twitter: @rolygate

PG Sensitivity

Throat irritation in new e-cigarette users
Sensitivity to PG
Sensitivity to glycerine
PG and VG safety issues
Glycols and glycerol are alcohols
Glycerine viscosity issues


Questions are sometimes raised about PG [1] sensitivity, leading to discussions about a change to VG - but this brings up its own issues, particularly regarding viscosity problems in carto tank systems. Some of these topics are examined below.

This article was written specifically for the benefit of e-cigarette store owners, one of whom brought up these issues; in general, shop owners/managers are keen vapers but not technical experts. There is no comparison between the relative size of the knowledge base for ecigarette vendors and tobacconists (vendors of cigarettes and tobacco): the amount of knowledge required by an ecigarette supplier is many times greater than that for a tobacconist. An ecig vendor needs at least two years of experience to reach a basic level of competency, since they are customer-facing staff who will have two major issues to contend with:

  • A continual volume of questions about the products, and about general vaping issues (two things that are interdependent: few vaping issues can be discussed without reference to the products).
  • A significant number of new users, who deserve the best advice.

In order to assist transition from smoking to vaping, a mentor requires knowledge of the many thousands of options in order to achieve the best success rate: zero-option vaping has a success rate of as low as 10% (that is, being restricted to one hardware & refill option), whereas full-option vaping may succeed in 75% of cases under optimal circumstances [2]. You can see that there is a massive difference between these two extremes.

For example, there are more than 7,000 refill variants offered by just one vendor; a few of these will work and most won't, for any smoker attempting to switch - and the optimal choices will be different for each smoker.

Perhaps at a later date there will be large chains of ecigarette stores, who will be able to run staff training courses; until then we may find a knowledge shortfall at the shop counter.

New ecig users experiencing throat irritation

Some smokers draw into the mouth first, then inhale to the lungs; some inhale direct to the lungs. We call the mouth-to-lung draw M2L as a short-form term, and the direct-to-lung draw is D2L.

First, ensure that persons who are new to ecigs do not inhale direct to the lungs, as many smokers do. Use the M2L method: draw the vapour into the mouth - hold - inhale (if required) [3]. Tobacco smoke contains anaesthetics / antitussives designed to reduce throat and lung irritation, ecig vapour does not. Despite the fact that vapour is simply a water-based mist, inhaling anything other than pure air causes a reaction, which can be mitigated by correct technique. All cases of 'lung irritation' are caused by incorrect technique, since an e-cigarette is not used like a cigarette (in multiple ways), at least for a new vaper.

Inhaling directly to the lungs is incorrect technique for beginners with an e-cigarette and can lead to irritation or coughing. While any/all variations of every aspect of ecig use may work for experienced users, who will certainly be able to use a D2L inhale if they wish (and many prefer it especially with RBAs, rebuildable atomisers), beginners need correct advice in order to transition successfully.

Not providing expert advice helps no one. Indeed, it would be easy to achieve extremely poor results with an ecigarette by isolating the beginner from correct advice (or even deliberately withholding it). This is clear from studies where deliberate minimising of information and restriction of product choice was employed, which generally results in success rates as low as 10% or issues such as 'lung irritation', neither of which have any relevance to correct use.

Sensitivity to PG

About 1 in 10 people are sensitive to propylene glycol, meaning that they suffer from a measure of upper respiratory tract irritation and/or excessive drying-out [4], but not a severe effect. Most will experience reducing symptoms if they persist with vaping, as they become tolerant to it.

However, it is important here to remember that virtually all new, unmentored ecig triallists will experience throat and/or upper respiratory tract irritation due to lack of a crucial piece of advice: vaporisers are not a tobacco cigarette and are preferably not used in the same way, by new users. A vaporiser is not a cigarette and cannot normally be used in the same way by a smoker new to vaping. With the correct advice, no irritation at all should be experienced. Irritation normally indicates poor advice or even lack of advice.

For some, the symptoms may persist, and there could be a benefit in changing to VG [5] or PEG-based [6] refills. Note that a sore throat is a common symptom of smoking cessation, though, and is experienced by a percentage of those who are successful in a smoking cessation attempt; thus, not all vapers' experiences are attributable to vaping - some are clearly attributable to smoking cessation.

About 1 in 100 people are very sensitive to PG in this way, and suffer from a painfully sore throat. If this combines with the 'quitter's throat' mentioned above, in infrequent cases, then an extremely sore throat can result.

About 1 in 5,000 people are highly sensitive to PG and experience skin problems ranging from dryness to severe itching.

Accuracy of interpretation
The above figures are anecdotal and were estimated from the frequency of reports on the giant vaping forums, some with 15 million posts, there being plenty of material to work with. To derive any meaningful data from clinical trials or RCTs on these subjects would require CTs on a massive scale and of great expense and with little realistic return value. With tens of millions of 'anecdotal' accounts, there comes a point at which it is no longer necessary to carry out CTs to get basic data: the situation is clear. Naturally, it is a popular subject of debate whether anecdotal evidence has any value or not, but scale is important with such issues: when a topic is not controversial and simply of informational value, several million answers is a good resource to work with. Where uncontroversial topics unlikely to be subject to much response bias are concerned, conclusions can be regarded as having significant value though not proven.

It is also worth keeping in mind that nothing can be proven with clinical trials: CTs and RCTs simply provide evidence, no matter how strong or how frequent; no amount of RCTs can provide any proof, especially outside the limited application of CTs: determining the efficiency and side effects of medical interventions. (CTs are useless for consumer products, and any conclusions are essentially anecdotal - all relative measures for such products will certainly be unusable.)

A fact is something that is clear from observation or, in the health area, when provided by unbiased national health statistics; at some point, enough observations constitute a fact, and when the total reaches a sufficient number of observations then a factual position has been reached. The highly engaged online community support for vaping provides us with, literally, millions of observations both for and against many aspects of the issues (vapers report everything whether positive or negative, with an especially vocal response to negative effects).

Sensitivity to glycerine

About 1 in 1,000 are sensitive* to VG / glycerine / glycerol (glycerol is the most accurate of the common names) and must avoid it. However at this point we don't know if these people are sensitive to a particular type of VG or glycerol itself, since there are at least 5 different ways of making it (animal source, single vegetable source, mixed vegetable source [7], synthetic source, biodiesel byproduct, and more). If it seems worth the effort, other types could be tried, although the main problem at this time is identifying the source type. In theory a synthetic glycerine should be problem-free and this is the basis for its growing use in many inhalable medicines.

* It is worth mentioning that the term 'sensitive' may be the best way of expressing issues with these types of products, as it is non-specific: 'allergic to' or 'intolerant to' can be used, but may be inaccurate. In particular, determining whether a sensitivity is correctly defined as an allergy or intolerance appears to have some pitfalls at present: some define allergy as a reaction to proteins, others consider it a reaction to plant or animal-origin materials ('organic' by one meaning of the word), others don't.

Another base type: PEG

PEG or polyethylene glycol is a viable alternative base especially as it presents flavours better than either PG or VG. It has been extensively used in prefilled cartos for this reason, especially when supplied from the Boge-Dekang factories. However it is the most usual base type to experience contamination issues with DEG, a toxic glycol made in the same manufacturing facilities from similar feedstocks and therefore liable to cross-contamination issues. It can occasionally be contaminated with EG (ethylene glycol) when supplied in low-quality grades.

If PEG400 can be located in a high-quality form and efficiently tested as absolutely free from DEG or EG, it is a good alternative base material [8].

PG and VG safety issues

Propylene glycol has one of the oldest and largest research data volumes of any regularly consumed non-food material (i.e. pharmaceuticals, medical products, and industrial products with human contact). It has at least 70 years of research and has been used and consumed for over 70 years.

The first large-scale clinical studies were conducted in the 1940s, when PG was used to mist the air in hospital wards. It was discovered that PG is one of the most powerful bactericidal and virucidal aerosols in existence: hospital wards where the air was PG-treated in studies over several years noted that communicable diseases were reduced by 95% compared to control wards with no PG mist, including for diseases transmitted by virus (e.g. influenza, acute coryza). All airborne pathogens are reported killed by a PG level of just 0.5ppm. This effect has still not been explained even today, since at such low levels it is presumed that a PG molecule cannot even come into contact with all the pathogens that are killed.

PG is one of the glycol family, of which there are many variants. Glycols are an alcohol, not an oil. Some glycols are highly toxic, some are in effect inert in the human body and therefore harmless. DEG or diethylene glycol is toxic, and kills by irreversibly destroying liver, kidney and other organ function. A large scale poisoning that resulted in more than 100 deaths in 1937, due to DEG inclusion in a medicine, caused the original foundation of the FDA in 1938 - so the FDA owes its existence to a toxic glycol.

However, some of the family are harmless: PG and PEG being two examples of non-toxic glycols. PG is not actually inert, it is treated by the body as a complex carbohydrate; some is metabolysed to lactic acid, and it is all excreted. It is reported to have one-third the intoxication effect of ethanol, the alcohol in drinks; but the amounts inhaled in vaping are far too low to show this effect (commonly about a teaspoon full daily - obviously too little for any effect of this type),

Because PG is harmless, and because it has a very low freezing point (like all glycols), it is used as an antifreeze in human-contact and animal-contact applications instead of the cheaper, toxic DEG, which is used where toxicity is not an issue. For example it is used for winterising RV and boat lavatory systems, where human contact may be a possibility, instead of the toxic but cheaper glycols - it is completely harmless and its pathogen-killing capability is also useful; farmers can add it to water troughs in winter, since it will stop the troughs freezing and is completely harmless even if drunk in quantity.

Because PG is in effect inert, and is one of the best excipients available (a diluent or solvent for other materials), it is used extensively in medicines for inhalation and injection. It has been inhaled in asthma inhalers for decades, for example; in the nebulisers used by lung transplant patients; and as the excipient for injectable medicines that are immiscible (cannot mix with water): diazepam and phenytoin are examples. It should be fairly obvious that if PG is considered inert, is injectable in large quantities, is inhaled by patients in the most fragile health and with the most serious lung conditions, is drunk by animals without any consequences, and has been used for medical applications for 70 years, then there are few if any reasons to consider its use as having any measurable risk. There are of course many who are ignorant of these facts or are prepared to lie about the issue for payment.

Another common use of PG is in air-conditioning plant. Because it has all the required attributes, and in fact performs exceptionally well in this regard, it is used as the treatment agent in the cooling water. It is harmless, inert, acts as an anti-freeze, is very well-dispersed as an aerosol, and is an exceptionally potent airborne pathogen killer. This is why Legionnaire's Disease is not a risk in well-maintained large buildings: the PG-treated water in the aircon plant kills all airborne and water-borne pathogens. Because PG is so well-dispersed as an aerosol, it does its work both in the the water phase and the aerosol phase. As PG has a very low freezing point, it also prevents the rooftop water treatment plant from freezing up.

Since it is obvious that the Legionnaire's Disease virus is dispersed in the air of a building with poorly-maintained aircon plant and can infect widely as a result, and PG disperses very well as an aerosol, then it is reasonable to assume that everyone in large buildings inhales PG in microscopic quantities and has done so for decades. We can probably assume that hospitals are very well maintained in this regard.

PG has no health concerns regarding its use in any form, whether ingested, inhaled or injected. Some people find it dries out the upper airway tract, especially when newly exposed; this generally recedes with familiarity. A very few are intolerant to it, as may be expected with any/all materials of any kind; they may find skin issues result from over-consumption. Of course, we cannot know the result of long-term high-volume inhalation, which could conceivably lead to reversible COPD stage 1 issues or the like. Compared to the reported certain death of perhaps as many as half of all continuing smokers, and the reported ten years average loss of life expectancy, this would be a statistically invisible issue of complete irrelevance.

We are only aware of one genuine PG-related health issue: cats have an unusual blood chemistry and high-volume ingestion of PG is reported to damage the blood cells. This is why there is no PG in cat food, as against dog food (PG is a powerful humectant, which here keeps the food moist and prevents drying-out). On the other hand, tens of thousands of vapers with hundreds of thousands of cats in their homes have reported absolutely zero health impact on their pets from '2nd-hand vapour'; so we can assume that the minuscule amount of PG in exhaled vapour has no health impact at all for cats, otherwise we would certainly have seen this reported on the giant ecig forums with millions of posts on these types of issues. It is also an additional checkmark for PG and vapour safety: even if a common animal's blood can theoretically be negatively affected by a vapour constituent, there is so little present even in closed rooms that it is not an issue, and this has been tested by millions of user-years (at 2015 we must now have over 50 million user-years of experience with vaping products).

Glycerine, or glycerol as it may perhaps be better termed, to reflect the fact it is an alcohol and not an oil (see the suffix -ol), can be obtained from multiple feedstocks. In the past, vegetable-based glycerine was the best choice for vaping, from coconut oil or palm oil, or a mix; the best available now is the pure synthetic form of pharmaceutical grade glycerol (e.g. Dow Optim).

Because PG has an initial drying-out effect on the throat or upper airway tract for some people, the manufacturers of inhalable medicines (such as asthma inhalers) increasingly use glycerine as well as or instead of PG. The health impact is assumed to be the same: essentially zero. Glycerine is treated as a carbohydrate by the body, metabolysed, and excreted. No lung issues are reported for medicinal use, when using the pure pharma grade of course.

Pulmonary oedema
ARDS, oedema, or 'wet lung' was originally reported as a concern among VG users, presumably unaware that inhalable medicines are moving over to this excipient to replace PG. Prof Polosa reports that there has never been any instance of such a condition associated with vaping and therefore it is no longer discussed as a potential issue [10].

Of course, such issues would be identifiable in patients exposed for decades to medicines using these materials (e.g. PG as used in asthma inhalers), so if they cannot be identified by now as serious negatives then it is safe to say these are non-issues.

General safety issues
PG (chemical name: propane 1,2,diol) and glycerol (chemical name: propane 2,3,triol) are alcohols, and are not vastly dissimilar molecules as can be seen by their proper names. Inhalation of alcohols cannot cause lipoid pneumonia, as has been alleged by the chemically illiterate.

However, some think that inhalation of glycerol may exacerbate a tendency to (ordinary) pneumonia in those with emphysema caused by smoking. Thus, there is a case to be made that those with irreparably damaged lungs and already at risk for pneumonia, as emphysema patients are, should avoid high-VG refill liquids with their ecig. Actually they should of course be using Snus as their THR product of choice.

PG, VG (and all similar widely-used compounds) come in a variety of grades:

  • Industrial grade
  • Agricultural grade
  • Food grade
  • Pharmaceutical grade

The difference is mostly the amount of contaminants, although the feedstock type can be a factor. Although it seems obvious that only pharmaceutical grade PG or glycerine must be inhaled, as other grades are guaranteed to be contaminated, this does not seem to have percolated down to some e-liquid manufacturers.

In the case of glycerine, there is now an especially-important caution: biodiesel byproduct glycerine must not be used for refill liquid manufacture. This is because such feedstocks increasingly contain jatropha plant glycerine, which is toxic: the phorbol esters of the jatropha plant are carcinogenic. Unfortunately, although this issue has been widely publicised with regard to pharmaceutical glycerine products, it is difficult to see how jatropha glycerine can be absolutely excluded from the supply chain unless only synthetic glycerine is used (as, realistically, it should be exclusively now - this is not the 19th century).

Glycols and glycerol are alcohols

These materials, as can be seen from their -ol suffix, are alcohols and not oils. No matter how 'oily' they look - viscous, thick, gloopy (especially undiluted glycerine) - they are not oils and do not act either chemically or on the human organism as oils.

Glycerine viscosity

The concept of VG being 'thick', that is to say gloopy, or more correctly too viscous, is erroneous: it means that it has not been sufficiently diluted. Glycerine must be diluted by 10% to 20% before it becomes optimal for vaping. It can be diluted by 5% to 20% with distilled water before flavourings are added. The final viscosity of a correctly-made glycerine-based refill liquid (aka 'e-liquid') is similar to a PG-based one - if not, it has been mixed wrongly and needs correcting. If it is too viscous then it needs further dilution.

For example a correctly-made VG liquid can certainly be used in a carto tank system, which is the most demanding of correct viscosity due to the small feed holes in the carto wall within the tank system. The precise viscosity of the final mix determines the exact size of the feed hole into the carto, just as it does for a PG-only or PG/VG mix. Different 'PG' liquids (i.e. PG/VG mixes, commonly around 80/20) may need the hole smaller or larger to avoid flooding or dry hits; this is normal. It's why no-button pulls are sometimes needed: sucking on the driptip without operating the on/off switch, in order to pull more liquid through a system with a feed hole too small for the refill viscosity. If no-button pulls are frequently needed, the liquid viscosity is too high and it needs diluting, or the feed hole needs to be enlarged.

A suitable diluent can be anything thinner than the VG element of the liquid, and could be alcohol, PG or distilled water. There is no downside to adding 10% water, or even more: it is still effectively nebulised [9].


References for all the statements made on this page may be found in the literature and are mostly cited on the References page here - see menu at right. A small selection are given in the Notes below.



[1] PG = propylene glycol, the most common ecig refill excipient (diluent or base material). PG is completely non-toxic and has multiple medical licenses for injection, inhalation and ingestion. It is regarded as an inert excipient although it is treated by the organism as a carbohydrate.
VG = vegetable glycerine, used with PG or by itself.
PEG = polyethylene glycol, another suitable excipient.

[2] If you give smokers one option when supplying an e-cigarette and refill system as an alternative to smoking, the success rate for 100% conversion and retention at 12 months measured by clinical trials is as low as 10% (90% of the participants relapse to smoking). If a very limited range of options is presented, the success rate can rise to around 30%. In contrast, if an expert mentor provides support, using the full range of e-cigarette and refill products that are actually available, a success rate as high as 75% is reported anecdotally. A success rate this high normally requires a family / friends / work colleague environment, as constant expert advice is needed.

Apparently such methods cannot be used in clinical trials, which must use one hardware and one or two refill products, and therefore cannot measure the proper range of real-world results. This is why real-world results are far better than clinical trials appear to show. They must be, after all, or ecigs wouldn't work and the vaping community would not exist. We estimate that clinical trial results for ecigs need to be multiplied by a factor of 3 in order to get a realistic real-world equivalent, and even then for a worst-case scenario.

As an aside, mentoring is needed for best success rates in any kind of smoking avoidance or cessation attempt, and switching to ecigs is no different. Mentoring at least doubles the success rate in all attempts of any kind. This is especially true of vaping, which in addition needs an expert vaper as mentor in order to achive the best possible outcome. The range of hardware and combinations of it, and the range of refill liquids, are both very large - and the permutations are almost limitless. This is why vaping is so successful - and why it doesn't work well at all if options are limited.

The multiple reasons why RCTs cannot be used to 'measure the efficacy' of consumer products and especially THR solutions require not just one page but multiple pages of their own to fully explain. If it is not obvious to the reader why clinical trials cannot be used to measure critical aspects of consumer products then the work of Dr CV Phillips is recommended reading. Beginners and especially scientists will appreciate the value of such evaluations by a senior scientist and professor with decades of experience in public health, economics and THR.

[3] An ecig can be used in 'cigar mode' if required - this means to draw into the mouth, and expel from the mouth or the mouth and nose. Inhalation is not required if the refill nicotine strength is sufficiently high, because much of the nicotine absorption from e-cigarette use, unlike cigarette use, is through the buccal and nasal membranes (mouth and nose) as well as the lungs. Therefore, provided that the refill strength is sufficiently strong, and perhaps if the user has previously-existing lung issues such as COPD, there may not be a need to inhale (although of course inhalation works better to replicate the satisfaction of cigarette smoking, together with full nicotine delivery). At any rate, cigar smokers are well-served by this method - as long as the refill strength is sufficient. With some hardware, the refill may need to be as high as 45mg strength (4.5%). We have seen users employ up to 60mg strength (60 mg/ml or 6% nicotine content) in order to avoid both over-use and relapse to smoking.

[4] But note that 'quitter's throat' is a common symptom of smoking cessation: a very sore throat, possibly due to the cessation of inhalation of the anaesthetics in cigarette smoke. It is more common than for example the 'quit zits' (a symptom of smoking cessation that results in skin eruptions even in those who never had acne in youth). There is a possible issue here with sufferers from both the 'quitter's throat' together with a PG sensitivity: this uncommon co-occurrence can result in a painfully sore throat.

PG or propylene glycol is the most common refill base liquid. Normally, it has additions of glycerine, flavouring and nicotine if required, to comprise the refill in total.

[5] We still use the term 'VG' or vegetable glycerine as a general term for glycerines suitable for vaping, though strictly speaking it is now incorrect. Originally, glycerine was only available from two source feedstocks: animal or vegetable, and the vegetable form was preferable. Now there are multiple types, and the best choice is a fully-synthetic form such as Dow Optim, since it avoids (in theory) all allergy or toxicity issues due to its synthetic derivation and purity.

Note that there is now a new contender, biodiesel byproduct glycerine, and this must be absolutely avoided due to the possibility of the vegetable source having contained Jatropha plant material, which is toxic; this plant is commonly used in biodiesel production.

For more information, search: <fda biodiesel glycerine jatropha> or similar.

[6] PEG or Polyethelyene Glycol is one of the three commonly-used base materials for e-liquid. Its principal use is in pre-filled cartos. The product used is PEG400, which defines the molecular weight, or more simply its exact physical form: PEG400 is a liquid of the correct viscosity. PEG comes in many forms including a waxy solid, powder and liquid.

[7] For example palm oil, coconut oil, or a mix of the two.

[8] We understand from the factories that the transport barrels are a likely source of cross-contamination; inefficient (or zero) cleaning protocols between use for DEG and then PEG.

[9] The technically correct term for the active element within an ecigarette is a nebuliser, not an atomiser: an atomiser is a device where liquid is turned into a spray by an air jet, using a venturi tube / pump.

[10] [private communication] Prof R Polosa is a pulmonary consultant, a specialist in airway obstructive diseases, and regarded as the authority on medical topics concerning the lungs and vaping.



created 2013-10-01
update 2015-06-01