H2O content in cigarette smoke= not stable

[thohry]

Hi all,

Our lab just start analyzing cigarette smoke for H2O and nicotin by GC. The cigarette is smoked by a machine then the smoke is trapped, extracted and injected to GC. The nicotin results are OK, but for H2O, the results are very unstable, normally very low than expected, sometime one tenth !.
Anyone have experience on this analysis could help ?

Some conditions of the analysis
Extraction solvent: 2-propanol
Column : porapak Q
istd : ethanol
Inlet: s/sp
Detector : TCD

Thanks for any input.

[Don_Hilton]

When you say unstable, are you talking about multiple ijections from on GC vial or across multiple smoking runs? While this procedure can be used routinely for many samples a day, the whole procedure must be set up carefully.

1) are you using conditioned cigarettes and are you smoking in a lab with controled temerature and humidity - such as specified by CORESTA?

2) are you attempting to reinject samples from the same GC vial? (This does not work well - samples begin to pick up water?)

3) How are you doing the calibration and are you correcting for background water that comes in IPA?

[thohry]

When you say unstable, are you talking about multiple ijections from on GC vial or across multiple smoking runs? While this procedure can be used routinely for many samples a day, the whole procedure must be set up carefully.

1) are you using conditioned cigarettes and are you smoking in a lab with controled temerature and humidity - such as specified by CORESTA?

2) are you attempting to reinject samples from the same GC vial? (This does not work well - samples begin to pick up water?)

3) How are you doing the calibration and are you correcting for background water that comes in IPA?

Thank you for replying me.
1. We tried to follow the CORESTA guidelines. The cigarettes are kept inside the conditioning for enought time. But it seem the lab conditions are not proper: the humidity too low (40%).
2. I have reinject samples from the same GC vial : the results are reproducible
3. I have subtracted the water content in IPA. The calibration curve is quite linear with good R2.

Would you think the lab humidity is the culprit ?

[Don_Hilton]

Smoking conditions affect smoke chemistry - water and a number of other compunds found in smoke. Be sure to have the room at specified condtions. And, be sure to check the smoking machine for correct puff volume and duration. The airflow across the cigarettes during smoking must be at the correct velocity. Placement of the cigarette filter into the pad holder must be as specified.

Avoid repuncturing the GC vials being used for water analysis. I do not remember the details, being only in and out of that lab to assist with GC issues, but I recall something like the pad extract being taken up in a syringe immediately up on opening of the extraction vial and injected into predried, precapped GC vials. Only one injection was taken from any vial. It seems to me that there was some effort taken to find a GC vial that gave adequate sample stability with that inital puncture in the cap. Exposure of the sample to air is the key. You need to avoid it - and even the puncture of the GC vial can be a problem. You can go as far as to do the extraction in a vial with a septum cap.

Also, pads must be removed from the holder and placed in the extraction vial without delay. You don't want them to equilibrate with the room.

[thohry]

Thank you very much Don-Hilton. You probably are very experienced on cigarette smoke analysis.
We will try to keep the room conditions stable. The problem is that: when it's too humid, it's OK. But when it is too dry, I can not make it more humid evenly within the room !

[Don_Hilton]

Tobacco companies and agencies monitoring the tobacco industry have rooms with dedicated air conditioning systems to maintain temperature and humidity. If you are exploring smoke chemistry to look for new and interesting compounds, you do not need to start in a conditioned laboratory. If you smoke two cigarettes side by side outside of the conditioned laboratory, and see that one gives more carbon monoxide than the other, that finding is valid, but the levels of CO found will not necessarily match the numbers generated when smoking under standard conditions. But the cigarettes need to be smoked at the same time and mixed across the smoking machine. If you are using a single port smoking machine, you are not going to be able to do this - so you have to take additonal steps ot be sure you have a reliable comparison.

[jdezeeuw]

HOw do you trap water? Most of the trapping materials used are very hydrophobic, meaning VERY little affinity (and retention) for water. Sampling times can be very short and are strongly related to temperature of the trap.

How do you know you trap the water?

jaap de zeeuw, Restek corporation

[Don_Hilton]

The measurement of water is not for determination of water produced during combustion, rather it is a factor used in the calculation of "tar" What is captured on the filter pad is determined ot be "tar," nicotine, and water. Subtract the other two from the weight of what you collect on the pad and you have "tar."