Material guide: grapefruit oil — the word grapefruit does not appear in its odor field

CAS 8016-20-4, 95 suppliers. The odor field reads dry, citrus, bitter, aldehydic, terpenic, fruity, tropical, orange. Eight words, and not one of them is grapefruit. The flavor field is a single word: grapefruit. The optical rotation explains why. At +91 to +96 that is the signature of limonene, which is most of what this oil is. The two trace components that actually make it smell of grapefruit sit in the database as separate items, and both are better supplied than you might expect: (+)-nootkatone has 104 suppliers, more than the oil's own 95, and grapefruit mercaptan has 64 with a strength field asking for 0.10% or less. A 2001 Munich study quantified 25 odour-active compounds in hand-squeezed grapefruit juice by stable isotope dilution and concluded that the typical sulfurous, grapefruit-like quality comes mainly from two thiols. In 2016 the same laboratory synthesised 34 derivatives for a structure-odour study, 18 of them reported for the first time, and recorded 1-p-menthene-8-thiol's odour threshold in air at 0.000034 ng/L, which they describe as among the lowest ever reported for a food odorant.

About a 4 minute read.

Too long, didn't read

Name: grapefruit oil, properly citrus paradisi peel oil, CAS 8016-20-4, 95 suppliers
Odor family citrus; descriptors: dry, citrus, bitter, aldehydic, terpenic, fruity, tropical, orange. No grapefruit
The flavor field is one word: grapefruit
Optical rotation +91 to +96, specific gravity 0.848–0.856
Longevity 264 hours, shelf life 12 months, flash point 43.89 °C
GHS: H226 flammable, H315 skin irritation, H319 eye irritation, H335 respiratory irritation
The regulatory field holds only FEMA GRAS and CoE

The word is not in the odor field

Put the two fields side by side.

Odor field: dry; citrus; bitter; aldehydic; terpenic; fruity; tropical; orange
Flavor field: grapefruit

Eight words against one, and not one of the eight is grapefruit. The closest is orange.

The vanilla extract piece showed the opposite shape: one word in the odor field against seven in the flavor field. This is the inversion, and it raises a question worth asking. Does grapefruit oil not smell of grapefruit?

The answer is in the next field.

Optical rotation +91 to +96 says what it mostly is

This oil's optical rotation is +91 to +96. That is a large number, and dextrorotatory.

The database's d-limonene is a strongly dextrorotatory monoterpene hydrocarbon and the main component of citrus peel oils. An oil whose optical rotation sits above +90 is effectively saying that most of it is limonene. Specific gravity 0.848–0.856 agrees; that is the order of magnitude of a monoterpene hydrocarbon.

And what does limonene smell of? The database gives it the odor family citrus, with citrus, orange and fresh at the head of the descriptors. Not grapefruit.

So the eight words are honest. The body of this oil smells dry, citrus, bitter and orange, and the grapefruit character is not in the body.

The two that make it grapefruit are better supplied than the oil

The grapefruit-related items in the database:

Item	CAS	Suppliers	Odor family	Strength field	Longevity
(+)-Nootkatone	4674-50-4	104	citrus	medium	236 hours
grapefruit oil	8016-20-4	95	citrus	medium	264 hours
grapefruit mercaptan	71159-90-5	64	citrus	high, 0.10% or less	105 hours
grapefruit flavor	(none)	31	(empty)	(empty)	(empty)
grapefruit oil folded	8016-20-4	27	citrus	(empty)	(empty)
grapefruit acetal	67674-46-8	27	fruity	medium	2 hours
Dihydronootkatone	20489-53-6	8	citrus	(empty)	(empty)

(+)-Nootkatone has more suppliers than the oil itself, 104 against 95. Grapefruit mercaptan has 64, an odor field reading "sulfurous, aromatic, grapefruit", and a strength field asking for 0.10% or less.

That table says one thing: the characteristic smell of this fruit is sold separately. The oil supplies the volume and the two trace components supply the identity, and the identity is easier to buy than the volume.

25 compounds in hand-squeezed juice, and the character comes from two thiols

In 2001 Buettner and Schieberle at the Technical University of Munich ran a quantitative study (PMID 11312864).

Using stable isotope dilution assays, they quantified 25 odour-active compounds in the fresh, hand-squeezed juice of White Marsh seedless grapefruits, then calculated each odour activity value as the ratio of its concentration in the juice to its odour threshold in water.

The most potent were the fruity esters ethyl 2-methylpropanoate, ethyl butanoate and (S)-ethyl 2-methylbutanoate, plus the fruity, sweet wine lactone, the grassy (Z)-hex-3-enal, and the metallic trans-4,5-epoxy-(E)-dec-2-enal.

The key sentence comes after that. The typical sulfurous, grapefruit-like odour quality was mainly due to the catty, blackcurrant-like 4-mercapto-4-methylpentan-2-one and the grapefruit-like smelling 1-p-menthene-8-thiol. Reconstitution experiments confirmed it.

Two things are worth a pause. First, the most potent compounds, the esters, are not the ones supplying the identity. A high OAV does not decide which fruit this is. Second, 4-mercapto-4-methylpentan-2-one is the same material that appeared in the 3-mercaptohexyl acetate piece, the one with 47 suppliers that circulates only as a 1% solution.

0.000034 ng/L, and changing one thing makes it worse

In 2016 Schoenauer and Schieberle from the same laboratory followed it into a structure-odour study (PMID 27121638).

They open with the number: 1-p-menthene-8-thiol was discovered as the key odorant in grapefruit juice several decades ago and contributes to the fruit's odour with an extremely low odour threshold of 0.000034 ng/L in air. In the authors' words, this value is among the lowest odour thresholds ever reported for a food odorant.

Then they modified the structure, synthesising 34 mercapto-containing p-menthane and 1-p-menthene derivatives plus several aromatic and open-chain mercapto monoterpenoids. Eighteen are reported for the first time in the literature, with their odour thresholds, odour qualities and analytical data supplied.

Two clear rules came out:

Hydrogenating the double bond clearly increased the odour threshold (made it less sensitive)
Moving the mercapto group into the ring always gave higher thresholds than thiols with the mercapto group in the side chain

All the tertiary thiols showed low thresholds, but the comparison found none that replaced the original.

It is a clean demonstration: change one place on a molecule and the sensory scale moves an order of magnitude. One formula, 217 materials was about a formula throwing structure away; this is the same point from the front. The detail of the structure is the whole thing.

Buying and storing

95 suppliers, easy to source. The care goes into physical data and regulation:

Shelf life is only 12 months and the flash point is 43.89 °C. Limonene-dominated citrus peel oils are all short-lived and flammable, the same bracket as juniper berry oil. Decant, date it, keep it cold and dark.
Citrus peel oils raise phototoxicity. Citrus phototoxicity covers the mechanism. For anything going on skin, check the current IFRA standards and local rules; the database's regulatory field holds only FEMA GRAS and CoE.
If you want "grapefruit" rather than "citrus", the oil alone may not get you there. Nootkatone and grapefruit mercaptan are sold separately, the latter used below ppm.
Optical rotation +91 to +96 and specific gravity 0.848–0.856 are two fields you can check yourself on arrival.
Four GHS lines: H226, H315, H319, H335. Ventilate and keep away from ignition sources.

→ Grapefruit oil in the database: full physical data, 95 suppliers and suggested blends.

What this article does not establish

"An optical rotation of +91 means limonene dominates" is inferred from the properties of monoterpene hydrocarbons; the database gives no percentage composition for this oil.
I read the absence of "grapefruit" from the odor field as the body lacking that character, but there is no account of how or by whom that field was filled.
The 2001 study measured hand-squeezed juice, not peel oil. Juice and cold-pressed peel oil differ in composition, so the direction of the conclusion carries over but the numbers do not.
0.000034 ng/L is a threshold in air, not in water or alcohol, and not a threshold for this oil.
The two structure-activity rules are results from those 34 derivatives, not general laws about thiols.