Shan, Pharmacology is a research discipline unrelated to Pharmacy and "state licensing". Although I make no great efforts to conceal my identity, I share the general unease that permeates this environment and prefer to retain some privacy. Where I studied and taught many years ago, and what I do with my time are not going to convince you of anything, and shouldn't. Let's see if we can reason together. We are mainly concerned with our exposure to 2,3-butanedione (diacetyl) and 2,3-pentanedione (acetyl propionyl), that are (oh so widely!) used as "buttery" flavoring agents, and both of which seem to cause damage when inhaled, in lab animals and humans. The principal investigator of the acetyl propionyl study you referenced just told me that a similar study on the third commonly used buttery flavoring molecule, acetoin, has completed and is currently in peer-review, i.e. should be available soon, but that she has not seen the data. However, there is reason to believe that it is specifically the diketone structure that is associated with this toxicity, so acetoin is not likely to be as toxic.
If we don't share certain fundamental beliefs, then surely there is nothing we can agree on. I hope that's not the case. Can we agree that the reason people passing a buttered popcorn stand don't just keel over is because there is a concentration of diacetyl at which acute toxicity is not seen, and another at which long term effects vanish as well, in other words that diacetyl, like all drugs and toxins, is bound to obey a dose-response curve that can be discovered? The three questions that I consider most relevant include what damage has been directly observed at what degree of exposure, what has been correlated statistically long term with what concentrations of these agents over how much time (both peak and average, and in conjunction with what other agents), and what level of exposure we are "enjoying" when we use e-liquids whose flavoring contains diacety (or Acetyl Propionyl).
As more information becomes available about the damage (or lack of damage) associated with different inhaled concentrations of flavoring agents, we will continue to see concentrations reported in Parts Per Million of inhaled air. What we need to get a handle on, as Panini said, is their concentration in the e-cig vapor, not in the liquid. Rather than argue about the methodology and conclusions of the few available studies, maybe we can work together to refine a model that might allow us to at least estimate the vapor concentration of flavoring compounds used in e-liquids? Diacetyl is just one that is getting attention at the moment. I rather believe similar issues will likely emerge for other flavoring agents that are mucosal irritants at high concentrations, so what we develop here will be useful long term.
"Vaping" is quite different from smoking. The flavoring maker FlavourArt has published the concentration of diacetyl present in many of their flavoring concentrates, which is very helpful, but when they start to talk about level of exposure from "vaping" they reveal a lack of familiarity with the way e-liquids are used. FlavourArt calculates that an e-smoker is exposed to "0.009 ppm" in an e-smoking session, based on 1.5% diacetyl being present in their Butter flavoring, 0.3% butter flavoring being used in e-liquid, and an e-cigarette containing 0.2 ml of e-liquid
Diacetile - Flavourart. La sartoria degli aromi
This figure is useless, as e-cigarettes aren't "consumed" the way they imply, and their figure doesn't represent a (relevant) concentration we might compare to concentrations reported in the emerging toxicology studies.
How can we realistically estimate the concentration of a flavoring compound that an e-cigarette user inhales? Here is what I came up with so far. The model is simplistic. I have no experience in e-cigarette inhalation studies, but then again I don't know if anyone else does either. I would be grateful if others would look this over with an eye for errors, mistakes, incorrect assumptions etc. Hopefully we can refine the model over time.
Let's say the compound of interest is diacetyl.
In a butter flavoring concentrate the manufacturer states that the diacetyl concentration is 1.5 % or 1.5 grams per 100 grams of solution.
1 milliliter or 1 cc or 1 gram of flavoring concentrate then contains:
1 * 0.015 = 0.015 grams of diacetyl, or 15 milligrams/ml
Let's say the proportion of the butter flavoring concentrate used to flavor some e-liquid (consisting primarily of propylene glycol and glycerin) is 0.3 % (with additional flavors adding up to e.g. 5% total). This is also the figure FlavourArt uses. We have
0.003 ml of flavoring in 1 ml of e-liquid
1 milliliter (roughly equal to 1000 milligrams) of e-liquid then contains
15 * 0.003 = 0.045 milligrams of diacetyl. So far we agree with FlavourArt.
The average person breathes 15 times a minute with a volume of 0.5 liters per breath, or 7.5 liters per minute. An e-cigarette is a personal vaporizer that is used to provide an on-demand stream of vaporized e-liquid. The mechanics of currently used devices dictate a specific 2 step aspiration technique. Let's say a "chain vaper" who deeply inhales, a heavy user, is "loading" every fifth breath with vapor, such that half the breath is mixed with vapor and the other half is "a chaser", and he observes that he goes through 1 ml of e-liquid in 8 hours.
I don't think these figures are wildly off; many users consume less, and some extreme vapers use more, or use variations in their "vaping" technique, but this seems like a reasonable first pass model. During that time, 8 hours,
7.5 * 60 * 8 = 3600 liters of air were inhaled
If we believe that one fifth of that volume was vapor-loaded half way, then, for the purpose of calculating the concentration in inhaled air, we can say that about
3600 / 5 / 2 = 360 liters
of air were actually loaded with vapor. In actuality the clean air "chaser" will almost instantly dilute the loaded half-breath, during inhalation, so if anything our calculations are (properly) biased high, towards the "worst case."
That's 0.36 cubic meters of vapor-loaded air inhaled over 8 hours.
If we assume no losses, no burned e-liquid, no mopped-up leaks, the concentration of diacetyl in that loaded inhaled air is then simply the diacetyl in the 1 ml of e-liquid depleted during that time, divided by the volume of the inhaled loaded air.
0.045 mg / 0.36 m^3 = 0.125 mg/cubic meter
The customarily referenced "ppm", or parts per million concentration of a substance in air is:
24.45 * mg/m^3 / mol wt
The molecular weight of diacetyl is 86. The concentration of diacetyl in the vapor-loaded air is then:
24.45 * 0.125 / 86 =
0.036 ppm
On some short draws it might be higher. If we believe that the loaded half-breath is immediately diluted by as much fresh air, the concentration that reaches lung tissue is about half of that:
0.036 / 2 =
0.018 ppm.
Whether this represents a hazardous level of a specific substance is a completely separate issue, but at least we have a ballpark figure on inhaled vapor concentration, and a first pass model for calculating the concentration of flavorings in inhaled e-liquid vapor, expressed in units compatible with toxicology studies. Hopefully this model can be refined over time.
Notes:
The ppm conversion equation can be found at
http://www.smarte.org/smarte/dynamic/resource/sn-units-of-measure.xml.pdf
The concentration of Acetyl Propionyl in a flavor where it is used in place of diacetyl, is roughly similar. Since the molecular weight is only a little higher, the "vape air" concentrations computed from this model also work out about the same. It's also true that the Acetyl Propionyl concentrations associated with the acute respiratory damage induced by diacetyl in lab animals, are similar. For all the practical purposes at hand, these compounds are more or less interchangeable.
