Material guide: dihydrocoumarin — one double bond fewer than coumarin, 40× the vapour pressure, 45 degrees lower melting point

About 5 minutes.

Short version

• Name: dihydrocoumarin (3,4-dihydrocoumarin, benzodihydropyrone), CAS 119-84-6
• Formula C9H8O2, MW 148.16; FEMA 2381, JECFA 1171, FLAVIS 13.009, CoE 535, FCC listed
• Odour: tonka type — sweet, tonka, coumarinic, coconut, herbal, cinnamon, balsamic, creamy, vanilla, spicy
• Strength: medium (smell at 10% or less), rank 2
• Substantivity: 228 hours; vapour pressure 4.0 mmHg
• Melting point 24–25 °C — the appearance field reads "colorless to pale yellow clear liquid to solid"
• Suppliers: 55
• cosmetic_uses: not used anymore
• Occurs naturally in white and yellow sweet clover, sour cherry and tarragon

One double bond, three fields change together

Coumarin and dihydrocoumarin differ by one hydrogenated double bond on the lactone ring. Put the two records side by side:

	Coumarin	Dihydrocoumarin	Change
CAS	91-64-5	119-84-6
Vapour pressure	0.1 mmHg	4.0 mmHg	40× higher
Melting point	68–74 °C	24–25 °C	~45 degrees lower
Substantivity	364 h	228 h	136 hours shorter
Suppliers	74	55
Odour type	tonka	tonka	same

Same odour type, and all three physical fields move.

They also move consistently: one double bond fewer, a ring that is no longer planar, worse intermolecular packing → the melting point falls, the vapour pressure rises, and substantivity follows the vapour pressure down.

(How reliably boiling point or vapour pressure predicts substantivity is measured in 1,560 materials have a boiling point and no substantivity, where the answer was correlated but not accurate enough. This pair is a same-family comparison, which is why it comes out so clean.)

4.0 mmHg is high for this database

There are 1,636 materials with ten or more suppliers and a readable vapour pressure. 4.0 mmHg ranks 207th — the top 12.7%.

What is genuinely rare is the combination with substantivity. Only 10 of the 1,636 (0.61%) have both ≥1 mmHg and ≥200 hours:

Material	Vapour pressure	Substantivity
Decanoic acid	15.00	336 h
BHT	15.00	400 h
(Z)-3-hexenal	11.20	240 h
Palmitic acid	10.00	400 h
Acetyl butyryl	10.00	220 h
Dihydrocoumarin	4.00	228 h
C-12 MNA	1.43	388 h
Lavandin oil	1.00	216 h

Read that table carefully. Vapour pressure measures how readily the neat substance evaporates; substantivity measures how long it can be smelled. They are not the same axis, and the substantivity field routinely loses its qualifiers (That "400 hours" for Ambroxan). What the table shows is that these two fields give inconsistent signals on these 10 materials, not that the materials are both volatile and persistent.

The practical reading is more conservative: do not use its vapour pressure to guess where it sits in a formula. By the fields it is a mid-to-late tonka note, and 4.0 mmHg would have you filing it as a top note.

Liquid or solid at room temperature, depending on your room

Melting point 24–25 °C.

The appearance field reads "colorless to pale yellow clear liquid to solid" — the database puts two states in one cell, because that range lands squarely on room temperature.

• Air-conditioned at 22 °C: solid or semi-solid
• An un-air-conditioned summer at 28 °C: liquid

This bites in practice: the same bottle looks different in different seasons, and you will suspect it has gone off. Telling apart what appears at the bottom of a bottle is Particles in the bottle — a melting point sitting on room temperature is the most harmless cause in that article.

(IPM sits near room temperature too, at 2–3 °C, and sets in winter. Opposite direction, same problem.)

Its regulatory status is cut in half

The regulatory field: JECFA, FEMA GRAS, CoE, FLAVIS, with FCC listed Yes.

The cosmetic_uses field: not used anymore.

One material, with the flavour line fully intact and the cosmetic line closed.

That is not a data-entry error. Of the 80 records marked "not used anymore", 26 still carry a FEMA number like this one and 54 carry none at all — and the pile without FEMA is chloroform, dichloroethane, phthalates, safrole and nitro musks. Cut by FEMA, those 80 split into two piles that mean entirely different things, which is Those 80 "not used anymore" materials.

So: you will still meet this one in a flavour context. Do not put it in a perfume.

That 2006 paper

Olaharski and colleagues published a paper in PLoS Genetics in 2006 that names it in the title:

"The flavoring agent dihydrocoumarin reverses epigenetic silencing and inhibits sirtuin deacetylases"

They screened environmental chemicals with a yeast heterochromatic derepression assay, asking whether humans are exposed to sirtuin inhibitors. The findings:

• Dihydrocoumarin disrupted heterochromatic silencing and inhibited both yeast Sir2p and human SIRT1 deacetylase activity
• In the human TK6 lymphoblastoid line it caused concentration-dependent increases in p53 acetylation and cytotoxicity
• Flow cytometry showed apoptosis increased more than threefold over controls

The paper itself describes it as commonly added to food and cosmetics — written in 2006.

Three things to be clear about:

1. This is yeast and cell lines, not a human risk assessment. It establishes that the molecule has this biochemical activity, not any health conclusion at any exposure.
2. I have no evidence that this paper caused its exit from cosmetics. The paper comes first and the cosmetic_uses marking after, but that is sequence, not causation, and I did not look into the actual basis for the removal.
3. It is still FEMA GRAS today. Twenty years after publication, the flavour line has not changed.

How to use it (in a flavour context)

• Evaluate at 10% or less.
• The coconut and creaminess are more pronounced than in coumarin. Alongside tonka and coumarinic, the odour field carries coconut, creamy, vanilla and cinnamon; the taste descriptions read "sweet, creamy, vanilla, coconut and milky".
• Do not treat it as a coumarin substitute. Same odour type, 40× the volatility, a different place in a formula.
• Mind the melting point sitting on room temperature. In winter, look at what state it is in before you weigh it.
• Do not use it in perfume or cosmetics.

What this doesn't establish

• I have not smelled it. The whole article is fields, database-wide counts and one paper.
• "One double bond fewer → lower melting point, higher vapour pressure, shorter substantivity" is an observation on this pair of records, not something I verified against more same-family comparisons.
• The vapour-pressure-versus-substantivity table mixes two different measurements, so it cannot be read as "these 10 are both volatile and persistent". It shows two fields disagreeing.
• Some vapour pressures are marked estimated, and I did not check which of these 10 are measured.
• The Olaharski paper is yeast and human cell lines, not a human study, and not a conclusion at any dose.
• I did not trace which regulation removed it from cosmetic use, or when.
• "FEMA GRAS is still current" is this database's field state; I did not check FEMA's own register.
• Acute toxicity: oral rat LD50 1460 mg/kg, intraperitoneal mouse 200 mg/kg, oral guinea pig 1760 mg/kg, from 1964 and 1974 literature. I did not compare these against current permitted food levels.