Material guide · 9
Material guide: phenethyl alcohol (PEA) — why rose water still smells of rose
· 14 min read
Phenethyl alcohol is among the most abundant compounds above fresh rose blossom. Because it dissolves readily in water, distillation sends it into rose water rather than keeping it in the oil. The last piece of the rose trio, with 110 suppliers.
During rose distillation, most phenethyl alcohol follows the water phase. What looks like a loss from the essential oil explains why rose water still carries a full rose smell. Geraniol and citronellol supply more of the structure; phenethyl alcohol keeps petal, honey and water in the same impression.
What matters
- Phenethyl alcohol is among the main constituents of fresh rose blossom scent (headspace: 8.4–33.9%). It dissolves in water, though, so distillation usually loses it to the water phase, collected in the rose water, and little stays in the oil. Rose water's rose character is directly tied to it.
- How standard is it as a floral? fMRI research uses it as the floral control condition. Dried petal plus honey plus a touch of bread; gentle, never harsh.
- 110 suppliers, cheap, stable. The trio (+ geraniol + citronellol) completes here.
The record
| CAS | 60-12-8 |
| Formula | C₈H₁₀O |
| MW | 122.17 |
| Odour | floral, rose, dried rose, sweet, fresh, bready, honey |
| Strength | medium |
| Substantivity | 32 h (neat, on a strip) |
| Specific gravity | 1.015–1.023 (25 °C) — heavier than water |
| Shelf life | 24 months or longer (cool, dark, sealed) |
| Registers | GRAS, JECFA, CoE, FEMA GRAS, IFRA |
Plain reading: CAS is the material's ID number. This is the smallest molecule of the trio (C₈ against C₁₀) and the only one heavier than water. It also comes from a different family: geraniol and citronellol are monoterpene alcohols out of nature's parts kit, while PEA is built on a benzene ring, an aromatic. That ancestry is why it skips the terpene oxidation road of the last three guides. At 32 hours it leaves earliest of the trio (geraniol 60 h, citronellol 56 h).
The rose water mystery
In 2025 Antonova-Nedeltcheva and colleagues profiled fresh white-rose (R. alba) blossom in Molecules using headspace solid-phase microextraction (plainly: smelling the air around the fresh flower directly, no distillation involved):
2-phenylethanol was the most abundant component (8.4–33.9%), followed by geraniol (12.8–32.5%) and citronellol + nerol (17.7–26.5%).
And then the same paper turns the story around:
Being highly soluble in water, the 2-phenylethanol is usually lost in the distillation process (or collected in the rose water), and its concentration in the rose oil derived by traditional hydrodistillation is low.
One of the richest constituents of the living flower runs off into the water during distillation. Three products, three outcomes:
- Rose oil (distilled): citronellol and geraniol dominate, PEA runs low. The Kumar measurements in the citronellol guide describe this world.
- Rose water (the distillation's water phase): the paper's words are that PEA is "collected in the rose water". Connecting that to rose water's rose character is our interpretive step.
- Rose absolute (solvent-extracted, no water involved): keeps its PEA.
One more instance of the batch variation article's "same flower, different extraction, different answer", and this time the why has a molecular explanation: solubility.
Why research uses it as a floral control
The Hedione imaging study placed phenethyl alcohol in the common-floral control condition. That makes it useful as a baseline; it does not make it the only standard or prove an absence of effects. Experimental roles are explained in Why smell research keeps using the same odorants.
Smelling it
Our record gives it 7 descriptors: floral, rose, dried rose, sweet, fresh, bready, honey.
- Evaluate at 10%. Neat it is stuffy; diluted, the dried-petal character appears. I write times on my strips out of habit, and with a material that leaves this early the habit earns its keep.
- "Bready" is not a typo: a faint, warm, yeasty background is what separates it from fresh-cut rose, and why it works just as well in gourmand and honey accords.
- The trio's division of labour, final version: geraniol brings sweetness and brightness, citronellol brings wax and body, and PEA lays the dried-petal-and-honey undercoat. Start 5:3:2 and push PEA up when you want more petal.
- It keeps comparatively well. None of the oxidation baggage of the last three guides applies here, though the usual storage habits still cost nothing.
The catalogue family
| Item | What it is |
|---|---|
| phenethyl alcohol (110 suppliers) | the material itself. Catalogues also write phenyl ethyl alcohol / 2-phenylethanol — same thing |
| phenethyl acetate (68) | the acetate — lighter, fruitier; a standard rose-accord brightener |
| phenethyl isobutyrate / isovalerate (46 / 39) | fruit-toned variants |
| phenethyl salicylate (33) | the salicylate — denser, balsamic |
Regulatory
Listed on GRAS, JECFA, CoE, FEMA GRAS, IFRA. As always: registers carry no limit values. PEA is regulatorily uncomplicated; for skin formulae consult the current IFRA Standards and the supplier's specification.
Buying it
- Three spellings, one material: phenethyl alcohol / phenyl ethyl alcohol / 2-phenylethanol. Trust CAS 60-12-8, the Hedione guide's old lesson. My own first order nearly went in as three separate materials.
- Order with the trio; one shipment brings home the whole rose skeleton.
- The bigger bottle is safe here: cheap, used at percent level as a floral undercoat, and not an oxidation-sensitive case.
→ Phenethyl alcohol in the database: full properties, supplier list, suggested pairings. → Geraniol | Citronellol | Phenethyl acetate | Search by odour: try "rose honey" for its neighbours.
References
D. Antonova-Nedeltcheva, A. Dobreva, K. Gechovska & L. Antonov, Volatile compounds profiling of fresh R. alba L. blossom by headspace – solid phase microextraction and gas chromatography, Molecules, 30(15), 3102 (2025). PMID 40807275
I. Wallrabenstein et al., The smelling of Hedione results in sex-differentiated human brain activity, NeuroImage, 113, 365–373 (2015). PMID 25797832