Lauric oils: coconut and babassu, and why they behave unlike everything else

Why coconut oil refuses to behave like other oils

A formulator who has learned the trade on sunflower and olive oil meets coconut oil and finds that the rules have changed. The jar that was a clear liquid in the warm kitchen is a white block on the workshop shelf. A soap recipe built around it behaves unlike an olive-oil recipe at every step, from the lye calculation to the feel of the bar. And the internet is full of confident, contradictory verdicts: coconut is the most comedogenic oil there is, coconut is a natural remedy for spots. Neither survives a look at the actual studies.

None of this is a property of "coconut". It is a property of the lauric group, to which coconut oil belongs together with babassu and palm kernel oil. The three are defined by a 12-carbon saturated fatty acid, lauric acid, which makes up roughly half of their fatty acids, and by an unusual escort of even shorter chains around it. Once the link between chain length and behaviour is clear, the group stops being surprising and becomes predictable. This article is the first spoke of our series on plant oils by fatty-acid group, and it deals with the group whose consistency changes at ordinary room temperatures.

Twelve carbons, and everything that follows from them

The Codex Alimentarius standard for named vegetable oils (CXS 210-1999) gives authentic-sample ranges for coconut oil: lauric acid (C12:0) 45.1–53.2 percent of total fatty acids, myristic (C14:0) 16.8–21.0, palmitic (C16:0) 7.5–10.2, oleic (C18:1) 5.0–10.0, and, unusually for a plant oil, meaningful amounts of very short saturated chains: caprylic (C8:0) 4.6–10.0 and capric (C10:0) 5.0–8.0. Linoleic acid, the doubly unsaturated acid that dominates sunflower and grapeseed, is almost absent at 1.0–2.5 percent, and with it goes most of the oxidation susceptibility that those oils carry. Babassu sits in the same family with 40.0–55.0 percent lauric, and palm kernel oil with 45.0–55.0.

Three consequences follow directly from this profile.

Melting near the temperature of a room

Saturated chains pack neatly and melt higher than unsaturated ones, but short saturated chains melt lower than long ones. Pure lauric acid melts at about 44 °C (PubChem lists 43.8 °C), well below stearic acid and well above the C8 and C10 acids that sit alongside it in coconut oil. The oil itself, being a mixture of triglycerides with several chain lengths, melts lower than the pure acid and over a range rather than at a point. In practice, as anyone who keeps a jar knows, that melting range sits around ordinary room temperatures: the same jar is solid in a cool British workshop and liquid on a summer windowsill, and the oil melts on contact with skin. That is the single most useful fact about the group. It explains the white block, it explains why a coconut-based balm behaves differently in January and July, and it is why formulators use coconut oil deliberately where a "melting" texture is wanted, such as cleansing balms and solid formats.

Almost nothing to oxidise

Lipid oxidation normally begins with the loss of a hydrogen atom from a carbon next to a double bond, so an oil with few double bonds offers few starting points. The Codex iodine value range for coconut oil, a measure of total unsaturation, is 6.3–10.6, against 118–141 for sunflower and 128–150 for grapeseed. Low unsaturation means a low oxidation risk from the oil itself; it does not exempt the finished product from stability testing, because everything else in the formula can still oxidise. Babassu and palm kernel share the property: their Codex iodine values are 10–18 and 14.1–21.0 respectively.

A lot of soap per gram

Saponification value measures how much potassium hydroxide is needed to saponify one gram of oil, and it rises as chains get shorter: every triglyceride has three ester bonds, but a short-chain triglyceride has a lower molar mass, so a gram of it contains more molecules and therefore more ester bonds to hydrolyse. Codex gives coconut oil a saponification value of 248–265 mg KOH per gram, against 188–194 for sunflower and 168–181 for high-erucic rapeseed. For a soap maker this is a practical number, not a curiosity: coconut needs noticeably more lye than the same weight of olive oil. Replace olive with coconut while keeping the original alkali amount and the bar comes out with excess unsaponified fat; make the reverse substitution and it comes out lye-heavy.

Comedogenicity: what was actually measured

The claim that coconut oil is "highly comedogenic" descends from one source more than any other: the rabbit-ear survey by Fulton published in the Journal of the Society of Cosmetic Chemists in 1989. It is worth knowing what that study did. Test materials were tested at 10 percent, one part material to nine parts propylene glycol, and 1 ml was applied to the inner surface of a rabbit's ear five days a week for two weeks, three rabbits per assay. Follicular keratosis was graded from 0 to 5; the paper treats 0 to 1 as not significant, 2 to 3 as borderline and 4 to 5 as positive. In the published table, coconut butter and lauric acid both receive a comedogenicity grade of 4, and so does cocoa butter. The study's own summary notes that medium-chain fatty acids were more potent than shorter or longer chains in producing follicular keratosis.

Two things follow. First, the model is a rabbit ear exposed to a 10 percent dilution of the raw material, repeatedly; a rabbit-ear result cannot be read directly across to human facial use. Second, the rating attaches to a raw material, not to a product. Draelos and DiNardo revisited exactly this gap in the Journal of the American Academy of Dermatology in 2006, using a modified Mills and Kligman assay on the backs of six human volunteers who were known to form comedones. Their conclusion was that finished products using comedogenic ingredients are not necessarily comedogenic. The study was small, and it does not prove that coconut oil is harmless on acne-prone skin; what it shows is that the rabbit-ear grade of an ingredient cannot be read across to a finished formula at its real use level.

There is a further irony that the group's chemistry produces. Free lauric acid, the very molecule graded 4 on the rabbit ear, is strongly antibacterial in the laboratory. Nakatsuji and colleagues (Journal of Investigative Dermatology, 2009) reported minimal inhibitory concentrations against Propionibacterium acnes, Staphylococcus aureus and Staphylococcus epidermidis more than fifteen times lower than those of benzoyl peroxide, with P. acnes the most sensitive of the three; applied to mouse ears, lauric acid reduced P. acnes numbers. This is a laboratory and animal study of the free fatty acid, not of coconut oil, and coconut oil is overwhelmingly esterified triglyceride rather than free acid. It says nothing about what a coconut-oil cosmetic does on human skin, and a cosmetic cannot be marketed for a skin condition in any case. What it does show is that "comedogenic" and "antibacterial" are not opposites; they describe different experiments on different forms of the same 12-carbon chain.

The practical reading for a formulator: none of these studies establishes a safe or a problematic use level for coconut, babassu or palm kernel oil in a finished cosmetic. They justify neither a blanket ban nor a blanket reassurance, and the only defensible answer for a specific product is a test of that product. The claim that survives scrutiny is neither "non-comedogenic" nor anything about spots; it is a tested formula.

In soap, lauric oils play by their own rules

Nowhere is the group more distinctive than in bar soap. Soap makers describe an all-coconut bar as quick to lather and harsh on the skin, and a coconut-plus-olive bar as the classic balance; that is craft experience, not a measured result, but the underlying soap chemistry has been studied. A 2025 paper in Langmuir re-examined the mixture soap makers call the "soap maker's eutectic", sodium laurate with sodium oleate, and found a broad depression of the Krafft temperature (the temperature below which a soap stops dissolving as micelles) of more than 15 °C at a sodium oleate weight fraction of about 0.5, traced to a distinct solid compound of the two soaps in a ratio of roughly 4 to 3. In plain terms: the lauric and oleic soaps together stay dissolved at lower temperatures than either does alone. That is a property of the soap solution; it does not by itself predict the lather or mildness of a finished bar, which still has to be made and tested.

The group also brings hardness: saturated chains give a firm bar, which is why soap makers reach for coconut, palm kernel or babassu when an olive-heavy recipe comes out soft. Palm kernel oil is used in the same role as coconut, and babassu, with a little more oleic acid in its profile, is described by soap makers as slightly milder there; the composition is consistent with that reading, though it is craft experience rather than a measured result.

The saponification value matters here too. Because coconut oil needs more alkali per gram, a recipe that swaps olive for coconut without recalculating the lye will be lye-heavy. Every soap calculator handles this; the point is to understand why the numbers differ rather than to trust the tool blindly.

Babassu, palm kernel and the family resemblance

Babassu oil comes from the kernels of a Brazilian palm and is, in composition, coconut oil's nearest relative. Codex ranges put its lauric acid at 40.0–55.0 percent, myristic at 11.0–27.0, palmitic at 5.2–11.0 and oleic at 9.0–20.0, with caprylic and capric acids present as in coconut. The slightly higher oleic share and slightly lower lauric share are consistent with what formulators report, that babassu feels a little softer and less drying than coconut, though that comparison is craft experience rather than a measured result. A 2020 characterisation in Food Research International of extra-virgin and virgin babassu oils confirmed that medium-chain fatty acids dominate; the two grades had the same physicochemical and thermal properties except colour and thermal stability, where the extra-virgin oil was lighter and more stable.

Palm kernel oil, from the kernel of the oil palm rather than its fruit, is the third member. Codex gives 45.0–55.0 percent lauric, 14.0–18.0 percent myristic, 6.5–10.0 percent palmitic and 12.0–19.0 percent oleic. It should never be confused with palm oil, which is a palmitic-oleic fat from the fruit pulp and belongs to a different group entirely. Palm kernel is also sold as fractions: palm kernel olein (the liquid fraction, Codex iodine value 20–28) and palm kernel stearin (the solid fraction, iodine value 4–8.5, lauric acid 52.0–59.7 percent). The fractions let a formulator dial the melting behaviour of the group up or down without leaving it.

A note on identity testing that matters for anyone buying these oils in the UK. Because lauric oils are the only common plant oils with substantial C8 and C10 content, Codex uses two old volatile-acid tests, the Reichert and Polenske values, as identity checks for them: for coconut oil the expected ranges are 6–8.5 and 13–18 respectively, with different bands for palm kernel and babassu. A supplier's certificate that reports these values gives you one more identity check to compare against the Codex ranges, alongside the fatty-acid profile and the other tests; none of them proves authenticity on its own.

Hair: where the short chain earns its keep

One of the better-documented advantages of the group has nothing to do with skin. Rele and Mohile, in the Journal of Cosmetic Science in 2003, compared coconut, sunflower and mineral oil as pre-wash and post-wash treatments, using a Taguchi design to cover the combinations. Coconut oil was the only one of the three that reduced measured protein loss from both undamaged and damaged hair; sunflower and mineral oil did not reduce it. The authors attributed the difference to composition: a triglyceride of lauric acid is small and linear enough to penetrate the hair shaft, whereas mineral oil has no affinity for hair protein and sunflower's kinked, unsaturated chains do not enter the fibre. It is one of the few cases where the chain length of an oil has been tied to a measured performance outcome.

On skin: a moisturiser, not a miracle

The clinical evidence for coconut oil on skin is more modest than either its fans or its critics suggest. Agero and Verallo-Rowell ran a randomised, double-blind trial (Dermatitis, 2004) in 34 people with mild to moderate dry skin, applying either extra-virgin coconut oil or mineral oil to the legs twice a day for two weeks. Both oils significantly improved skin hydration and surface lipids; there was no significant difference between them in transepidermal water loss or skin pH; and the authors' conclusion was that coconut oil is as effective and safe as mineral oil as a moisturiser. That is a useful, honest result. It supports coconut oil as an occlusive emollient in body care and it gives no support to claims that it "repairs" or "heals" anything, language a UK safety assessor or claims reviewer would strike out in any case.

Where lauric oils belong, and where they do not

Putting the chemistry and the evidence together gives a fairly clear map of the group's territory.

The table also explains a material that often confuses newcomers. Caprylic/capric triglyceride, often sold as "fractionated coconut oil", is a triglyceride of the C8 and C10 acids; commercial material may be obtained by fractionation or by re-esterification and may derive from coconut or palm kernel oil. It is liquid at room temperature and it is not a lauric oil at all: the lauric acid that defines the group is absent. Everything in this article about melting range, soap and hair applies to whole coconut oil and does not carry over to this material.

Refined or virgin

Virgin coconut oil, usually cold-pressed, keeps its characteristic odour; refined, bleached and deodorised coconut oil is nearly odourless and paler. For the fatty-acid behaviour described here the two are the same material, because refining does not change the triglyceride backbone. The choice is about odour, colour and the quality limits your supplier can meet. Codex quality factors allow an acid value of up to 4.0 mg KOH per gram for cold-pressed and virgin oils against 0.6 for refined, and a peroxide value of up to 15 milliequivalents of active oxygen per kilogram for cold-pressed and virgin against 10 for refined. A virgin oil is therefore allowed to arrive with more free fatty acid and more primary oxidation than a refined one, which is worth knowing before the word "virgin" decides the purchase; the finished product's shelf life is a separate question answered by stability testing.

What to check before you buy

• Lauric acid 45–53 percent for coconut, 40–55 for babassu, 45–55 for palm kernel, with C8 and C10 present. If lauric acid is missing and only C8 and C10 remain, you are looking at caprylic/capric triglyceride, not a lauric oil.
• Iodine value in the Codex band (coconut 6.3–10.6). A figure outside the band means the material does not conform; ask the supplier before assuming why.
• Saponification value 248–265 for coconut; use the batch figure, not a generic one, in soap calculations.
• Reichert and Polenske values where offered: the short-chain fingerprint of the group.
• Acid and peroxide values against the Codex quality limits for the grade you ordered.
• Odour and colour appropriate to the grade: a "refined" oil that smells strongly of coconut, or a "virgin" oil that smells of nothing, is a question for the supplier.

Coconut oil is not difficult. It is simply different, and the difference is a 12-carbon chain. Read the group rather than the name and its behaviour in the jar, the bar and the bottle becomes something you can predict before you melt the first gram.

Sources

• Codex Alimentarius. Standard for Named Vegetable Oils, CXS 210-1999 (amended 2015): fatty-acid composition (Table 1), iodine and saponification values (Table 2), Reichert and Polenske values and quality factors (Appendix). FAO PDF
• PubChem. Lauric acid (CID 3893), melting point. PubChem
• Fulton J. E. Comedogenicity and irritancy of commonly used ingredients in skin care products. Journal of the Society of Cosmetic Chemists, 40, 321–333, 1989. SCC Media Library
• Draelos Z. D., DiNardo J. C. A re-evaluation of the comedogenicity concept. Journal of the American Academy of Dermatology, 54(3), 507–512, 2006. PubMed 16488305
• Nakatsuji T., Kao M. C., Fang J. Y., Zouboulis C. C., Zhang L., Gallo R. L., Huang C. M. Antimicrobial property of lauric acid against Propionibacterium acnes: its therapeutic potential for inflammatory acne vulgaris. Journal of Investigative Dermatology, 129(10), 2480–2488, 2009. PubMed 19387482
• Hu J., Anachkov S. E., Moaddel T., Carnali J. O. Reexamining the Enhanced Solubility of Sodium Laurate/Sodium Oleate Eutectic Mixtures. Langmuir, 41(3), 1568–1576, 2025. PubMed 39792245
• Bauer L. C. et al. Physicochemical and thermal characterization of babassu oils (Orbignya phalerata Mart.) obtained by different extraction methods. Food Research International, 137, 2020. PubMed 33233140
• Rele A. S., Mohile R. B. Effect of mineral oil, sunflower oil, and coconut oil on prevention of hair damage. Journal of Cosmetic Science, 54(2), 175–192, 2003. PubMed 12715094
• Agero A. L., Verallo-Rowell V. M. A randomized double-blind controlled trial comparing extra virgin coconut oil with mineral oil as a moisturizer for mild to moderate xerosis. Dermatitis, 15(3), 109–116, 2004. PubMed 15724344