Determine VLE of acetonitrile-toluene with SHS-GC

[nick.s]

I'm a student studying vapor liquid equilibria of different mixtures with static headspace gas chromatography. Now I'm working with the mixture of acetonitrile-toluene. In order to check my results I need to compare them with literature data.

I analysed different molar fractions of acetonitrile-toluene from 0.1-0.9 of acetonitrile, respectively toluene. I used a sample volume of 5ml in the vial. And thermostated at 70 °C (because they also used this temperature in literature) during 15 min (I already checked whether the equilibrium was reached and indeed it was). Then the detection happend with a FID.

In order to set up a Y-X diagram of acetonitrile and toluene. Since I've made the solution I know the molar fraction in liquid phase (X). But I have some problems to determine the molar fraction in vapor phase (Y). In the chromatogram I have 2 peak areas (from acetonitrile and toluene). If I need to calculate to molar fraction of 1 compound in vapor fase, is it scientifically correct to only use the peak areas ?

For example: component A : peak area 10 000, component B peak area 20000
Then the Y for compent A is: 10 000/(10 000+20 000)= 0.33 ?

Because if I do this my experimental data does not match the literature data. But what I do notice is that if I add a constant value of 0.22 to each molar fraction in vapor fase, the experimental data does match the literature data (but I don't have any explanation for that, so I can't do this).

Now I was wondering if sombody is seeing something that I am forgotten or did wrong. I also found different things concerning activity coefficiënts and response factor but I don't seem to find anything to associate it with the peak area.

I hope someone can help me.

Regards

Nick

[Peter Apps]

The missing term in your calculations is the response of the detector per mole for each of your two compounds.

By far the most straightforward way of determining this is a liquid injection of some of the mixtures that you already have, with a very high split ratio. Say for arguments sake you use the 50:50 mix. If the response per mole is the same for each compound the two peak areas will be equal. Divide the area for compound A by the area for compound B and you have a correction factor for the areas that you got from your headspace experiments.

Peter

[nick.s]

Is it also applicable with azeotropes then ? Because if we use the 50/50 mixture at 70°C that there is for example more acetonitrile in vapor phase then toluene, that the peak area will be bigger for acetonitrile then toluene ?

So could it be its not only the response of the detector but also the properties of the azeotrope mixture that will determine the peak area for a 50/50 mixture for example ?

Nick

[Peter Apps]

Hi Nick

Certainly any interactions between the components of the liquid phase will influence the composotion of the vapour phase - that is what your experiment is all about after all.

But if you read my post carefully you will see that I am not suggesting that you analyse the vapour phase - I suggest a liquid injection, in other words the classical gas chromatography sample introduction technique with a microsyringe and a vapourising inlet.

Peter

[nick.s]

Oh sorry, I see. I read it to quickly.

Thank you for the information!

Nick

[nick.s]

The results with the response factors for the organic mixtures are great.

I had only one question left. I'm now working with a mixture of water and ethylacetate. Since FID is not capable to detect water, I cannot determine the response factor. Because no peaks are found on the chromatogram and I cannot compare the peak areas of the 2 compounds.

Is there any solution to it, or do I have to use another detector?

Regards

Nick

[DSP007]

In general water - flame off. Therefore, the peak of water to the FID you will not see or will see the negative.
And once again, the chemistry can not be fooled!
When entering the vapor phase you must always bear in mind the fundamental laws of physical chemistry, including the formation of azeotropes .

[Peter Apps]

Hi Nick

Thanks for the feedback, good to hear that it is working.

Certainly you cannot detect water with an FID, and so if you have to measure the water in the vapour phase you will need a different detector. Unfortunately you will need a different column also.

As long as you can see the ethyl acetate peak you can measure vapour phase ethyl acetate vs liquid phase ethyl acetate, which might get you part way to where you need to be.

Peter