Material guide: phenethyl alcohol — a median dose of exactly 10.00%, matching what Arctander wrote sixty years ago

The median dose of this material is 10.00%.

Not 9.8, not 10.3. I took the 274 formulas in a 954-formula public corpus that use it, sorted the doses, and the middle one lands on a round 10.00.

Arctander, in his materials book, puts the rate at which it enters compositions at "5-10-20% or sometimes much more".

That book came out in 1960. Six decades later, a body of formulas from entirely different sources has its median sitting on the middle of his three numbers.

The short version

• Appears in 274 of 954 formulas (28.7%), third most common, behind linalool (338) and benzyl acetate (317).
• Median dose 10.00%, interquartile range 4.0%-13.0%. 51% of formulas take it above 10%, 47 go above 20%, the heaviest reaches 64.4%.
• It is the largest ingredient by weight in 48 formulas (solvents excluded), with a median rank of 4.
• The database rates odour strength medium and substantivity 32 hours at 100%. Weak but long — that is why it can be dosed this heavily.
• Water solubility 22,000 mg/L, absurdly high for a fragrance material. Rose water is made of this.
• Roses build the molecule by two routes, and the dominant one changes with the season (Hirata 2016).
• Natural material is expensive because the compound is toxic to the microbes producing it (Bernardino 2024).

About 8 minutes.

How common, and how heavily dosed

First the position. In the 954-formula corpus, solvents excluded:

Material	Formulas	Share	Median dose
linalool	338	35.4%	5.93%
benzyl acetate	317	33.2%	5.00%
phenethyl alcohol	274	28.7%	10.00%
coumarin	225	23.6%	2.34%
citronellol	220	23.1%	5.94%

It appears less often than the two above it and weighs more every time it does. Among the ten most common materials, its median dose is the highest.

The distribution:

Dose	Formulas	Share of those using it
< 1%	9	3%
1-4%	54	20%
4-10%	71	26%
≥ 10%	140	51%
≥ 20%	47	17%

More than half of the formulas that use it take it past 10%.

That is rare in this series. We ran the numbers and most materials have a median dose below 1% — things at the level of irone and damascenone get counted in hundredths of a percent. Phenethyl alcohol is a different animal: it is the volume of the formula.

In 48 formulas it is the heaviest single ingredient. Median rank 4, so even when it is not first it is usually near the front.

Why it takes such a heavy dose

The database rates its odour strength medium. And as we've written, the strength label runs opposite to how much you can use: the stronger a material, the smaller the share it can occupy.

Substantivity is 32 hours at 100%. Weak, but lasting thirty-two hours — that combination sets its role.

Arctander is blunt about it: low cost, versatility, general acceptability on odour, and excellent stability all speak strongly in favour of what he calls a relatively weak odorant. He adds that it is almost never out of place in a composition, whether floral, balsamic, oriental, mossy, herbaceous, or aldehydic.

You can see that claim in the corpus. It turns up in 28.7% of the formulas, and those formulas run right across the genres.

The physical properties cooperate:

Property	Value
CAS	60-12-8
Formula	C₈H₁₀O, MW 122.17
Boiling point	219-221 °C
Melting point	−27 to −25.8 °C
Flash point	102 °C
Shelf life	24 months or longer
Suppliers	110
GHS	H302 (harmful if swallowed), H319 (causes serious eye irritation)

A 102 °C flash point is kind to shipping and storage, which is not the norm among the materials we've covered — cineole flashes at 48 °C.

What usually stands next to it

I ran co-occurrence: within the 274 formulas that use phenethyl alcohol, which materials appear more often than their baseline rate across the whole corpus.

Material	Co-occurrences	Lift
rhodinol	29	2.59×
phenylacetaldehyde	42	2.57×
cinnamic alcohol	63	2.24×
phenethyl acetate	33	2.21×
cyclamen aldehyde	38	2.07×
hydroxycitronellal	88	1.90×
geraniol	97	1.76×

That table is essentially the shopping list for a classical rose base. Rhodinol, geraniol, citronellol, phenethyl alcohol: the combination settled in the late nineteenth century. Hydroxycitronellal and cinnamic alcohol push it toward lily of the valley and balsam.

The second row is worth pausing on. Phenylacetaldehyde is the oxidation state of phenethyl alcohol, one step away. And as the next section shows, that step is the last one roses take when they build phenethyl alcohol. Perfumers put two molecules that sit adjacent on a biosynthetic pathway into the same formula, and mostly not for that reason.

Rhodinol, at the top of the lift table, only co-occurs 29 times. That combination is about specificity: people who use rhodinol almost always use phenethyl alcohol too, while most people who use phenethyl alcohol never touch rhodinol.

How a rose builds the molecule

Hirata and colleagues reviewed the biosynthesis of 2-phenylethanol in rose flowers in Bioscience, Biotechnology, and Biochemistry in 2016.

The method is clean: feed rose flowers deuterium-labelled phenylalanine (L-[²H₈]phenylalanine) and use GC-MS to follow where those deuteriums end up. The number of labelled atoms surviving tells you which steps the molecule went through.

The paper reports that the feeding experiments and subsequent GC-MS analysis revealed the hypothetical biosynthetic intermediates to [²H₈]-2PE, and that biochemical and genetic analyses elucidated the principal pathway.

Their recent finding is the one worth keeping: a season-specific pathway that produces [²H₇]-2PE from [²H₈]phenylalanine — one deuterium fewer, meaning a different route.

In their words, this is a unique example where the dominant pathway to a specific compound changes with the seasons.

The same rose builds the same molecule differently at different times of year. That has no direct bearing on formulation, and it does offer a candidate explanation for something people meet in practice: batch variation in rose absolute from one region and one cultivar may not come only from extraction conditions.

(We covered compositional swings in rose oil in the citronellol piece; this is a second angle on the same problem.)

Why the natural material costs what it does

Bernardino and colleagues reviewed biotechnological 2-phenylethanol production in Molecules in 2024.

They state it plainly: the compound can be obtained naturally from various flowers, but chemical synthesis is the most used route to meet market demand.

For a "natural" label, the route is different: microbial biotransformation of L-phenylalanine into 2-PE via the Ehrlich pathway, in yeasts and bacteria. Material made that way can be considered natural.

Then comes the bottleneck:

Due to the toxicity of the aroma to the producing microorganism, low production yields are typically obtained.

The yeast is stopped by what it is making. That is why yields stay low and costs stay high. A large part of the review deals with getting around it — producing strains, cultivation optimisation, strategies to mitigate product toxicity, and low-value feedstocks.

That sentence explains something else too. Phenethyl alcohol is itself used as a preservative in pharmaceutical products. Its toxicity to microbes is a known property rather than a side effect; in a fermenter, that property simply works against the manufacturer.

(This is the same class of thing as the ethanol antimicrobial point in today's other piece about white particles: a perfume already contains several things that suppress microbes.)

A practical quirk: it dissolves in water

The database gives water solubility as 22,000 mg/L at 25 °C (21,990 estimated, 22,200 experimental).

That is 2.2%. Preposterously high for a fragrance material — most are counted in single-digit mg/L.

This explains rose water. After rose distillation the layers separate: the oil phase is rose otto, and what stays behind in the aqueous phase, the rose water, is largely phenethyl alcohol. That water phase is traditionally re-extracted, because what it holds is worth recovering.

In practice:

• If you are making a water-containing product (hydrosol, mist, toner), phenethyl alcohol is one of the few fragrance materials that goes in without a solubiliser.
• If you are buying rose otto, note that its phenethyl alcohol content is low, because the material went into the water. That is partitioning, not adulteration.
• If you are making an alcoholic perfume, this property does not affect you, since ethanol dissolves it anyway.

What this doesn't establish

Arctander's "5-10-20%" matching my median of 10.00% is a nice coincidence rather than a validation. 74% of those 954 formulas come from patents or otherwise published sources, and their authors have very likely read the same book. These are not two independent observations.

The 10.00% median is a weight share of the concentrate, not of a finished product. Finished perfume is usually diluted to 10-20%, so what reaches skin is roughly 1-2%.

Co-occurrence lift describes appearing together, not interacting. Phenylacetaldehyde turning up alongside phenethyl alcohol says nothing about whether they reinforce each other in the air; that needs sensory testing. The biosynthetic adjacency is something I noticed afterwards, not an intention I can attribute to the formulators.

The Hirata paper is a review, not a primary experimental report. The deuterium-labelling detail lives in the papers it cites, which I have not read individually.

I cannot connect the seasonal pathway to specific absolute batch differences. It is a mechanistically plausible inference, not a measured association.

"Toxic to the producing microorganism" is the review's description, not something I measured. Inhibitory concentrations vary by strain; the review compiles them and I have not carried any number across.

The database lists both an estimated and an experimental value for that 22,000 mg/L solubility. They agree closely, and I have not measured it myself.

The GHS classification is what the database records — H302 and H319 — not a complete regulatory determination. Work from the SDS for your own batch and the current IFRA standard.

References

H. Hirata, T. Ohnishi, N. Watanabe, Biosynthesis of floral scent 2-phenylethanol in rose flowers, Bioscience, Biotechnology, and Biochemistry, 80(10), 1865-1873 (2016). PMID 27297332. doi:10.1080/09168451.2016.1191333

A. R. S. Bernardino, C. A. V. Torres, J. G. Crespo, M. A. M. Reis, Biotechnological 2-Phenylethanol Production: Recent Developments, Molecules, 29(23), 5761 (2024). PMID 39683919. doi:10.3390/molecules29235761

Related: citronellol, geraniol, how big is the palette, the starter palette map.