rounded peaks in amine analysis

[chemistry girl]

Hello. I have been using SPME on GC/MS in an attempt to form a standard curve for triethylamine, and heptylamine. However, my peak shapes are very rounded, and they tail a bit. They have been doing this for as long as I have been doing my research, but I only recently found out that this was a problem. I have tried changing SPME fibers (both to a new fiber and a new fiber coating). I have tried using a split-injection. Thus far, nothing has altered the peak shapes. Please, if you could help me to discover a solution, I would be quite thankful!

[Don_Hilton]

Amines are not easy. Without seeing any pictures it is a bit of guesswork - I am goint to guess that your column is the problem. If you check the column catalogs, you will find that there are stationary phases made for analysis of amines.

Let us know what GC and inlet conditions you are using, a description of the samples you are using (like concentration ranges), and a picture of a chromatogram and we may be able to give you more detailed help.

[chemistry girl]

The GC is 5890 series 2. I was using 99.99% pure standard diethylamine and triethylamine. I used a tri-mix (1:1:1) of three different amine standards, thus the concentration of each was 33%. I also tried diluting with various organic solvents, but none worked well for me.

I don't know my inlet information. I will try to figure out how to load a picture.

[Don_Hilton]

Are you putting the fiber into the mixture or sampling the vapors above it? If you are sampling the liquid, you are probably overloading the column big time. If you are sampling failry concentrated vapors, you may still be overloading the column.

Tell us how you are sampling the amines. Vial, method of exposing the fiber, temperature and time.

[chemistry girl]

I am sampling the vapors in the headspace of the sample at room temperature. I expose the fiber for a minute. I have tried shorter exposures (down to 5 seconds), with similar results. The area beneath the peak decreases, but the shape remains the same.

[Don_Hilton]

What column (dimentions, stationary phase), temperatures (Inlet, oven), flow rate, inlet liner, detector?

[AICMM]

chemistry_girl,

Amines are really hard so I wish you luck. The more info you provide the more people here can help. Having said all that, you are concentrating (SPME) a mix which is already concentrated (33%). I would suggest you try something a completely different. Try injecting just one microliter of the headspace you are currently sampling (using a normal syringe) and see if the rounding goes away. I don't expect the tailing will even at this concentration but the rounding might. I would try this both splitless and split (separate runs.)

Best regards,

AICMM

[chemistry girl]

My column is 30mx0.25mm with CP-sil 8. Inlet temp at 200, detector at 300. Initial temp at 45, with a slow ramp to 210. I have played with the ramp a bit, but to no avail.

I have tried a split injection, and it did not work (there was two peaks, and one retained the rounded shape). Today, I tried a direct injection of the headspace, but my limits of detection were insufficient - it did not register. A SIM method may help with this, but I believe that I will eventually need to do all of my work on SPME. Thanks for your encouragement.

I tried sampling a 2% solution diluted with methanol, and this gave me a nearly imperceptible peak - still rounded. I don't think that overloading is an issue.

Thank you for your assistance!

[Peter Apps]

We really need to see chromatograms - "rounded" can mean a lot of things. Instructions on posting chromatograms are in a sticky at the top of the GC page.

In the meantime, try increasing the inlet temperature.

Are you able to make up amine solutions in volatile solvents and do ordinary liquid injections ?

Peter

[chemistry girl]

I will work on getting the chromatograms posted.

The problem with direct injection is that my actual research is going to be with substances that will be dificult to analyze in that manner. I have considered using a solvent to extract the amines from the substances that I will be analyzing, but I was hoping to avoid that step, as I have no idea how to do it!

[Peter Apps]

Sometimes to do troubleshooting you have to use techniques that you know will not work with real samples in order to find and fix a problem with e.g. an inlet or a column. Once you have that problem sorted, you go back to the technique that you want to use on the samples.

For the liquid injections make up a solution in a voaltile organic solvent that will give about 50 ng on the column - so if you are injecting 1 ul and splitting 30:1 you need 1500 ng/ul. The set up for split injections at 30:1 split ratio and inject.

Peter