Headspace GC

Chromatography Forum: GC Archives: Headspace GC
Top of pagePrevious messageNext messageBottom of pageLink to this message  By Anonymous on Sunday, January 7, 2001 - 02:42 am:

For Headspace GC , is it always necessary to dissolve the sample in the solvent used.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Ron on Monday, January 8, 2001 - 05:46 am:

No, but it is necessary to have enough surface area to release volatiles into the headspace. A powder will give better results than a solid block.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By C.Z. on Tuesday, January 16, 2001 - 05:09 am:

Adding volatile solvents is not too good an idea.
When you add solvents the HS concentration of analytes is quenchend, so you loose sensitivity.

Surface is another thing of importance, but it is not as simple as stated by Ron.

If the surface is smaller, you just need longer equillibration times.

If you don't measure in an equilibrium state, you have to use an autosampler, in order to keep sampling conditions constant and void systems with gastight syringes !


Top of pagePrevious messageNext messageBottom of pageLink to this message  By John Hinshaw on Tuesday, January 16, 2001 - 08:09 am:

There are a number of accepted HS techniques to deal with solid samples. Adding a displacer solvent to a powder assists with the transfer of volatiles into the headspace, but careful selection and evaulation is required.

Multiple headspace extraction (MHE) will often reduce or eliminate matrix problems as well.

If you'd like to discuss this further, please contact me by email.

John Hinshaw


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Dafna Stern on Friday, June 22, 2001 - 06:55 am:

Trying to determine the best conditions to detect volatiles with GC in a certain atmosphere through the analysis of the HS of dilutted known solutions of these volatiles, I do not know if the volume of the sample will affect the final result. i.e.: if a 10 ml vial sample will give a better result than a 5 ml vial of the same dilutted solution? Thank you


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Ron on Friday, June 22, 2001 - 08:54 am:

If the vials are filled with the same percentage sample, 5 ml in a 10 ml vial vs. 2.5 ml in a 5 ml vial there will be no difference in sensitivity. If you are comparing different percentages of filling, 2.5 ml in a 5 ml vial vs. 7.5 ml in a 10 ml vial, the answer depends on the partitioning. Analytes that partition into the headspace will show higher sensitivity with more sample in the vial. Analytes that remain primarily in the solvent will have a smaller increase in sensitivy when the volume of sample is increased.

This is very simplified, so basically the best way to tell is to experiment. Using a larger vial does not guarantee better sensitivity.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Travis on Monday, August 27, 2001 - 07:22 pm:

It would seem to me that a smaller volume in a smaller vial would reach or approach equilibrium most rapidly. A major consideration in headspace is temperature. Smaller volumes will reach temp quicker. Volume of gas in headspace is important because you want enough to provide room for pressurization gas. Also need enough headspace from static vial pressure and vial pressurization to fill your sample loop. For this reason, vial size should be large enough to allow the best loop fill volume. The sample loop volume is very important for sensitivity and low RSD's.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Ron on Tuesday, August 28, 2001 - 05:38 am:

Actually one of the most important factors for reaching equilibrium quickly is vial agitation. A 20 mL vial heated with agitation will reach equilibrium more rapidly that an unshaken 10 mL vial. In a well designed headspace sampler there is not that much difference between equilibration times for 10 or 20 mL sample vials.

I usually use 10 mL vials because they are cheaper, take less room to store, use less sample, and generate less waste. There are some applications which require 20 mL vials, but most people I know have switched over to 10 mL vials when possible.

For most applications the 10 mL vials do meet the criteria listed above for adequate volume, etc. The larger sampling loops (>2 mL) do require a larger vial.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By donna sirenski on Wednesday, December 19, 2001 - 05:01 pm:

I am using hdsp-GC to analyze 1,2-hexanediol in solid samples (samples are more like a solid block than a powder and we expect >50% of the target analytes in the samples). There may be water in the solids as well. The slope of the logarithmic plot varies significantly for several aliquots of sample, while it is quite consistent for a total vaporization standard. I am concerned that the pressure in the sample vial is too high and may be causing a leak in the cap. Are there disadvantages of lowering the equilibrium temperature (from 200C)?


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Ron on Thursday, December 20, 2001 - 06:12 am:

Some headspace samplers allow you to check the static vial pressure of a heated vial. That would be a good starting point if your instrument allows this. The main disavantage of lowering the temperature would be lower recovery and lower detector response. You might also consider shortening the equilibration time if possible.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Rob on Thursday, December 20, 2001 - 07:31 am:

Solid block samples are notorious for poor quantitative reproducibility. Analytes have to diffuse through the solid matrix to reach the surface, and then move into the vapor phase. Unless the distance that each molecule has to diffuse is both short and consistant, the reproducibility will be very poor. Differing aliquots of block samples will have significantly different mean diffusion path lengths. My advice is to grind your sample down into a powder, the finer the better. However, this has to be done very carefully to avoid losing analyte during the grinding, or creating analyte due to heating while grinding. Although few will have the equipment, I have had good results using a cryogenic mill. The sample is cooled with liquid nitrogen. Besides minimizing losses, the LN2 also improves the grind by making the sample much more brittle.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Ron on Thursday, December 20, 2001 - 07:58 am:

Excellent point, Rob, and one I wish I had mentioned. If the sample is a powder, then it is possible that both the equilibration time can be shorter and the equilibration temperature can be lower. Both will help decrease the possibility of leaks. You might want to consider adding an internal standard if possible to compensate for or at least determine leakage.

I agree thoroughly with the cryogenic grinding to preserve the sample characteristics during the preparation.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Rodney George on Friday, December 21, 2001 - 08:12 am:

John Hinshaw's message from January is still of vital import. Evaluation Evaluation Evaluation.

You can estimate sometimes very closely to the actual content of volatiles in solid samples but never assume you have measured accurately without a lot of confirmation work of other techniques to confirm your HS values. Crystalline samples are infamous for holding volatiles and NOT releasing them unless dissolved. They can soak for a month and still not release all their volatiles.

As shown in my paper in AC of June 1997, bigger vials and larger samples do not mean more accurate or more precise answers. Smaller samples do equilabrate quickly and thus give fewer troublesome side reactions to deal with. Total evaporization as noted is preferable to soaking powders or solids in liquids to 'extract' residual volatiles. (How do you know you got it all?)

I have reproduced analyses from a library of different dissolution solvents, aqueous acids, bases.

Some narcotics only dissolved in 50/50 DMAc/water!

Dissolution IS prefered. Remember, it is not necessary to dissolve the sample at RT, it has to dissolve at temperature in the sealed vial. Solubilities go up with temperature.

Sometimes your solid sample itself will melt at higher temperatures and thus need no dissolution solvent. Testing Zinc stearate for volatiles comes to mind.

For confidence in results, dissolve the sample, (smaller is better) but if not, as Mr Hinshaw has stated,
evaluate evaluate evaluate
and confirm your answer by some other technique such as spectroscopy.

Rodney George


Posting is currently disabled in this topic. Contact your discussion moderator for more information.