Dear chromatographers, I desperately need all help I can get here! We are using HP1100 at 200nm. We are validating the method where we have normally got about 250-300 mAU*s response for a normal standard concentration. I changed the lamp, which was not dead yet, to a new one, and we started to get only 100 mAU*s response! I also changed the old lamp back and got the same poor 100 mAU*s response. The lamp changing procedure is quite simple; just replace the lamp and recalibrate. Lamp intensity (energy) values have been ok all the time. Agilent says that this has nothing to do with the lamp and we all of a sudden got high solvent base and loss our response. I know that 200nm is cumbersome to use and I believe something has gone wrong in changing the lamps. Any ideas, what this could be??

Is this a VWD or a DAD ? If DAD, make sure you remove the flowcell prior to running cals or tests. If VWD, it looks at the ref diode. If you run a test injection w/ lamp A and only change to lamp B, and the problem persists, you have a problem somewhere else. Run all the tests available on the detector. Does the holmium test look OK. Being at such a nasty lambda, double check solvents and modifiers...

Thank you! Detector is VWD and I have ran holmium-, intensity-, calibration- and cell tests. Holmium and intensity were well in limits. Calibration showed no deviations (after recalibrating the lamp). Cell test gave a result 1,11. This is an intensity ratio: sample/reference, water in the flow cell. There is no acceptance limits for this test and I don't have a reference in my maintenance log (haven't done this before). High cell test result indicates a dirty flow cell but I have heard this problem usually appears only with buffers (I have an acetonitrile/water/tetrahydrofuran gradient, polycarbonate as a sample matrix). I don't know if this is a normal result or slightly elevated?? Just to be sure I flushed the flow cell with tetrahydrofuran and methanol (this didn't improve the cell test though).
I lost my standard response when I started to change lamps. All the test results looks ok. Of course it is possible that some other problem appeared at the same time...The method is just developed and validated in a different lab. We don't know much of the ruggedness yet. I guess we just have to try different lots of solvents and maybe a new column? Any other ideas??

Could you please explain this?... "we all of a sudden got high solvent base and loss our response" ... I don't know what you are trying to explain with this statement.

Check your response with other analytes? Have they changed also?

We have cleaned flow cells by removing the window and placing it in acid (remove it from the brass housing)

John B,
Problem is we have only one analyte in the method. This is relatively small and broad peak. Peak elutes in a region where we have some solvent front caused by gradient (baseline is not zero). Lots of things absorbs in 200nm and problem might be that small changes in solvent qualities gives us a high baseline to the region of interest in the chromatogram. High solvent front reduces the analyte response which means the method is not sensitive and precise anymore.
At the moment I actually believe there is nothing wrong in our detector/lamp (all tests being passed and ok) and this is a matter of solvent quality. I tried water from a different system but this didn't make any difference. I'm waiting to receive a new batch of acetonitrile to try.

Thanks a lot

Anon,

I think you are on the right track with you last comment--sounds like you're dealing with a method that has the potential to become a real bear. Can you post the specifics of your method and compound...maybe there is a better method out here? Just a thought. Good luck.

You should change something so that your analyte is not in your solvent front. One example might be the starting conditions of your MP gradient.

Anon, John B,

Thanks a lot! This is a validated method and I'm just transferring and re-validating on our site. It's been a challenging task. Considering the difficulty of the matrix and analyte we are happy with the method. I'm sorry I can't give you any details of the method, it's a property of another company. Can't go changing the gradient either. I have few options to play with: temperature / system void volume. I'm trying to use these as much as I can to move my analyte away from the solvent front.

In terms of UV detection at 200nm there seems to be quite a lot of variability in acetonitrile batches. One batch is good another one is bad. Anybody out there having similar experiences? What do you do; order few bottles to try and if it is ok order your warehouse full?? Doesn't sound too robust thing to do but we might need go to that.

Strange, I am just looking, once again, into checking solvents before using in a chromatography. Solvent impurity is certainly existing. One of the more infamous cases: So called certified, very expensive 1-propanol had a strong fluorescence with strong variations among four bottles bought at one time. The vendor refused to take them back (contamination by us could be absolutely excluded, already on basis of the contaminantīs spectrum).
What to do: In isocratic HPLC one can usually ignore it. In the future, though, I donīt want to tolerate sloping baselines, unexpected baseline shifts, and who knows what.

We use ACS N,N-dimethylformamide (DMF) as extraction solvent in several of our assays and competitive product analyses of consumer products. A few years ago we noticed peaks by GCMS in the DMF never before observed, found by running a blank; this DMF was from the same vendor, and a call to supplier assured us that it only needed to meet ACS specs and that the impurities (1% maximum) were not guaranteed to be consistent, so we changed suppliers. My guess is that for solvents such as this that there are just a handful of manufacturers, others just buy from them and repackage, possibly purify; about 15 years ago there was a DMF shortage at most vendors, which led me to believe this.