Formaldehyde release in e-cigarette vapor

The New York Times story explained in detail


Dr Farsalinos

A study to be published in Nicotine and Tobacco Research was featured in the New York Times and has generated a lot of interest. The article mentioned that e-cigarette vapor can be the source of carcinogens, depending on the heating process.

The article is true and expected. We know that thermal degradation can lead to the release of toxic chemicals. And we know that formaldehyde, acetaldehyde and acrolein have been found in vapor. There is nothing new to it. However, this study found that levels may approach those present in tobacco cigarettes. Of course there some inaccuracies in the NYT article, such as that nicotine gets overheated (which means nothing).

Herein, I present with more detail the results of this study. Researchers used an EGO Twist battery (variable voltage) and a top-coil clearomizer (with unknown resistance, thus unknown wattage delivery). At 3.2 and 4.0 volts, formaldehyde levels were 13-807 times lower compared to tobacco cigarettes!! At 4.8 volts, formaldehyde levels were increased by up to 200 times, and reached to levels similar to tobacco cigarettes.

The main criticism to this study is that in my opinion it is highly unlikely that a top-coil atomizer like the one used in this study would be used at 4.8 volts. At a resistance of 2.2 Ohms that would represent 10.4 watts of energy delivery to the atomizer. I tried 10 watts with an EVIC battery in a Vivi Nova top-coil atomizer (for a clinical study i perfomed few months ago), and many vapers were unable to use it due to the dry puff phenomenon. Unfortunately, the researchers did not measure and could not provide any information about the resistance of the atomizers, thus it is unknown how much energy was delivered to the atomizer. In my opinion, this is crucial. Moreover, it is very important to examine new-generation (rebuildable or bottom coil) atomizers at similar conditions, since it is more likely for vapers to use such advanced atomizers for high-wattage vaping. I am certain that, due to better liquid resupply to the resistance and wick, the results will be much more favorable.

Another important point is that, although formaldehyde levels can be similar to tobacco, several other toxic chemicals are completely absent from e-cigarette vapor. For example, acrolein was completely absent although they used liquids with glycerol as the main ingredient. In fact, glycerin-based liquids had much lower formaldehyde levels in vapor compared to PG or PG/VG liquids, suggesting that they are much safer to use. As a general remark, finding few chemicals at similar levels does not mean that the risk is equivalent to tobacco cigarettes. Of course, all this information was not presented in the NYT article.

Concerning the remarks about dripping, we should admit that dripping does not allow the user to see how much liquid is present in the atomizer. The same happens with cartomizers. We currently do not know whether the elevation in formaldehyde levels happens just at the time of dry puff phenomenon, or it happens earlier (before being detected by the vaper). Clearomizer-type atomizers (also called tank systems) seem to be the future in e-cigarette use, giving consumers the ability to know when they need to resupply the atomizer with liquid.


Share Button


0 #44 john r walker 2014-07-22 22:41
Quoting Dr Farsalinos:
It seems that most of the problem originated from PG, not VG. Since there was no acrolein, you are right that the temperatures did not elevate to very high levels..

Yes, PEG (which performed very well ) is stable to about 300C.

A aspect of the test that could have been most unrealistic could be this : the ecigs were kept constantly horizontal and therefore there would be much less turbulence and mixing, of the small portion of the liquid in close contact with the heating element, than if the ecig was waving around in a hand .
0 #43 john r walker 2014-07-22 06:30
Sorry electrics is really not my area.
0 #42 Dr Farsalinos 2014-07-22 06:07
I think it is more appropriate to refer to wattage instead of voltage....
0 #41 john r walker 2014-07-22 06:06
Dr Farsalinos
I have been carefully reading the report (should have done that, first, sorry :sigh:
The report contains a finding that is encouraging :

"An interesting finding of our study is that no toxic carbonyls were detected in a single sample with reduced content of VG and PG. In this product (A6), the primary solvent was polyethylene glycol (PEG). It would suggest that PEG-based e-liquids might have reduced toxicity from decomposition products."

And their plain quite well performing VG control solution, contained %10 water i.e has a much lower boiling point.

Personally the only real attraction of the bigger batteries is they last longer, 3.8 volts is plenty of power.
0 #40 Dr Farsalinos 2014-07-22 05:35
It seems that most of the problem originated from PG, not VG. Since there was no acrolein, you are right that the temperatures did not elevate to very high levels..
0 #39 john r walker 2014-07-21 23:48
Dr Farsalinos sorry to go on :-)

The marked difference between the boiling point for VG @ %90 concentration: 138C, and VG @ % 98 concentration: 290C,
( ) Suggests to me that all things being equal VG diluted with %10 distilled water might possibly be the solution?
0 #38 john r walker 2014-07-21 23:31
The complete absence of acrolein , combined with the spike in levels of formaldehyde, suggests that the temperature reached, at 4.8 volts , over 2.8 seconds duration, was bellow 200+ C , but high enough to cause some of the PG to break down into Lactic acid (and traces of other similar organic acids). And that the heating then continued long enough for some incomplete oxidation of these breakdown products to happen.

Re The result for PG and VG mixes: the vapor point for glycerine when diluted by %10 with water is about 130C - bellow the temp needed for serious problems, with either PG or VG.
However if the glycerine was more concentrated (than %90) then its vapor point rises steeply to about 200+C . That could be a problem with a mix of too pure VG, with PG.

Pragmatic suggestions are do not go over 4 volts and/or do do go past about 2 seconds (and possibly stick to VG with %10 water.)
0 #37 john r walker 2014-07-20 03:33
Another strange thing is that the Dow information on the thermal breakdown products of propylene glycol do not list formaldehyde
0 #36 John r walker 2014-07-19 21:36
There seems to be a marked non- linear or emergent quality to the study's results. I think it would be wise to ask a chemical engineer or someone like that about the chemistry and physics that would be required to produce such a result. The results seem quite strange.
+2 #35 Dr Farsalinos 2014-07-18 22:17
You are right. I also asked the authors to provide some kind of explanation, but they could not.
We will have to wait until someone else repeats the same study...

Add comment

Security code

Disclaimer: all presentations in this site are personal opinions of the authors or website owners. By reading and using the information provided, you agree that we are not held responsible for your choices and you are personally responsibility for using or sharing this information. No recommendations are made for using the electronic cigarette.