The Propylene Glycol Question: What's Actually in Vape Juice, and Should You Worry?

E-liquid—the fluid that becomes aerosol when heated in a vaping device—typically contains four ingredients: propylene glycol (PG), vegetable glycerin (VG), flavoring compounds, and nicotine. Each of these ingredients is familiar, extensively studied, and generally recognized as safe—in other contexts. Propylene glycol is used in theatrical fog machines, asthma inhalers, and as a food additive. Vegetable glycerin is a common ingredient in cosmetics, food, and pharmaceuticals. Flavoring compounds are the same ones used in food, candy, and beverages. Nicotine is the same molecule delivered by NRT products that have been used safely for decades. The simplicity of the ingredient list is one of the strongest arguments for vaping's reduced-risk profile compared to the 7,000+ chemicals in cigarette smoke. But 'safe to eat' is not the same as 'safe to inhale after heating to 200°C hundreds of times per day for decades.' The propylene glycol question—and its extension to the other e-liquid ingredients—is: what happens when these familiar compounds are used in a way they were never designed for?

Propylene glycol is the most studied of the e-liquid base ingredients, and the evidence for its inhalation safety is more reassuring than critics acknowledge but less definitive than advocates claim. PG has been used in theatrical fog machines and aviation emergency training simulators for decades, with occupational exposure studies showing respiratory irritation at high concentrations but no evidence of long-term lung damage at typical exposure levels. The FDA classifies PG as 'generally recognized as safe' for ingestion, and it's used as a carrier in some pharmaceutical inhalers. The concentrations of PG in e-cigarette aerosol are within the range of occupational exposure limits for fog machine operators, though the exposure is more frequent and sustained for daily vapers. The concern—legitimate but not supported by current evidence—is that decades of daily PG inhalation might cause cumulative respiratory effects that short-term studies haven't detected. The precautionary principle says 'we don't know, so be careful.' The harm-reduction calculus says 'even if there's a risk, it's almost certainly smaller than the risk of continued smoking.' Both statements are reasonable. Neither is definitive.

Vegetable glycerin is less studied as an inhalant than propylene glycol but has a similarly benign toxicological profile in the available research. VG is a sugar alcohol that's widely used in food, cosmetics, and pharmaceuticals. When heated to vaping temperatures, VG can degrade to form low levels of carbonyl compounds—formaldehyde, acetaldehyde, acrolein—that are respiratory irritants and potential carcinogens. The levels of these compounds in e-cigarette aerosol are orders of magnitude lower than in cigarette smoke (typically 90–99% lower, depending on the compound and the device), but they're not zero. The formation of carbonyls increases at higher temperatures—a phenomenon called 'dry puff' that vapers typically avoid because the taste is unpleasant. The practical implication is that device design matters: lower-power, temperature-controlled devices produce fewer carbonyls than high-power devices pushed to their thermal limits. The risk is not eliminated, but it's minimized by appropriate device use.

Flavoring compounds are the most chemically diverse and least studied category of e-liquid ingredients. Hundreds of different flavor molecules are used in e-liquids, each with its own toxicological profile when heated and inhaled. The most well-characterized concern involves diketones—diacetyl, acetyl propionyl, and acetoin—which are used in some buttery and creamy flavorings and are associated with bronchiolitis obliterans ('popcorn lung') when inhaled at high concentrations in occupational settings (microwave popcorn factories). Most reputable e-liquid manufacturers have removed diketones from their products in response to this concern, though enforcement is inconsistent and the concern was arguably overblown relative to the actual risk (cigarette smoke contains much higher levels of diketones than e-liquid, and smoking is not associated with popcorn lung). Other flavoring compounds, including cinnamaldehyde (cinnamon), vanillin (vanilla), and benzaldehyde (cherry/almond), have shown cellular toxicity in laboratory studies at high concentrations, but the relevance of these in vitro findings to real-world vaping is uncertain. The flavoring question is the most unresolved dimension of e-liquid safety.

Nicotine, in the concentrations used in e-liquids, is the best-characterized ingredient from a toxicological perspective, and its risks are well-understood: addiction, cardiovascular stimulation, and developmental effects on fetal and adolescent brains. The nicotine in e-liquid is pharmacologically identical to the nicotine in NRT products that have been used safely by millions of people for decades. The primary difference is the speed of delivery—inhalation delivers nicotine to the brain faster than transdermal or oral absorption, which increases addiction potential but also increases user satisfaction and smoking cessation effectiveness. The nicotine concentration in e-liquids varies widely, from 0mg (nicotine-free) to 50mg/mL or higher in some products (particularly nicotine salt formulations). Higher concentrations deliver a more cigarette-like nicotine experience, which improves cessation effectiveness but may also increase addiction potential. The dose-response relationship between nicotine concentration, cessation success, and addiction severity is one of the most important and least-studied questions in vaping science.

The cumulative risk of e-liquid inhalation—the combination of PG, VG, flavorings, and nicotine, heated and inhaled thousands of times over years—cannot be directly measured because the exposure duration in the population is still too short. The longest-duration vapers have been using products for roughly 15–20 years, which is not long enough to detect the cancers and chronic respiratory diseases that typically emerge after 20–40 years of smoking. The absence of a clear disease signal in this population is reassuring—if vaping were as harmful as smoking, we would be seeing disease by now—but it's not definitive. The most reasonable interpretation of the available evidence is that long-term vaping is likely to carry some health risk (because inhaling anything other than clean air carries some risk) but that this risk is dramatically lower than the risk of continued smoking (because most of the harm of smoking comes from combustion products that vaping eliminates). The propylene glycol question, and the broader question of e-liquid safety, will not be definitively answered for another decade or two. In the meantime, the precautionary calculus for current smokers is clear: the known, severe risk of smoking substantially outweighs the unknown, probably small risk of vaping.