Material guide: BHT, the antioxidant everybody tells you to add, and the database gives it an odour type

In short: the only material I have written up whose purpose is not its smell. And it has one, which is worth knowing.

Fields

Field	Value
CAS	128-37-0
EINECS	204-881-4
First synonym	BHT
Formula	C15H24O, MW 220.355
Database category	antioxidants and preservatives for fragrance and flavor
Appearance	colorless crystals (est)
Assay	95.00 to 100.00
Melting point	70-73 C
Boiling point	265 C @ 760 mmHg
Flash point	> 93.33 C
logP	5.30 (est)
Water solubility	0.6 mg/L @ 25 C (exp)
Solubility	alcohol
Odour type	phenolic
Odour	phenolic; camphoreous
Strength	low (rank 1)
Substantivity	400 hours (20% in dipropylene glycol)
Odour description	at 100.00%: mild phenolic camphor
Occurrence	garlic bulb; rue oil China @ 0.05%
Suppliers	52
Regulatory	FEMA GRAS, FEMA 2184
Cosmetic uses	antioxidants; fragrance
Oral toxicity	rat LD50 890 mg/kg
GHS	H410 very toxic to aquatic life with long lasting effects (the only one)
Note	a di-tert-butyl phenol with antioxidant properties; used in cosmetics, foods and pharmaceuticals

It has an odour type

This is the most overlooked and most worth knowing thing on the record.

When BHT comes up on a forum, what gets discussed is its function. But the database files it in the same fields as every other aroma material, so it has an odour type, an odour field, a strength and a substantivity.

Odour type: phenolic. Odour: phenolic, camphoreous.

The strength field says low at rank 1, the same band as phenethyl salicylate and benzyl benzoate. So it will not overpower anything.

But the substantivity field says 400 hours (20% in dipropylene glycol), the ceiling value in this library.

Low strength plus not leaving means it accumulates in the base. Somebody on the forum said he puts 0.1% into every material and every bottle of alcohol. The cost of that is not one batch's smell; it is an extra layer of phenolic under every formula.

(Whether it is detectable at 0.1% I did not test. The odour description field is tagged at 100%.)

It is a solid

Melting point 70 to 73 C, colourless crystals.

Using it means making a solution first, and with a water solubility of 0.6 mg/L (measured) and logP 5.30 it is strongly lipophilic. The solubility field lists only alcohol.

In practice: make a stock in ethanol; do not try to sprinkle it into a formula.

It is itself a phenol

BHT's full name is 2,6-di-tert-butyl-4-methylphenol. It is a hindered phenol, and the antioxidant mechanism sits on that phenolic hydroxyl: it takes up a radical and becomes a relatively stable phenoxyl radical, and the chain reaction stops there.

(Which is also why the odour type is phenolic. Function and smell come from the same structure.)

The same mechanism explains why tocopherol, vitamin E, is also a phenol. Somebody on the forum uses both BHT and tocopherol and, asked why, said there was no reason. They are the same class of antioxidant with overlapping mechanisms.

One GHS statement, and it is environmental

H410: very toxic to aquatic life with long lasting effects.

No skin statement, no inhalation statement, no carcinogenicity statement. Just that one.

The precautionary statements point the same way: avoid release to the environment, collect spillage, dispose of contents to an approved waste disposal plant.

Oral toxicity is rat LD50 890 mg/kg. Against the carrier solvents I have written up: isopropyl myristate > 16,000, dipropylene glycol 14,850, benzyl benzoate 1,700-2,800. It is below all three.

(That does not make it unsafe to use. It has FEMA GRAS and FEMA 2184 and is an approved food additive. It is simply not in the same band as the solvents.)

It occurs in nature

The occurrence field has two entries: garlic bulb, and rue oil from China at 0.05%.

No practical use in itself, but it makes a good example: a material treated as the archetype of a synthetic additive occurs naturally.

(How unreliable that classification is in this database is computed in The synthetic versus natural argument my database cannot settle: 306 materials say "not found in nature" in one field and are tagged natural in another.)

Does adding it work

I have no answer here.

Hagvall and colleagues established in Contact Dermatitis in 2008 that the problem is real: terpenes in lavender oil exposed to air oxidise at the same rate as the pure compounds oxidising alone, produce the same oxidation products, and the sensitising potency rises accordingly. The other constituents of a whole essential oil did not protect its main ones.

But that paper measured the absence of natural protection, not the effect of added BHT.

I could not find a controlled study of added antioxidants in fragrance materials. So I can tell you what BHT is and what adding it costs, and not whether it works. (The full discussion is in Should I add BHT.)

What this doesn't establish

• I have not smelled it. This is database fields plus one paper.
• "The only material whose purpose is not its smell" refers to the hundred and six I have written up, not a library-wide scan. (After getting an "only one" claim wrong recently, the scope is stated up front.)
• Whether the phenolic is detectable at 0.1% I did not test. The odour description field is tagged at 100%.
• The 400-hour substantivity was measured at 20% in dipropylene glycol and is not comparable with a 0.1% use level.
• The hindered-phenol mechanism is textbook chemistry; I did not verify its behaviour in a fragrance matrix from a paper.
• "BHT and tocopherol overlap in mechanism" is inferred from structure; I did not look for studies on whether combining them is synergistic or antagonistic.
• Hagvall 2008 measures sensitisation rather than odour degradation. Related, but not the same endpoint.
• I have no evidence at all about the effect of added antioxidants. That is the largest gap.
• I did not run corpus statistics. (This is not the kind of material that appears on a formula sheet anyway.)