A study published in late December 2015 in the The Journal of the American Association for Aerosol Research, Aerosol Science and Technology, by researchers from the University of California, Irvine, aims at evaluating the quality of the emissions of e-cigarettes and compared their figures to combusted reference products.

Novel PTR-MS analysis proves efficiency compared to the old gaseous or liquid phase chromatography techniques

To achieve this goal, they develop a novel analytical approach based on real-time measurements of volatile organic compound (VOC) concentrations and their size spectrum. The machine, a proton transfer reaction time-of-flight mass spectrometer (PTR-MS), is being used to measure concentrations of VOC by NOAA in atmospheric and stratospheric studies, and has also been tested by Imperial, a British Tobacco company, in the framework of a comparative study of their products with concurrent ones. Size spectra were obtained from another analysis but on the same sample.

E-cigarettes emit particles and VOCs

The US study by S. Blair and her co-authors received the support of the National Cancer Institute of the National Institutes of Health and the Family Smoking Prevention and Tobacco Control Act, and was funded by the National Science Foundation. It confirms an increase of VOC emissions with increasing tar content and the success of charcoals filter to retain those compounds. The presence of menthol in e-cigarette brands also seems to increase the VOC content compared to their non-menthol equivalent, as the likely result of the addition of this flavor in the formula. It also confirms that e-cigarettes emit particles in similar amounts as “light” tobacco cigarette (probably the lightest “light” cigarette available).

Particle concentration emitted by e-cigarettes is not different from “light” cigarettes

In brief, the researchers sampled the air emitted from normalized puffs of e-cigarettes and tobacco cigarettes and analyzed particles and VOC contents. VOCs and particles were screened by their instruments and compared for two types of e-cigarettes and 9 types of classical ones among which tar levels ranged from 1.67 mg/cig (light) to 25.0 mg/cig (unfiltered) and nicotine content from 0.16 mg/cig to 1.7 mg/cig. Nicotine content of the two e-cigarettes remained around 0.55 mg/cig and the only difference between the two types were that one contained e-liquid made from propylene glycol (e-cigarette 1) while the other one was vegetal glycerin (e-cigarette 2). No indication is given with respect to the brands used but the authors mention “Kentucky Reference cigarettes” since most of them were employed in other standardization by the University of Kentucky in Lexington.

Acetaldehyde, acetone, acrolein, methanol and acetonitrile

Among the emission products, acetaldehyde, acetone, acrolein, methanol and acetonitrile concentrations were investigated. The latter only occurs in tobacco cigarettes, in relation with the nitrogen present in tobacco (pigments and proteins). Methanol is mainly associated to the pyrolysis of the pectin present in tobacco leaves during smoking and should be a minor constituent of the VOC emissions of e-cigarettes. Acetaldehyde, acetone, and acrolein may result from the oxydation or the thermal decomposition of sugars, cellulose, pectin, triglycerides, and glycerol and ere found in all the samples in various concentrations. More specifically, acetaldehyde, acrolein, and acetone were found in the e-cigarette studied, supporting the evidence of the electrochemical oxidation of vegetable glycerin.


Is it better vaping PG, VG or a mix ?

The authors notice that e-cigarette 2 had ~10 times more acrolein than e-cigarette 1 in the chamber experiments and hypothesize that the cartridges of e-cigarette-1 and -2 were from different brands which might have contributed to this difference. They alternatively suggest a loss because of the permeability of the Teflon tube for this compound. In the graphs, only the data of the e-cigarette 2 are compared to the other cigarettes. However, the two e-liquids also differ from their composition: PG for -1 and VG for -2. It is noticeable that any puff of a tobacco cigarette emits more VOCs than an e-cigarette, even if the e-liquid employed during the test contains 100%VG.

No conclusion with regard to toxicity

Since the toxicity is dose-dependant for each constituent, the authors have not been able to infer any potential toxicity of any of the different cigarettes tested and suggest further investigation in this domain.

Further research is needed…

The authors showed that amount of VOCs and particle concentration increases with puff number for conventional cigarettes but not for the e-cigarette 2, which had no puff number dependence over less than 10 puffs (at 1 puff/min).

Fig. 8: VOC and particle content of e-cigarette-2 as a function of puff number during continuous use with a single cartridge (the battery was fully recharged at the beginning of vaping and at each interval (indicated by a line)
Fig. 8: VOC and particle content of e-cigarette-2 as a function of puff number during continuous use with a single cartridge (the battery was fully recharged at the beginning of vaping and at each interval (indicated by a line)

Their figure 8, showing the integration of data from a continuous vaping experiment however raises a few comments with regard to the reproducibility of emission results with e-cigarettes. The authors noticed that that volatiles were not emitted with a consistent delivery rate and link the decrease to either the depletion of the battery or the depletion of the cartridge.

  • The concentration of particles follows a reproducible pattern between two battery recharges but do not seem to be divided by 2 over the experiment, as observed for acrolein and acetaldehyde.
  • The change over time (along with puff increment on the X-axis) of acrolein is mirrored by acetaldehyde which suggests that similar processes are underlying their production. An overall decrease along the experiment for both compounds is observed and may be attributed to a continuous change of electrochemical oxydation settings.
  • Acetone and methanol were coupled after the puff #230 and decoupled over the first half of the experiment.

An unclear methodological background

It is regrettable that no information was provided with respect to the type of e-cigarette used by the team. The technical properties of the device may provide insights for the interpretation of the results:

  • the availability of a temperature-controle mode on the device,
  • the type of metal used for the coil (Ti, Ni, Cr, Ag, Fe…),
  • the performance of the chip to regulate the power as the battery is getting depleted.

To go further

The experiment that appears on figure 8 most likely reflects the use of an e-cigarette by a vaper. Two ways remain to be explored when looking at the evolution of the concentrations. They address the evolution of the quality of the coil and its impact on the emission of potentially harmful VOCs.

  • A loss of matter from the coil occurs on the long-term that decreases the production of acrolein and acetaldehyde by electrochemical oxydation (maybe a selective loss of one of the catalysers of the reaction?). Such loss of matter is expected during use even if it is not expected to be of significant health concern for smokers switching to e-cigarette use, according to Konstantinos Farsalinos, after a litterature review on this topic. As pointed out by Konstantinos Farsalinos and Pedro Carvalho, the material the resistors are initially made of is not suitable for a direct contact with the constituents of e-liquid, further research may help finding the most appropriate alloy to be used in e-cigarettes.
  • A change in coil temperature that depends on power regulation by the device itself may be superimposed to the long-term trend. Such changes in temperature could explain the quasi-reproducible pattern of acrolein and acetaldehyde on the one hand, and of particles in the other hand, between two recharges of the battery.
  • It remains unclear why methanol and acetone do not follow the same patterns.

Blair, S. L., Epstein, S. A., Nizkorodov, S. A., & Staimer, N. (2015). A Real-Time Fast-Flow Tube Study of VOC and Particulate Emissions from Electronic, Potentially Reduced-Harm, Conventional, and Reference Cigarettes. Aerosol Science and Technology : The Journal of the American Association for Aerosol Research, 49(9), 816–827.