Chromatography Forum

Good day, colleagues!
I would like to ask you one (probably) easy question. I realize an analysis of Chlorogenic acid by GCMS. I derivatize the acid with MSTFA, dilute it with dichloromethane, and then inject the mixture.

I noticed, that the area of late eluting peaks is not stable. It can change quite strongly (1,5 times) from injection to injection. Early peaks have the same integral from injection to injection - everything is fine here! For peaks in the middle of the chromatogram, the area also tends to change, but not so much (1-2%). And late eluting peaks change quite a lot, as I have already noted.

Thus, there is an obvious dependence of the area change on the thermostat temperature. The injection is in Splitless mode (300'C). The Oven program is 50'C (0.7 min) and then 10'C/min to 310'C (5 min). Chlorogenic acid is eluted at about 310'C. The signal is clear and has good intensity. But reproducible is poor.

I see, that the answer is very easy, but I can't find the problem. Why the strength of the effect is so dependent on the Oven Temperature? Can you help me?

Thank you!

Why dilute with dichloromethane? I've made trimethylsilyl derivatives for decades and typically use N,N-dimethylformamide or pyridine if dilution was needed.

Do you really need to start at such low initial temperature? I suspect more of a GC issue than an MS issue.

Why dilute with dichloromethane? I've made trimethylsilyl derivatives for decades and typically use N,N-dimethylformamide or pyridine if dilution was needed.

Do you really need to start at such low initial temperature? I suspect more of a GC issue than an MS issue.

Hi,
Thank you for your reply.

In order to use pyridine, I have to know that it does not contain even trace amounts of water, which is quite difficult in my case. Dichloromethane is less harmful, although it is difficult to compare.
However, thank you for your comment.

The lower temperature point according to the Splitless injection rule: the initial temperature should be less 10-20'C than the boiling point of the choosing solvent.

Yes, I agree with you and think so too - the problem in GC.
But I can't find the weak point.

We derivatized and assayed consumer products samples containing over 80% water, after dilution with dimethylformamide or pyridine right out of the bottle. As long as we had excess derivatizing agent, had no problems.

Did this for decades, with both GC and GCMS.

Why dilute with dichloromethane? I've made trimethylsilyl derivatives for decades and typically use N,N-dimethylformamide or pyridine if dilution was needed.

Do you really need to start at such low initial temperature? I suspect more of a GC issue than an MS issue.

Hi,
Thank you for your reply.

In order to use pyridine, I have to know that it does not contain even trace amounts of water, which is quite difficult in my case. Dichloromethane is less harmful, although it is difficult to compare.
However, thank you for your comment.

The lower temperature point according to the Splitless injection rule: the initial temperature should be less 10-20'C than the boiling point of the choosing solvent.

Yes, I agree with you and think so too - the problem in GC.
But I can't find the weak point.

The rule for injecting near the boiling point of the solvent is only necessary if you are trying to resolve early eluting components near the boiling point of the solvent. For something that elutes at such a high temperature you can easily begin the oven at 100C or even higher. I have one injection with dichloromethane that I do with a starting temperature of 150C and the peaks look great because they elute after 250C.

If you can, start the run at 150C and extend to 325-330C and see if that improves the peaks.

Also be sure that there are no problems with the oven closures in the back, if they do not close consistently then the higher temperatures will not be as accurate. I have experienced this before.

Are you using a constant flow or constant head pressure method for the column?

Thank you, for your reply, James.
It's a good idea about oven closures in the back of GC. I will try to check. However, I think that the difference in temperature influences the RT precision. The peak shape is OK, I have only a difference in the area. However, thank you for your suggestion.

Thank you for your comment about the initial point of the oven temperature. It's very useful for me.

Yes, I use a Constant Flow mode.

For how long is the split closed ?

Peter

We derivatized and assayed consumer products samples containing over 80% water, after dilution with dimethylformamide or pyridine right out of the bottle. As long as we had excess derivatizing agent, had no problems.

Did this for decades, with both GC and GCMS.

Hi,

If you don't mind saying, what products were you derivatising, which agent, and what procedure did you follow?

I currently derivatise BPA, and Phthalic acids using BSTFA. It would simplify my prep if I could derivatise aqueous samples directly without evaporating first...

Something like this for liquid products containing up to 85% water; solid matrices were blended with solvent instead, then filtered.

Weigh 0.25 g well-mixed liquid sample

Mix and dilute to 10ml with DMF or pyridine; filter a portion (if cloudy) into a beaker.

Transfer 500 microliters to an autosampler vial; add 250 microliters BSTFA, cap, and shake well. Measure volumes as accurately and consistently as possible; total volume needs to be amenable to autosampler needle depth, but ratio of sample to BSTFA needs to be 2:1 or less to ensure excess BSTFA for samples high in water.

Make up standards similarly.

Capillary GC using non-polar capillary column.