Beginner perfumer #116: optical rotation, the one field you can verify yourself after buying

Only 359 materials in the database (0.9%) carry an optical rotation, but the information density in those 359 is high. The deterpenation ladder is perfectly monotone across five citruses: lemon goes from +57~+65.5 to -8~-10 when deterpenated, sweet orange from +94~+99 to +78~+91 folded and +11~+24 terpeneless, lime from +34~+47 to -9~+2 folded and -9~-11 terpeneless, grapefruit from +91~+96 to +15~+22. The substantivity field confirms the same mechanism independently: sweet orange runs 140 hours whole and 212 deterpenated. Following that line turned up two errors. laevo-menthol has 114 suppliers and an optical rotation of -1 to +2, while every other laevo- or dextro- material in the database agrees with its own name, including menthol's own acetate ester at -70 to -75. And lime oil expressed reads +35.00 to -41.00 where its Mexican sibling reads +41.00 to +35.00; that minus sign is a typo.

About a 5 minute read.

The short version

Only 359 materials (0.9%) carry an optical rotation
But it is the one field in this database you can verify yourself after buying
The deterpenation ladder is monotone across five citruses (terpene fraction > whole oil > folded > terpeneless)
The substantivity field confirms it independently: sweet orange runs 140 hours whole and 212 deterpenated
43% all positive, 33% all negative, 24% spanning zero
Two errors found: laevo-menthol reads −1 to +2 (114 suppliers), and lime oil expressed reads +35.00 to −41.00

Why this field is different

Most of the fields I have audited over dozens of rounds cannot be checked by you. Substantivity takes a month, the odour field is somebody's subjective description, and 96% of vapour pressures carry (est).

Optical rotation is not like that. One polarimeter, one sample tube, five minutes. And what it measures is molecular three-dimensional structure, which is where odour actually comes from and one of the harder things to fake.

The catch is scarcity: 359 records out of 42,225.

Five citruses, one ladder

Last round I found that lemon's rotation flips sign on deterpenation. This round I lined up all five citruses:

	Terpene fraction	Whole oil	Folded	Terpeneless
Lemon	—	+57 ~ +65.5	—	−8 ~ −10
Sweet orange	+92 ~ +105	+94 ~ +99	+78 ~ +91	+11 ~ +24
Lime	+45 ~ +57	+34 ~ +47	−9 ~ +2	−9 ~ −11
Grapefruit	+95 ~ +100	+91 ~ +96	+62 ~ +73	+15 ~ +22
Bergamot	+44 ~ +55	+8 ~ +24	—	—

Every row is monotone. Terpene fraction highest, whole oil next, folded lower, terpeneless lowest.

The reason is that paragraph of notes from last round: citrus oils are over 90% d-limonene, and limonene's specific rotation runs around +100. Remove it and the oxygenated remainder cannot move the needle.

The bergamot row is the interesting one. Its whole oil reads only +8 to +24 because bergamot is the citrus with a low limonene fraction and a high linalyl acetate fraction. The rotation reads that compositional difference straight off.

The substantivity field confirms it independently

If the ladder is real, then removing the most volatile component should also lengthen the substantivity. It does:

Material	Optical rotation	Substantivity
orange oil terpenes	+92 ~ +105	4 hours
sweet orange peel oil	+94 ~ +99	140 hours
sweet orange peel oil terpeneless	+11 ~ +24	212 hours
grapefruit oil terpenes	+95 ~ +100	12 hours
grapefruit oil	+91 ~ +96	264 hours

Rotation down, substantivity up. Two entirely separate fields pointing at one mechanism.

I have spent several rounds checking one record's fields against each other. This is the upgraded version: across materials, across two fields, across five fruits, all consistent.

24% span zero

Of the 359, eighty-seven (24%) have a range covering both signs. Most sit close to zero:

Optical rotation	Suppliers	Material
−3.00 to +2.00	198	geraniol
−2.00 to +2.00	135	linalool
−2.00 to 0.00	128	clove bud oil
−2.00 to +2.00	96	(E)-anethol

That is reasonable. Geraniol has no stereocentre, commercial linalool is racemic, and clove oil's main constituent eugenol has no stereocentre either. A range spanning zero means the material does not rotate, and the spec is only allowing for measurement error.

But one entry does not belong on that list

laevo-menthol (#24093, 114 suppliers, CAS 2216-51-5) is registered as:

-1.00 to +2.00

A material whose name declares laevorotation, with a spec that permits it to rotate right.

I pulled every material in the database whose name carries laevo- or dextro-:

Material	Optical rotation	Agrees with its name
laevo-carvone	−62 ~ −57	✓
laevo-limonene	−118 ~ −108	✓
dextro-limonene	+87 ~ +102	✓
laevo-linalool	−15 ~ −19	✓
laevo-bornyl acetate	−35.5 ~ −39	✓
laevo-menthyl acetate	−70 ~ −75	✓
laevo-menthol	−1 ~ +2	✗

Sixteen of seventeen agree; menthol does not, and it has the most suppliers of any of them.

There is also an internal argument needing no outside data: laevo-menthyl acetate is the acetate ester of l-menthol and reads −70 to −75. An ester cannot rotate strongly while its parent alcohol does not rotate at all.

And a third clue: in the same database, dextro,laevo-menthol (the racemate, CAS 89-78-1, 38 suppliers) has an empty rotation field.

The −1 to +2 looks like it landed on the wrong record. It belongs to the racemate, not to the laevo form.

And one minus sign is a typo

The lime oil group:

Material	Optical rotation	Suppliers
lime oil	+34.00 to +47.00	39
lime oil distilled	+34.00 to +47.00	31
lime oil expressed	+35.00 to −41.00	34
lime oil expressed mexico	+41.00 to +35.00	17

lime oil expressed reads +35.00 to −41.00, a 76-degree range crossing zero.

Its Mexican sibling reads +41.00 to +35.00, descending but positive at both ends.

Same oil, same process: one is +35 to +41 and the other is +35 to −41. The minus sign is spare.

(This is the third single-record error I have caught by lining up a family of materials side by side. The first two were the boiling point and melting point fields and the aldehyde C-numbers.)

How you can use it

Measure any deterpenated or folded oil you buy. It should read clearly below the whole oil, and negative for lemon and lime. If it matches the whole oil, it was not deterpenated.
Measure whole citrus oils too. Sweet orange should read +94 to +99, grapefruit +91 to +96. A low reading usually means deterpenated oil or solvent was blended in.
A single compound named laevo- or dextro- should carry the matching sign. One record in this database does not; your bottle should not follow it.
Materials whose range spans zero need no measurement. They do not rotate.

Professional practice is far finer than this. Authenticity work now uses the enantiomeric ratio — separating the two mirror-image molecules on a chiral column and measuring their proportions — rather than a whole bottle's net rotation. Hong and colleagues measured the enantiomeric composition of several citrus species by MDGC-MS in 2017 (PMID 28444796), and in 2023 Cucinotta and colleagues went further and measured chiral isotopic fractionation in lemon oil (PMID 37757528).

A polarimeter gives you the crude cheap version. It will catch "this was never deterpenated"; it will not catch "this lot has 5% blended in."

What this doesn't establish

I measured no optical rotations. Everything here compares field values.
"Limonene's specific rotation runs around +100" is a general figure. The database's dextro-limonene reads +87 to +102, which is that material's specification, not a literature value for the pure substance.
I did not consult outside sources on laevo-menthol. The argument is entirely internal: a comparison against sixteen sibling materials, against its own acetate ester, and against the racemate's empty field. All three point the same way, but I have no authoritative source.
The same goes for lime oil expressed. My basis for calling the minus sign a typo is the other three records in its group, not an original document.
The five-citrus ladder has gaps. Lemon and bergamot have no folded records, which does not mean the market lacks them.
Only 0.9% of materials carry a rotation, so this whole approach works on those 359 only.