Beginner perfumer · 6
IFRA limits the finished product, not the material — and when a beginner needs to care
· 16 min read
IFRA limits rest on a quantitative risk assessment for skin sensitisation, and they are set per product category — the same material has completely different ceilings in a candle and in a face cream. What the limits actually restrict, why they force structural changes to a formula, and when someone working only on smelling strips can ignore them.
Beginners react to IFRA in one of two ways: ignore it completely, or get too scared to use anything. Both come from the same misreading, the idea that IFRA bans materials.
What IFRA actually restricts is the concentration in a finished product, and the ceiling depends on the product category.
How the limits are derived
In 2008 Api and Vey set out the method in Regulatory Toxicology and Pharmacology: RIFM and IFRA adopted dermal sensitization Quantitative Risk Assessment (QRA) as the core strategy for primary prevention of sensitisation in consumer products, implemented through the 40th Amendment to the IFRA Code of Practice (May 2006).
The key detail: the approach sets exposure limits across 11 product categories. The paper walks through citral as its worked example, showing how an acceptable exposure level is derived.
So "what is the IFRA limit for this material?" is an incomplete question. You have to ask which product category first. Leave-on facial products, body lotion, a candle and a detergent differ enormously in skin contact area, contact time, and whether the product is rinsed off.
Which is why an online claim like "material X is limited to 0.02%" is missing half the sentence.
Why the limits exist
Numbers help. The 2016 EDEN Fragrance study by Diepgen and colleagues interviewed 12,377 people from the general population in Sweden, the Netherlands, Germany, Italy and Portugal, and patch tested a random sample of 3,119:
| Prevalence | |
|---|---|
| Positive to any allergen in the European baseline series | 27.0% (95% CI 25.5–28.5) |
| Fragrance Mix I | 1.8% (95% CI 1.4–2.3) |
| Fragrance Mix II | 1.9% (95% CI 1.5–2.5) |
Prevalence was significantly higher in women than in men.
1.8% and 1.9% are not alarming numbers, but this is the general population rather than a dermatology clinic. Across Europe that works out to several million people with a fragrance contact allergy. That is what IFRA is for. Sensitisation is cumulative and permanent: once sensitised, a person reacts at any dose, for life. Toxicity was never the point.
QRA goes after the induction step, the process of becoming sensitised; elicitation in someone already sensitised is a separate problem. That is why the whole framework is built around exposure.
Why it forces structural changes
Three reasons.
-
The limit applies to the finished product, not to your concentrate. If the ceiling is 0.1% in the finished product and your fragrance concentrate is 10% of it, you may use up to 1% in the concentrate. Beginners routinely confuse the two and end up either over the limit or needlessly far under it.
-
The restricted materials tend to be the high-impact ones, which is exactly the problem. Molecules with higher sensitisation potential are often also the loudest and most characterful. When a ceiling bites you cannot simply dial the material down; that leaves a hole, and you have to substitute or restructure.
-
Change the category and the arithmetic starts over. What you balanced for a body mist may be over the limit in a face cream, and no change of packaging fixes that.
What our data has — and what it does not
The part that matters most is what the records leave out:
Our compiled records note whether a material carries a register number — FEMA, JECFA, CoE, FLAVIS. They contain no limit values at all, and they do not cover IFRA, because we hold no structured IFRA data.
An earlier version of this article showed an IFRA coverage column. It was wrong and has been removed: that field was produced by searching each scraped page for the string "IFRA", which matches supplier blurbs and navigation, so it measured nothing about the material. For the actual limits, consult the current IFRA Standards; the Code of Practice is amended periodically, so use the official source.
What the data does show is which end of the palette the flavour registers cover, and that turns out to be enough to be useful. Across the 4,620 materials with an odour description and a listed supplier:
| Odour family | FEMA number | Any flavour register | n |
|---|---|---|---|
| Musk | 17% | 21% | 42 |
| Woody | 21% | 30% | 271 |
| Balsamic | 49% | 57% | 176 |
| Floral | 50% | 57% | 626 |
| Citrus | 56% | 61% | 259 |
| Fruity | 68% | 78% | 499 |
| Nutty | 74% | 81% | 81 |
| Caramellic | 76% | 76% | 67 |
| Sulfurous | 84% | 93% | 170 |
The line is sharp and it runs from musk to sulfur: a musk has a one-in-five chance of being a registered flavouring, and a sulfurous material has a five-in-six chance. Two thirds of the whole palette carries such a number, so the interesting part is where the exceptions cluster.
The sulfurous, nutty and caramellic end of the database is mostly flavour material. Go digging there and you will often find no supplier willing to sell a small quantity. Nothing has been left out; those materials were mostly not intended for fragrance in the first place.
My own first order included a couple of materials from that end of the table, chosen because the names sounded interesting. On a strip they smelled like snack seasoning, and I never touched them again. If you are making perfume, start from the families at the top of this table — the low-flavour-coverage end — and save yourself the detour.
So do you need to care?
Three situations.
Smelling strips only, for yourself: no. IFRA governs consumer products, and a strip involves no skin contact. Spend this stage training your nose.
Anything that goes on skin, including your own: yes. Sensitisation is cumulative and irreversible, and you are the first person at risk. The classic beginner mistake happens exactly here: dabbing the neat concentrate on a wrist to smell it.
Giving it away or selling it: absolutely, and according to the rules where you sell. The EU has written parts of the IFRA framework into law; other regions differ. At that point you have outgrown this article. What you need is the supplier's formal specification, the safety data sheet, and the current IFRA Standards.
One practical suggestion
Check early. Do not wait until the formula is finished.
The previous piece argued for adding the slow materials early, and regulation works the same way. Spend three weeks perfecting a structure, then discover the lead material is three times over its ceiling, and most of those three weeks are gone. Know the ceiling from day one and you can design so the formula still stands when that material shrinks.
A limit is a design constraint, to be known before you start alongside a material’s strength and its volatility.
References
A. M. Api & M. Vey, Implementation of the dermal sensitization Quantitative Risk Assessment (QRA) for fragrance ingredients, Regulatory Toxicology and Pharmacology, 52(1), 53–61 (2008). PMID 18635300
T. L. Diepgen et al., Prevalence of contact allergy in the general population in different European regions, British Journal of Dermatology, 174(2), 319–329 (2016). doi:10.1111/bjd.14167
Next: olfactory fatigue. Why you stop smelling something, how long recovery takes, and how that decides how many materials you can evaluate in one sitting.