TL;DR @ the bottom.
I would love to hear some more thoughts on this from the educated folk that frequent this sub. I recently came across some information that leads me to draw some odd conclusions.
##The Topic
The recent topic of the presence of diacetyl in vendor eliquids has been a major talking point. Interestingly enough, after completing my thorough research it appears to me as though acetyl propionyl and acetoin and other similar substances do not turn into diacetyl via the aging process when combined with nicotine as has been hypothesized by many on this forum. Folks have gone so far as to blame vendors for bad testing by sending juice that was 0mg nicotine for testing and saying that wasn't adequate. In reality, the true inadequacy may be the manner in which these juices are being tested.
Many folks are trying to blame the eliquid vendors and the flavor suppliers for providing false information about the contents of their flavors. In reality, it is quite possible that at normal room temperatures, these flavors DO NOT contain diacetyl.
##The "New" Information
Now what I found interesting: This information was included in an email I received from Medicine Flower when inquiring about diacetyl in some of their flavors.
> Simply put, a-acetolactate, acetolactic acid, etc constitute precursors for diacetyl through processes potentiated by increased temperature such as vaping and gas chromatography tests. Testing could induce multi-level transformations of flavor content due to its being subjected to temperature exceeding 450F.
Please note that most vapors don't exceed 450F unless running super low builds on mech mods...or suicidal TC freaks =)
A little more research yields:
> Injection Port Temperature
> The temperature of the GC injection port must be high enough to vaporize a liquid specimen instantaneously. If the temperature is too low, separation is poor and broad spectral peaks should result or no peak develops at all. If the injection temperature is too high, the specimen may decompose or change its structure. If this occurs, the GC results will indicate the presence of compounds that were not in the original specimen. Source
##The Conclusion
Perhaps Gas Chromatography batch testing under standard models (temperatures) are not an adequate way for the E-Juice industry to be conducting their testing especially if those tests increase temperatures above recommend use for vaping.
In the end, while this may seem like a moot point, it really isn't, ESPECIALLY if you are utilizing a temperature control set up and can keep your temp below the threshold at which these compounds transform into diacetyl.
##Experts Please Chime In
I would love to hear from folks in the chemistry business that are knowledgeable on Gas Chromatography testing and the temperatures involved during those testing processes (specifically relating to the temperature required to test nicotine and diacetyl levels).
I would also love to know more about the temperature at which a-acetolactate, acetolactic acid, acetyl propionyl and acetoin begin this multi-level transformation process that may yield the result of a positive test for diacetyl and if typical vapors exceed this temperature threshold.
This webpage will explain the GC testing process relating to temperature in more detail..
TL;DR - Perhaps vendors and flavor suppliers aren't lying to us. Their flavors do NOT contain diacetyl, and instead, diacetyl is formed from molecules similar to a-acetolactate, and acetolactic acid as temperatures increase above 450 Fahrenheit.
Adding a disclaimer than I'm not a chemist, my credibility amounts to a bit of college chemistry/self teaching, and will leave this here until someone with more credibility fills in the gaps or otherwise corrects these assumptions:
My basic premise is that because acetolactate (likely a contaminate/unintended byproduct in certain flavorings after extractions, haven't found literature on it that touts its use as a flavor additive) through non-enzymatic redox reaction turns into diacetyl, when a gc/ms test is preformed, this catalyzes the reaction and turns up readable diacetyl levels. Though contrary to the Medicine Flower diction, this reaction is not necessitated by high temperature and naturally occurs at some rate. Meaning that regardless of testing, as a juice ages, those containing acetolactate may end up with diacetyl over time. I don't think the situation is that it takes 450 degrees or a very high amount of heat to cause these reactions to occur, they only accelerate it, in the same way aging a juice is accelerated from heat as it speeds up reactions.
Again could be totally wrong, don't want to spread misinformation, just a hunch. Happy to be proved wrong/correctly informed if misspoke on this.
Sources:
http://morebeer.com/brewingtechniques/library/backissues/issue1.2/fix.html (read the Oxidation column, on the nonenzymatic part)
https://beersensoryscience.files.wordpress.com/2010/11/untitled.png http://www.whitelabs.com/sites/default/files/Diacetyl_Time_Line.pdf
Looking at your sources, it appears to me that diacetyl in beer is produced (and somewhat reduced) by yeast and/or bacteria. It's not a case of the precursors spontaneously transforming into diacetyl. This would not apply at all to e-juice, which should be pretty close to sterile due to the humectant property of both VG and PG, along with the bacteriostatic effect of PG.
Of course I'm open to correction on this, too. I'm certainly no expert.
Yeah, it's not clear that it's used as a flavour chemical at all. It may appear in natural extractions but "natural" does not mean "safe" and these extractions are inherently riskier for vaping precisely because it is hard to control what's in them.
TFA doesn't have it listed as an ingredient in any of their flavours: http://shop.perfumersapprentice.com/specsheetlist.aspx?cas=71698-08-3
Great information, I will read through those articles when I have some time.
As I said, I was trying to draw conclusions from limited amounts of new information that I had received knowing full well that I'm not the expert on such things.
I think the clue is in who owns Enthalpy and who is pushing for the whole market to test all liquids produced.
I have been dubious of this for a long time. The hysteria in the community is incredible and facts get lost in the panic.
In my humble (and unpopular) opinion, it all comes down to $$$$.
I think this kind of thinking though throws out the baby with the bathwater, because saying this is saying well it's all bunk because money. I don't think that's helpful to the conversation, unless you have very specific examples of what's actually going on. money is a factor in all business, so decided as the downfall of the testing process is ignoring what business is all about, making something that people need for a price the market will bear. People need testing, and it's done for a price the market will bear. Perhaps we need independent testing, but to save money is the problem is ignoring the fact that businesses are in business.
Aren't they tied in tight with AEMSA? I remember reading something about The Vapor Chef sending in a 30ml bottle split into two 15ml bottles, one with their brand, the other with a made up company, and the two tests came back radically different.
I am not a chemist or in any way qualified to speak authoritatively.
Here are a couple tidbits I've been told:
- Global Labs offers a more comprehensive testing using a smoking machine. Kim, one of their techs, explained it to me. Basically they're mimicking cigarette testing; a smoke machine inhales the vapor from a device and atomizer of your choice, and they create a vapor concentration similar to the tobacco smoke concentrate used to test cigarettes. I'm sure it varies highly depending on the setup you select (you send it to them), but on first blush it seems like it would be much more indicative of what's actually being inhaled. It's also roughly 8 times more expensive.
- GC/MS gives only one "view" of the liquid – what it looks like only when analyzed using that particular method. It may or may not reflect the actual constituents of the aerosol we inhale.
- A GC/MS tech in the perfume industry indicated that acetyl propionyl has a very similar "peak" on a GC/MS graph to propylene glycol, as well as 1,3-propanediol, and it can be sometimes difficult to distinguish the "signatures" of those molecules when interpreting test results.
Not an expert either although from what I remember from Chem is it all comes down to method developemnt to assist in validating peaks. What you’d want in a method is for somebody to make standard solutions of each compound at and near the target concentrations. Then they would inject each standard solution into the GC and build a calibration curve for each material.
For example, if you are expecting to see diacetyl at 15-25 ppm, you might make standard solutions at 5, 15, 30, and 50 ppm. That allows you to build a calibration curve to show how the detector reports each of these concentrations and allows you to have confidence in the reported value. You would build a cal curve for each component.
You also need to find the right capillary GC column that is able to separate the individual components. The cal curve must be built using that column, and is unique to the machine to be used. Here lies part of the problem in my humble opinion. No one shares the method development...
You're very right, and this aligns with what Global Labs shared with me. They have calibrations for the -acetyls, -aldehydes, glyoxals, and other molecule groups.
I want to make clear that I'm not speaking for their organization in any capacity, simply sharing what I've learned from them. Of all the e-liquid testing facilities I've interacted with, Global Labs is without question the most in tune with the reproducibility of their analyses, their methods, and their limitations.
Great information here.
When dealing with these calibrations, does the temperature used in the GC test vary based on the components being tested for?
Well a temperature program is used. It is hard to explain from what I remember but it is essentially a 'ramping on temp' to better space the peaks and the elution of that compound from the column. And the max temp depends on if the column is polar or non-polar. If the column was a fixed temp then you would have several peaks at once as many enter the mobile phase. If the column is maintained at a low temperature for the duration of a sample run, the first peaks to elute will likely be well-spaced, but the components staying on the column longer will find themselves bound to the stationary phase for longer periods of time; this results in large band-broadening, and long run times. With temperature programming, this effect is overcome by maintaining a low temperature for a short period of time, and increasing the temperature to help force out the longer-‘sticking’ compounds. This changes the retention time compared to a isothermal run, but if the same temperature ‘ramp’ is used, the elution times will remain constant for each component.
Yes, this is how analytical chemistry works. :P Analysis of volatiles through GC-MS is very hairy--not like a lot of other kinds of analyses. For instance, doing analysis for trihalomethanes, I might expect a correlation of 0.990, whereas doing an analysis for something like copper or ammonia I would expect 0.9999+, or I've just been very lazy about making my standards (or maybe there is some extreme interference in the analytical method). It's more difficult to perform, and sampling techniques are more critical. And these standards are EXPENSIVE and do not have good shelf lives in general because of their volatility.
Good to know that there is already a player in the game offering a more reliable test, even if it is 8 times the cost...
Thank you fizz!
Well, I should mention that I discussed with Kim the myriad variables that would affect the results: atomizer material, coil material, wicking medium, power level, temperature, airflow...it's endless. The results would be basically meaningless without establishing some sort of baseline for comparison.
Global Labs is tooling up to support a higher volume of these sorts of tests. Last I spoke with them they'd just purchased a second smoking machine, so they seem committed to the methodology as a potential "best testing practice" (my quote, not theirs) moving forward.
> due to its being subjected to temperature exceeding 450F
slowly looks down at current 480F setting on mod
even more slowly looks down at current 535F setting on mod
This does also raise a valid point - does the TC measure the coil or the vapour? How much temp drop is there in the path from coil>mouth. Could make an appreciable difference.
If I didn't feel any health effects from vaping diketones, I probably wouldn't be worried and I'd find this information encouraging. Unfortunately that's not the case, and it likely is related to having smoked for 20 years. Whenever I vape liquids that have diketones, my chest tightens up, I'm short of breath, and I get very wheezy. The testing I've done on myself are such that if I have just a touch of diketones (like a little bit of TFA banana bread in a mix), it's not bad, but it is noticeable. But if I try something like Mustard Milk (OG version), it becomes very irritating and I can't make it through a tank without having to stop vaping all together for a couple hours.
I don't think that vaping diketones will have zero health effects on people, and if asked, I'd recommend that people avoid them. But vaping is harm reduction, and if a person is not feeling the effects, they have every right to vape them. I am not on a crusade to stop anyone from enjoying tasty liquids. I'd just like people to understand the risk of the unknown that we're facing. Not being able to breath is very scary, and to live out the elder years of life having difficulty breathing sounds like it would be a living hell.
>Please note that most vapors don't exceed 450F
VG has a boiling point of 554° F, PG's is 370°F. So you are definitely getting above 450 unless you do "all PG", which is not a common thing.
Hmmm, so temp control mods running 420F aren't vaporizing VG? I don't think what we are doing is boiling...I dunno though, I'm still no chemist...
You are changing a liquid to a gas. I'm pretty sure that's the definition of boiling, but I'm no chemist either.
There's some discussion elsewhere in your thread about the accuracy of the temp displayed on a TC mod.
makes sense to me =)
Perhaps there's no difference in when inhaled via vaping versus GC testing, I wasn't sure hence the request for those chemists out there. All I know is that the chemical structure is probably to change when undergoing testing /shrug
Maybe there is a lab who will perform analysis by HPLC rather than GC. I'm not that deep into this, but I know some of you are and may not be aware of such instrumentation, since everyone seems to think that GC is the best application for these molecules because we vaporize them normally when there's really no basis for that as far as I've seen (and furthermore there's too much contention about the GC results we do get trickled down to us).
Acetyl propionyl is an almost identical molecule to diacetyl. It's considered equally risky to inhale, concerns over it are nothing to do with the potential to produce diacetyl. See: http://www.cdc.gov/niosh/docket/archive/docket245.html
Acetoin is made using diacetyl as a precursor, it isn't 100% diacetyl-free to begin with. See: http://shop.perfumersapprentice.com/flavorsworkshop/custard.html
It's a misconception that high power = high temperature. None of us vape at a temperature high enough to burn the wick. Around 420F for cotton, 550F for silica, with some variation depending on the specific materials.
While we do need more research into what happens at vaporisation temperatures, diketones are widely used in the flavouring industry and acetoin is a common substitute for diketones. We know for sure that some widely used flavours contain them because TFA and Capella say they do. Some of FlavourArt's diketone-containing flavours have also been sold in the vaping market. There's no mystery about how they might have got there.
Farsalinos tested vapour as well as juices in his study of diketones and found a very close match between what was in the juice and what was found in the vapour. It was only a very small number of samples using only one concentrate so not enough to make broad-based claims. Summary here: http://ukvapers.org/Thread-Diketones-in-e-juice-study-by-Farsalinos-et-al-summary-of-results