Benzene Calibration Curve for GC-FID Analysia

[N.Jane]

Hello,

I have to measure the concentration of benzene gas from my reactor using GC-FID. For that, I need to prepare the standard curve of peak area vs concentration. The problem is now, I don't have the gas standard and I only have liquid benzene. Can anybody explain to me how do I use the benzene liquid to make a standard curve and the calculation of the concentration?

[rb6banjo]

That is tricky business. This is how they make gas-phase dilutions at Restek:

https://blog.restek.com/?p=37152

Perhaps there's something there that you think you could adapt to accurately vaporize a known amount of benzene.

I've tried many different ways of injecting liquid materials (like benzene) into evacuated Restek 3-liter canisters (very similar to what the Restek scientist does by affixing the septum to the evacuated canister) and I can never get reproducible analysis results that give me a "warm-and-fuzzy" feeling that I've done it correctly.

Good luck! If you come up with something, please post it so I (perhaps all of us) can benefit.

[muGC]

Purchase a certified calibration standard from a company (like DCG Partnership). If you have an Agilent GC, they sell an automated dynamic blending system using EPC to dilute calibration standards. Then you could build a calibration curve. Easier than having multiple bottles of different concentrations. See slide 20:
https://www.agilent.com/cs/library/slidepresentation/Public/Houston_Seminar_Carrier_Gas_BackFlush%20and%20CFT_Faster_GC_Part_2.pdf.

[eg@aaclab]

Like the last person said buy a certified gas standard from DCG, Airgas, Scotty etc that is at least as high as you want to calibrate to. If you don't have a diluter etc just buy a 100cc good quality glass syringe and dilute the standard down in the syringe using Helium....its simple cheap and works perfectly..if using liquid standards make a heated injection port using ss tubing and something that looks like an injection port that has a septa in it...make sure its slightly warm or you'll loose some of the Benzene to surfaces....like the other person said it's tricky to get something perfect this way....

[mckrause]

You can build a curve of peak area vs mass of benzene by injecting uL volumes of your liquid benzene (probably diluted in a suitable solvent). Once you know the response versus mass you can measure the mass in a known injection volume of your gas phase samples, then calculate both ppmw and ppmv benzene in the gas.

[Peter Apps]

If you go the liquid injections route you must do splitless injections. The fraction of the actual sample that gets transferred to the column during a split injection is not exactly the same as the split ratio.

Peter

[N.Jane]

That is tricky business. This is how they make gas-phase dilutions at Restek:

https://blog.restek.com/?p=37152

Perhaps there's something there that you think you could adapt to accurately vaporize a known amount of benzene.

I've tried many different ways of injecting liquid materials (like benzene) into evacuated Restek 3-liter canisters (very similar to what the Restek scientist does by affixing the septum to the evacuated canister) and I can never get reproducible analysis results that give me a "warm-and-fuzzy" feeling that I've done it correctly.

Good luck! If you come up with something, please post it so I (perhaps all of us) can benefit.

I read somewhere they use the same approach, but using Tedlar bag instead of canister. Still, it's a tricky business

[N.Jane]

You can build a curve of peak area vs mass of benzene by injecting uL volumes of your liquid benzene (probably diluted in a suitable solvent). Once you know the response versus mass you can measure the mass in a known injection volume of your gas phase samples, then calculate both ppmw and ppmv benzene in the gas.

That is what i was thinking. Using diluted liquid benzene (I plan to use methanol, what do you think?), then do some calculation to correlate the concentration of gas benzene to the liquid one. But I can't find anywhere to justify this method, which makes me not so confident to use this method.

[rb6banjo]

I would be uncomfortable with this procedure. Seems like you have to make too many inferences about transfer of material to the column. There must be a better way.

[rb6banjo]

This doesn't talk about benzene specifically but the results from this study may deter you from trying to expand the benzene into a Tedlar bag and going from there:

http://scholar.google.com/scholar_url?u ... cholaralrt

[AICMM]

N Jane,

Do you expect a high concentration of benzene or very low? You might consider a permeation system if you have access to a good balance and lots of compressed air. Or a diffusion tube. I can send some literature your way if you are interested, aicmm at flash dot net

Best regards,

AICMM

[JI2002]

The preferred way of doing this is to buy a gas standard and do static dilution
as shown in restek video or dynamic dilution with a dynamic dilution system (best).
It should not be too expensive for a customized benzene standard in nitrogen.
If you want to prepare a benzene gas standard from neat benzene,
I would suggest making it in a glass bulb of known volume:
Flush the glass bulb with nitrogen to remove potential contamination
Cap the glass bulb with septum
Inject 1 ul benzene to the glass bulb
Place it to an oven at no higher than 40c to facilitate vaporation
Remove it from the oven and let it to cool down

The standard prepared this way should be fairly close to the expected value,
But likely to be on the low side due to adsorption. Glass bulb should perform
better than tedla bags.