Importance of vial volume in headspace analysis

[mojo]

We have a headspace analyser that can only use 10 mL or 20 mL vials. However, a customer has supplied a method where 22 mL vials are to be used. Would this 2 mL difference in volume be significant? The sample volume is approx' 2 mL.
If it is significant, would it be reasonable to adjust the sample prep' to 91% of original the original volume, i.e. 1.82 mL instead of 2 mL (20 mL = 91% of 22 mL).
Thanks in advance.

[Peter Apps]

As long as you run standards and samples inthe same vials this should not be a problem - if the method is sensitive to headspace volume (which some of them are) it should be using larger sample volumes than 2ml. Also the 20 ml and 22 ml are nominal volumes - most 20 ml HS vials do contain about 22 ml, just as 2 ml autosampler vials hold 1.8 ml and 4 ml vials hold 4.8 ml.

Peter

[mojo]

Thankyou Peter.

[zokitano]

I think that you should consider whether your analytes have high or low solubility in the liquid phase.
If your analytes have low solubility in the liquid phase you should use sufficient amount of your sample (liquid phase) to ensure greater volume of the gas phase in the vials (to ensure the evaporation of the analytes in the gas phase). Or you might need to add salt to your sample to increase the "evaporation" of your analytes in the vials during the thermostating or equilibration of the vials.
Remember that phase ratio (volume of gas phase/volume of liquid phase) is very important when running HS analyses

Best regards

[Peter Apps]

Hi Zokitano

Can you explain how increasing the volume of the headspace in the vial (by decreasing the amount of sample) will increase the concentration of volatiles in the headspace ?

Taking the two extremes of aqueous alcohols and BTEX by headspace:

the alcohols partition weakly into the headspace. For very roughly equal volumes of sample and headspace, there is much more (by a few orders of magnitude) alcohol in the liquid than in the gas phase, therefore the concentration in the gas phase depends on the concentration in the water and the partition co-efficient (mainly determined by temperature) but not on the volumes of the phases.

The BTEXs (benzene, toluene, ethyl benzene, xylenes) partition strongly into the head space, depleting the concentration in the water. The mass in the headspace depends on the partition co-efficient (as for the alcohols) but also on the volume of the headspace (a bigger volume allows more analyte to partition out of the water) and on the mass of analyte in the original water sample. A large headspace implies a small sample containing a low mass of BTEX, which is lowered still further by partition into the large headspace. Thus a small sample with large headspace gives a lower headspace concentration than a large sample with a small headspace.

The outcome is that alcohols in water is robust to small changes in sample (and vial) volume, and can be run with small samples in large vials, while BTEX in water is sensitive to sample and vial volume and needs as big a sample as possible while still allowing enough headspace to be taken off by the sampler.

Peter

[zokitano]

Hi Peter

It is true that the amount of the analyte in the headspace is directly proportional with its concentration in the sample (and the amount of the sample). But when we consider two analytes with different partition coefficients in, for example water, the one with larger partition coefficient (small partition in water) will mainly remain in the gas phase. The vapor pressure of that analyte will be dependent on its concentration in the liquid phase, partition coefficient and the volume of the gas phase (headspace volume)
The Rault law says that the vapor pressure of the dissolved analyte over its solution is directly proportional to its mole fraction in the solution.
As you said, alcohols have better solubility in water than BTEXs. BTEXs will tend to evaporate in the headspace (result=higher concentration in the head space). Unlikely, alcohols will tend to stay in solution. When using water solutions with same concentration of alcohols from one side and BTEXs from another, you will need to put larger volume of the alcoholic sample in the vial (example 20mL vial) than the volume of the BTEXs' sample to obtain the same sensitivity.
I agree with you about your statements, and I think that I hadn't express myself clearly in the first post.

Best regards
Zoran