By: Arber Frakulli, Toronto ON, Canada
Faculty Advisor: Dr. Deborah Drake
Affiliation: Medical University of the Americas (MUA)
Revised Date of Submission: October 16, 2017
Article Word Count: 4,907 words

Hypothesis: At significantly high coil temperatures within E-Cigarettes (>150°C), which is dictated by the voltage and power applied, there is formation of carbonyl breakdown products released in the aerosolized vapor due to thermal decomposition of glycerol and propylene glycol.

Abstract: Word count = 249

E-cigarettes emerged in 2003 and have since exploded in popularity. They have been marketed as a safer alternative to conventional cigarettes and as a tool for quitting smoking. The relatively recent mainstream success garnered by E-cigarettes means that little research is available to investigate these claims.

An NCBI database search on recent journal articles focused on the level of exposure of E-cigarette users to carbonyls and their potential contribution to cancer development were identified and summarized. The null hypothesis is that no carbonyl compounds are liberated within the aerosolized vapor produced by E-cigarettes. The alternative hypothesis is that at significantly high coil temperatures within E-Cigarettes (>150°C), which is dictated by the voltage and power applied, there is formation of carbonyl breakdown products released in the aerosolized vapor due to thermal decomposition of glycerol and propylene glycol.

Based on results compiled, there exists considerable variability in the level of aldehyde formation based on the brand of E-cigarette device used, as well as variability amongst different models within the same brand. While earlier studies showed a lower aldehyde profile for first generation E-cigarettes relative to conventional cigarettes, newer more sophisticated E-cigarettes with increased power outputs, have been demonstrated to increase the level of aldehyde production.

It is evident that at coil temperatures exceeding 150°C, significant production of aldehydes occurs which may surpass the levels produced through combustion of conventional cigarettes. The carbonyls released are known or potential carcinogenic agents in humans that may initiate the process of cancer production or propagation.

Keywords: E-cigarette, Vaporizer, Vaping, Carbonyl, Aldehyde, Carcinogen, Safety


Lung cancer claimed 1.6 million lives worldwide in 2012 alone (Stewart & Wild, 2014). Not only is smoking implicated in lung cancer, but it is also responsible for other smoking-related diseases such as emphysema and chronic bronchitis. It is estimated that the cost of lost productivity as a result of smoking is nearly $100 billion a year in the US (Smith, 2008). Although there is evidence that smoking rates have declined or tapered off in the developed world, WHO has demonstrated that consumption of tobacco is still on the rise in the developing world (World Health Organization, 2009). According to the World Bank, it is estimated that there will be 1.5 – 1.9 billion smokers by 2025 (WHO, 2003). As a result of health organizations worldwide campaigning and educating consumers on the risks of cigarettes, many have tried various methods of smoking cessation with varying rates of success.

Smoking cessation methods are bountiful but their effectiveness has remained controversial, even after many years in the market. The methods most often used to quit smoking range from quitting “cold turkey” (i.e. without any assistance), using nicotine replacement therapy or relying on medications such as bupropion and venlafaxine. These methods have varying success rates but are not always appealing to the consumer. With this in mind, I hope to elucidate the safety of one relatively new method of smoking cessation, namely the Electronic-Cigarette (EC). There are several kinds of EC designs but the basic premise they all have in common is that they super-heat an e-liquid solution (made of varying proportions of propylene glycol & glycerin) to temperatures of 100 – 200°C within their chamber. This results in the production of an aerosolized vapor which the user then proceeds to inhale thus simulating the act of smoking a cigarette. The act of inhaling from EC devices is colloquially referred to as “vaping”.

The US e-liquid market alone is estimated to grow to $8.26 billion USD by 2025 (Research and Markets, 2015). There are various explanations for the expanding growth in the E-cigarette sector including the claims that they are cheaper, smell-free and most importantly, healthier than conventional cigarettes. Additionally, the popularity of EC may also be due to the sustainability of the bio-behavioral feedback loop of the oral fixation associated with cigarettes which is simulated by EC (Barbeau et al, 2013). These products are marketed by their manufacturers as a much safer alternative to smoking. However, these advertisements may be misleading because not only has this method of smoking cessation only recently gained popularity, but it has yet to be rigorously studied and proven to be a safer alternative. Furthermore, the efficacy of EC as an aid in quitting smoking is still questionable and needs further study.

While some studies have demonstrated that EC are safer than smoking cigarettes, there is mounting evidence that they are not as safe as advertised by the manufacturers. Additionally, their long-term health effects are largely not known. This is because the EC is a relatively new product in the smoking cessation arsenal and has yet to be properly scrutinized scientifically. Long-term monitoring of chronic use of EC would help shed some light on these questions.

There is mounting evidence to support that the use of EC exposes users to a variety of different carcinogenic compounds. These include carbonyls, heavy metals as well as TSNAs (Tobacco-Specific Nitrosamines) (WHO, 2015). Two of the most important carbonyl compounds liberated by EC, as well as regular cigarettes, are formaldehyde and acetaldehyde. Both glycerin (GL) and propylene glycol (PG), solvents used in e-liquids, have been shown to generate carbonyls via decomposition at high temperatures (Paschke et al, 2002). A further complication is the fact that it has been demonstrated that carbonyls may also be present in some e-liquids itself (Farsanilos et al, 2014).

Formaldehyde is categorized as a Group 1 known carcinogen by the International Agency for Research on Cancer (IARC, 2017). The IARC has concluded that sufficient evidence exists for the association of formaldehyde with nasopharyngeal cancers as well as strong but insufficient evidence for a link in leukemia mainly due to a low number of human studies (IARC, 2017). It has been demonstrated that formaldehyde-DNA adducts occur in the lymphocytes of smokers (Wang et al, 2009).

Acetaldehyde on the other hand is categorized as a Group 2B possible carcinogen in humans by the IARC. In addition to forming DNA adducts, acetaldehyde is responsible for causing point mutations in the HPRT1 (Hypoxanthine Phosphoribosyl Transferase 1) locus as well as binding to glutathione (essential for anti-oxidant and free-radical quenching) and proteins required for DNA cytosine methylation and DNA repair (Seitz and Stickel, 2007). A study on patients with Fanconi anemia (genetic defect in DNA repair proteins) showed that they were more susceptible to acetaldehyde thus supporting the evidence that acetaldehyde interferes with DNA repair machinery (Mechili, 2008).

DNA adduct formation of carcinogens, like formaldehyde and acetaldehyde, may cause the incorporation of wrong bases during DNA replication. Not only can adduct formation lead to an increased chance of mutagenicity, but these effects may be long-lasting and perhaps heritable. A study by Chen et al has demonstrated, in a study on smoking cessation, where leukocyte DNA of smokers was measured for two weeks while actively smoking and then again after four weeks of quitting, 30% of the individuals did not show a decrease in the number of acetaldehyde-DNA adducts by the end of the study (Chen et al, 2007). However, long-term chronic exposure is required for a critical amount of mutations to accumulate in order to lead to tumor progression and there does exist a long latency period (Vogelstein and Kinzler, 2004). In addition, the complete cancer formation process has been demonstrated to require more than the initial DNA damage (Wu et al, 2016). There is much to learn but it still stands that there exists a dose-related effect of carcinogens on DNA adduct formation, mutation rate and neoplastic transformation (Maher & McCormick, 1984).

With each puff of an EC, current is passed through a heating element (usually a nichrome wire coil). This process produces heat in turn causing the e-liquid to form an aerosolized vapor which is subsequently inhaled by the user. The temperature achieved by the heating coil is dependent on the power output of the EC which is determined by the voltage of the battery as well as the resistance of the coil (which thus dictates the amount of current passing through the coil). Increasing the voltage or power used by the EC, as well as decreasing the resistance of the coil (which allows more current to pass through the coil), causes a subsequent increase in the temperature of the coil. The temperature of the coil is also affected by the airflow through the EC apparatus, the level of the e-liquid within the EC chamber, the puff duration and the interpuff duration (time between puffs). Studies have shown that heating coils can reach temperatures of up to 350 °C (Balhas et al, 2014; Schripp et al, 2013; Talih et al, 2015).

Based on current studies published, coil temperatures seem to be an extremely important predictor of the levels of carbonyl species produced within the aerosolized vapor of EC. Hence, this review will focus on the level of exposure of EC users to carbonyls and their potential contribution to cancer development.

The null hypothesis is that no carbonyl compounds are liberated within the aerosolized vapor produced by ECs at any temperature. The alternative hypothesis is that at significantly high coil temperatures within ECs (>150 °C), which is dictated by the voltage and power applied, there is formation of carbonyl breakdown products due to the thermal decomposition of glycerol and propylene glycol.


This review employed preclinical experimental studies found through PubMed. The article search employed the MeSH (Medical Subject Headings) Database index with the search strings used being: “Electronic Cigarettes/adverse effects” OR “Electronic Cigarettes/mortality” in addition to (“Formaldehyde” OR “Acetaldehyde”) AND “Electronic Cigarettes”. Articles were selected based on experimental studies performed in order to identify and/or quantify the levels of aldehydes (specifically Formaldehyde and/or Acetaldehyde) formed from the vaporization process of e-liquid EC solutions.

Based on the search criteria listed above, a total of 400 articles were found using both Boolean operator search strings listed above. Of these, ten articles were identified that specifically identified and quantified the level of carbonyls produced by ECs. One of the articles simulated the EC vaporization process through the use of a stainless steel tubular reactor (Wang et al, 2017); the rest of the articles used at least one EC for their methodology. All articles chosen were published within the last five years.

The majority of the studies employed the CORESTA protocol for the detection of aldehydes using High Performance Liquid Chromatography (HPLC) (CORESTA Recommended Method No 74, 2014).


With respect to first-generation EC, it was observed that the average formaldehyde and acetaldehyde levels produced was much lower than conventional cigarettes (cigarette:EC ratio – 9:1 and 450:1 ratio of formaldehyde and acetaldehyde formation) (Goniewicz et al, 2013). In another study on the first-generation EC by Uchiyama et al, a large variability between the level of formaldehydes formed was seen when evaluating 363 different EC devices from 13 different brands (Uchiyama et al, 2013). Four of the brands did not generate any aldehydes whereas the other 9 generated variable amounts; the maximum amount of aldehydes observed for 10 puffs was equivalent to approximately 2 cigarettes worth (Uchiyama et al, 2013).

Gillman et al demonstrated a positive correlation between the power level of EC used (from 5.2 W to 25.0 W) and the quantity of aldehydes formed by the vaporization process of the EC (Gillman et al, 2016). One of the EC devices studied was found to exceed the aldehyde production of a conventional pack of cigarettes even at the lowest power setting and another EC device exceeded aldehyde production of a pack of cigarettes only at the highest power tested; the three other devices had lower aldehyde yields than a pack of cigarettes (Gillman et al, 2016).

Farsanilos et al also demonstrated a significant amount of aldehyde production for one of the two EC devices they were investigating, but only at the two highest powers tested (9W and 10W), and minimal aldehyde production at the lower powers tested; minimal aldehyde production was also seen at all powers tested for the second device (Farsanilos et al, 2015).

A separate study using NMR spectroscopy showed that hemiacetals containing formaldehyde were formed in the process of vaping (Jensen et al, 2015). At the highest voltage tested (5.0 V), the level of formaldehydes formed via 10 puffs of the EC was equivalent to the yield of approximately two cigarettes, whereas at low voltage (3.3V), no formaldehyde-releasing agents were observed (Jensen et al, 2015). Kosmider et al also demonstrated similar findings when observing voltages ranging from 3.2V to 4.8V (Kosmider et al, 2014). It was also reported that e-liquid solutions with the highest proportion of PG produced the most amount of aldehydes during vaporization and hence may be more susceptible to decomposition at higher temperatures.

In regard to puffing topography, it was observed that the formation of aldehydes is specific to the second part of a typical ten second puff period and hence does not occur continuously throughout the process of puffing (Hutzler et al, 2014). In addition, it was seen that the aldehyde release is associated with lower liquid levels within the cartridge/tank of the EC but occurs before the liquid is fully exhausted (Hutzler et al, 2014). Low aldehyde production was seen at coil temperatures under 100 ˚C but significant amounts were observed at temperatures greater than 150 ˚C (Hutzler et al, 2014)

Using a device-independent investigatory method (i.e. stainless steel tubular reactor) in order to simulate the vaporization process that takes place within the atomizer of an EC, Wang et al demonstrated that below 215°C, few aldehydes were produced, whereas above 215°C, aldehyde production increased precipitously (Wang et al, 2017). They also demonstrated that the thermal decomposition of GL produced more carbonyls than that of PG at temperatures of 270°C (Wang et al, 2017). At an average EC usage rate, a typical EC user would be exposed to approximately two cigarettes worth of formaldehyde per day if coil temperatures are maintained at 215°C but would dramatically rise at temperatures above that level (Wang et al, 2017). Moreover, no additional carbonyls were seen with the addition of various flavourings and nicotine to the solutions being vaporized (Wang et al, 2017).

Device aging, due to repeated use of the same device without maintenance/cleaning was found to lead to an increase in aldehyde production by 60% (Sleiman et al, 2016). This was important because it may simulate more realistically the conditions a typical EC user would be subjected to with long-term usage of an EC device.

The only study found on direct drip atomizers demonstrated a very high level of formaldehyde production, much higher than previously reported with conventional EC and cigarettes (Talih et al, 2015). Consecutive short-term puffs between non-puffing intervals were shown to greatly increase the production of formaldehyde and acetaldehyde (Talih et al, 2015). A maximum coil temperature of 350 ˚C was recorded (Talih et al, 2015).

The general findings that were reproduced by the studies reviewed were that with increasing Power/Voltage applied, the EC vaporization process showed an increase in the temperature of the metallic heating coil and consequent increase in formation of carbonyl species.


With the growing number of EC users as well as its growing social acceptance in the non-scientific community, a careful and analytic approach must be undertaken to clarify the potential dangers underscoring use with respect to the short- and long-term health effects of EC. As demonstrated by recent surveys, people perceive EC as safer alternatives and less harmful than cigarettes with respect to lung cancer and pregnancy (Baeza-Loya et al, 2014). Passive vaping has also become a concern, especially with the public’s perception of EC being less harmful (Schripp et al, 2013).

Based on the results compiled in this review, it is clear to see that there exists considerable variability in the level of aldehyde formation based on the brand of the EC device used as well as variability amongst different models within the same brand. Early models of EC did not have an option of changing the power or voltage applied to the resistor and hence operated at a fixed, pre-determined temperature. Newer and more sophisticated EC have the option of changing the voltage or power applied to the resistor and hence give more control to the user as to what temperature they may like to vape at. While earlier studies showed a lower aldehyde profile for first generation EC when compared to conventional cigarettes, newer EC with increased power outputs have been demonstrated to cause an increased level of aldehyde production. The problem that arises with this increased level of control in the hands of EC users is that they may unknowingly be vaping at sufficiently high temperatures that may lead to a significant formation of carbonyl species. As revealed by the studies referenced within this review, an increased level of power output leads to an increase in temperatures reached by the heating coil which subsequently leads to an increased thermal decomposition of PG and GL into carbonyl compounds.

With respect to the study by Gillman et al, it was found that the less efficient an EC was (i.e. the less aerosol it produced), the more aldehydes were present in the aerosol (Gillman et al, 2016). This phenomenon is counter intuitive since it is often stated within the vaping community that the higher the power/voltage used in vaping, the higher the temperature reached by the heating coil and thus the higher the subsequent volume of aerosolized vapor produced; taking this thought one step further, and synthesizing it with the studies reviewed, the higher the temperatures reached by the coil, the higher the carbonyl content of the aerosolized vapor should be. More studies are necessary in order to elucidate the link between the efficiency of EC, based on the amount of aerosol produced, and the level of aldehydes liberated.

Additionally, there is general consensus that different e-liquids produce aerosolized vapors with variable aldehyde profiles mostly based on the proportions of the solvents used (PG and GL). Two studies offered contrasting results on whether PG or GL was the dominant solvent responsible for carbonyl production. Kosmider et al found that solutions having higher ratios of PG:GL formed more carbonyl compounds when vaporized (Kosmider et al, 2014). Alternatively, Wu et al found that significantly more formaldehyde was formed, at 318 °C, from solutions with higher ratios of GL:PG (Wang et al, 2017). It was also reported that levels of carbonyls generated from solutions containing both PG and GL were not the sum of the separate PG and GL vaporization product profiles which suggests that complex reactions may be occurring during the process of vaporization (Wang et al, 2017). Of note, polyethylene-glycol containing e-liquids produced no detectable carbonyl compounds when vaporized (Kosmider et al, 2017). Further investigation into the potential use of polyethylene-glycol as an e-juice solvent is required in order to confirm this phenomenon. Additionally, in order to better understand how different proportions of PG and GL affect the carbonyl content of aerosolized EC vapor, more research must be undertaken.

There appears to be some contradiction within the scientific community with Wang et al finding that flavorings within e-liquids do not contribute to additional carbonyls (Wang et al, 2017). This is in contrast to a separate article that demonstrated that the thermal decomposition of certain flavors may lead to additional carbonyl formation (Khlystov and Samburova, 2016). The role of flavor additives and its potential role in the formation of carbonyls requires further study.

Farsanilos et al argues that although high levels of aldehydes are produced as higher levels of wattage or voltage are used for vaping, a strong unpleasant taste is detected by the EC users thus leading the users to avoid vaping at these higher power levels (Farsanilos et al, 2015). This is termed as the “dry puff” and is due to the overheating of the e-liquid. Using noxious stimuli as a deterrent to prevent EC users from vaping at dry puff conditions is problematic because this is entirely a subjective experience. Another potential drawback is that the dry puff (or vaping at conditions close to the dry puff) may be masked by strong flavor additives within the e-liquid which may lead to users being unaware of the high levels of aldehydes being produced.

With the exception of the paper by Farsanilos et al, all of the studies being reviewed had no human EC users involved concomitantly testing the products being investigated. If we assume that the dry puff technique is an effective deterrent to vaping at power levels high enough to cause a significant level of carbonyl production, it can be argued that these conditions found within the reviewed papers are not generalizable to the average EC user if it is assumed that dry puffing would have deterred them from vaping at those power levels. Research should thus also include an EC user component blinded to the vaping conditions they are testing in order to further elucidate the viability of dry puffs acting as a deterrent.
With respect to “Direct Dripping” atomizers, only one study was found that investigated the level of toxicants released. Direct drip atomizers (DDA) work via a small amount of e-liquid being dripped onto an exposed heating coil and are touted for their high vapor production. Because of the nature of DDAs having a limited capacity for holding e-liquid, it is likely that the coils dry out quickly, subsequently becoming too hot and thus producing high amounts of carbonyls as well as dry puffs. This was demonstrated by Talih et al where it was noted that formaldehyde production by DDAs was greater than even conventional cigarettes and that the coil reached a maximum temperature of 350 ˚C (Talih et al, 2015). Making sure that there is always enough liquid around the coil would be the best way to remedy this hazard. However, the paper by Hutzler et al seems to cast doubt on the aversive stimulus supplied by the dry puff when e-liquid levels get too low since the production of aldehydes was seen to increase well before the liquid itself was fully exhausted (Hutzler et al, 2014). Nevertheless, aldehyde occurrence was still found to correlate with lower liquid levels within the cartridge (Hutzler et al, 2014).

The concept of device aging is also an important parameter to keep in mind when investigating EC vapors. Since most EC users do not clean their devices before each use, residue, colloquially referred to as “coil gunk”, may be a significant source of additional aldehydes (Sleiman et al, 2016). This phenomenon of device aging, and the deposition of coil gunk, is important to investigate in the future as it is more consistent with, and representative of, normal daily use of EC devices by the average user.

The general limitations that limit generalizability of these studies to all EC:

  1. Wide variability amongst different brands of EC.
  2. Wide variability amongst different models (or even between different devices of the same model) of the same EC brand.
  3. Wide variability with regards to the ratio of PG:GL solvent base and flavor additives used for e-liquid production.
  4. Potential sources of contaminants being present within the e-liquid itself.
  5. Lack of a standardized method to generate vapors in smoking machine regimen in order to best model the vaping behaviors of the typical EC user.
  6. Lack of human EC user participation with regard to potential dry vaping that may be occurring at higher voltage and power levels used.
  7. The resistances of the coils used within the EC investigated were not stated in some of the studies. The lower the resistance of the coil, the higher the current passing through the coil and thus the higher the temperature achieved for any given voltage applied. Thus, although some studies may be investigating EC at similar voltage levels, there may be a dramatic difference between the currents created if the coils used had a different resistance.

Based on the review of all currently available literature on EC, it is evident there exist conditions that lead to a significant production of aldehydes which may surpass the level of conventional cigarettes. There is a clear positive correlation between the voltage or power level that the EC is operated at, and the temperature reached by the heating element, and the subsequent level of carbonyl species generated within the aerosolized vapor inhaled by the EC user. The carbonyls released are either known or potential carcinogenic agents in humans which can initiate the process of cancer production or further propagate it. Based on these findings, the null hypothesis of no carbonyl compounds being liberated within the aerosolized vapor produced by EC is rejected. The alternative hypothesis stating that at sufficiently high coil temperatures within EC (>150 ˚C), there is significant formation of carbonyl breakdown products caused by the thermal degradation of glycerol and propylene glycol is not rejected.

Possible changes that may have a direct benefit to user safety and health, in addition to minimizing passive vaping in non-users are:

  1. Requiring FDA approval of all EC devices after thorough testing to minimize potential toxicant formation or address any other safety concerns.
  2. Regulating the sale of EC and their products and taxing them like conventional cigarettes and perhaps using the money raised to perform longitudinal studies to better reveal the long term health effects of EC.
  3. Outlawing indoor use of EC in public places to prevent passive vaping.
  4. Limiting the temperatures of the heating coils themselves (perhaps by limiting the voltage or power and coil resistance).
  5. Installing safety mechanisms in order to automatically shut the EC device off if a certain threshold temperature is reached.


Baeza-Loya, S., Viswanath, H., Carter, A., Molfese, D. L., Velasquez, K. M., Baldwin, P. R., … Salas, R. 2014. Perceptions about e-cigarette safety may lead to e-smoking during pregnancy. Bulletin of the Menninger Clinic78(3), 243–252.

Balhas Z., Talih S., Eissenberg T., Salman R., Karaoghlanian N., Shihadeh A. (2014). Effects of user puff topography and device characteristics on electronic cigarette nicotine yield. Presented at the 20th Annual Meeting of the Society for Research on Nicotine and Tobacco (SRNT), February 5–8, 2014, Seattle, WA. POS4-57.

Barbeau, A.M., Burda, J., Siegel, M. 2013. Perceived efficacy of e-cigarettes versus nicotine replacement therapy among successful e-cigarette users: a qualitative approach. Addiction Science & Clinical Practice, 8(1), 5.

Chen L, Wang M, Villalta PW, Luo X, Feuer R, Jensen J, Hatsukami DK, Hecht SS. 2007. Quantitation of an acetaldehyde adduct in human leukocyte DNA and the effect of smoking cessation. Chem Res Toxicol 20:108–113.

CORESTA Recommended Method N° 74: Determination of Selected Carbonyls in Mainstream Cigarette Smoke by HPLC. Paris, France: CORESTA; 2014. Retrieved from:

Farsalinos, K. E., Spyrou, A., Tsimopoulou, K., Stefopoulos, C., Romagna, G., & Voudris, V. 2014. Nicotine absorption from electronic cigarette use: comparison between first and new-generation devices. Scientific reports4.

Farsalinos, K., Voudris, V. and Poulas, K. (2015). E-cigarettes generate high levels of aldehydes only in ‘dry puff’ conditions. Addiction, 110(8), pp.1352-1356.

Gillman, I., Kistler, K., Stewart, E. and Paolantonio, A. (2016). Effect of variable power levels on the yield of total aerosol mass and formation of aldehydes in e-cigarette aerosols. Regulatory Toxicology and Pharmacology, 75, pp.58-65.

Goniewicz, M., Knysak, J., Gawron, M., Kosmider, L., Sobczak, A., Kurek, J., Prokopowicz, A., Jablonska-Czapla, M., Rosik-Dulewska, C., Havel, C., Jacob, P. and Benowitz, N. (2013). Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tobacco Control, 23(2), pp.133-139.

Holliday, R. K. 2016. E-cigarette vapour is not inert and exposure can lead to cell damage. Evidence-based dentistry., 17(1), 2-3.

Hutzler, C., Paschke, M., Kruschinski, S., Henkler, F., Hahn, J. and Luch, A. (2014). Chemical hazards present in liquids and vapors of electronic cigarettes. Archives of Toxicology, 88(7), pp.1295-1308.

International Agency for Research on Cancer (IARC). 2017. Agents classified by the IARC (Monographs, Volumes 1–119). Geneva, Switzerland: International Agency for Research on Cancer

Retrieved from:

Jensen, R., Luo, W., Pankow, J., Strongin, R. and Peyton, D. 2015. Hidden Formaldehyde in E-Cigarette Aerosols. New England Journal of Medicine, 372(4), pp.392-394.

Ji et al. 2016. Characterization of electronic cigarette aerosol and its induction of Oxidative stress response in oral Keratinocytes. PloS one., 11(5), 1371.

Kosmider, L., Sobczak, A., Fik, M., Knysak, J., Zaciera, M., Kurek, J. and Goniewicz, M. (2014). Carbonyl Compounds in Electronic Cigarette Vapors: Effects of Nicotine Solvent and Battery Output Voltage. Nicotine & Tobacco Research, 16(10), pp.1319-1326.

Kosmider et al. 2016. Cherry-flavoured electronic cigarettes expose users to the inhalation irritant, benzaldehyde. Thorax., 71(4), 367-377.

Khlystov A., Samburova V. 2016.Flavoring Compounds Dominate Toxic Aldehyde Production during E-Cigarette Vaping. Environmental Science & Technology 50 (23), 13080-13085

Lauterstein et al. 2016. Frontal cortex Transcriptome analysis of mice exposed to electronic cigarettes during early life stages. International journal of environmental research and public health., 13(4), 417.

Madsen et al. 2016. A cancer that went up in smoke: Pulmonary reaction to e-cigarettes imitating Metastatic cancer. Chest., 149(3), 65 – 67.

M.E. Counts, M.J. Morton, S.W. Laffoon, R.H. Cox, P.J. Lipowicz. 2005. Smoke composition and predicting relationships for international commercial cigarettes smoked with three machine-smoking conditions. Regul. Toxicol. Pharmacol., 41 pp. 185-227.

Maher VM, Mccormick JJ. 1984. Role of DNA lesions and repair in the transformation of human cells. Pharmacol Ther 25:395–408.

Mechilli M, Schinoppi A, Kobos K, Natarajan AT, Palitti F. 2008. DNA repair deficiency and acetaldehyde-induced chromosomal alterations in CHO cells. Mutagenesis. 23:51–56.

Paschke, T & Scherer, Gerhard & Heller, W.D. 2014. Effects of Ingredients on Cigarette Smoke Composition and Biological Activity: A Literature Overview. Beitr Tabakforsch Int. 20. 107-247.

Research and Markets. 2015. U.S. E-Liquid Market Focus on Type, Origin and Distribution Channel – Estimation & Forecast, 2015-2025. US: BIS Research.

Schripp, T., Markewitz, D., Uhde, E. and Salthammer, T. 2013, Does e-cigarette consumption cause passive vaping? Indoor Air, 23: 25–31.

Seitz HK, Stickel F. 2007. Molecular mechanisms of alcohol-mediated carcinogenesis. Nat Rev Cancer. 7: 599-612.

Sleiman, M., Logue, J., Montesinos, V., Russell, M., Litter, M., Gundel, L. and Destaillats, H. (2016). Emissions from Electronic Cigarettes: Key Parameters Affecting the Release of Harmful Chemicals. Environmental Science & Technology, 50(17), pp.9644-9651.

Smith, H. 2008, September 10. The high costs of smoking. Retrieved from MSN Money:

Stewart, B., & Wild, C. 201). World Cancer Report 2014. Lyon (France), Geneva (Switzerland): WHO Press, World Health Organization.

Talih, S., Balhas, Z., Salman, R., Karaoghlanian, N. and Shihadeh, A. (2015). “Direct Dripping”: A High-Temperature, High-Formaldehyde Emission Electronic Cigarette Use Method. Nicotine & Tobacco Research, 18(4), pp.453-459.

UCHIYAMA, S., OHTA, K., INABA, Y. and KUNUGITA, N. (2013). Determination of Carbonyl Compounds Generated from the E-cigarette Using Coupled Silica Cartridges Impregnated with Hydroquinone and 2,4-Dinitrophenylhydrazine, Followed by High-Performance Liquid Chromatography. Analytical Sciences, 29(12), pp.1219-1222.

Valentine et al. 2016. The effects of alcohol-containing e-cigarettes on young adult smokers. Drug and alcohol dependence., 159, 272-276.

Vogelstein B, Kinzler KW. 2004. Cancer genes and the pathways they control. Nat Med10:789–799.

Wang M, Cheng G, Balbo S et al. 2009. Clear differences in levels of a formaldehyde-DNA adduct in leukocytes of smokers and nonsmokers. Cancer Res, 69: 7170–7174.

Wang, P., Chen, W., Liao, J., Matsuo, T., Ito, K., Fowles, J., Shusterman, D., Mendell, M. and Kumagai, K. (2017). A Device-Independent Evaluation of Carbonyl Emissions from Heated Electronic Cigarette Solvents. PLOS ONE, 12(1), p.e0169811.

World Health Organization. 2003. Past, current and future trends in tobacco use.Washington DC: The International Bank for Reconstruction and Development / The World Bank.

World Health Organization. 2009. WHO/WPRO – Smoking Statistics. Manila, Philippines: WHO Regional Office for the Western Pacific.

Wu S, Powers S, Zhu W, Hannun YA. 2016. Substantial contribution of extrinsic risk factors to cancer development. Nature 529:43–47.

Yu et al. 2015. Electronic cigarettes induce DNA strand breaks and cell death independently of nicotine in cell lines. Oral oncology., 52, 58-65.