Material guide · 165
Material guide: 3-mercaptohexyl acetate (3-MHA) — 65 of the database's 85 thiols are filed under something foul, this one is filed fruity
· 24 min read
CAS 136954-20-6, 35 suppliers, C8H16O2S. The database holds 257 materials with mercapto, thiol or sulfanyl in the name; 85 of them carry an odor family, and 65 of those sit in the sulfurous, meaty, onion or rotting-vegetable end. Only three are filed fruity, and only one of the three is actually buyable: the other two have 3 and 7 suppliers. Read further down the same record, though, and the odor field says passion fruit, blackberry, blackcurrant bud while the flavor field says sulfurous, roasted, onion, durian. One molecule, two fields, two different materials. The difference hides in each field's measurement conditions: the odor was smelled at 0.10% in DPG, while the flavor field's usage note reads tropical fruit at 0.5 to 1 ppm. A 2000 Bordeaux study showed that passion fruit stores no aroma at all, only an odourless cysteine conjugate. A 2003 Belgian study synthesised 21 mercapto esters for sensory comparison and concluded this family sits at the higher-threshold end of the thiols. And the record contradicts itself: the main fields give a boiling point of 186 °C, a note further down says 240 °C.
About a 5 minute read.
Too long, didn't read
- Name: 3-mercaptohexyl acetate, also 3-sulfanylhexyl acetate, 3-MHA, CAS 136954-20-6
- C8H16O2S, MW 176.28, 35 suppliers; FEMA 3851, JECFA 554, FLAVIS 12.234
- Odor family fruity; descriptors: floral, fruity, pear, tropical, passion fruit, blackberry, raspberry, blackcurrant bud
- The flavor field is another material entirely: sulfurous, roasted, meaty, onion, durian
- Longevity 181 hours @100%, strength high, evaluate at 0.10% or less
- Natural occurrence: passion fruit, wine. Filed as natural/synthetic
- Regulatory field lists JECFA, FEMA GRAS and FLAVIS only. No IFRA entry
- Known impurity: 3-mercaptohexan-1-ol, the alcohol left when the acetyl group goes
Of 85 thiols with an odor family, three are fruity
Start with the family. The database holds 257 materials with mercapto, thiol or sulfanyl in the name, and 85 of them carry an odor family:
| Odor family | Count |
|---|---|
| sulfurous | 52 |
| meaty | 8 |
| fruity | 3 |
| herbal | 3 |
| 18 other families, 1–2 each | 19 |
Add up sulfurous, meaty, onion, alliaceous, cabbage, radish and burnt and you get 65 of 85, or 76%. That is the general impression people have of thiols, and the database confirms it.
The three fruity ones: hexyl 3-mercaptobutanoate (3 suppliers), 3-mercapto-3-methyl-1-butyl acetate (7), and this one, 3-MHA (35). In other words, exactly one fruity thiol is genuinely on the market.
Another angle on how unusual that is. Seven thiols have 35 or more suppliers:
| Material | Suppliers | Odor family |
|---|---|---|
| fish thiol | 78 | sulfurous |
| 2-methyl-3-tetrahydrofuran thiol | 61 | meaty |
| 3-mercapto-2-butanone | 48 | sulfurous |
| ethyl methyl mercaptopropionate | 48 | sulfurous |
| 4-methyl 4-mercaptopentan-2-one 1% solution | 47 | sulfurous |
| 3-mercapto-2-pentanone | 36 | sulfurous |
| 3-mercaptohexyl acetate | 35 | fruity |
Six foul, one fruit.
The same record: passion fruit in one field, onion and durian in the next
The most interesting thing about this material is not how it compares to others. It is inside its own record.
Odor field: floral; fruity; pear; tropical; passion fruit; blackberry; raspberry; currant bud
Flavor field: sulfurous; fruit tropical fruit; passion fruit; roasted; meaty; onion; durian
Passion fruit appears in both. But the flavor field adds onion, roasted, meaty and durian. Same CAS, same single record.
The difference is written in the small print of each field. The odor description opens with "at 0.10 % in dipropylene glycol" — 0.10% in DPG, the condition it was smelled at. The flavor field's usage note reads "tropical fruits at 0.5-1 ppm; and for roasted savory notes in chicken, pork." 0.5 ppm is 0.00005%, two thousand times thinner again than the smelling condition.
Neither entry is a mistake. One molecule was recorded by different people at different concentrations in different media. Both records are right; neither states in the field which square it is standing on. A record that supports two readings at once came up in a different form in when two fields hold the same number.
The practical reading is direct: dilution here is not a working preference, it is the switch that decides what the material becomes. The database recommends 0.10% or less, and of the 2,451 materials that carry a strength field, only 45 ask for 0.10% or less. That is under 2%, and only 3 of the 257 thiols make the list. The threshold discussed in what strength should I smell it at has a bottom square, and this material sits in it.
Passion fruit stores no aroma, it stores an odourless precursor
Tominaga and Dubourdieu, at the Faculty of Oenology in Bordeaux, went looking for the origin of the 3-MHA family in passion fruit juice in 2000 (PMID 10898639).
Their result reads as follows. 3-Mercaptohexan-1-ol and 3-mercaptohexyl acetate are "already known to contribute to the aroma of passion fruit", and two further thiols were identified in the fruit for the first time. The heart of it is in the second half: taking the fruit's nonvolatile extract and adding a cell-free extract with beta-lyase activity (EC 4.4.1.13) released the thiols in vitro. They then identified the precursor itself as S-(3-hexan-1-ol)-L-cysteine.
Their own wording is worth keeping: this release "strongly suggests" that the volatile thiols are present in the fruit in combined form.
Put in perfumery terms: what the fruit holds is an odourless amino acid conjugate, and the aroma only appears once an enzyme cuts the bond. It also explains why Sauvignon Blanc juice smells of so little and the finished wine does not. The thing doing the cutting is yeast.
21 mercapto esters compared: this branch is the easy-going one
The second paper is from Vermeulen and Collin at the brewing and food engineering unit of the Université catholique de Louvain, 2003 (PMID 12769535). They used combinatorial chemistry to synthesise 21 mercapto acetates in one go, two of which — 3-MHA and 3-methyl-3-mercaptobutyl acetate — had already been detected in foods, mainly passion fruit and wine.
The sentence worth copying out is the conclusion: compared with other polyfunctional thiols, mercapto esters show relatively high BE-GC-LoADS values, with very diverse descriptors.
That acronym has to be unpacked. LoADS is the lowest amount detected by sniffing, so a higher number means a less sensitive detection, meaning more material is needed before anyone notices. What the sentence says, then, is that the mercapto ester branch is not as fierce as the other polyfunctional thiols.
I read it backwards the first time. The word "high" had me writing "unusually potent" before I unpacked the acronym and found the direction reversed. It does not contradict the 0.10% recommendation above: 0.10% is the strictest 2% of the whole database, but within a family priced in parts per trillion, this is one you can stand relatively close to.
The record disagrees with itself about the boiling point
Worth noting in passing. The main fields say:
- Boiling point 186.00 °C @ 760 mmHg
- Flash point 214 °F TCC (101.11 °C)
At the tail of the taste description field, next to the impurity note, it says "Boiling point 240 °C. Flash point 73 °C."
One record, two sets of numbers, 54 degrees apart on the boiling point and 28 on the flash point. Which set is right is not something I can settle. What can be settled is that the main-field set carries its conditions (@760 mmHg, TCC closed cup) and the note does not. The principle from what est actually tells you applies here as well: a number travelling with its method can be compared, a number travelling alone is a hint.
Anyone shipping or clearing customs should ask the supplier for the SDS rather than use either set from this page.
The neighbours in the blackcurrant bud corner
The end of its odor field reads currant bud. Materials carrying that same descriptor with 20 or more suppliers:
| Material | CAS | Suppliers | Odor family | Longevity (hr @100%) |
|---|---|---|---|---|
| buchu leaf oil | 68650-46-4 | 53 | herbal | 176 |
| black currant bud absolute | 97676-19-2 | 49 | spicy | 344 |
| 4-methyl 4-mercaptopentan-2-one 1% solution | 19872-52-7 | 47 | sulfurous | — |
| delta-damascone | 57378-68-4 | 42 | floral | 220 |
| 3-mercaptohexyl acetate | 136954-20-6 | 35 | fruity | 181 |
| cassis specialty | — | 34 | floral | — |
| cassis pentanone | 23550-40-5 | 24 | sulfurous | — |
The third row is worth stopping on. Across all 42,225 database records, only 6 items carry "solution" in the name, and 4-MMP is one of them, with 47 suppliers selling it. When a material circulates in catalogues only as a 1% dilution, that fact is itself an operating instruction: nobody intends for you to touch the neat material.
3-MHA is catalogued neat. That is the substantive difference between it and its neighbour, and it is worth remembering more than the difference in the odor field. Where buchu's sulfur note comes from has its own article.
Buying and storing
35 suppliers, which counts as well supplied for a thiol. A few practical points:
- No IFRA entry in the regulatory field. What is there is JECFA 554, FEMA GRAS 3851, FLAVIS 12.234, with the EFSA evaluation under FGE.08 and structure class I. That combination describes a flavour material. If you are putting it in something that goes on skin, check the current IFRA Standards yourself. An empty field is not the same as no restriction.
- The known impurity is 3-mercaptohexan-1-ol, the alcohol form with the acetyl group gone. Acetates hydrolyse, so what is in the bottle drifts towards the alcohol over time, and the two smell different. Decanting and cold storage after opening are worth the trouble.
- Vapour pressure 0.049 mmHg (est), longevity 181 hours: second among thiols with a longevity record, behind 2-mercaptopropionic acid at 400. A thiol is not automatically a top note.
- Water solubility 284.5 mg/L (est), logP 2.10 (est), specific gravity 0.991–0.996. The weighing and dilution routine is in its own article; at the 0.10% order of magnitude you want a two-step dilution.
→ 3-mercaptohexyl acetate in the database: full physical data, 35 suppliers and suggested blends.
What this article does not establish
- For the gap between the two descriptor fields I offered an explanation, not a proof. 0.10% in DPG and 0.5–1 ppm in food are different measurement conditions, but I have no controlled comparison run by one person under both.
- Vermeulen and Collin measured 21 mercapto esters against each other, not 3-MHA's rank among all thiols; the comparison group in the paper is "other polyfunctional thiols".
- Tominaga's paper is about passion fruit, not wine. The Sauvignon Blanc sentence is mine, drawn from the same mechanism. There is a large literature on it, but it is not either of the two papers cited here.
- I did not settle which pair of boiling points is correct, only which pair carries its measurement conditions.
- Vapour pressure, logP and water solubility all carry (est) — they are calculated, not measured.