Very few flavor-vendors KNOW what's in the bulk flavoring their upstream sources offer them. They are "flavor-blenders" more than "flavor-makers". They buy a "Rich Vanilla" with NO idea what's in it, and they blend it with maybe a "Caramel", sweeteners and a "Milk" into say a tasty "Carpentella's Vanilla Custard".
Their supplier might or might not know what is in the "Caramel" they sell to Carpentella in gallon jugs, and this ignorance continues up the chain until you reach chem labs that most often protect their trade secrets, and don't disclose how they blend their bulk flavoring.
FlavourArt is a rare company that makes most if not all their flavors, and they are willing to say which use diketones, and which do not. To me that's very valuable info, in many ways, because if they had to use diketones to make a specific flavoring, it's very very likely everybody else had to as well. That tells me what to avoid, no matter where is comes from, and, frankly, no matter what cloud of BS they surround it with. When a company says "diacetyl-free", it doesn't say there's e.g. no acetyl propionyl in their flavors, and that's just as toxic as diacetyl, it's another diketone. A vanilla custard, or a buttered cake are classical examples of blended flavors to avoid, because it's very likely that diketones are present. That's the official stance of the admins on this forum.
As to knowing for sure, especially in less obvious cases, for most flavor-blenders (who are not as close to the chemistry as say FlavourArt) it would come down to running some GLC/MS tests on suspect bulk flavoring, testing for say a dozen of the known diketones. But if they run this test on batch #2344 of the "Caramel", and it costs $200, will they have to retest batch #2345 a week later? And of course the same applies to many of their other bulk-purchased flavoring that, albeit by their taste, suggest diketones might be present. Even if the flavor-blender cares, the cost and hassle adds up. There's no point in vague statements like "this company is diacetyl-free" or "that company uses diacetyl".
Finally at the juice-mixer level, where flavoring is mixed with VG-PG-Nicotine base, the level of awareness and concern take a dive, since the end-users don't understand or care that diketones are not a "maybe toxic over 20 years" issue. Diketones can kill lab animals in a few hours at levels that are say 10 times what people who use a lot of flavoring are inhaling. We don't yet have real measurements, but if the models I've tried are right, that's the ball-park. Are you feeling safe inhaling something that at only 10 times greater concentration, is demonstrably deadly?
The way I look at it is simple: anything rich and milky is very likely to contain diketones, because nobody has yet blended such flavors without the use of diketones. So I avoid them completely. No Vanillaey-Caramelly-Cake-CandyBar stuff for me. I didn't switch to vaping to breathe THAT stuff.
Yes, it means a lot of yummy juice is off-limits. I'd rather have working lungs.
I DIY and I don't use over 10% flavoring ever, and most often I consider 5% the limit. I've found FA maple and a simple vanilla to be very useful to blend "yummy" flavors. And I use fruit and "tobacco-like" flavors a lot, with some Sucralose sweetener. I'm undecided on Ethyl Maltol, I'm experimenting. The toxicology testing so far suggests it's pretty safe, but it seems harsh to me somehow. Maybe it's a good throat hit.
As to the juice's base, personally I don't use PG, because side effects I've observed in a lot of people, myself included, are "indigestion", sleepiness and dehydration. These are not serious health problems, it's nothing like the toxicity of diketones or Diethylene Glycol, you can take some acid-reducers like Famotidine (Pepsid AC) or Ranotidine, drink more water and coffee. But I've converted to all-VG, and feel better. VG makes more vapor, PG wicks better and lets the flavor come through stronger. I thin my VG with a little water and/or alcohol, and it wicks fine. But I don't consider it a health risk to vape all-PG e.g. when I have trouble sleeping.
VG has one known issue, and that's breakdown to a harsh/toxic compound called Acrolein above 280 degrees C. That normally does not occur if the atomizer isn't burning juice, but a badly made atomizer can have glowing coil sections on which it could happen. You can spot that "burning grease" taste a mile away, and I toss any atomizer that emits it. But as long as juice isn't burning, VG/PG ratios I think come down to what people feel better with. We will know more in a few years.
