Chromatography Forum

Can i inject an aqueous solution without sample pretreatment (directly) into the injector of a GC-FID instrument? I know that it is not a common work. However, I should do it. I want to know the effects of the injection of water on the performance of each part of the instrument, including injection port, capillary column and FID detector of GC instrument.
For example, i think that large amounts of water can damage stationary phase of GC column. In a similar way, carrier gas should be dried to prevent damage of column.
I also hear that flame ionization detectors may be flame-out in some cases.

Is there any paper or application note discussing about this subject? Thanx

Analysor,

There are lots of posts on this subject as well in the forum. Start by searching the forum on water and FID and you will see plenty of discussion on this subject.

Suffice it to say, if you have to you have to but it is fraught with problems. And, yes, it is very easy to blow out your FID (especially, in my experience, the 5890 series FID's.)

If you still have questions after you have looked at all the other posts let us know.


Best regards,

AICMM

With a capillary column, much less chance to extinguish the FID flame.

We inject aqueous solutions all the time, for hand sanitizer assays for ethyl alcohol, on PEG and -624 type columns. We don't have to use high temperatures for this though. We keep the injection volume 0.5ul or less, as water expands a ton.

There are some discussions about water injection on Restek's site (the Blog). The most information I've been able to find is from clinical labs. Possible problems:
Injection: as little as possible due to water's expansion. Also the problem of everything else in the sample deposited on the liner/column. Must use glass wool to aid evaporation.
Column: water doesn't destroy the column, but the phase must be matched to water's polarity, ie wax type column or possibly blood alcohol column. Also, carrier gas should be high purity regardless the application.
Detector: I haven't seen any problems, unless using a nonpolar column, the water beads up below about 80 degrees and can put out the flame.

I have had best results mixing the water sample 1:1 with an alcohol, peak shapes and calibration much better.

Good luck, and any way to avoid water the better.

There are some discussions about water injection on Restek's site (the Blog). The most information I've been able to find is from clinical labs. Possible problems:
Injection: as little as possible due to water's expansion. Also the problem of everything else in the sample deposited on the liner/column. Must use glass wool to aid evaporation.
Column: water doesn't destroy the column, but the phase must be matched to water's polarity, ie wax type column or possibly blood alcohol column. Also, carrier gas should be high purity regardless the application.
Detector: I haven't seen any problems, unless using a nonpolar column, the water beads up below about 80 degrees and can put out the flame.

I have had best results mixing the water sample 1:1 with an alcohol, peak shapes and calibration much better.

Good luck, and any way to avoid water the better.

Thank you for your response
can you explain how can glass wool aid evaporation?

More surface area for the liquid injection to evaporate from.

Rod

Another thing to consider, if using a split injection for water, is liner discrimination. We have had issues with some solvent assay where water is present. It is more apparent if the solvent being analysed for forms an azeotrope with water. If you prepare the calibration standard without water present and then test a wet sample against it then the results will tend to be low. Putting water into the standard can help a bit, however very small changes in the amount of water present in the injected solution will still have an effect.

GCguy