peaks are a bit larger (not 3 times larger if equilibration time is 3 times longer).
I wonder if apart from the water the contamination could come from somewhere else...

[seamoro]

prepare 6M nitric acid and flush it through the lines of the hplc. That will remove any crap in the HPLC.....problem solved!

[JJG]

If your MPA is 100% HPLC Grade Water, and you are adjusting your starting conditions with 10%MPB (Methanol), then I recommend just taking a new bottle of HPLC grade water and putting your mobile phase line straight into the bottle. This way, you can eliminate glassware as the source of contamination.

If you have any sinker frit, or other filters for line A, you should change those too.

[mbreslav]

I think that the problem on both of your systems is contamination of the pump heads.

[saioa]

Thanks for the answers.
And how would you clean the pump heads?

[Mattias]

It can just as well be the quality of the ACN or MeOH that is not good enough. When you run at 10% ACN in the start, any hydrophobic junk in the ACN will also get trapped on the head of the column. You run a very steep gradient, and if you also collect data at low wavelengths I think you will always see some peaks in the end.

[AMacD]

Hello saioa,

Were you able to resolve your contamination problem? I am having a similar one and it's very frustrating. I'm seeing a peak in my no injection runs near the end of my gradient (at 100% ACN) and I know from looking at the MRM transitions it is a few benzodiazepines in my method (oxazepam, temazepam, and lorazepam) causing this peak. I have run the three blank gradient test and the response increases with increasing equilibration time. My aqueous mobile phase is 0.1% acetic acid and I have tried two different bottles of acetic acid, flushing the line with IPA, trying formic acid, using the other line we use for our aqueous solvents, and preparing the mobile phase from two different sources of water and the peak is still there. I am at a loss. Is there anything that worked for you? Any help you can provide would be greatly appreciated.

[ym3142]

Let's admit that there is NO pure water in this world.



Most of the commercial pure water is received from MilliPore line;


A very pure water deteriorates quickly in storage, and more quickly after it is open.



did anyone try a method with no equilibrium step? so the method will be run under non-equilibrium status but so that Aq contamination can be avoid or minimized.

[boydjg]

Hi, I too have been having a problems with contamination. I keep getting two peaks with absorbance maximums at 245 and 250 nm respectively that elute at about 30% ACN/water from C18 column. The problem is not in the column or ALS. I have tried changing A and B solvents about six times, removing the sinker frits, everything I can think of. We have also cleaned the pump head and changed the seals (it's an old workhorse HP 1050). And we ran a liter of IPA/MeOH/water 1:1:1 through the system over the weekend.

Running solvent with no column, I can see the 245 and 250 peaks when I shift from 100% water to 100% ACN. It runs at the solvent front and it's very strong absorbance. But there is nothing in line, no column, no injector, so where is the chromophore accumulating? If I run 50/50, I can see some aborbance at 250 nm in the PDA scan. If I zero balance the detector at 100% ACN the absorbances I see at 100% water are: 220 nm = 0 mAu, 254 nm = 8 mAu, 280 nm = 8 mAu.

I suppost after all this it can still be the water but I have tried milliQ water (18.2 MOhm) fresh off of the purification system, as well as two bottles of freshly opened HPLC water. I just can't track it.

I will try the nitric acid wash next.

FYI, I don't run exquisitly sensitive experiments so these are not small peaks, but huge ones. They just appeared out of nowhere about a week ago. Any newsuggestions are welcome.