Injecting water sample to GC-FID?

Basic questions from students; resources for projects and reports.

8 posts Page 1 of 1
Dear Sir or Madam

Hope you are very well. Please help.

We bought a new system from Thermo called GC-FID. My colleagues always bring water samples to test for volatile organic compounds from water sources.

Now, what I know that if water enters to GC-FID that will break it , am I right? I am worry that we might loss the system but how I can convenice them , is there is paper strongly warn of injecting water into GC-FID capillary column

Note they just *filter* bottled or collected clean water and injected. No organic solvent present.

I am really disappointed hopefully I am wrong

Best Regards
You won't break it. The system just may not perform well for you. Water is not a very nice solvent for gas chromatography. Mostly, it's because it is because water does not interact well with the stationary phases typically used for GC analysis. Generally, you will not get well formed analyte bands at the head of the column after injection. They don't get better as the analyte travels down the column. Also, injection of large amounts of water can extinguish the flame detector when it comes out.

You won't break it but if you can keep from putting it into your system, it (and you) will be much happier.

What are your analytes? Are you developing a method for this analysis for the first time or are you trying to implement a method that has been published?
Thanks rb6banjo for spending valuable time in answering my question with deep knoweldge.

They measure hydrocarbons 6 or 10 analytes at low carbon number. They said that they use EPA method but I think the EPA method is written for LC not GC. I am not really a ware of their specific project but I just saw 100% water samples being injected to FID and wander that may affect the detector, you are right high injection volume may distinguish the flame.

Many thanks
The more information you can supply to us, the more we can help you. We are "shooting in the dark" if you can't provide specific descriptions of the analysis, sample preparation scheme, analytes, etc.

Hydrocarbons are not easily determined by HPLC. The typical light detectors that are used don't work well if the analyte doesn't have a chromophore. You might get something on molecules that have benzene rings but that's about it. I'd bet it's a GC method of some type.
An FID detector is about the most robust part of a GC, and the least likely to suffer damage from water injections. Injecting water without special conditions will surely mess up the column and the deactivation of the inlet though, and should be avoided if possible. For light (how light ?) hydrocarbons in water by far the best method is equilibrium headspace analysis.

Peter Apps
If you don't have acces to a Headspace autosampler (sample-oven; heated syringe, ...) You can try to do a liquid-liquid extraction. this way you can simply inject a liquid solvent and you already have all the tools to do that.

- take 100ml of water-sample in e.g. a 150ml bottle/a separation funnel
- add NaCl or MgSO4 to saturate the sample ( exact amout is not critical as long as it is the same in each sample).
- add 4ml dichloromethane
- shake or stirr very well.
- alow the CH2Cl2 to sink to the bottom
- remove the CH2Cl2 with a pipette. collect the CH2Cl2 from the separation funnel
- put the solvent phase in a test tube (without water !!)
- add a little dry Na2SO4 and vortex to remove any disolved water from the solvent.
- centrifuge or allow the Na2SO4 to precipitate.
- transfer the clean solvent to a vial -> ready for injection.

This extraction is often done with hexane or pentane, but since you are analysing hydrocarbons they will be your analytes, so CH2Cl2 is a good alternative.
If they are citing an EPA method they are most likely citing SW-846 method 8015 which is at: ... /8015c.pdf

Direct aqueous injection is one of the techniques listed. Be aware that the expansion volume of water is twice that of DCM and larger than other organic solvents. This means that expansion in the inlet will limit your injection volume under the most common inlet conditions. 0.5uL injections are normal.
Thanks indeed my dear for you all for all these suggestions. I can not thank you enough.

8 posts Page 1 of 1

Who is online

In total there are 4 users online :: 0 registered, 0 hidden and 4 guests (based on users active over the past 5 minutes)
Most users ever online was 200 on Fri Nov 10, 2017 4:15 pm

Users browsing this forum: No registered users and 4 guests