Chromatography Forum

By fangwang on Thursday, August 19, 2004 - 09:45 pm:

Dear all,
I am developing method with a derivatization process. My compound reacts with the reagent mixture quite well giving highly fluorescent derivative. However I have a problem that the blank reagent gives a peak at the same retention time with my derivative as well. The higher concentration of the reagent, the bigger this peak. I tried to optimize the derivatizing condition but with the lowest concetration of reagent I can use, the blank still gives this peak. Have any of you ever developed an analytical method which the blank also shows a peak as me? What will you do in this situation?
I am waiting for all of your suggestion. Thanks in advance.

-------------------------------------------------------------------------------------------------------
By HW Mueller on Thursday, August 19, 2004 - 11:17 pm:

Handle it like any other separation: Change chromatographic conditions until the peaks separate. What is your derivatization reagent? It may be possible that you can also suppress the interfering peak.

-------------------------------------------------------------------------------------------------------
By fangwang on Thursday, August 19, 2004 - 11:51 pm:

Thanks HW Mueller,
I tried many ways to separate:changed the buffer, the ratio of ACN and MeOH, changed the column but couldn't success. It seems from my reagent (benzylamine/dimethylformamide and potassium ferricyanide) so I want to ask whether I can live peacefully with it?

-------------------------------------------------------------------------------------------------------
By HW Mueller on Friday, August 20, 2004 - 02:54 am:

No experience with this agent, so maybe someone else can comment on it..are you forming an imine of?
If I understand you correctly, namely that something from the derivatization reagent interferes with the derivative you can, of course, not neglect it. It will most likely be variable so that you can not subtract it. I had a problem like this with FMOC and an amino acid, the peak (just a trace, actually) "followed" my derivative peak no matter what I did. I now think that there was a contamination of derivative, somewhere, giving carryover. (No chance to prove this, the interest in this analysis has long since passed).

-------------------------------------------------------------------------------------------------------
By fangwang on Friday, August 20, 2004 - 05:38 am:

Dear HW Mueller,
My problem may be the same as you; now even when I inject the mobile phase, the peak appears as well. I thought that I contaminated the system with my derivative so I flushed system with IPA overnight but it is still there. And you are right, it is very variable so I can't do the substract blank. I don't know which part of the system was contaminated. What did you do before when you thought that was a contamination? Your suggestion is really appreciated!

-------------------------------------------------------------------------------------------------------
By HW Mueller on Friday, August 20, 2004 - 07:32 am:

We didn´t think about this at the time, but since then it was found that the most tricky carryover can be caused by the rotor/stator of the Rheodyne injection valve (other analytes). In the worst case this had to be opened (separated) and cleaned. Also, for some years we have cleaned syringes by removing the piston and drawing solvent through the syringe, and washing the piston separatly after every injection (if different samples are injected). Autosamplers might have different problems. Any movable injection components, contacted by sample, can produce carryover.

-------------------------------------------------------------------------------------------------------
By Consumer Products Guy on Friday, August 20, 2004 - 07:48 am:

We once had to take apart the entire autosampler plumbing, and flush/clean all components manually, to remove a contaminant. Can you temporarily plumb in a manual injector and try that, see if same happens?

-------------------------------------------------------------------------------------------------------
By Uwe Neue on Friday, August 20, 2004 - 03:09 pm:

If indeed it is a rotor/stator problem, it can sometimes be fixed by using different construction materials. If this is a possible explanation, see what is available for your injector!

-------------------------------------------------------------------------------------------------------
By fangwang on Saturday, August 21, 2004 - 01:11 am:

Thanks all of you!I will try to use a manual injector to see if the same happens.
Dear Consumer Products Guy, if it is really something in autosampler, I will take part the autosampler and clean so could you please suggest me the best way (solvent) to clean it? Thank you and wish all of you a happy weekend!

-------------------------------------------------------------------------------------------------------
By fangwang on Monday, August 23, 2004 - 02:19 am:

Dear Consumer Products Guy,
I connected my system to another manual injector and found out that I did contaminate the autosampler. So now I really need to clean my autosampler, I am working with many samples so autosampler will save my time a lot. Could you please help me the way to clean it properly? Your help is greatly appreciated!

-------------------------------------------------------------------------------------------------------
By Consumer Products Guy on Tuesday, August 24, 2004 - 08:00 am:

Fangwang: here's what I did: I took apart the rotor valve, sonicated in ACN what parts I could. Other parts I swabbed with H2O, then IPA, then ACN. I replaced the rotor seal with a new one. I'd replace the needle seat as well. Personally, I did attach 1/16 inch fittings to a glass syringe (I cut a 1/16 inch ss needle) and attached using unions, etc., and physically pumped through water, then IPA, then ACN to clean the tubing, but suppose you could re-assemble and let the pump to similar for you. Remember the bypass tubing when you do this, as well.

-------------------------------------------------------------------------------------------------------
By fangwang on Thursday, August 26, 2004 - 05:24 am:

Thank you vey much Consumer Products Guy,
I worked out where the dirty was! It was in my needle seat, I took it out and cleaned it with range of solvents (we don't have a new one to replace) and the peak has gone! I am really happy now, thanks again!

-------------------------------------------------------------------------------------------------------
By Consumer Products Guy on Thursday, August 26, 2004 - 08:44 am:

OK. Now that I remember, I think we flushed stuff like this backwards when possible, that's why we did manual flushing.

Fangwang,

Have you checked to make sure there are no impurities in your mobile phase? Are you using HPLC grade solvents? Does the peak in question change when one of your solvent concentration changes? This is where I would look first before taking apart my instrument.

koby

Koby, how does an increase in reagent concentration cause an increase in a peak derived from the mobile phase?
Generally, mobile phase "dirt" produces a rise in baseline (if the dirt is detected), a peak is only produced if one disturbs the equilibrium.

Sorry for no reply for long time, I am too busy after solving out my problem.
Koby, it was not in my mobile phase, it was contaminated in the needle seat by my derivative when I injected with huge scale (using prep column). I cleaned the needle seat and it disappeared (actually, it took me almost 3 weeks to work out: I also tried different HPLC grade solvents, sparged my mobile phase with He...all logical and even inlogical things).

HW, I guess Koby meant the "ghost" peaks in gradient elution. If the ghost peaks form from contaminants in mobile phase, firstly cumulated then eluted, by the way as Koby depicted the ghost peaks can be observed to be changed.

best regards,
kiknos

Kiknos,
usually ghost peaks refer to analyte peaks obtained when blanks are injected. Ok if the mobile phase is contaminated with analyte and you have some disturbance at the head of the column you may get a ghost.