Ghost peak during the determination of aldehydes using dnph

[Houtuh]

Hell everybody,

I'm a student from the netherlands, so don't mind my english.
I have a problem:

I´m analysing aldehydes en ketones using dnph cartridges and hplc. For several months i had have normal chromatograms but the last 2 weeks i have a ghostpeak in my chromatogram. I dont know where it is coming form.
I´m using a waters HPLC, pump, dual absorbance detector, autosampler and degasser. The eluens i am using is A: Water/ACN/THF and B: Water/ACN. I´ve tested the eluens by injection but this gave no ghost peaks. I replaced my precolumn, this diden´t work, i even replaced my column, but this also diden´t work.
The weird thing about it is that when im analysing an standard(15 components calibration standard) i get a normal chromatogram with no ghost peaks. So im thinking the problem must be in the cartridges,
But cartridges from another batch also have the same problem.

It is not a peak from previous injections, this i already tested.

Has anybody any idea what to do???

I´m runnig out of options.

Please help!

Thanks in advance

Remco van Houten

[Mark Tracy]

Are these small peaks offset about half a minute from the main peaks? The DNPH derivatives have cis-trans isomers, and the minor isomer is only a few percent. You can see an example here: http://www1.dionex.com/en-us/acclaim_li ... 28434.html I was able to confirm this by LC/MS.

[Houtuh]

Thanks for your reaction,
but that is not the problem. I already had that information but inmy case it is a large ghost peak about 2 to 3 times the sizes of by example formaldehyde. The place of this peak can be a little bit different but it always around the retention time between formaldehyde,acetaldehyde and acetone.

Has anybody any idea??

[juddc]

If you only have the problem with your samples and not your standards, then it sounds to me like the issue may indeed be sample related. If you have a highly retained species that reacts witht he DNPH, then that might yield the problem you describe. What's your sample? Is your method isocratic or a gradient?

To test whether the problem is related to the cartridges, try doing the reaction without the cartridges. I'll typically add 1 mL of sample to 450uL of DNPH solution (1 mg/mL DNPH in 1.6N HCl), incubate about 5 min @ ambient, neutralize with an appropriate quantity of NaOH (680uL of 0.1N NaOH followed by 1000uL 0.1M phosphate @ pH7), then filter and inject.

Best,

Chris

[HW Mueller]

Your standard consists of hydrazones, which are thus not run through the reaction cartridges? Anyway, I am surprised that you didn´t have any bothersome extraeneous peaks earlier. I always had some peaks wandering all over when I attempted to analyze ascorbic acid via osazones.

[Uwe Neue]

I am recommending to do a blank derivatization, i.e. without any analytes. This would pinpoint any sources in your reagents, including the solvents.

[Houtuh]

Thanks all for your reactions,

But my problem is that i'm using the cartridges to collect air from building material. So ik can't test without the cartridges because i need these. My method is gradient. Starting with ACN/water/THF 30/60/10 and after 10 min ACN/Water 60/40.

[HW Mueller]

Nobody said not to use the cartridges, just don´t pass any aldehydes through......

[Houtuh]

I understand that,

When i receive a box of cartridges there is an certificate of analysis in the box. Formaldehyde, acetaldehyde and acetone are always presented in the cartridges but in small amounts. When i analyse a blanc cartridge(by taking a cartridge and desorb it with ACN) and analyse this with HPLC, i still get the ghost peak. So I though the problem must be in the cartridges. After this i have ordert an new box of cartridges from a different batch, but i still kept having the ghost peak. But suddenly after a while for some reason the ghost peak dissapeart so i thought that the problem was solved, but after about 5 series there it was again. At about the same time/place. So if the problem was in the cartridges, the why did i get a good chromatogram for about 5 series with the same cartridges?

[HW Mueller]

How about a robustness problem, like the one I mentioned above?

[Uwe Neue]

So it is either in the cartridges, or in the acetonitrile. If it is in the acetonitrile, it could be in anything that the acetonitrile is in contact with. Try to sort out, what you did different for the 5 samples that did not have the problem.

[Houtuh]

I agree, i came to the same conclusion.

Maybe its a combination of both.
I'm going to try some things now, let you now later what came out of it.

Regards

[Houtuh]

Hello everybody,

The latest update:

We have found the problem!

It was in a contaminated batch HPLC-water.

We checked this by analyzing other batches HPLC-water and there we dindn't find the ghost peak. The ghost peak is a reaction between an aldehyde in the HPLC-water that reacts with the DNPH.

Thanks all for your help

Regards