Do e-liquids undergo thermal decomposition (pyrolysis) when vaped?
E-Cig aerosols are typically generated at temperatures of 100–250°C, which is predicted to cause pyrolysis of the e-liquid vehicle (
162) and may also induce breakdown of other e-liquid constituents. Recently, formaldehyde has been detected in E-Cig emissions (
77). However, these data have been disputed (
55). Part of the problem lies in deciding which temperature the e-liquid is heated to during the experiment vs. what occurs during actual vaping. For example, Jensen et al. found significant amounts of formaldehyde (∼380 μg/10 puffs) in the emission from a tank-style E-Cig device when the battery voltage was set at 5.0 V, with no formaldehyde being detected when a lower voltage (3.3 V) was used (
77). Because the power consumption/electrical resistance of the coil was not quoted by Jensen et al., it will be hard to see how this observation transfers to other E-Cig devices. That is, the power generated by the heating coil cannot be determined purely by the quoted voltage since it also depends on the current, and the temperature reached by the e-liquid is dependent on the power output of the heating element. Thus, for reproducibility, it may be useful for researchers to quote the power output of their E-Cig device in addition to the puff profile used. Farsalinos et al. have reported that E-Cig users do not use this higher voltage setting, and they also proposed that E-Cigs only produce formaldehyde in “dry puff” conditions (
55), where a dry puff refers to the scenario where there is little liquid on the atomizer coil and temperatures get higher than would be seen with sufficient liquid, leading to the potential for increased pyrolysis. However, acrolein and other carbonyls have also been found by other investigators both in neat e-liquids and in E-Cig aerosols that were generated by unmodified E-Cig devices (
133), suggesting that the occurrence/production of these compounds may be more common than originally suspected. Interestingly, neat glycerin does not pyrolyze at 900°C. However, when diluted, significant amounts of acrolein were produced following pyrolysis of glycerol (
28). Similarly, these aldehydes are known to be released from vegetable oil (of which glycerol is a major component) when it is heated during cooking, even to 180°C, which is close to temperatures reported for E-Cigs (130–350°C) (
146). For example, the acrid smell that occurs when oil is burned on a stove is from acrolein (
13,
29). Similarly, the chemical decomposition of sugars also causes the release of aldehydes, including acrolein (
144).
It has been proposed that E-Cig users tend to avoid the bitter taste that is associated with release of aldehydes during overheating/dry puffing and that, in actual E-Cig users, aldehyde exposure never actually happens (
55). However, during the aforementioned practice of dripping, where the e-liquids are placed directly on the coil, it is possible that significant pyrolysis occurs. Certainly, cigarettes can produce a harsh taste that is concomitant with the production of significant amounts of acrolein, formaldehyde, and other aldehydes, along with many other toxicants (
144). However, this relatively unpleasant taste is soon overcome in new smokers due to the power of the nicotine drive (
136) and due to cross-desensitization of transient receptor potential ankyrin subtype 1 (TRPA1) channels in sensory neurons (
19). Therefore, it is also possible that E-Cig users will “learn” to overcome any unpleasant taste due to increased aldehyde production if the nicotine drive is great enough. It is also worth pointing out at this point that many flavors are themselves aldehydes, including anisaldehyde (sweet), cinnamaldehyde (cinnamon), and isovaleraldehyde (nutty). The effects of these flavors on pulmonary surfaces are not known. However, their potential inclusion in e-liquids may increase overall aldehyde exposure to the lung. Indeed, cinnamaldehyde is present in some e-liquids (
14) and activates TRPA1 (
108), suggesting that they may exert effects on the lung. Similarly, activation of this ion channel in sensory neurons in the airways of rodents by unsaturated aldehydes has previously been shown to trigger neurogenic inflammation (
7) and to inhibit the CFTR ion channel (
4), suggesting that a higher aldehyde burden may indeed be toxic to the lung. However, the degree of adverse effects will likely depend on dose ranging and whether aldehydes are actually generated in sufficient quantities during real vaping conditions to trigger these responses.
In addition to aldehydes, Lerner et al. also found that E-Cig aerosols generated from two separate devices produced oxidants and reactive oxygen species (OX/ROS) (
94). Because the amount of OX/ROS changes with time as smoke matures, these data suggest that freshly produced E-Cig aerosols may be more potent than “aged” E-Cig aerosols, which has important implications for their study. Indeed, with regular cigarette smoke, different biological effects are seen with freshly produced vs. aged smoke, with aged smoke often being less biologically potent, which has previously been attributed to the decline in OX/ROS over time (
74). Furthermore, because OX/ROS are highly reactive, they may also react with other components in the E-Cig aerosol, further changing its chemical composition. Indeed, Sussan et al. demonstrated that E-Cigs contain 1011 free radicals/puff, which is about 100 times less than is seen in regular cigarettes (
142), but still likely to exert significant biological effects (
45).