Material guide · 123
Material guide: methyl salicylate — the only oral-woman record in the database, and the paper that actually matters measured skin
· 17 min read
CAS 119-36-8, 88 suppliers, FEMA 2745. Wintergreen is this molecule — the odour field reads wintergreen, minty, sweet, root beer, phenolic. Its toxicity field lists eight rodent LD50s and then one line: "oral-woman LDLo 355 mg/kg", the only record labelled woman among the database's 156 human entries. But an ingestion figure is not what a perfumer should be reading. Martin 2004 measured the route that matters: eight subjects wearing eight commercial patches for eight hours reached a plasma peak of 29.5 ng/mL for methyl salicylate. Bell 2002 supplied the risk multiplier — broken skin plus an occlusive dressing. And its cosmetic_uses field carries an unusual value: denaturants. It is used to denature alcohol.
About 6 minutes.
Short version
- Name: methyl salicylate, the principal constituent of wintergreen oil, CAS 119-36-8, C8H8O3
- FEMA 2745, JECFA 899, FLAVIS 09.749, CoE 433, FCC listed Yes
- Odour type: minty; odour: wintergreen, minty, sweet, root beer, aromatic, phenolic, camphoreous
- Strength: medium (smell at 10% or less); substantivity only 8 hours; vapour pressure 0.0343 mmHg
- Assay: 99.00 to 100.00 — a one-point range, unusually clean
- Suppliers: 88
cosmetic_uses: denaturants; fragrance; perfuming agents; soothing agents- GHS: H302, harmful if swallowed (acute toxicity Category 4)
Its cosmetic_uses field carries an unusual value
Most fragrance materials read "fragrance" or "perfuming agents" in cosmetic_uses. This record has two more:
denaturants; fragrance; perfuming agents; soothing agents
A denaturant is something added to alcohol to make it undrinkable.
In Which grade did you buy I measured that ethanol carries seven grades, two of which are "completely denatured" and "specially denatured". This material is one of those denaturants. It has an odour, it is irritant, it is cheap, and it is very hard to separate from alcohol again.
So if you buy denatured alcohol, there may be a little wintergreen in it that you did not put there.
The ingestion figure, and the route that actually matters
The toxicity_oral field lists eight rodent LD50s — mice at 1110 to 1440, rats at 887 to 3049 mg/kg — with authors and years attached (Ohsumi 1984, Giroux 1954, Jenner 1964, Givaudan 1982…).
And then one more line:
oral-woman LDLo 355 mg/kg: LUNGS, THORAX, OR RESPIRATION: RESPIRATORY STIMULATION; GASTROINTESTINAL: NAUSEA OR VOMITING; BEHAVIORAL: COMA
That is the only record labelled oral-woman among the database's 156 human toxicity entries. (The full list is in 156 records carry human toxicity data.)
But that is a swallowing scenario, and a perfumer's question is about skin.
Martin 2004 measured the skin route
That is exactly what Martin and colleagues did in the Journal of Clinical Pharmacology in 2004. Their opening states the gap:
Camphor, menthol, and methyl salicylate occur in numerous over-the-counter products. Although extensively used, there have been no estimates of human exposure following administration via dermal application, and there is little information about the pharmacokinetics of those compounds.
The method is direct: three groups of eight subjects (four male, four female) applied 2, 4 or 8 commercial patches for 8 hours, with plasma assayed by gas chromatography.
The results:
| Camphor | Menthol | Methyl salicylate | |
|---|---|---|---|
| 8-patch Cmax | 41.0 ± 5.8 ng/mL | 31.9 ± 8.8 | 29.5 ± 10.5 |
| 4-patch Cmax | 26.8 ± 7.2 | 19.0 ± 5.4 | 16.8 ± 6.8 |
Eight patches over eight hours reach 29.5 nanograms per millilitre in plasma.
This is an answer with a number attached, and it is dermal, human and dose-graded. It is far closer to your situation than the ingestion figure — though a patch is still not a perfume, since patches are designed to promote absorption and perfumes are not.
(Camphor and menthol appear in the same experiment, and menthol has its own article here. Three materials in one study is an unusually direct cross-comparison for this database.)
Bell 2002 supplied the risk multiplier
Bell and Duggin reported a case in Emergency Medicine in 2002: a 40-year-old man became acutely unwell after herbal skin treatment, and the diagnosis was salicylate toxicity.
The mechanism they identify is the point:
Transcutaneous absorption of the methyl salicylate was enhanced in this case due to the abnormal areas of skin and use of an occlusive dressing.
The presentation was tinnitus, vomiting, tachypnoea and a characteristic acid-base disturbance.
Two multipliers: broken skin, and occlusion.
That has a direct formulation implication. How much a material is absorbed through intact skin is not how much is absorbed through eczema, a wound, or skin under a covering. And a perfume's wearer does not check their skin first.
(The paper also notes that methyl salicylate is widely available in over-the-counter topical analgesics and Chinese medicated oils.)
Wintergreen oil and the isolate
In the same database:
| Record | CAS | Suppliers | Odour type |
|---|---|---|---|
| Methyl salicylate (isolate) | 119-36-8 | 88 | minty |
| Wintergreen oil | 68917-75-9 | 56 | minty |
| Wintergreen flavor | — | 17 | — |
Wintergreen oil's occurrence field reads Gaultheria procumbens, and that oil's principal constituent is methyl salicylate.
Two records, 144 suppliers, essentially one odour. Choosing the isolate or the oil is a decision about cost, regulation and "natural" labelling, not about smell.
Its concentration in a natural can be very high
One figure in the occurrence field stands out: cassie absolute @ 94.40%.
That is not a trace constituent — in that absolute it is nearly everything. Cassie absolute has 30 suppliers in the same database, with an odour field reading fresh, sweet, spicy, clove, violet, green, weedy.
Other sources listed include bilberry, cassia bark, white wine, tea, porcini, Bourbon vanilla, clary sage and cherry.
How to use it
- Evaluate at 10% or less, as the field recommends.
- It is a top-to-middle material with only 8 hours. Do not expect it to hold the back end.
- Its direction is wintergreen and root beer, not mint. The odour type reads minty, but it is the salicylate kind of minty — unrelated to menthol's cooling, which travels the TRPM8 route this molecule does not.
- Keep it out of anything that will meet broken skin. Bell's mechanism is unambiguous.
- Mind occlusion. Patches, coverings and ointments raise absorption.
- Check IFRA. This material is restricted, and the toxicity field will not give you that number.
What this doesn't establish
- I have not smelled it. The whole article is fields, database-wide counts and two papers.
- Martin 2004 used commercial patches, not perfume. A patch is engineered to promote dermal absorption, so 29.5 ng/mL cannot be carried across to a perfume sprayed on skin. That is the biggest extrapolation limit here.
- That study had 24 subjects in three groups of eight.
- Bell 2002 is a single case report, involving an unregistered practitioner and an occlusive dressing — an extreme scenario, not ordinary use.
- The oral-woman LDLo is ingestion data, and I have not converted it into any skin dose, because that conversion does not hold.
- I did not investigate which denaturing formulations actually use it, despite the
cosmetic_usesvalue. - The cassie absolute figure of 94.40% carries no method or sample provenance, and natural compositions vary by nature.
- I did not check the current IFRA standard for this material.