Use of electronic cigarettes has grown exponentially over the past few years, raising concerns about harmful emissions. This study quantified potentially toxic compounds in the vapor and identified key parameters affecting emissions. Six principal constituents in three different refill "e-liquids" were propylene glycol (PG), glycerin, nicotine, ethanol, acetol, and propylene oxide. The latter, with mass concentrations of 0.4-0.6%, is a possible carcinogen and respiratory irritant. Aerosols generated with vaporizers contained up to 31 compounds, including nicotine, nicotyrine, formaldehyde, acetaldehyde, glycidol, acrolein, acetol, and diacetyl. Glycidol is a probable carcinogen not previously identified in the vapor, and acrolein is a powerful irritant. Emission rates ranged from tens to thousands of nanograms of toxicants per milligram of e-liquid vaporized, and they were significantly higher for a single-coil vs a double-coil vaporizer (by up to an order of magnitude for aldehydes). By increasing the voltage applied to a single-coil device from 3.3 to 4.8 V, the mass of e-liquid consumed doubled from 3.7 to 7.5 mg puff(-1) and the total aldehyde emission rates tripled from 53 to 165 μg puff(-1), with acrolein rates growing by a factor of 10. Aldehyde emissions increased by more than 60% after the device was reused several times, likely due to the buildup of polymerization byproducts that degraded upon heating. These findings suggest that thermal degradation byproducts are formed during vapor generation. Glycidol and acrolein were primarily produced by glycerin degradation. Acetol and 2-propen-1-ol were produced mostly from PG, while other compounds (e.g., formaldehyde) originated from both. Because emissions originate from reaction of the most common e-liquid constituents (solvents), harmful emissions are expected to be ubiquitous when e-cigarette vapor is present.

This study characterized emissions from IQOS, a heated tobacco product promoted as a less harmful alternative to cigarettes. Consumable tobacco plugs were analyzed by headspace GC/MS to assess the influence of heating temperature on the emission profile. Yields of major chemical constituents increased from 4.1 mg per unit at 180 °C to 6.2 mg at 200 °C, and 10.5 mg at 220 °C. The Health Canada Intense smoking regime was used to operate IQOS in an environmental chamber, quantifying 33 volatile organic compounds in mainstream and sidestream emissions. Aldehydes, nitrogenated species, and aromatic species were found, along with other harmful and potentially harmful compounds. Compared with combustion cigarettes, IQOS yields were in most cases 1-2 orders of magnitude lower. However, yields were closer to, and sometimes higher than electronic cigarettes. Predicted users' daily average intake of benzene, formaldehyde, acetaldehyde and acrolein were 39 μg, 32 μg, 2.2 mg and 71 μg, respectively. Indoor air concentrations were estimated for commonly encountered scenarios, with acrolein levels of concern (over 0.35 μg m^-3) derived from IQOS used in homes and public spaces. Heated tobacco products are a weaker indoor pollution source than conventional cigarettes, but their impacts are neither negligible nor yet fully understood.

E-cigarettes likely represent a lower risk to health than traditional combustion cigarettes, but they are not innocuous. Recently reported emission rates of potentially harmful compounds were used to assess intake and predict health impacts for vapers and bystanders exposed passively. Vapers' toxicant intake was calculated for scenarios in which different e-liquids were used with various vaporizers, battery power settings and vaping regimes. For a high rate of 250 puff day^-1 using a typical vaping regime and popular tank devices with battery voltages from 3.8 to 4.8 V, users were predicted to inhale formaldehyde (up to 49 mg day^-1), acrolein (up to 10 mg day^-1) and diacetyl (up to 0.5 mg day^-1), at levels that exceeded U.S. occupational limits. Formaldehyde intake from 100 daily puffs was higher than the amount inhaled by a smoker consuming 10 conventional cigarettes per day. Secondhand exposures were predicted for two typical indoor scenarios: a home and a bar. Contributions from vaping to air pollutant concentrations in the home did not exceed the California OEHHA 8-h reference exposure levels (RELs), except when a high emitting device was used at 4.8 V. In that extreme scenario, the contributions from vaping amounted to as much as 12 μg m^-3 formaldehyde and 2.6 μg m^-3 acrolein. Pollutant concentrations in bars were modeled using indoor volumes, air exchange rates and the number of hourly users reported in the literature for U.S. bars in which smoking was allowed. Predicted contributions to indoor air levels were higher than those in the residential scenario. Formaldehyde (on average 135 μg m^-3) and acrolein (28 μg m^-3) exceeded the acute 1-h exposure REL for the highest emitting vaporizer/voltage combination. Predictions for these compounds also exceeded the 8-h REL in several bars when less intense vaping conditions were considered. Benzene concentrations in a few bars approached the 8-h REL, and diacetyl levels were close to the lower limit for occupational exposures. The integrated health damage from passive vaping was derived by computing disability-adjusted life years (DALYs) lost due to exposure to secondhand vapor. Acrolein was the dominant contributor to the aggregate harm. DALYs for the various device/voltage combinations were lower than-or comparable to-those estimated for exposures to secondhand and thirdhand tobacco smoke.