As the e-cigarette industry gets to grips with the requirements of regulation such as the EU’s Tobacco Products Directive (TPD), it’s a good time to take a look at the practical implications on e-cig testing for brands and manufacturers.
The safety, quality and efficacy concerns of the developing regulatory environment put new focus on the constituents of the e-liquid and the inhaled vapour. Which elements could cause concern, and how should we test for them?
When e-cigarettes arrived onto the mass market about a decade ago, little was known about their health profile. They were promoted to the public as being safer and healthier than traditional cigarettes. The basis for these claims was that the nicotine solution lacked the harmful substances found in combustible cigarettes. However, by 2011, some results from scientific assessments in the UK and U.S. suggested the contrary.
The primary ingredients in the nicotine solution are propylene glycol and/or glycerine, and nicotine. However, analyses of the ingredients and emissions also revealed detectable levels of known toxic substances such as diethylene glycol, tobacco-specific nitrosamines, and tobacco-specific impurities. The primary ingredients, additives and the hardware have been implicated as the cause of the contamination.
More recent laboratory studies have revealed the presence of contaminants such as carbonyls, volatile organic compounds, heavy metals and silicate. Some studies suggest overheating of the heating element may give rise to toxic substances. One study (Williams et al, 2013) found metal particles greater than 1 micrometre (0.001 mm) in the emissions from an e-cigarette marketed by a leading manufacturer. Citing poor product design and manufacturing, the study suggested that solder joints, components in the cartomiser, and wires contributed to the presence of heavy metals in emissions.
Scientific laboratories around the world have since conducted a variety of tests on multiple e-cigarette brands and types, and it appears that a safety profile of the product is emerging.
The conclusions of these scientific laboratory studies play a part in certain aspects of the Publicly Available Specification (PAS) for e-cigarettes, sponsored by the trade body ECITA and developed with the British Standards Institute.
The PAS provides guidance for e-cigarette producers, and it indicates that there will be a need for similar laboratory studies on the e-liquid and emissions to build safety and quality profiles of products to be marketed. Furthermore, the PAS suggests that the provision of a toxicological risk assessment based on the results of e-cigarette testing will be required.
How will that be conducted, and what will it mean?
The process and the analytical chemistry techniques used when testing a two-piece e-cigarette – comprising a cartridge filled with nicotine solution and an atomiser or battery unit – can be complex and involve many steps. Not all of them have been presented here.
The first stage involves preparing a sample of the e-liquid and emission for analysis. The nicotine solution is absorbed by a mesh, which surrounds the heating (“atomising”) element. After removal from the cartridge, the mesh is placed in a glass beaker containing a solvent such as methanol. The contents are stirred and the nicotine solution is drawn from the mesh into the solvent. Analysis is performed on a sample taken from this mixture.
“Smoking machines” used for testing tobacco products simulate a person inhaling on a cigarette and capture the emission. These are typically used for traditional cigarettes but can be adapted for e-cigarettes. If a machine is unavailable, standard laboratory equipment can be assembled to the same end.
The machine consists of a pump and a port, which are controlled via a computer interface. This configures the puff volume (PV), puff duration (PD) and puff interval (PI). A regime defined by the International Organization for Standardization (ISO) for testing traditional cigarettes has the following configuration: PV 35 cubic centimetres, PD 2 seconds, PI every 60 seconds.
However, this is insufficient for testing e-cigarettes because there is a slight delay between the onset of the draw by the user and the activation of the heating element. Therefore, a modified version is required for e-cigarettes, such as PV 70 cubic centimetres, PD 1.8 seconds.
The mouthpiece of the e-cigarette is inserted into the port of the smoking machine. The pump is activated, creating a draw and triggering the e-cigarette into producing an emission. This passes through a tube and is captured on a solid material (or in a liquid solution), which is then prepared for analysis.
Perhaps the most important technique in sample analysis is chromatography. Able to separate the components of a mixture by exploiting their differing physico-chemical properties to provide qualitative and quantitative analysis, it is used by many industries to analyse environmental samples, crude oils and medicines.
There are several types of chromatography available, but the most commonly used for sample analysis are gas chromatography (GC) and high performance liquid chromatography (HPLC).
In GC, the sample is converted into a gas and passed through a tube, where the compounds separate according to their volatility. The most volatile substances have lower boiling points and move through the tube quicker. Those compounds with higher boiling points move slowly.
Particle size is another factor influencing the rate at which a compound moves through the tube. An instrument at the end of the tube detects the rate and intensity at which they arrive. Compounds can be identified by the time they are detected and quantified by the intensity. Results are displayed in a graph called a chromatogram.
In combination with GC, a technique called mass spectroscopy enables the identification of the chemical substances. The combination of these two procedures is abbreviated to GC-MS.
In HPLC, a tube containing silica is used, along with a solvent other than water. Ultraviolet light is then employed to detect the separating components of a mixture. This is technique is abbreviated to HPLC-UV.
Toxicological risk assessment
Records of known toxins go back to around 1500BC. Since then, the Industrial Revolution, war, medical tragedies and experimental laboratory studies have rapidly populated the database of known toxins and new ones have been added every year.
Toxicological information from animal testing and human studies is used to understand the toxic nature of a chemical and establish a level of exposure that is considered safe for humans. It should be noted that “safe” levels are not necessarily risk-free, but at a level of risk that is acceptable in society.
What all this means for e-cigarettes is that the toxicity of chemical substances does not have to be tested afresh. It will already be known whether or not a particular substance is dangerous. The question is whether it is present in the e-liquid at all, and at what concentration.
A toxicological risk assessment can begin once the toxins in the sample have been identified and quantified. This will determine whether the level of exposure of toxins produced by the e-cigarette to be marketed is considered “safe”. The assessment will compare the level of exposure of toxic substances found in the e-cigarette with a benchmark.
This benchmark may, for example, be the workplace exposure levels at a manufacturing site where the same toxins have been found. It acts as a reference point of measurement to establish a tolerable exposure level. However, some assumptions and estimations will need to be made because the way toxins are inhaled via an e-cigarette is very different to inhalation at a workplace. The risk assessment will take this into consideration.
What is a poison?
Toxicology can be defined as the study of poisons. To define a poison is a complex issue, because a chemical substance can be without harmful effect at a low dose but harmful at higher doses. Therefore, quantity is key. The toxicity of metals illustrates the importance of quantity in toxicology. Low levels of intake of iron, copper and zinc are essential to the diet. However, at high doses, these metals can cause adverse reactions.
Toxicity may also be acute (causing ill effects through high-level, short-term exposure) or chronic (having an effect after prolonged exposure). Where e-cigarettes and e-liquids are concerned, acute and chronic toxicity are both possible. There is a risk of acute toxicity if the nicotine solution is ingested or comes into contact with skin. By contrast, daily use of an e-cigarette over a period of years will expose the user to small but cumulative amounts of toxin, a case where chronic toxicity might apply.
Based on knowledge from historical laboratory studies of e-cigarettes, the laboratory may seek to identify and quantify the following toxic substances:
The following table gives an overview of the toxicology of the potential toxins found in e-cigarette emissions.
|compound||potential toxic effect||comments|
|Carbonyl compounds – formaldehyde, acetaldehyde, acrolein.||Cytotoxic, carcinogen, irritant, genotoxic.||Carbonyls may be the cause of mouth and throat irritation, the most commonly reported side effect of e-cigarette use.|
|Volatile organic compounds – toluene, benzene.||Carcinogen, neurotoxic, irritant.||Used in industry. These are commonly found in indoor environments.|
|Tobacco-specific nitrosamines.||Carcinogen.||Major cancer-causing agents in traditional cigarettes.|
|Heavy metals – tin, lead, cadmium, mercury, chromium, nickel and iron.||Carcinogen, irritant.||Naturally occurring elements found in the environment.|
|Diethylene glycol.||Central nervous system depression, kidney failure and death in extreme cases.||Presence in e-cigarette emission can be due to low- grade contaminated propylene glycol or vegetable glycerin.|
|Diacetyl.||Decline in respiratory function and bronchiolitis obliterans.||Considered safe if ingested as food flavouring, but not via inhalation.|
Ingestion vs inhalation
Reasons for observed differences in the toxicity of ingested and inhaled substances are quite a complex matter. The properties of the different forms (such as solid, liquid, and aerosol) in which toxic substances are transported into the human body vary. Anatomical, chemical and biological differences between the gastrointestinal tract and the lung are also a factor.
Importantly, the strategic positioning of the liver between the intestinal tract and the blood circulatory system means toxins from the venous blood from the stomach and intestine are subject to detoxification before entering the bloodstream (“first pass” detoxification). Blood then circulates through the body and makes a second pass through the liver, allowing for further detoxification.
But inhaled toxins are not subjected to the “first pass” detoxification. Any inhaled toxins pass first into the bloodstream, and undergo detoxification by the liver only after circulating the entire body.
Ultimately, one cannot conclude that a chemical substance that has a low order of toxicity when ingested will necessarily be equally non-toxic when inhaled. This is illustrated by the case of the approved food additive diacetyl. Based on knowledge from animal and human studies, it is considered safe for consumption via ingestion. Diacetyl is found naturally in food and is used as a flavouring agent to give a product a creamy or buttery flavour.
Yet inhalation of this substance through occupational exposure (such as at a flavouring manufacturing site) has been linked with bronchiolitis obliterans: a severe lung disease referred to as “popcorn lung”. Animal testing in rats has confirmed the potential issue of inhalation toxicity of this substance – concluding that inhaled diacetyl can cause the death of the cells in the airway.
E-cigarette producers have been aware of the hazards of using diacetyl. After the safety issues of inhaled diacetyl came to light, some manufacturers indicated they had removed it from their e-liquids. But in some cases manufacturers were substituting it with a structurally similar chemical – acetyl propionyl – seemingly unaware that this can be as toxic.
Recently, a study designed to test e-cigarette liquids and emissions for diacetyl and acetyl propionyl concluded that these additives were present in a substantial proportion of the samples tested (Farsalinos et al, 2014). Moreover, the authors reported that the levels of these substances were greater than those considered safe.
One study (Williams et al, 2013) demonstrated that e-cigarettes could produce an aerosol of particles of nanoscale proportions. Particle size is a characteristic of importance in toxicology. Nanoscale particles (less than 0.0001mm across) are potentially more toxic than their larger counterparts because of their increased surface-area-to-mass ratio. A larger surface area means more of the substance is available to undergo chemical reactions, potentially leading to increased toxicity.
Nanoparticle production is significant as the particles can achieve deep penetration in the lung (to the alveoli sac) and become deposited there, creating the risk of adverse effects due to interactions with the cells and tissue. Nanoscale particles can also travel to other organs including the brain, and they even have the potential to cross cell membranes into a cell.
In conjunction with the toxicological risk assessment, the results from the qualitative and quantitative analysis of e-liquid and emissions will shed some light on the safety of the e-cigarette under scrutiny. However, more robust scientific data is required in many areas.
There is a lack of long-term toxicological data (from animal and human studies) on the inhalation of the toxic substances found in e-cigarettes. A risk assessment of an e-cigarette must rely on limited toxicological inhalation data, assumptions and estimations. Therefore, there is some level of uncertainty in the risk assessment. Until long-term toxicological data on e-cigarette use in humans becomes available, the limited data available will have to suffice.
Another area in need of research is typical e-cigarette user behaviour. Sufficient data is required to establish how the average user draws on the product (in terms of puff volume/duration/interval). This information could be used to calibrate the smoking machine more accurately for e-cigarette testing. If the puff calibration does not simulate the “average” user, the test results from a scientific study could be seen as invalid.
This data could enable the industry and regulators to develop and agree on an acceptable testing puff configuration for e-cigarettes (and subsequent amendment to the ISO 3308 standard for smoking machines, to include an e-cigarette testing standard). This can only improve the accuracy of the testing and therefore provide a better understanding of its safety.
It is clear that poor product design and manufacturing can be causes of contamination of the e-liquid and emissions. One way to improve the safety profile could be to substitute problematic materials of the hardware for less toxic or more inert ones.
Here, particular attention should be paid to materials that make contact with the solution. Where possible, materials with a low level of chemical interaction with the components of the nicotine solution should be used. For example, substitution of the metal mesh that facilitates the vaporisation of the e-liquid with a glass fibre mesh may result in fewer toxic emissions.
Product developments since the first-generation e-cigarettes and e-liquids have led to some improvements in product safety. For example, many e-liquids on the market are claimed to contain pharmaceutical-grade propylene glycol and vegetable glycerin. This reduces the risk of contamination with a chemical known to be lethal in humans, diethylene glycol.
However, any gains in safety may be undermined by the addition of flavourings that have unknown or potentially toxic inhalation profiles. Synthetic flavourings and the potentially more hazardous (due to their complex composition) natural flavourings are an important safety issue. It may be that the vast range of flavours that exist today will be consolidated to a handful of synthetic flavours, where the inhalation toxicology is better understood and the level of toxicity is acceptable.
There have been some technical developments that appear to address concerns raised by some of the scientific studies. As previously mentioned, some studies suggested that overheating of the heating element led to the formation of toxins. In an apparent response to this concern, electronic units are available (to be installed into e-cigarettes either at the manufacturing stage or by consumers) that can purportedly control the temperature of the heating element. Again, however, any gain in safety may be offset by other factors relating to the e-liquid or the other components of the device.
The implementation of standards in the manufacturing process (“good manufacturing practice”, or GMP) is likely to lead to improvements in e-cigarette products. The principles of GMP ensure consistent production at appropriate levels of quality.
Indeed, the PAS indicates that regulatory bodies may favour the adoption of GMP (as it is applied in the food industry) for e-cigarette production in order for the manufacturing process to consistently yield a product with appropriate quality. Importantly, the adoption of GMP would address the issue of variable product quality between manufactured batches of the same brand, as observed by one study (Goniewicz et al 2013).
Ultimately, regulation will demand improvements to the safety of an e-cigarette. This means manufacturers will need a complete understanding of the toxicology, chemistry and function of the components that comprise the device and nicotine solution, and which elements are a cause for concern.
As the body of scientific research grows, the more complete our understanding will be. This knowledge, combined with advances in technology and adoption of GMP, will very likely yield a better-quality, more effective and safer e-cigarette.
Goniewicz M., Hajek P., McRobbie H. “Nicotine content of electronic cigarettes, its release in vapour and its consistency across batches: regulatory implications”. Addiction 2013; 109: 500-507.
Williams M., Villarreal A., Bozhilov K., Lin S., Talbot P. “Metal and silicate particles including nanoparticles are present in electronic cigarette cartomizer fluid and aerosol.” PLoS One 2013; 8: e57987.
Farsalinos K., Kistler K., Gillman G., Voudris V. “Evaluation of electronic cigarette liquids and aerosol for the presence of selected inhalation toxins.” Nicotine & Tobacco Research doi:10.1093/ntr/ntu176.