« IFRA Reveals its Toxicological “Evidence” against Melissa Oil | Main | Robertet Reveals its Evidence on Melissa Oil to Cropwatch »

May 31, 2009

IFRA’s Proposed 44th Amendment. More Grief.

Copyright © Tony Burfield, May 2009

Update on Melissa Oil

You may recall the recent Cropwatch posting to Aromaconnection on a proposed IFRA restriction for Melissa oil, and the non-availability of the relevant evidence in the public domain. Following separate Cropwatch requests to the holders of the privately-held information (Robertet & RIFM), Catherine Gadras, in charge of the regulatory and safety department of Robertet, Grasse, has mailed promising to forward a summary to Cropwatch by 15th June 2009, in respect of the LLNA and HRIPTs  tests that have been conducted on behalf of Robertet, ‘in order to allow the use of this EO for perfumery use’. This is a welcome development. RIFM have not, as yet, either replied or acknowledged the request.

Further Points on IFRA’s Proposed 44th Amendment

We are looking below at three further proposed IFRA Standards under the forthcoming 44th Amendment, which have recently been circulated to IFRA membership groups for comment, with a 3rd June 2009 deadline.

Vanillin – Some Brief Notes

The first consideration is a proposed new IFRA Standard for vanillin. Readers will be aware that amongst flavour & fragrance ingredients, vanillin is possibly the most important aromatic aldehyde, with its easily recognisable & attractive powdery sweetness. It is available as a costly natural product via isolation from the vanilla pods of Vanilla planifolia G. Jacks, in which it occurs at up to 23,000 ppm, & via various biofermentation routes from natural starter materials (e.g. Rhodia have a microbiological biotranformation process using ferulic acid from rice bran). The production of vanilla itself was estimated at 2,000 tons in 2001 (Biolandes 2001), with 70% of the total production going to the US & Canada. Production rose to 3,600 tons in 2008 (Manceau 2009), but there are problems ahead, including pricing & compositional issues for vanilla from Uganda & Papua New Guinea, the effects of Madagascar’s political crisis, and from the damage caused by the fungi Fusarium oxysporum & Phytophthora spp. infecting Malagasy vanilla vines (see Gleason 2009; Manceau 2009). This is sufficiently serious that Dominiques Roques of Biolandes (through Gleason 2009) estimates a 1,200 ton/annum vanilla production loss from Phytophthora infection. Perfumery ingredients produced from Vanilla spp. (absolute, oleoresin, tincture, oil, CO2 extract etc.) are too familiar to describe in detail here. Vanillin also occurs as a minor component of a number of essential oils (e.g. star anise, clove bud & asafetida oils), and in absolutes, and balsams (e.g. Peru balsam, benzoin Siam, benzoin Sumatra).

The production volume of the cheaper & more easily available synthetic vanillin (which has previously run at approx 1% or less of the price of natural vanillin) has been estimated at about 6,000 tons/annum, & the material has been historically prepared from feedstocks such as guaiacol, catechol, ortho-dinitrochlorobenzene & lignin; nowadays synthetic vanillin is mainly derived from guaiacol and glyoxylic acid. Opdyke (1977) previously found vanillin to be relatively non-toxic, non-irritant & non-sensitising. The OECD SIDS report on >99% pure vanillin (20.08.1996) concluded that in animal tests, vanillin was sensitising in 5 out of 10 studies, but was not sensitising in the only test conducted under GLP. Vanillin was also said to be non-sensitising at 2% in maximisation tests carried out on 25 human volunteers.

According to information seen by Cropwatch, the true situation may be even more complex, since in trials with human volunteers >99% pure vanillin ex lignin was found to be non-sensitising, whereas vanillin ex guaiacol. or via the former ortho-nitrochlorobenzene process, provoked sensitising reactions in some individuals. Vanillin prepared from certain natural sources may also be slightly sensitising [of the 110 separate Vanilla spp., only 3 are cultivated: V. planifolia G. Jacks (Bourbon or Indonesian vanilla), V. tahitensis Moore (Tahitian vanilla), and V. pompona Schneide (Guadeloupe vanilla; vanillons; W. Indian vanilla). Eighty percent of vanilla production occurs in Madagascar; other producing areas include/have included Uganda, Papua New Guinea, Comoros & Reunion (the latter producing vanilla “Bourbon”), Java, Tahiti, Martinique, India (production hit by Fusarium infection), Sri Lanka, Tanzania & the Seychelles].

Cropwatch believes that there is more to learn about the alleged weakly sensitising properties of vanillin, and the effects of minor impurities, just as was about coumarin, although this has still to be recognised by the legislators.


clip_image002 clip_image004

Vanillin – Uses in Perfumery

Vanilla occupies a important position in perfumery, having been widely employed in formulations, especially as a key ingredient in orientals, for more than a century. First use in Jicky (Guerlain1889) was followed by Narcisse Noir (Caron 1912), Shalimar (Guerlain 1925), Old Spice (Shulton 1937), Opium (Yves St. Laurent 1977) & Lagerfield (Lagerfield 1978), Vanillin has also featured in more recent orientals like Joop! Femme (Parfums Joop 1987). Vanillin is also employed in florientals, & in modern perfumes like JP Gaultier's le Male, & in the class of Vanilla fragrances themselves which were very popular in the mid ‘nineties e.g. Vanilla Fields Coty 1993). Today vanillin is also key material in sweet foody type perfume notes e.g. toffee, chocolate and berry notes, such as strawberry.

Vanillin under IFRA’s 44th Amendment

Proposed Limitations for vanillin in the finished product under the QRA system fan out as follows:

Category 1 0.03 % (Lip products, toys, insect repellents)

Category 2 0.04 % (Deodorants/Antiperspirants)

Category 3 0.17 % (Hydroalcoholic Products for Shaved Skin, Eye Products, Men’s Facial Cream & Balms, Tampons)

Category 4 0.50 % (Hydroalcoholic Products for Unshaved Skin, Hair Styling Aids & Sprays, Body Creams)

Category 5 0.26 % (Women’s Facial Cream/Facial Make-up, Hand Cream, Facial Masks, Wipes/Refreshing Tissue for Hands, Face, Neck, Body)

Category 6 0.80 % (Mouthwash, Toothpaste)

Category 7 0.08 % (Intimate Wipes, Baby Wipes, Insect Repellent (intended to be applied to the skin)

Category 8 1.10 % (Make-up Remover, Hair Styling Aids Non-Spray, Nail Care)

Category 9 5.00 % (Shampoo, Rinse-Off Conditioners, Bar Soap, Feminine Hygiene Pads & Liners)

Category 10 2.50 % (Detergents, Hard Surface Cleaners, Diapers, Toilet Seat Wipes)

Category 11 “Should not exceed the usual concentration of the fragrance compound in the finished product”. (All Non-Skin or incidental skin contact products)

IFRA’s newly proposed restrictions under the 44th Amendment for the extremely weak sensitiser, vanillin, seem to be largely based on three reports, two of which are internal RIFM reports (and one of which is only in draft form). These are not freely in the public domain. These are as follows:

Basketter D.A., Wright Z.M., Warbrick E.V., Dearman R.J., Kimber I., Ryan C.A., Gerberick, G.F., White I.R. (2001). “Human potency predictions for aldehydes using the local lymph node assay.” Contact Dermatitis, 45, 89-94.

RIFM (Research Institute for Fragrance Materials, Inc.), 1970. Maximization study with vanillin. RIFM report number 1760, October 7. (RIFM, Woodcliff Lake, NJ, USA).

RIFM (Research Institute for Fragrance Materials, Inc.), 2009. Human repeated insult patch test. DRAFT REPORT. (RIFM, Woodcliff Lake, NJ, USA).

It is impossible for fragrance companies to approve or make comment on the scientific robustness of the evidence for making these restrictions, if they cannot see all the evidence. It would seem important therefore IFRA/RIFM to make these studies available in the public domain, especially since the newly reported evidence flies in the face of previous conclusions about the sensitising potential of vanillin. It is slightly unclear, too, whether the newly proposed IFRA Standard just refers to deliberately added vanillin in fragrance compounds, or to the total vanillin content of the fragrance (i.e. including contributions from vanillin-containing natural materials).

As a final point, many have written in to Cropwatch pointing out that the toxicological investigation / restriction of components which are found in natural complex materials, is being pointedly pursued, whereas the toxicology of closely related & commercially available synthetic materials is being ignored. In this particular case, no mention is made of the any investigation of closely related synthetic, ethyl vanillin. Good to see people are thinking for themselves, but previous investigators [e.g. Patlewicz et al. (2001) & Basketter et al. (2001)] found ethyl vanillin to be non-sensitising, which may rather deflate the argument! Other investigations which show a similar lack of breadth in the selection of natural & synthetic ingredients to investigate, include the studies made by Hagvall et al. regarding possible mechanisms for dermal sensitisation by linalol & geraniol (see updated Cropwatch article at http://www.cropwatch.org/The Trouble with Oxidation of Essential Oils.pdf). We have also been treated by the academics concerned, via the trade press & websites dealing with health matters, to opinions about what these studies indicate for the users of cosmetics containing linalol- & geraniol-rich essential oils. Regarding the linalol studies, to our knowledge no investigation has been made of the widely-used & closely related synthetic, ethyl linalol, and many have concluded, rightly or wrongly (& bearing in mind their reported remarks in the press) that these researchers are riding on an anti-naturals ticket. Cropwatch considers a more likely explanation is that the academics concerned have a limited experience of the cosmetics trade & the available choices of commercial aromatic ingredients.

Estragole (methyl chavicol)

The draft document showing the IFRA proposal for the restriction / prohibition of estragole to 0.02% in fragrance compounds looks like an unfinished piece of work. The grounds cited for the restriction / prohibition, are those of alleged carcinogenicity, but, somewhat surprisingly, no supporting evidence or references are supplied in the circulated draft of the new Standard.

The restrictions, if applied to the total estragole content of a fragrance compound, including naturally-occurring estragole from natural ingredients and not just to added estragole, will severely impact on the use of those essential oils in which estragole naturally occurs in cosmetic products. These include star anise (to 6.4%), exotic basil (to 90%), fennel sweet (to 6.4%) and tarragon (to 82%), as well as more minor amounts in bitter fennel, cananga & ylang ylang oils & absolutes, and the oils from certain Pinus spp. The point was also made by Cropwatch at the SCS Symposium (Burfield 2009) that limitations on substances like safrole, methyl eugenol & estragole have already had significant effects on the fragrance styles entering the marketplace - traditional aromatic masculine fougères and rich spicy notes are very difficult to achieve at the so-called ‘safe’ levels for these materials. There is little prospect of substitution either – the contribution of estragole, for example, to the odour profile of naturals and finished fragrances, is virtually irreplaceable. So here we have another prospect of IFRA further restricting the art of the possible in the fragrance art with the progressive introduction of their restrictive Standards.

So what is the evidence? Animal experiments using high doses of estragole have led to its classification as a possible weak genotoxic hepatocarcinogen (SCF 2001). Other expert committees have come to different conclusions. The FEMA Expert Committee concluded that dietary exposure to estragole did not constitute a cancer risk, and ventured that a non-linear relationship exists between dose, profiles of metabolism, and covalent binding of estragole to protein and DNA (Smith et al. 2002). We in the aroma industry do not need to be caught in the crossfire of differing toxicological opinions anymore – rather we need firm evidence that this same situation (of zero cancer risk) does not similarly apply to bio-available estragole from the application of estragole-containing fragrances to human skin.


Continuing the potential damage to the usage of natural aromatic products, IFRA are also introducing a new Standard limiting benzaldehyde concentration in fragrance compounds. Benzaldehyde is, of course, the major component in bitter almond oil, and is used to create almond and cherry notes in perfumes & flavours. Because of its pungency and odour character, it is also used in reodourants perfumes. Benzaldehyde is a minor component of many other natural products, including cinnamon leaf oil; cassia oil; cassie, narcissus & champaca absolutes; some cistus oils; clove oils & rosewood oil. Natural benzaldehyde is available from peach, cherry & plum stone processing, and via biofermentation routes e.g. starting from natural cinnamaldehyde ex cassia oil.

The grounds for the proposed restriction of benzaldehyde in perfume compounds by IFRA are based on the alleged weak sensitising properties of benzaldehyde, for which three references are quoted by IFRA:

Basketter, D.A., Wright, Z., Gilmour, N.J., Ryan, C.A., Gerberick, G.F., Robinson, M.K., Dearman, R.J., Kimber, I., 2002. “Prediction of human sensitization potency using local lymph node assay EC3 values.” The Toxicologist, 66(1-S), 240.

RIFM (Research Institute for Fragrance Materials, Inc.), 1973. Maximization study with benzaldehyde. RIFM report number 1802, October 11a. (RIFM, Woodcliff Lake, NJ, USA).

RIFM (Research Institute for Fragrance Materials, Inc.), 2009. Human repeated insult patch test. DRAFT REPORT. (RIFM, Woodcliff Lake, NJ, USA).

Again, the clincher for many of us in being able to judge the robustness of the scientific evidence necessitates the public availability of the draft RIFM report listed above.


In conclusion, these three IFRA proposals appear to be incompletely assembled and over-hastily produced. As we previously noted, until we know any further judgment from the EU legislators on the acceptability of the corporate-science styled QRA technique (following the SCCP’s severe criticisms in SCCP/1153/08), it would seem expedient to hold back on the implementation of this further set of IFRA Standards, if only to avoid unnecessary industry costs. Any communication on these matters from the authors of documents cited above, from RIFM or from the EU Cosmetics Commissioner, will be circulated by Cropwatch.


Basketter D.A., Wright Z.M., Warbrick E.V., Dearman R.J., Kimber I., Ryan C.A., Gerberick G.F. & White I.R. (2001) "Human potency predictions for aldehydes using the local lymph node assay." Contact Dermatitis 45(2), 89-94.

Biolandes (2001) – figures quoted in Biolandes Letter No 30 July 2001.

(Burfield 2009) – see http://www.cropwatch.org/Legislators & Natural Aromatics on PowerPoint.ppt

Gleason J. (ed.) (2009) “The state of vanilla: challenges & opportunities.” Perf & Flav. 34, 20-22.

Manceau M. (2009) “Thugs, Bugs & Vanilla.” Perf & Flav. 34, 24.

Patlewicz G., Basketter D.A., Smith C.K., Hotchkiss S.A.M. & Roberts D.W. (2001) "Skin-sensitization structure-activity relationships for aldehydes." Contact Dermatitis 44(6), 331-336.

Smith R.L., Adams T.B., Doull J., Feron D.J., Goodman J.I., Marnett L..J., Portoghese P.S., Waddell W.J. et al. (2002) “Safety assessment of alkyloxybenzene derivatives used as flavouring substances –methyl eugenol & estragole” – FCT 40,851-870.

Posted by Tony Burfield on May 31, 2009 in Essential Oils/Plant Extractions, Perfumery, Regulatory Issues, Safety/Toxicity | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference IFRA’s Proposed 44th Amendment. More Grief.:


The comments to this entry are closed.