Material guide · 163
Material guide: maple furanone — recommended at 0.01% or less, with an empty strength rank
· 17 min read
CAS 698-10-2, 67 suppliers, C7H10O3. Its strength field reads very high and recommends evaluating at 0.01% or less, a dilution asked for by only 9 materials in the whole database, and this is the best-supplied of them. Yet its strength_rank field is empty. Checking why turned up something systematic: that field has three levels while the adjective has four. Low maps to rank 1, medium to rank 2 and high to rank 3 without a single exception, and all twelve very high materials carry no numeric rank at all. The twelve strongest materials in this database do not exist in the strength rank column. It has three other problems: two boiling points at 0.50 mmHg (102–103 and 83–86) with neither marked est; a melting range of 31 to 35 °C identical to shoyu furanone's, which shares its molecular formula but not its CAS number; and vapour pressures 353-fold apart between the two.
About a 4 minute read.
The short version
- Name: maple furanone, systematically 5-ethyl-3-hydroxy-4-methyl-2(5H)-furanone (abhexon), CAS 698-10-2
- C7H10O3, molecular weight 142.15, 67 suppliers; FEMA 3153, all four regulatory listings
- Odour type caramellic; odour: maple, sweet, fruity, caramellic, fenugreek, brown sugar, nutty, chicory, praline, butterscotch
- Strength field reads
very high, recommending 0.01% or less. Only 9 materials in the database do - And its
strength_rankfield is empty - Two boiling points at 0.50 mmHg, and a melting point identical to another material's
The strength rank has three levels; the adjective has four
The odor_strength field reads:
very high ,recommend smelling in a 0.01 % solution or less
and strength_rank is empty. I took that for an omission until I checked the database:
| Adjective | Records | Numeric rank |
|---|---|---|
| low | 7 | all rank 1 |
| medium | 431 | all rank 2 |
| high | 352 | all rank 3 |
| very high | 12 | all with no rank |
The first three map one to one without a single exception, and the fourth has no number.
Which means the twelve strongest materials in this database do not exist in the strength rank column.
That has consequences. When I measured how the strength rank relates to real dose, rank 1 came out at a median 8.07%, rank 2 at 2.69% and rank 3 at 0.76%. That analysis never contained the strongest group at all, because they carry no rank.
Only nine materials ask for 0.01%
The distribution of recommended evaluation concentrations:
| Recommendation | Records |
|---|---|
| 0.01% | 9 |
| 0.1% | 45 |
| 1% | 204 |
| 10% | 527 |
| 50% | 12 |
Those nine:
| Suppliers | Material | Odour type |
|---|---|---|
| 99 | 2,3,5-trimethyl pyrazine | nutty |
| 67 | maple furanone | caramellic |
| 41 | caramel furanone | caramellic |
| 28 | (Z)-4-heptenal | green |
| 24 | propyl mercaptan | alliaceous |
| 16 | isopropyl mercaptan | alliaceous |
| 13 | ortho-cresol | phenolic |
| 9 | denatonium benzoate | — |
| 8 | meta-cresol | phenolic |
Two mercaptans, two cresols, two furanones, one pyrazine and one green aldehyde. The loudest group in the database, and maple furanone is the second best-supplied of them.
(denatonium benzoate is a denaturant bitterant, the most bitter substance known rather than the
most odorous. It appears here because
the strength field mixes in things other than smell.)
Its boiling point field cannot be resolved
102.00 to 103.00 °C. @ 0.50 mm Hg | 83.00 to 86.00 °C. @ 0.50 mm Hg
One pressure, two values, twenty degrees apart.
I swept for this last round: 208 materials are in this
state and nine in ten resolve through the (est) marker by keeping the unmarked one.
Neither of this record's values is marked, putting it among the six that do not resolve.
Its melting point is identical to another material's
The mp field reads 31.00 to 35.00 °C.
And shoyu furanone (#36501, 46 suppliers, CAS 27538-09-6) reads 31.00 to 35.00 °C.
Exactly the same.
Both share a molecular formula (C7H10O3) and a molecular weight (142.15), but they are different compounds with different CAS numbers. Two different compounds do not share an identical melting range. One was copied.
I swept the database: 40 pairs share a molecular formula and an identical melting range while
carrying different CAS numbers. Most are one substance registered twice (anethol and
(E)-anethol, cinnamic acid and (E)-cinnamic acid), which is fine. A few are genuinely
different compounds: besides this pair, myrcene and gamma-terpinene both read −10 °C.
And their vapour pressures are 353-fold apart
| maple furanone | shoyu furanone | |
|---|---|---|
| CAS | 698-10-2 | 27538-09-6 |
| Formula / MW | C7H10O3 / 142.15 | C7H10O3 / 142.15 |
| Melting point | 31–35 °C | 31–35 °C |
| Vapour pressure @25 °C | 0.000034 | 0.012 |
| Substantivity | 400 hours @1% | 168 hours |
| Strength | very high (no rank) | high (rank 3) |
| Suppliers | 67 | 46 |
Both vapour pressures carry (est), and they are 353-fold apart.
I have just written that two estimates agreeing is not evidence. This is the other face of it: two estimates 353-fold apart means at least one fell outside the estimator's working range.
Incidentally, the two logP columns are one column
Looking at this record turned up something else. Its logp reads 0.90 (est) and its xlogp3
also reads 0.90 (est).
I swept the database: of the 18,450 records with both, 18,402 (99.7%) carry identical numbers, with a median difference of 0.00.
They are not two independent estimates; they are one column duplicated. If seeing a material's logP and xlogp3 agree makes you feel better about the value, that reassurance is not real.
What it actually is
Systematically 5-ethyl-3-hydroxy-4-methyl-2(5H)-furanone, commonly abhexon: the ethyl homologue of sotolon, the fenugreek lactone.
Its occurrence field: blackberry, blueberry, coffee, raspberry, soy sauce.
Its notes field is one line: Prod. in protein hydrolysates.
"Soy sauce" and "protein hydrolysates" are the same fact stated twice. Furanones of this class form in Maillard chemistry and protein breakdown, which is why one compound turns up in soy sauce, coffee and maple syrup alike.
In 2012 Collin and colleagues analysed the main odorants of Jura flor-sherry wines in J Agric Food Chem, and the relative contributions of sotolon and abhexon are that paper's subject (PMID 22117650). RIFM published a full safety assessment for CAS 698-10-2 in 2024 (PMID 38016534).
Zero in the corpus
Across 954 formulas, maple furanone appears zero times. Its relatives
strawberry furanone appear 4 times and furaneol 5.
That follows. This is a flavour material (FEMA 3153) and the corpus holds perfume formulas. Maple and fenugreek are not directions perfumery goes.
What this doesn't establish
- "Very high has no numeric rank" comes from comparing the whole database, not from any statement by it. Those twelve may simply be unfilled.
- I did not determine which melting point is correct. I can only say two different compounds should not share a range.
- I did not judge which of the two vapour pressures is right. Both are estimates.
- "
logpandxlogp3are one column" is inferred from 99.7% agreement; I have not seen the program that produced them. - My explanation for denatonium benzoate's presence in that table is speculation.
- I did not resolve the two boiling points. This is one of the six unresolved cases from last round.