disappearing peak

[A66642]

I'm trying to set up a dopamine analysis method using a very old (about 25 years) BAS HPLC. It is built of of model PM-48 pump, LC-4B amperometric detector and Rheodyne 7125 manual injector. At the moment I'm using a 300x39 mm 10 um microBondapak C18 column. It's not going to be the optimal one for my purposes but those columns were available in the lab at the moment. I was hoping to test that the system works otherwise ok before purchasing a new, smaller column.

Mobile phase consists of 15 % MeOH, 0.1 M citrate buffer pH 4.5, 0.6 mM octanesulphonic acid and 1.2 mM EDTA. Flow rate is 1 ml/min We usually have mobile phase constantly circulating in the system (at low 0.2 ml/min flow rate when not in use) so as to cut down the time needed for equilibrating.

The latest problem with this instrument is the weird dopamine peak that comes and goes. I injected dopamine standards and managed to see a peak that seemed to be dopamine (retention time around 10 min). Shape of the peak was good. However, when injected again after some time (10-20 min) the peak had vanished. I suspected that dopamine was eluting this time right after the solvent front but could not be sure. Third injection produced no peaks at all other than the solvent front. The sample was kept in ice at all times so it's very unlikely that it would have degraded between the analyses. I was using a very concentrated standard so if there would have been a peak it would have been clearly visible. I redid this experiment on another day and again experienced an oddly "bouncing" peak. If I just let the instrument run and chartwriter register without injections or if I do a blank injection there are no peaks.

Solvent front peaks always appear 2.4 min after the injection, so the injections seem to be succesful. I also tested the pump and it seems to be working ok. Where is the problem? Injector still failing somehow? Column issue? Detector hiccups?

I really appreciate if somebody could help me with suggestions what could be checked or improved. Thank you!

[mardexis]

The first thing I would do would be to remove the chromatography. It could be so many things going on. Replace the column with a union, preferably a frit also and try a few injections of standard directly to the detector. If you can reliably produce the same height/area injection of standard then at least you know that your injector and detector are working properly. You ought to dilute your standard by a factor of 5 to 10 or so to get the peak height in the ballpark of what you get with your column.

[HW Mueller]

A66642, you didn´t tell us whether the peak you see on a subsequent experiment is from the same standard or from a freshly prepared one.

The experiment suggested by mardexis is done via area only, which should be the same whether column is there or not if the flow rate is the same. Also, there should be no other substances present which are detected and no system peak.

[A66642]

Thanks for your replies!

A66642, you didn´t tell us whether the peak you see on a subsequent experiment is from the same standard or from a freshly prepared one.

When I repeated the experiment on another day I was using a freshly prepared standard.

I'm going to carry out the experiment that mardexis suggested. I'll keep you posted about my findings.

[A66642]

Ok, I tried injecting standard straight into column. Since I assumed that the injection itself would produce a peak of some kind, I first injected water samples to adjust sensitivity and scale. (I only have an old-fashioned chartwriter, so there's no possibility to play with the scale afterwards. That's why I also did not measure peak areas but peak heights.)

The height of water peaks seemed to vary from one injection to another. When I injected dopamine standard (5x more diluted than the one I used in the previous experiments) the peak quite unexpectedly diminished. I'm planning to try again tomorrow with more concentrated standards.

[HW Mueller]

You mean to say that you injected directly into the detector?
Though I think your experiment is less than semi-quantitative it shows that your sample is in a less disturbing solvent than H20 and possibly that analyte has diminished. Also, if it is as you mentioned that a freshly prepared st shows up (with the column), it appears that you indeed have a problem with standard stability.

[A66642]

Yes, I replaced the column with a union and then injected the samples. My standards were actually prepared in Milli-Q water, same stuff as I used for blank injections. Dopamine should keep several hours in aqueous solution when refrigerated and this experiment was carried out within 0.5 h after the standards were prepared.

[A66642]

Ahh... meant to write "straight into detector". What a stupid mistake, sorry!

[HW Mueller]

Now that you noticed how easy it is to make a mistake, it seems to me it might be best for you to forget the "without column" experiment and instead concentrate on keeping your samples stable.
Years ago I also had stability problems with endogenous amines and tried to circumvent that via fluorescence derivatization. The project was scuttled when the medical interest waned, and I don´t remember how far I got. However, I do remember that at least one company brought a HPLC kit on the market, so it seems with a fair amount of discipline it should be possible to get reasonable estimates.

(Experiments without a column can require a lot of tinkering to be meaningful, for instance, it might be possible that you get more air injected with the water than with the sample in water).