Opinion Document to the IFA: a Brief Safety

Guidance on Essential Oils.

Copyright © Tony Burfield June 2004.

Version 4d Updated Sept 2004.

Essential oils are the steam-distilled volatile compounds obtained from aromatic vegetable matter. They are concentrated materials, and not intended for direct oral consumption. Keep essential oils out of reach of children, storing in cool/cold conditions in the absence of light, in clearly labeled, tightly stoppered bottles with integral droppers.

General Comments.

Safety has been defined before as freedom from danger, injury or damage (Burfield 1999). Although many essential oils are potentially hazardous materials, if handled in the appropriate manner, the risk involved in their use can be very small. So therefore, most commercially offered essential oils are safe to use for the purpose intended in a domestic/professional/clinical environment, if correctly used according to the producers specific directions or the recommendations of appropriate professional bodies. Detailed information on hazards, risks, emergency & first aid measures & detailed toxicological data can be located on the suppliers’ MSDS’s. Toxicological information on many individual essential oils in the related field of perfumery can be found on the IFRA website www.ifraorg.org. Tisserand & Balacs (2000) have written a useful book on The Safety of Essential Oils, and Martin Watt has produced a Safety Data Manual called Plant Aromatics  - details of which are found on his website http://www.aromamedical.com/paper.html

Use the appropriate personal protection for the situation in which you are working e.g. protective coat, goggles, safety gloves etc. Determine what is appropriate for your situation via a Safety Audit for your working environment (if in doubt consult your local Health and Safety Department). Remember that in any situation, you are your own Safety Officer! You have a duty of care for safety matters both to yourself and to others around you!

N.B. It is not generally appreciated that spilled essential oils wiped up with tissue/rags can autoxidise rapidly, especially in sunlight, posing a distinct combustion hazard. This has lead to a number of serious fires within the essential oil trade in recent years. All oil-containing waste should be placed in a closed metal bin, and the contents safely disposed of outside the building, on a nightly basis.

Methods of administration.

Essential oils can be administered orally, topically, vaginally, rectally or by inhalation, and different doses, degrees of absorption, metabolic processes and biodegrative outcomes between oil constituents occur between these methods. English aromatherapy practice has - up to now - been principally connected to massage (and therefore topical administration), and in many countries the National laws may not legally permit oral (or yet vaginal or rectal) administration except by a medically qualified person. Nevertheless any professional aromatherapy organisation with the word “International” in its title should have a working knowledge of all these areas. Aromatherapy massage using essential oil concentrations from 1-3% v/v in fixed vegetable oil is not unusual in UK practice.

Oils that are suspected carcinogens.

A carcinogen is a chemical which may give rise to tumor production, which is an unrestrained malignant proliferation of a somatic cell, resulting in a progressively growing mass of abnormal tissue. Do not confuse with mutagens which are substances which may cause heritable defects arising from their action on mammalian germ cells: here tumor formation may result from their action on somatic cells via cellular disruption. Whereas many mutagens are carcinogens, not all carcinogens are mutagens. Teratogens are different again, being substances that interfere with the normal development of either the embryo or foetus in utero, giving rise to abnormalities in the neonate.

The following carcinogenic essential oils should not be used in aromatherapy:

Birch tar oil crude (polynuclear hydrocarbons) Betula penda CAS No. 85940-29-0
Cade oil crude (polynuclear hydrocarbons) Juniperus oxycedrus CAS No. 8013-10-3
Some Croton oils such as C. tiglium & C. oblongifolius
Calamus oil (b- asarone type) Acorus calamus CAS No. 84775-39-3
Sassafras oils (safrole) Sassafras albidum CAS No. 84787-72-4
                                Ocotea cymbarum Brazil CAS No. 91770-39-7
                                Cinnamomum porrectum China

In the interests of balance, it should be mentioned that Guba (2002) has (bizarrely) argued that calamus should not be banned by the authorities, arguing that high dose levels are required to promote tumours in rats. N.B. The EC limits b-asarone levels in food to 0.0001g/Kg. IFRA limits cis- & trans-asarone in consumer products to 0.01%. There have been worries that estragole (aka methyl chavicol), a genotoxic carcinogen with a DNA-toxicity similar to safrole, might give rise to risk in aromatherapy practices. Drinkwater (1976) reported that estragole (methyl chavicol) has been previously found to be a rodent carcinogen and there was no safe “no effects” level. The German Bundesrat have made it illegal to market flavourings and foodstuffs containing added estragole after June 30th 2001. A draft position statement by the Working Party on Herbal Medicinal Products July 2003 made an opinion on estragole in herbal medicinal products according to http://www.emea.eu.int/pdfs/human/hmpwp/033803en.pdf They report that the limit of estragole in food flavourings recommended by the CEFS (2000) was 0.5mg/Kg, but that there is no data for chronic, acute and sub-acute exposure. They conclude that exposure from estragole consumed in herbal medicine poses no significant cancer risk, but exposure to estragole for children, pregnant and breast-feeding women is to be avoided.

Repeated exposure to methyl eugenol-containing oils might also pose a presently unquantifiable carcinogenic risk to aromatherapists (see separate section below).

Oils that are Toxic.

Acute oral (single dose) & dermal limit tests conducted by RIFM constitute much of our knowledge of essential oil toxicity; there is only a much smaller body of information on chronic- (6-30 month duration), sub-chronic- (up to 90 days), inhalatory - and immuno - toxicity. Some oils with LD50 values of less than 1g/Kg are recommended by IFRA not to be used in perfumery (Boldo, Mustard, Calamus, Chenopodium etc).

The following oils are recommended by the author not to be used in aromatherapy on toxicity grounds. The substances in brackets represent particular items of toxicological concern: 

Almond oil bitter* (hydrocyanic acid) Prunus amygdalus CAS No. 90320-35-7
Armoise oil (thujones) Artemisia herba-alba  CAS No. 84775-45-1
Boldo leaf (ascaridole) Peumus boldus  CAS No. 84649-96-7
Calamus oil (b-asarone type) Acorus calamus CAS No. 84775-39-3
Chenopodium oil aka Wormseed (ascaridole) Chenopodium ambrosioides CAS No. 89997-48-8
Croton oils with known toxicological properties, such as C. tiglium & C. oblongifolius
Horseradish oil (allyl & phenylethyl isocyanates) Amoracia rusticana CAS No. 84775-62-2
Lanyana oil (thujones) Artemisia afra 
Mustard oil (allyl isocyanate) Brassica spp. esp. B. nigra & B. juncea CAS No 90064-15-6
Parsley herb oil (dill apiole) Petroselenium crispum  CAS No. (USA) 8000-68-8
Pennyroyal oil (pulegone) Mentha pulegium  CAS No. 90064-00-9
Perilla oil (perilla ketone – lung toxin) Perilla frutescens  CAS No. 90082-61-9
Savin oil (sabinyl acetate) Juniperus sabina CAS No. 68916-94-9 – see below
Sassafras oil (safrole) Sassafras albidum  CAS No. 84787-72-4 – see below
[Savoury oil, summer Satureja hortensis is classified T- toxic in many inventories]
Tansy oil (thujones) Tanacetum vulgare CAS No. 84961-64-8
Wintergreen oil (methyl salicylate) Gaultheria procumbens CAS No 90045-28-6
Wormwood oil (thujones) Artemisia absinthium CAS No. 84929-19-1 

* Almond oil FFPA is normally traded = almond oil bitter Free From Prussic Acid (hydrocyanic acid).

Note that Sassafras oils (as safrole) are controlled under the Controlled Drugs (Scheduled Substances used in Manufacture)(Intra-Community Trade) Regulations 1993 and subsequent EU Directive 3677/90 as amended by Council Regulation 900/92 as a Category 1 substance. In other words, Sassafras oil cannot be bought or traded without registering with the Home Office (UK).

Other oils such as hyssop oil Hyssopus officinalis (pinocamphones) CAS No 84603-66-7, and parsley herb oil Petroselinum crispum CAS No. USA 8000-68-8 may also present some level of risk, which dramatically rises if taken orally. Oils such as wintergreen, calamus, almond oil bitter, pennyroyal and armoise are labeled R22: harmful if swallowed. Oils containing high levels of camphor (e.g. camphor oil brown Cinnamomum camphora) are not used in aromatherapy but Sage oils such as Dalmatian Sage Salvia officinalis may contain up to 30% camphor. In spite of the moderately low oral toxicity of camphor to rodents, Tisserand and Balacs (1995) recommend that Sage oils are not to be used in aromatherapy. It may be useful to remind here that the FDA in 1980 limited the level of camphor in consumer products to 11% in topical lotions and moth repellents and totally banned the sale of camphorated oil, camphor oil and camphor liniment (Burfield 1999).

In addition essential oils may contain harmful levels of pesticides, heavy metals, dioxins, polychlorinated biphenyls etc., and if sold as food flavourings, levels of these items are restricted (see EU Directive 88/388/EEC). 

N.B. In the interests of balance, Watt (see website info above) argues that Pennyroyal oil is safe to use in Aromatherapy practice; Guba (2002) argues that it is safe for inhalation purposes. However the principle use for pennyroyal oil is arguably as an insect deterrent, rather than as an aromatherapy massage oil. 

Oils that are Photo-toxic.

Photo-toxicity is light-related irritation, and involves percutaneous penetration & bio-distribution of a light-activated substance in the dermis, followed by skin exposure to light of the right wavelength and intensity. Therefore if photo-toxic oils are applied to the skin, and exposure to bright light/UV lamps/sunshine (especially at 312 to 320 nm wavelength) occurs over the next 12- 24 hours, photo-toxic contact dermatitis effects may subsequently occur, due to re-radiation of energy from the inherent furanocoumarin content or other constituents of the oils:

Ford (1991) lists these materials which cause photo-toxic contact dermatitis:

Angelica root oil Angelica archangelica  CAS No. 84775-71-7
Bergamot oil expressed Citrus aurantium ssp. bergamia  CAS No. 89957-91-5
Cumin oil Cuminum cyminum  CAS No. 84775-51-9
Fig leaf absolute Ficus carica  CAS No. 90028-74-3
Lemon oil cold pressed Citrus limon CAS No. 84929-31-7
Lime oil expressed Citrus aurantifolia  CAS No. 90063-52-8
Orange oil bitter Citrus aurantium CAS No USA: 68916-04-1
Rue oil Ruta graveolens CAS No. 84929-47-5
Tagete (oil & absolute) Tagete spp. CAS No. 91770-75-1
Verbena oil Lippia citriodora  CAS No. 85116-63-8

Other oils may also be problematic as they also contain bergaptenes:

Amni visnaga oil Amni visnaga
Grapefruit oil expressed Citrus paradisi  CAS No 90045-43-5
Mandarin oil cold-pressed Citrus reticulata  CAS No: 84929-38-4
Opoponax qualities Commiphora erythrea  oil, absolute, resinoid: CAS No. USA 9000-78-6
Parsley leaf oil Petroselinum crispum CAS No. USA 8000-68-8
Tangerine oil cold-pressed Citrus reticulata  CAS No. 93686-22-7

…or because they contain the photo-toxic compound methyl N-methyl anthranilate:

Petitgrain Mandarin oil Citrus reticulata var. mandarin CAS No: 72968-50-4

….or from the thiophene content (tagete oil above– also see addenda).

The IFRA Standard recommends that for fragrances not washed off the skin and subsequently to be exposed to bright light, the total level of bergapten (5-methoxypsarolen) should not exceed 15 ppm. Typical bergapten contents for a number of the above photo-toxic oils are set out on the IFRA website www.ifra-org.org.

Citrus oils FCF (furanocoumarin free) is sometimes recommended for aromatherapy, but these oils (usually distilled from the expressed oils) are organoleptically merely pale shadows of the original materials, and keep less well.

Methyl eugenol containing oils.

Methyl eugenol has been identified as a potent rodent carcinogen, and occurs in several essential oils, a few being:

Bay oil West Indian (Pimenta racemosa) CAS No. 85085-61-6………. to 12.6%
Basil oils (Ocimum spp.) CAS No. 84775-71-3 …. some chemotypes to 65%
some Melaleuca oils (e.g. Melaleuca bracteata) ………………………… to 50%
Nutmeg oil (Myristica fragrans) CAS No. 84082-68-8  ………………….. to 1.2%
Pimento oils (Pimenta officinalis) CAS No. 84929-57-7  ………………… to 15%
Rose oils (Rosa spp.) CAS No. 90106-38-0  …………………………………. to 3.0%

Burfield (2004) reviews the topic & discusses possible risks to aromatherapists.

N.B. Some aroma companies now offer low-methyl eugenol products (e.g. Charabot currently offer rose absolute with a 200 ppm max. methyl eugenol content).

Oils that cause Sensitisation.

Sensitisers can be dermal or respiratory. More usually in aromatherapy a sensitiser is a substance that causes dermatitis only after alteration (sensitization) of the skin by previous exposure to that substance. It involves the immune system; the following steps (permeation, metabolism, hapten production, antigen production) typify the skin sensitization process*:

1. The chemical permeates the dermis, and undergoes bio-distribution. Permeation is related to lipophilicity & molecular volume.
2. It is either metabolised by cutaneous enzymes or other processes to form a reactive metabolite, or often may be chemically modified through the reaction of UV light, or remains unchanged. The rate of conversion of pro-haptens to haptens is important in sensitisation. For example for the weak sensitiser eugenol is oxidised to a highly reactive ortho-quinone hapten (in mouse skin**) according to
Lepoittevin & Mutterer (1998):



3. These so- produced reactive haptens bind to dermal proteins (haptens are often electrophilic and can bind covalently with -NH2 groups and -SH groups on proteins, modifying the protein, which when presented to the immune system, will react with antigen-presenting cells in the dermis).
4. The Langerhans cells react with the allergen (the hapten-protein complex). They then migrate to the thymus.
5. The Langerhans cells teach T-cells to recognise the allergen and when they leave the thymus they are sensitised.
6. When they encounter the allergen they release lymphokines.

* After Lepoittevin & Mutterer (1998) & Burfield (1999).

** The author does not condone animal testing in any shape or form.

Under EC labeling laws, skin sensitisers are labeled Xi, R43. According to the IFRA Hazards Working Group opinion (June 2004), “quenching phenomena” effects can still be taken into account (according to Section 3.3 of the EU Dangerous Preparations Directive 88/379/EEC); however quenching phenomena effects between eugenol and cinnamic aldehyde are now unsupported according to the Notification No 4. of 38th Amendment to the IFRA Standard.

The following oils are responsible for severe sensitisation effects and should not be used in aromatherapy massage:

Cassia oil (cinnamic aldehyde, coumarin) Cinnamomum cassia CAS No.  84961-46-6                                                
Cinnamon bark oil
(cinnamic aldehyde) Cinnamomum zeylanicum CAS No. 84649-98-9
Costus oil, abs, concrete
(sesquiterpene lactones) Saussurea lappa CAS No. 90106-55-1
Elecampne oil
(sesquiterpene lactones) Inula helenium  CAS No.(USA) 1397-83-7
Fig leaf absolute
Ficus carica  CAS No. 90028-74-3
Massoia bark oil
(massoia lactone) Cryptocarya massoia CAS No.(USA) 85085-26-3
Melissa oil
(citral) Melissa officinalis  CAS No. 840082-61-1
non-IFRA compliant Oakmoss  Evernia prunastri  CAS No. 90028-68-4
Pseudoevernia furfuracea  CAS No. 84696-53-7
& Cedarmoss Evernia furfuracea qualities (resin acids)
qualities Commiphora erythrea CAS No. (USA) 9000-78-6
Oxidised oils
especially from Pinaceae (Pinus & Cypress spp.) and Citrus oils (hydroperoxides)*
Peru balsam & oil
Myroxylon pereirae CAS No. (USA) 8007-00-9
Styrax qualities
Liquidamber spp. CAS No: 232-458-4
                     resinoid: CAS No. (USA) 8046-19-3; oil: 8024-01-9

Verbena absolute & oil
Lippia citriodora  CAS No. 85116-63-8
Tea absolute 
Camellia sinensis  CAS No. (USA) 84650-73-4
Turpentine oil
Pinus spp. CAS No. (USA) 8006-64-2

*IFRA recommends for example that oils from the Pinaceae e.g. Fir needle oil Canada Abies balsamea should have a peroxide value of less than 10 millimoles of peroxide per litre.

Some fragrance houses internally restrict the use of bay laurel oil (Laurus nobilis) in their fragrances because of customer sensitization issues.

Predictably also these essential oils marketed to aromatherapists are also sensitisers:

Backhousia citriodora oil (high citral/citronellal content)
Inula graveolens
(sesquiterpene lactones)

N.B. Some aroma concerns offer low sensitiser ranges of oils where sensitizing constituents have been selectively removed by moved techniques such as spinning cone distillation.

The 38th Amendment to the IFRA Standard requires that manufacturers add 0.1% BHT or a-tocopherol to high-linalol containing oils such as bois de rose (rosewood), coriander and ho oils to prevent the build up of the skin sensitising substance linalyl hydroperoxide.

High-linalol containing oils:

Coriander oil Coriandrum sativum  CAS No. 84775-50-8
Ho oil
Cinnamomum camphora  var. linaloolifera
      & Cinnamomum camphora var. glavescens etc. CAS No. 91745-89-0

Kuromoji oil
(Linda umbellata, L. membranacea etc.)
Linaloe oils,
Indian & Mexican origins (Bursera aloexylon, B. simaruba & B.
etc.) CAS No. 92874-96-9

Petitgrain oil
terpeneless Citrus aurantium ssp. aurantium CAS No. 72968-50-4
Rosewood oil
aka Bois de Rose oil Aniba rosaedora  CAS No. 83863-32-5

Oils that cause Irritation.

An irritant is an agent which can cause cell damage if applied in sufficient concentration and for a long enough period. Immunological processes are not involved, and the chemical insult releases a potent vasodilator called histamine from mast cells producing erythma and increased vascular permeability, accompanied by eventual migration of polymorphonuclear leucocytes to the area. Dermatitis can follow without prior sensitization. Those with fair skin are more easily irritated, but the irritant reaction can also be shown to decline with increasing age, and to increase with increasing temperature, such that irritant dermatitis may only occur in some individuals in summer. The irritant must exceed a certain threshold to produce a reaction, but the dose response curve is less acute for allergens (Burfield 1999). Unlike sensitization, irritation reactions fade when the insult is removed.

The following oils listed below can cause irritation effects. It will be noted that a number of phenolic oils are contained in the list. Oils such as clove & Litsea cubeba are classified as R38: irritating to the skin.

Bay oil West Indian (Pimenta racemosa) CAS No. 85085-61-6
Clove oils
(stem, leaf, bud) Syzygium aromaticum  CAS No. 84961-50-2
May Chang oil
aka Litsea cubeba  CAS No. 90063-59-5
Melissa oil
Melissa officinalis  CAS No. 840082-61-1
Origanum oil
Origanum vulgare & other spp. CAS No. (USA) 8007-11-2
Pimento berry & leaf oils
Pimenta officinalis CAS No. 84929-57-7
Summer Savoury oil
Satureja hortensis  CAS No. 84775-98-4
Winter Savoury oil
Satureja montana  CAS No. (USA) 8016-68-0
Tagetes oil
Tagete spp. CAS No. 91770-75-1
Tea tree oil
Melaleuca alternifolia CAS No. 85085-48-9
Thyme oil
Thymus spp. CAS No. 84929-51-5
Turpentine oil
Pinus spp. CAS No. (USA) 9000-64-0

Aspiration Hazards.

Materials which can cause lung damage on ingestion are labeled Xn, R65, S62. This applies to essential oils and oil blends containing over 10% hydrocarbons (based on supplier information/analysis) and with a kinematic viscosity of less than 7 x 10-6 m2/sec and with a surface tension of 33mN/m @25°C (effectively most essential oils with hydrocarbon contents between 10% and 90%). Examples of oils labeled R65 are: 

Bergamot oil Citrus aurantium ssp. bergamia 55% total hydrocarbons
Cedarwood oils
(Cedrus spp.) …………………….60% total hydrocarbons
Copaiba oil
(Copaifera spp.)………………………..80% total hydrocarbons
Ginger oil
(Zingiber officinalis)…………………….90% total hydrocarbons
Manuka oil
(Leptospermum scoparium)………..70% total hydrocarbons.

Allergens according to the EU 7th Amendment to the Cosmetic Act.

An allergen is any substance that can trigger an inappropriate immune response, or allergy, in susceptible people. Common allergens include animal fur, dust, pollen and certain foods or medications.


Fragrance allergy is believed to occur in the general population at a level of  around 1-2% (Nielsen & Menné 1993), and which may be much higher - for dermatology patients - some workers indicate to 7-8%. A way of looking at this is that there is therefore a pool of pre-sensitised individuals already in existence who will react if an allergen is applied to their skin above a certain dose, this being individual to them. There is a requirement within the EU to label retailed cosmetic products (which includes fragrances) that contain fragrances which show concentrations of 26 identified allergens above a certain (very low) limit, according to product. This was derived largely from an SCCNFP opinion on fragrance allergy in cosmetics and non-food products intended for consumers in 1999, which originally listed these 26 allergens (16 of which are natural and are found in essential oils). These materials are alleged to cause skin sensitivity, or to be harmful in other ways, and many essential oils may be involved by this legislation – in fact many of the essential oils commonly used by aromatherapists to massage into the skins of their subjects will contain levels of several of the 16 allergens identified in SCCNFP opinion. These allergens will be applied to the skin in normal aromatherapy practice at concentrations which are considerably more (often by a factor of over 10 times) than the 0.01% limit identified for labelling under the 7th Amendment to the EU Cosmetics Act. A list of these natural allergens is set out below, with corresponding levels found in some essential oils (according to TB) given in brackets.

para-Anisyl alcohol (found in Vanilla tahitensis beans at 0.6% Ehlers & Fister 1997)
Benzyl alcohol
(to 4.5% in peru oil; in ylang ylang oils to 0.5%)
Benzyl benzoate
(to 78% in peru oil; also in ylang ylang & cinnamon leaf oils)
Benzyl cinnamate
(to 0.8% benzoin resinoid)
Benzyl salicylate
(to 5% in ylang ylang oil III; also in Cananga oil)
Cinnamic aldehyde
(to 88% in cassia oil & in cinnamon bark oil)
Cinnamyl alcohol
(54% in styrax oil)
(to 75% in lemongrass & Litsea cubeba oils; also in Melissa, Backousia citriodora etc)
(to 43% in geranium oil Chinese; also in citronella oils)
(to 65% in tonka bean absolute; in deertongue resinoid)
(> 90% in clove & pimento oils; to 85% in cinnamon leaf oil)
(isomers to 4.5% in neroli oil; 1% in rose)
(to 90% in palmarosa oil, also in geranium oils)
(<0.5% in ylang ylang oil extra)
(to 96% in sweet orange oil; & generally in citrus, mint & pine oils)
(98% + in rectified ho oil; also in coriander, linaloe, rosewood oils etc.)

Comment. This brings us into an interesting area. Palmarosa oil (Cymbopogon martinii) for example has a geraniol content of 80% - 90% and previously been found to be non-sensitising by RIFM. It is postulated by Leopoittevin & Mutterer (1998) that geraniol acts as a pro-hapten (see above) and is oxidized to the hapten geranial, which may be  responsible for sensitizing/allergenic effects. However there is an on-going discussion as to whether there were impurities or oxidation products in the synthetic items used for toxicological testing in the case of benzyl salicylate, coumarin and linalol at least, since pure items do not give these reactions. We thus may have a situation where erroneous (extrapolated) conclusions have been made by the dermatologists conducting the tests. This situation leads to doubts about the true allerginicity of essential oils containing these substances. But for now, aromatherapists should be aware of the issue of allerginicity and review their treatment protocols in the light of this information.

Hand dermatitis in Massage Therapists.

Crawford et al. (2004) investigated the self-reported and symptom based

prevalence of hand dermatitis in aromatherapy massage therapists in a 12-month study finding it to occur at 15% and 23% respectively (compared to reported rates of 2-12% in the general population). The study found significant associations between the reporting of dermatitis and use of aromatherapy products in massage oils and having a history of atopic dermatitis. Dorsal hand dermatitis (49%) as opposed to palmar dermatitis was most prevalent, the latter being typically associated with allergic contact dermatitis Interestingly the therapists themselves cited frequent hand-washing as the most significant factor.

Oils that have not undergone formal safety testing.

Many essential oils used in aromatherapy in a more pan-global context have not undergone formal safety testing, or if they have, the information is not in the public domain. Examples include those oils from Amni visnaga or from Catnip (Nepeta spp.), and even more universally from familiar oils such as Niaouli (Melaleuca quinquenervia) or Ravensara (Ravensara aromatica). Further, many oil chemotypes used in aromatherapy such as Rosemary oil verbenone type or Helichrysum italicum ssp. serotinum also remain untested.

Pregnancy & Effects on the Reproductive System.

Many essential oils are currently under examination for genotoxic effects. We already know that sabinyl acetate is embryotoxic, fetotoxic, teratogenic and abortifacient and is found in the following oils, which should not be used in aromatherapy:

Plectranthus fruticosa oil
Spanish sage oil
Salvia lavandulaefolia  CAS No. 90106-49-3
Savin oil Juniperus sabina  CAS No. 90046-03-0

Parsley herb and leaf oils and some chemotypes of the seed oil Petroselinum crispum contain dill apiole to 20% which is severely hepatoxic and high dose  levels have (endangering the subject) been used to procure an abortion.

Vitex agnus-castus berry oil.

Although some other essential oils may show a weakly estrogenic effect, Vitex agnus-castus berry oil is sometimes used for ‘hormonal balancing’ by aromatherapists: especially in post- and peri-menopausal women. There is evidence that diterpenes in the oil cause circulating female hormone levels to change, sometimes dramatically. The oil should therefore only be used under medical monitoring & supervision (Lucks B. 2003-4; Sorenson J. 2003).

Exposure to essential oils generally should be avoided/minimized during pregnancy, especially during the first trimester. Many of the components of essential oils, once they appear in the bloodstream, are probably capable of crossing the placenta. Since we are largely unsure of de-toxicification routes in foetal development, no “expert” is capable of guaranteeing 100% safety following short or long-term exposures to essential oils. 

Vegetable Massage Oils.

It is sometimes forgotten that skin sensitivity can be caused by vegetable carrier oils, just as it can by essential oils. Further, the aromatherapist must always be alert to possible subject allergy to nut oils (almond, arachis etc).

Interaction with medication.

Certain essential oils can cause problems with subjects taking medication, including those taking anti-coagulant and anti-depressive drugs. Until you have consulted your physician, or have otherwise sought expert advice, avoid undue exposure to essential oils.


As far as is known, this document has been assembled from current data sources which are believed to be accurate and reliable, but the author does not claim that the information above is complete, and this data is supplied without warranty, either expressed or implied, regarding its’ correctness or accuracy. It is the user's responsibility to determine safe conditions for the use of the essential oils described above, and to assume liability for loss, injury, damage or expense arising from improper use of these products.


Aka: also known as

CAS No: Chemical Abstracts Services number

CEFS: Committee of Flavour Experts

IFRA: International Fragrance Research Association

FCT: Food & Cosmetics Toxicology

LD50: The dose (usually expressed in g/Kg) that kills 50% of a group of matched animals (that were administered different doses).

MSDS: Material Safety Data Sheet

Quenching: a phenomena whereby in predictive human skin sensitisation human testing, the expected sensitisation effects of a fragrance allergen on skin contact are reduced or nullified by the simultaneous presence of another chemical.  

RIFM: Research Institute for Fragrance Materials

Note: all CAS Nos. quoted above for essential oils are in the form of a EU CAS No (= EC No.); if these were not available, a CAS Nos. (USA) was quoted.


Bertrand F. et al. (1966) “Skin sensitization to eugenol and iso-eugenol in mice: possible metabolic pathways involving ortho-quinone and quinone methide intermediates” Chem Res Toxicol 10, 335-343.

Burfield T. (1999) Safety Lecture to Aromatherapy Organisations Council. July 1999 available at: http://www.users.globalnet.co.uk/~nodice/new/magazine/magsafetylecture.htm

Burfield T. (2004) Cropwatch 3 – see http://www.users.globalnet.co.uk/~nodice/

Crawford G. et al. (2004) “Use of Aromatherapy Products and Increased Risk of hand Dermatitis in massage Therapists”. Arch Dermatol 140 (Aug 2004) 991-996.

Ehlers D. & Fister M. (1997) “Compounds of Vanillons (Vanilla pompona Schniede)” J. Essen Oil Res. 9, 427-431.

Ford R. (1991) The toxicology and safety of fragrances. In Muller PM, Lamparsky D. (eds) Perfumes, Art, Science & Technology, pub. Elsevier New York pp441-463.

Guba R. (2002) “Toxicity Myths: the Actual Risks of Essential Oil Use” pub. Centre for Aromatic Medicine (2002).

Leopoittevin J.P. & Mutterer V. (1998) “Molecular Aspects of Fragrance Sensitisation” in P.J. Frosch, J.D. Johansen, I.R. White (eds) Fragrances: Beneficial & Adverse Effects pub Springer-Verlag 1998.

Lucks, Barbara (2003-4): aromatherapy newsgroup mails.

Nielsen N.H. & Menné T. (1992) “Allergic contact sensitization in an unselected Danish population. The Glostrup Allergy Study, Denmark.” Acta Dermatol

Venereol 120, 33-36.

Sorenson, Janina (2003): aromatherapy newsgroup mails.

Tisserand R. & Balacs T. (2000) Essential Oil Safety – a Guide for Health Professionals. Churchill-Livingstone.

Addenda: Note on tagete oil photo-toxicity: Example of how experimental findings are translated into an IFRA Standard:

J. Meynadier (1983) at the 9th Essential Oils Congress in Singapore presented a paper showing that the photo-sensitisation properties of tagete oil is due to the thiophene content . RIFM’s subsequent unpublished findings regarding photo-toxicity (NOEL was believed to be 0.05%) for a long time lead to an IFRA guideline on the basis of powerful photo-sensitising effects, limiting the usage of tagete oil’s usage in perfume compounds to 0.25% under the 20% rule, where skin is exposed to sunlight (excluding bath prepns. etc). More recently Letizia C.S. & Api M.A. (2000) working with oils and absolutes derived from South African and Egyptian species of Tagetes minuta and Tagetes patula, published their findings on tagete oil’s powerful photo-sensitising effects confirming the NOEL at 0.05% - see Letizia C.S. & Api M.A. (2000) The Toxocologist 54(1) 397). Now IFRA Standards recommend that (by rephrasing data without reference to “the 20% rule”) tagete oil is limited to 0.01% final concentration in consumer products for application to skin exposed to sunshine, excluding soaps, bath preparations and other products which are washed off the skin.