Using Watersamples for GC/MS

[Absolutdude]

Hello,

I want to analyse Watersamples using a GC/Ms. I heard often times that water is really bad for GC/MS but I havent really find out , why this is by now.

Sure water has a huge expansionsvolume in the liner. But as long as you only inject a small volume und calculate the expansionsvolume it should be all right.

Can anybody help me?

THank you

[CE Instruments]

Water is a horrible solvent for GC. Yes expansion volume is one issue, another is inorganics and non volatiles that disolve in water. Then steam is very good at attacking your stationary phase and creating active sites.
For water based samples the best solution is headspace the second best is extraction, for ultra low levels purge and trap and only if none of the above are applicable direct injection.

If using a PTV or SSL use a water liner, consider using a pre column for inorganics etc. Carry out your routine maintenance regularly and you should be fine.

[Absolutdude]

thank you for your quick response.

I do a destillation for sample preperation. So inorganics or non volatiles shouldn't be a big problem.
But I recognised something in the liner that looked like lime-scale so maybe this is a problem...

I just have a split/splitless injector and unfortunately no headspace sampler.

[CE Instruments]

Can you let us know what you are looking for ? A simple solvent extraction might prove as accurate and remove the water phase ?

[Absolutdude]

Im am looking for acetaldehyde in drinkingwater. The destillation is for concentrating the acetaldehyde.


I have a problem with the decrease of my sensitivity.
If I get you right ,this could come from using water in the system and creating active sites...right?


"I work with an GC/MS (Agilent GC 78990A + MS5975c) and I recognised that there is a decrease of sensitivity over a sequenz.
If I measure one sample 10 times (with 10 different analytes in it) every run shows a little bit less Area of every compound than the run before.
If I wait one day and do the same thing, the Area is higher than the 10th run the day before, but still not as high as it was in the beginning.

So I think that is a sign for active sites in the system.
I read a lot and tried to fix the problem, but so far nothing helped:

- I took 10 different vials accept of one (because I thought, maybe there is a vacuum after the first injection)
- Changed the syringe
- Changed Liner, Septum, gold Seal
- Changed split vent trap
- Trimmed the column for 10 cm
- Checked the MS , but tunes looked okay
"

[rb6banjo]

What is the "typical" concentration of AA in your distillate?

Have you tried headspace solid-phase microextraction? I have found the carboxen/pdms (black fiber) to be particularly good for low-molecular weight species like AA. You can play games with the ionic strength (adding salt) to drive the AA into the headspace for an even better detection limit if you need it. You might not even need the distillation.

[Absolutdude]

Typicall concentration after destillation is 100 to 200 µg/l.
100 µg/l is nearly my detection limit. Which I think is quite high...

I haven't tried headspace solid-phase microextraction.

[rb6banjo]

That's pretty dilute in the distillate. You probably won't be able to eliminate that step.

Mostly, we look for it at single-digit µg/mL concentrations in our products. I've never tried to salt it into the headspace. It still might work with your distillate.

I agree with the others. Water in gas-chromatographic systems should be minimized. I prefer to never inject water directly.

[Bigbear]

If you are using water as a solvent, you are activating your source over time. Tha "rest" over nite rids the water.

[dblux_]

You may try to increase sensivity through derivatisation with PFBHA:
http://cfpub.epa.gov/si/si_public_recor ... Presented=

[Absolutdude]

@dblux_: thanks for that suggestion. Maybe I will trie this method later...

@Bigbear: Sorry, but I don't understand it... How activates water my source? And what does "Tha "rest" over nite rids the water" mean?
The reason is probably my poor english...

Thanks again

[dblux_]

@dblux_: thanks for that suggestion. Maybe I will trie this method later...

If your GC is equipped with ECD then don't think twice.
This derivatisation is simple and very effective. You will substantialy improve method sensitivity.

[Absolutdude]

No I don't have an ECD, but it could also work with MS, right?

[dblux_]

No I don't have an ECD, but it could also work with MS, right?

In fact PFBHA has 5 atoms of fluorine in a molecule, that's why ECD is so sensitive for it.
Never checked how sensitive is MSD for PFBHA.

[Bigbear]

Sorry for the poor typing. When the MSD sits unused over night the residual water will pump away. Those of us doing purge and trap methods with Agilent 5973 and 75's see the diminishing IS and surrogate areas durring a sequence.