Cropwatch No: 3b
Methyl eugenol-containing essential oils.
© Tony Burfield May 2004
Worries about possible
risks due to the methyl eugenol content of natural materials – herbs,
essential oils - have surfaced in the recent past but there is a dearth of
information on the subject directly available in the public domain to
aromatherapists or complementary health practitioners. The following feature
is an attempt to add some background information to this subject.
The warm, musty-mild-spicy odoured aromatic compound Methyl Eugenol (aka eugenol methyl ether, or 4-allyl-1,2-diomethoxybenzene) is prohibited from being directly added as an ingredient to fragrances intended for retailed cosmetic products, due to worries about its’ potential carcinogenicity.
As it occurs naturally in
many essential oils and extracts, the addition of these ingredients is not
restricted outright, but on provision that the methyl eugenol content does
not exceed the following concentration in the following finished products
according to the IFRA standards (see www.ifraorg.org/):
Oral hygiene products
Non skin (as defined on
of 0.02% for the starred items applies to the concentration in the
In effect this means that there is an obligation on ingredient suppliers, under the requirements of due diligence, to supply information to customers, to make sure that they receive the necessary information in order for them to comply with the above requirements of the IFRA Standards. To spell this out in more detail, reporting the methyl eugenol content of the specific batch of the ingredient will then allow the customer to further calculate final levels of methyl eugenol appearing in the finished product. It is difficult to see how many small essential oil suppliers, without resort to internal analytical expertise, will be able to perform this function. Additionally, it is relatively easy to find plants for sale on the Internet, who’s essential oils contain high levels of methyl eugnol e.g. Black tea tree plants can be ordered at http://www.hotkey.net.au/~macs_oils/plant01.htm. No warning about the potential toxicity of methyl eugenol is presented.
It has long been established that methyl eugenol occurs in essential oils such as Canadian Snake root, Bay, Citronella, Laurel, Emodia, Fennel, Betel, “Brisbane Sassafras”, Pimento, Hyacinth etc., and its occurrence often coincides with the additional presence of eugenol (Poucher 1991). And so, purely as a guide, here below is presented a “snapshot” guide to the reported methyl eugenol content of several further essential oils.
Published data on Methyl Eugenol
Contents of Essential Oils.
1. FEMA have published data to members on methyl eugenol contents of essential oils (no geographic origins specified).
The BFA on 12.02.02 circulated BEOA data from 09.11.01 on the
methyl eugenol content of a number of analysed commercial oils. Oils were
classified by botanical name (no chemotypes were distinguished) and by
origin. There are no particular surprises, although methyl eugenol contents
on rose otto seemed low-ish compared with other published data, and the
range of methyl eugenol contents of the 23 basil oils (all apparently from
Egypt) was relatively large. No data on fennel oil (identified by the EU
Scientific Committee on Food as a dietary source of methyl eugenol) was
The BEOA data document makes comment that expert analysis of genuine
essential oils shows how widely essential oils vary in composition, and
makes comment that the BACIS commercial data-base of essential oils shows
methyl eugenol contents of 258 oils, that some of this data is misleading,
and not representative of genuine high volume essential oils used in
IFRA data on methyl eugenol contents of essential oils, as
presented on the IFRA website
in May 2004 does not define the plant source species, the geographical
origins of oils or any chemotype information. A document circulated by IFRA
(to members only – not in the public domain – but most of the information
the same as on the IFRA website) on April 6th 2004 lists 21
essential oils, again giving no botanical identification, only giving
geographic origins for two types of oils (citronella and rose), and giving
chemotype information for basil only. As has been observed previously by
this author, the standard of botanical reporting in IFRA documents, and in
EU legislation leaves a great deal to be desired.
A list of plants containing methyl eugenol, duplicating the species
names of many of the entries below, can be found on the Agricultural
Research Services data-base at
Table I - Various References re: Methyl Eugenol content of EO’s.
N.B. Question marks in the above table
appear when the author has had to make an intelligent guess at the botanical
origin of the oil because the original source failed to reveal it.
Remarks on the Toxicity of Methyl Eugenol.
methyl eugenol to the structure of safrole, a known carcinogen, have not
Following the RIFM/FEMA workshop in May 2000, the FEMA
expert panel issued a report entitled “Safety Assessment of
Allylalkoxybenzene Derivatves Used as Flavouring Substances – Methyl Eugenol
and Estragole”. This included a description of the 2-year bioassay with
methyl eugenol by the National Toxicity Program (NTP) whose aim was to
establish the carcinogenic potential of methyl eugenol regardless of route
of administration. The NTP study found that the present exposure to
methyl eugenol from food (mainly intentional addition of essential oils,
spices and spice isolates) presents no hazard to human health. The
report finds that although very high doses are carcinogenic, they are such
that they must have first induced a hepatotoxic effect. We also
subsequently learned from the RIFM/FEMA workshop write-up, that one serving
of pesto contains from 10-100 times the average daily human consumption of
methyl eugenol, although even this level was 100 times lower than the lowest
dose forcibly given to animals in the NTP assay.
The NTP technical study on the 2 year toxicology and carcinogenisis studies on methyl eugenol in F334/N rats and B6C3F1 mice was published in July 2000. It showed clear evidence of carcinogenic activity of methyl eugenol in the tested rodents, and can be viewed at http://ehis.niehs.nih.gov/ntp/docs/tr491/tr491abs.pdf
The German Bundesrat decided on May 11th 2001 not to market flavourings and foodstuffs containing added methyl eugenol (or methyl chavicol) after June 30th 2001, although this ruling did not apply to methyl eugenol naturally present in flavourings or foodstuffs.
EU’s Scientific Committee on Food expressed an opinion on methyl eugenol on
26.09.01, which can be viewed at
The committee remarked that methyl eugenol is a multi-site, multi-species
carcinogen, being both genotoxic and carcinogenic. Average
human intake from diet of methyl eugenol amounted to 13 mg/person/day and
the 97.5th percentile was 36 mg/person/day (on a body weight basis these
values correspond to 0.19 and 0.53 mg/kg bw/day, respectively).
The committee was unable
to establish a safe exposure limit.
Subsequently IFRA decided to severely restrict the limits of methyl eugenol in finished fragranced products in 2001 (36th Amendment to the Code of Practice).
Low methyl eugenol rose oil has been commercially offered by a small number of aroma houses. Removal of the methyl eugenol content by high vacuum fractional distillation seems to adversely affect the typical rose character in products offered. Removal of the methyl eugenol content by spinning band or spinning cone distillation may be more satisfactory, but production time is at a premium on this expensive technology. Rose oils naturally very low in methyl eugenol are known in Eastern Europe and further East, but the quality is very poor to actually unacceptable for most purposes, even before methyl eugenol removal.
As a closing comment, it is hard to see why the aromatherapy and cosmetic industries are “led by the nose” on the choice of available commercial rose qualities utilised, which merely reflect historical perfumery trade uses. It has previously been established that rose absolutes from varieties of garden roses can demonstrate beneficial cosmetic properties (Étienne et al. 2000) whereas a conventional commercial rose absolute showed none of these effects. Further, it is likely that certain of these other varieties will only present a fraction of the methyl eugenol levels encountered in conventionally sourced rose ottos and absolutes.
Methyl Eugenol in Aromatherapy.
The author is unable to find any detailed advice given by professional aromatherapy organisations to members on this issue, on a par with that put by IFRA for its membership in the perfumery profession. Harris (2002) has reviewed the position of methyl eugenol in aromatherapy practice in the light of IFRA restrictions in the fragrance industry. It is worth exploring a number of points.
Harris notes that the IFRA have published a list of essential oils (e.o.’s)
with methyl eugenol contents, commenting that these figures only pertain to
oils used in the fragrance industry. Harris instead quotes e.o. data from
Lawrence (1998-2002). However IFEAT have previously criticised the use of
Lawrence’s data (specifically over the separate 26 allergens issue), as they
maintain it is relates only to experimental data and does not relate to the
composition of commercial oils. In any case, in the real world, the e.o.’s
distributed by many (but certainly not all) aromatherapy oil suppliers are
identical to those distributed by the fragrance industry.
Harris further maintains
that “the average aromatherapy treatment regime consists of 5-10 sessions,
given at most once per week, generally with the essential oils employed
being changed during this regime according to the improvement of the
client…”, and goes on to state, “those most at risk from methyl eugenol are
the aromatherapists themselves”, but does not investigate exposure of this
most “at risk” group in any satisfactory detail. Harris further mentions
avoidance of high methyl eugenol containing oils by therapists, and talks of
using “3 drops in a blend” – which, as several professional therapists have
privately commented, “is not
Since aromatherapy is a poorly paid profession, many professional aromatherapists are obliged to work extended hours, and may have to perform 6-8 massages per day, 5-6 (or more) days per week. Further, a whole body massage may well be carried out with 20-50 mls of massage oil containing 2-2.5% e.o., although some practitioners apparently have been known to use even higher concentrations (Guba 1998). Unknown amounts of methyl eugenol are therefore absorbed by the therapist throughout the week, via skin absorption through the hands, and by inhalation of vapour. Harris doesn’t mention the fact that diet is additionally adding to the therapist’s body burden of methyl eugenol.
The above factors may eventually allow a more realistic calculation of daily human body loading from methyl eugenol for aromatherapists, but interpretation of the data revolves around interpretations of the NOEL (no-effects) level in the longer-term and appropriate safety factors (IFRA used a factor of 1000 X). Since aromatherapeutic treatments such as whole body massage are vastly different from animal dosing studies, drawing direct conclusions about possible toxicological effects is distinctly risky. Further, it is already known from human liver microsomal preparations that metabolism rates by human cytochrome P450 isozymes for methyl eugenol varies more than 37-fold (Gardner et al. 1997) suggesting a wide range of serum concentrations will occur in the general population following methyl eugenol exposure.
Meanwhile Schecter et al. (2004) have produced a study on human consumption of methyl eugenol and its elimination from serum under a mandate from the National Toxicology Program of the US Department of Health and Human Services. In particular the team investigated the consumption of methyl eugenol from a brand of gingersnaps, found to contain a relatively high concentration of methyl eugenol at 3.3mg/g (a number of other foodstuffs containing lower concentrations of methyl eugenol are also listed in the article & cigarette tobacco’s were identified as another possible source of methyl eugenol exposure!). Serum peak levels of methyl eugenol were found to be within range of a concurrent study of 213 non-fasting subjects in the third Nutrition Examination Survey (NHANES III, 1988-1994). However in this latter study, the authors found that methyl eugenol levels in the blood of the general US population were higher than expected (but the highest concentration found, 390pg/g, was still 2000 X lower than the lowest dose used in the NTP rodent studies referred to above). Nevertheless, as Schecter et al. remark, the significance of the elevated levels with respect to any toxicological consequences, still remains to be determined.
It may well eventually
turn out that a working aromatherapist, constantly using basil and rose
oils, and with a fondness for pesta and flavoured cigarettes is more likely
to be hit by a meteorite than to contract a toxicological problem due to
daily methyl eugenol exposure from all these routes. Its just that it would
be nice to think that those entrusted with a duty of care towards working
people in our society were actively investigating this topic. The situation
being as it is, assessments on this topic are more likely to be made by
self-educated laymen, than by formerly qualified toxicologists – and to this
end, Cropwatch has written to some toxicologists for some learned
opinions on this matter. Any replies will be published in further editions
of this organ.
Professor Arnold Schecter (see reference above) kindly read my piece on methyl eugenol above and hinted from the tone of the article above that I might have understated the risk slightly, commenting further as follows:
“What my work
followed during a year I worked at NIH was that methyl eugenol is
extremely carcinogenic to rodents and causes cancers in rats and mice, two
species, and in multiple tissues. The human levels may or may not be of
concern, both those we reported and the higher levels we alluded to in the
general US population, so high for unknown reasons. ME does not occur by
itself in humans but in combination with many other toxic chemicals so
potential human health effects might be from ME alone or in combination
And further: “In
combination with other carcinogens (methyl eugenol) might be harmful at
lower levels than those derived dosing animals with one chemical only.
Many chemicals in our bodies.”
I take these points
on board, and suggest even louder now, that the aromatherapy profession
needs to take this issue seriously, perhaps appealing for outside help to
more properly evaluate the risk.
for printer friendly version
BFA: British Fragrance
Federation of Essential Oils and Aroma Trades
IFRA: International Fragrance
Table Data References:
Aurore, G. S. Abaul, J.
Bourgeois, P. Luc, J. (1998) “Antibacterial and Antifungal Activities of
the Essential Oils of Pimenta racemosa var. racemosa P. Miller
(J.W. Moore) (Myrtaceae).” J. Essential Oil Res. 10(2),
BEOA: British Essential Oils Association 9th Nov 2001 – data
reproduced by kind permission.
Brophy JJ: Brophy JJ (1999)
“Potentially Commercial Melaleucas” in Tea Tree – the Genus Melaleuca
eds. Ian Southwell & Robert Lowe. Harwood Academic Publishers.
Brophy et al: Brophy et al.
(1999) J Essen Oil Rec 11, 327-332.
Duke: Duke J (?) from
Chemicals and their Biological Activities in: Peumus boldus MOLINA (Monimiaceae)
– Boldo – see
EOS: “Essential Oil Safety” Robert
Tisserand & Tony Balacs Churchill-Livingstone 1996.
F & P: Franchomme P. & Peneol D (1995) “l’Aromatherapie
Exactement” pub. Jollois. R.
Guba R (1998) “Toxicity Myths –the Actual Risks of Essential Oil Use.” Centre for Aromatic Medicine 1998.
information as at 01.05.2004
IFRA: Annex 1 IFRA Standards.doc April 6, 2004.
IS: Ian Southwell (1999) “Tea
Tree constituents” in Tea Tree – the Genus Melaleuca eds. Ian
Southwell & Robert Lowe. Harwood Academic Publishers
Kam: Kameoka H. (1993) “The
Essential Oil Constituents of Some Useful Plants from China” in Recent
Developments in Flavour & Fragrance Chemistry –Proceedings of the 3rd
Int. Haarman & Reimer Symposium Pub. VCH NY 1993.
Lawr.: Lawrence BW (1989) EO’s 1981-7 Allured Publ.
Lawr. a: Lawrence BM et al.
(1985) Perf & Flav 10(6), 56-58 Dec 1985-Jan 1986
Mazza G. (1983) “GCMS
Investigation of Volatile Components of Myrtle Berries” J. Chromatog.
SCIB: Zhu Lianfeng et al. (1993) Aromatic Plants & Essential Constituents South China Inst of Botany, Hai Feng Publishing Co.
Shiva et al: Shiva MP, Lehri
A, Shiva A. (2000) Aromatic & Medicinal Plants pub IBD 2000.
TB: Tony Burfield (2000) Natural Aromatic Materials: Odours and Origins pub. AIA Tampa.
TBb: Tony Burfield
TBc: Tony Burfield & Sylla
Sheppard-Hanger (2002) “Basil Oils Monograph” AIA UK 2002.
TQ: trade suppliers
questionnaire (IFF 2003)
Poucher (1991) Poucher’s Perfumes, Cosmetics and Soaps - Vol 1 The Raw Materials of Perfumery 9th edn. Blackie Academic & Professional.
Zhu Liangfu et al. (1993)
Aromatic Plants & Essential Constituents South China Inst of Botany.
Étienne et al. (2000) “New
and unexpected cosmetic properties of perfumes. Effects upon free radicals
and enzymes induced by essential oils, absolutes and fragrant compounds.”
International Journal of Cosmetic Science 22, 317-328.
Gardner et al. (1997)
“Cytochrome P450 mediated bioactivation of methyleugenol in Fisher 344 rar
and human liver microsomes.” Carcinogenesis 18, 1775-1783.
Harris B. (2002) “Methyl
eugenol – the current bete noir of aromatherapy”. Int. J. of Aromatherapy
Lawrence B.W. Progress in Essential
NHANES III 1988-94 National
Centre for Health Statistics (1994). Plan and Operation of the Third
National Health & Nutrition Examination Survey, 1988-94. Series 1:
Program & Collection Procedure No 32.
Schecter A et al. (2004)
“Human Consumption of Methyleugenol and Its Elimination from Serum”
Environmental Health Perspectives 112(6), 678-680.