Synthetic and natural esters: the emollients that fix a heavy balm
Caprylic capric triglyceride, coco caprylate, C12 to 15 alkyl benzoate and five more, compared on viscosity, spreading value and use levels of 5 to 40 percent.
An ester is the cheapest way to change how a balm feels without changing what it does. Swap eight points of a heavy seed oil for a light ester and the same wax load, the same butter and the same occlusive film go on with half the drag. This page gives the eight esters worth keeping on a shelf, the three numbers that predict how each one behaves, and the levels at which each stops helping.
Caprylic/capric triglyceride is the reference light emollient at 25 to 35 mPa s, and everything else is judged against it. Coco-caprylate, dicaprylyl carbonate and isoamyl laurate are thinner and faster at 5 to 9 mPa s. C12-15 alkyl benzoate and cetearyl ethylhexanoate are thicker, more polar and wet pigment. Use esters at 5 to 40 percent of a balm.
What a cosmetic ester actually is
An ester is what you get when an acid and an alcohol join and lose a molecule of water. Every fat you already use is one: a triglyceride is three fatty acids esterified to glycerol. The materials on this page are the same chemistry run deliberately, with the acid and the alcohol chosen for the property you want rather than for whatever a seed happened to contain.
That is the whole reason to use them. A plant oil hands you a fixed bundle of chain lengths, colour, odour, minor components and oxidation risk, and you take the bundle or leave it. An ester hands you one defined molecule with a known viscosity, a known spread and no odour, and against the oils ranked in carrier oils compared it usually costs less than what it replaces. The trade is that you lose the minor fraction: no tocopherols, no sterols, no phytochemistry, nothing to write on the front of a label.
Three structural choices explain almost all of the behaviour. Chain length sets viscosity and spread, and shorter is thinner and faster. Branching lowers the melting point and stops the molecules packing, which is why isononyl and isoamyl esters stay liquid and feel silky. Ester group density, the number of ester oxygens relative to molecular weight, sets polarity, and polarity decides pigment wetting, gloss and what will dissolve. Two of the eight are not strictly esters: dicaprylyl carbonate is a carbonate diester and behaves like a slightly more polar ester, and the trade counts it in the same family.
The eight, side by side
Read the viscosity column first, because it is measured to a real standard and transfers between suppliers. Read the spreading column as a band, not a figure.
| Ester | Viscosity (mPa s) | Spreading (mm2/10 min) | Polarity | Use level |
|---|---|---|---|---|
| Caprylic/capric triglyceride | 25-35 | 700-900 | Medium | 5-40% |
| Coco-caprylate | 5-9 | 1400-1700 | Low | 5-30% |
| Dicaprylyl carbonate | 6-9 | 1100-1400 | High | 5-25% |
| Heptyl undecylenate | 6-10 | 900-1200 | Medium | 3-15% |
| Isoamyl laurate | 5-8 | 1400-1800 | Low | 5-25% |
| Isononyl isononanoate | 5-8 | 700-900 | Low to medium | 3-20% |
| C12-15 alkyl benzoate | 18-25 | 400-600 | High | 3-15% |
| Cetearyl ethylhexanoate | 15-25 | 400-600 | Medium | 5-25% |
Viscosity is a genuine measurement, normally kinematic viscosity by ISO 3104 converted with the density, or a rotational reading. Spreading value is the area a fixed quantity covers on a standard substrate in ten minutes, and although the concept is standard the substrate, dose and timing are not, so a figure from one supplier cannot be compared with a figure from another. The bands and their meaning are set out in balm texture science. What you can take from the column is the ranking, which is stable: coco-caprylate and isoamyl laurate spread furthest, the benzoate and the cetearyl ester least.
Viscosity and spreading value run roughly inverse to each other but not exactly, and where they disagree the spreading value wins on feel. Isononyl isononanoate is as thin as coco-caprylate in the bottle and covers half the area on skin, which is why it reads as silky rather than as vanishing.
Caprylic/capric triglyceride, the reference point
At 25 to 35 mPa s it is the material every other ester on the list is described against, and it is the one to buy first. It is a genuine triglyceride, three C8 and C10 fatty acids on a glycerol backbone, made by fractionating coconut or palm kernel oil to the short saturates and re-esterifying. Molecular weight is around 470 to 555, calculated from the structure, against roughly 880 for an ordinary C18 seed oil triglyceride, and that difference is the whole reason it feels lighter.
It is colourless, odourless, fully saturated, liquid at refrigerator temperature and cheap. In a balm it does one job better than anything else at the price: it cuts the greasy phase short without reducing the film, which is why it carries a fifth to a quarter of a typical cleansing balm formula. The distinction between the whole oil and this fraction causes more failed substitutions than any other naming problem on an ingredient list, and it is worked through in coconut oil in balms. On a label it is Caprylic/Capric Triglyceride, never "fractionated coconut oil", which is a supplier's trade description.
Its limits are worth stating. It has no cling, so a formula built on it alone wears off fast. Its medium polarity makes it a mediocre pigment wetter next to a benzoate. And because it is derived from coconut or palm kernel, palm-free and certified-supply questions apply to it exactly as they do to the parent oil, which is a sourcing conversation rather than a formulation one.
The plant-derived light esters
Three materials cover the ground between caprylic/capric triglyceride and a silicone, and all three are made entirely from renewable feedstocks.
Coco-caprylate is coconut alcohol esterified with caprylic acid, molecular weight around 300 to 340. At 5 to 9 mPa s and a spreading value in the 1400 to 1700 band, it is the closest plant-derived approach to the fast, dry break of a volatile silicone, and it is the standard replacement when a formula has to drop cyclomethicone. Watch the INCI: Coco-Caprylate and Coco-Caprylate/Caprate are different grades and the second is slightly heavier. The silicone comparison it is usually asked to win is set out in silicones in balms.
Dicaprylyl carbonate is two octyl groups joined through a carbonate, C17H34O3, molecular weight 286.5. It spreads almost as fast as coco-caprylate but is markedly more polar, which makes it the one light emollient that also dissolves things. If you need a fast, dry carrier that will still keep menthol, camphor or a salicylate in solution, this is it, and the solubility arithmetic is in solubility in anhydrous formulas. Its carbonate group is usually made from dimethyl carbonate, whose origin decides whether a given grade qualifies as natural, so ask.
Heptyl undecylenate comes from castor oil at both ends: undecylenic acid from the pyrolysis of ricinoleic acid, and heptanol from the aldehyde that pyrolysis produces alongside it. Molecular weight is 282.5. It is light, slightly polar, and carries a faint characteristic odour that some noses pick up in a fragrance-free product. It is also the only one of the eight with a double bond, which matters below. It is the shortest chain length of the group at C11 on the acid side, and for anyone building around the chain length restrictions discussed in malassezia-safe balms that places it awkwardly, whereas the C8 and C10 esters sit safely.
Dry finish and pigment wetting
The other three are chosen for what they leave behind rather than how fast they go.
C12-15 alkyl benzoate is a benzoic acid ester of a C12 to C15 alcohol cut, so it is a mixture rather than a single molecule, molecular weight around 290 to 330. It is the most useful of the eight and the most misunderstood. It is thick, low spreading and polar, and it dries down to a light, non-tacky finish that almost no other emollient of that viscosity achieves. Its aromatic ring gives it a refractive index near 1.48 against roughly 1.45 for caprylic/capric triglyceride, so it adds gloss, and it wets iron oxides better than any of the aliphatic esters here. In a tinted stick it does most of what castor oil does with none of the tack, which is why so many colour formulas contain both. It has a mild antimicrobial effect at the levels used, which is a side benefit and not a substitute for anything.
Isononyl isononanoate is a branched C9 acid on a branched C9 alcohol, C18H36O2, molecular weight 284.5. The double branching is the point: it is thin in the bottle but spreads only moderately, so it feels silky and substantive at once, and it leaves a dry, powdery finish. It is fully synthetic, from petrochemical butene, and there is no plant route to it.
Cetearyl ethylhexanoate pairs a long, straight C16 to C18 alcohol with a short branched acid, molecular weight around 370 to 400. It is the most oil-like of the three, with enough substantivity to replace part of a seed oil rather than just thin it, and a soft dry-down. If it looks unfamiliar on an old formula sheet, it is the material formerly listed as Cetearyl Octanoate, renamed under the INCI dictionary without any change to the substance, which is the sort of trap covered in INCI names explained.
Polarity, and the four things it decides
Polarity is measured in the trade as interfacial tension against water in mN/m, with a lower number meaning a more polar oil. Some suppliers publish an inorganic to organic balance index instead. Neither is comparable across data sheets, but the ranking is robust to method, and it is the ranking that formulates.
Pigment dispersion. Iron oxide and titanium dioxide surfaces are polar and hydrophilic. A non-polar oil cannot wet them, so the particles stay agglomerated, the colour reads weak and streaky and the pigment settles in the pot. A polar ester displaces air from the surface and keeps the particles apart. This is the mechanism behind streaky or mottled colour and the reason a tinted balm needs a polar fraction of at least 10 percent of its oil phase, whether that comes from castor or from a benzoate. Which pigments need most help is covered in colourants and tints.
Gloss. Shine needs a smooth continuous film with a refractive index that differs from the substrate. Aromatic esters sit higher than aliphatic ones, so a benzoate raises gloss where a carbonate at the same level does not, and the effect is largest on a lip where the substrate is wet and the film is thin.
Whether a wax gel stays clear. A clear oleogel needs the wax to stay dissolved above its melting point and to crystallise into a fine network on cooling. Polar esters are poor solvents for hydrocarbon waxes, so a paraffin or microcrystalline wax carried in a high benzoate load will haze or throw a coarse crystal. Polar waxes with free acids and esters of their own, beeswax most of all, tolerate polar esters much better. If a stick that should be translucent has gone cloudy, the wax and ester polarities are mismatched, and oleogels and wax networks explains the network side of it.
What dissolves. Polar esters dissolve polar actives, crystalline materials and organic UV filters that non-polar oils will not hold, which is why a menthol or camphor formula that recrystallises in the tin is usually short of polar solvent rather than over-dosed on the active.
Where the percentage goes, from 5 to 40
Esters are not a fixed-dose additive. The same material is a corrective at 5 percent and the base of the formula at 35, and what changes as you climb is predictable.
| Ester level | Drag | Cling and wear | Afterfeel | What goes wrong |
|---|---|---|---|---|
| 3-8% | Slightly reduced at the start of the rub | Unchanged | Unchanged | Too small to fix a real drag complaint |
| 8-15% | Clearly easier to spread | Marginally shorter | Noticeably less greasy | Nothing. This is the corrective band |
| 15-25% | Low throughout | Shorter wear, needs reapplying | Dry, quick, close to a lotion | Balm may read as thin or insubstantial |
| 25-40% | Minimal | Short unless a cling agent is added back | Almost nothing left after two minutes | Wax softens, tins slump, product feels like it does nothing |
Two practical consequences. First, esters soften a balm, because you are replacing a viscous or semi-solid fat with a thin liquid. Above about 20 percent expect to add a point of wax back, and check the result against balm too hard or too soft. Second, a high ester load buys lightness by giving up cling, and if you want both you have to put the cling back deliberately with castor oil, a small polymer fraction or a heavier ester, not by cutting the light one.
If a balm drags, do not cut wax first. Move 8 to 12 percentage points of the formula out of the heaviest oil and into caprylic/capric triglyceride, keeping wax and butter identical, and make the two versions side by side. The comparison isolates the variable, and the full diagnostic ladder is in drag and poor glide.
Oxidative stability, and what it is worth
Seven of the eight have no carbon to carbon double bonds at all, so they have no allylic hydrogens and no bis-allylic hydrogens, and autoxidation has nothing to start on. That is a categorical difference from a seed oil rather than a matter of degree: the mechanism set out in rancidity and oxidation does not apply to them. Typical shelf lives quoted by suppliers are two to three years for an unopened, well-stored container, and the practical failure mode is contamination or moisture pickup rather than rancidity.
Heptyl undecylenate is the exception, and it is an instructive one. Undecylenic acid carries one terminal double bond, which by calculation from the structure gives the ester an iodine value near 90. That would look alarming next to a butter at 40. It is not, because the double bond sits at the end of the chain with no second double bond anywhere near it, so there is no bis-allylic methylene and the oxidation rate is closer to that of oleic acid than to linoleic. Iodine value counts double bonds and is blind to how they are arranged, which is exactly the trap it always sets.
What the stability buys is specific. It lets you carry a small percentage of something genuinely fragile, rosehip or sea buckthorn or an infused oil, without the base setting the pace of the whole formula. A balm that is 30 percent ester and 5 percent rosehip has a much lower blend oxidation index than one that is 35 percent sunflower, and the arithmetic of that is what the shelf life calculator does. It does not remove the need for an antioxidant on the fragile fraction, covered in vitamin E and antioxidants.
Naming, natural origin and the certification rules
"Synthetic" and "natural" describe the feedstock, not the chemistry. Esterification is the same reaction whether the acid came from a coconut or from a cracker, and the finished molecule does not remember. What the certification schemes assess is where the atoms came from and how the reaction was run.
ISO 16128 parts 1 and 2 give the arithmetic: an ingredient made by an allowed process from renewable feedstock is "derived natural", and its natural origin index runs from 0 to 1 in proportion to the natural fraction of the molecule by molecular weight. COSMOS-standard goes further, requiring the process itself to appear on a permitted list and capping the petrochemical moieties allowed anywhere in a finished product. Ecocert Greenlife has certified natural and organic cosmetics to COSMOS since 2017, so a supplier saying "Ecocert approved" in current documentation almost always means COSMOS approved. The broader landscape is in natural and organic certifications.
| Ester | Acid from | Alcohol from | Natural origin position |
|---|---|---|---|
| Caprylic/capric triglyceride | Coconut or palm kernel | Glycerol, plant | Fully renewable, COSMOS grades common |
| Coco-caprylate | Coconut | Coconut | Fully renewable, COSMOS grades common |
| Dicaprylyl carbonate | Carbonate, from dimethyl carbonate | Coconut or palm kernel | Depends on the carbonate route, ask the supplier |
| Heptyl undecylenate | Castor | Castor | Fully renewable, COSMOS grades available |
| Isoamyl laurate | Coconut or palm kernel | Fermentation | Fully renewable, COSMOS grades available |
| Isononyl isononanoate | Petrochemical, C4 route | Petrochemical, C4 route | Synthetic, no plant route |
| C12-15 alkyl benzoate | Benzoic acid, usually petrochemical | Coconut or palm kernel | Part renewable, generally not COSMOS approved |
| Cetearyl ethylhexanoate | 2-ethylhexanoic acid, petrochemical | Plant cetearyl alcohol | Part renewable, grade dependent |
None of this is a safety ranking, and it should not be sold as one. It is a compliance and marketing constraint, and a formula that is 100 percent natural origin can still be a poor product. If your positioning rests on it, get the certificate with the batch and file it with the record. If it does not, choose on viscosity, spread and polarity and ignore the column.
Where esters are the wrong answer
An ester will not fix a formula that is short of structure. If a balm slumps, feels thin or will not hold a shape, adding a fast-spreading liquid makes every one of those worse, and the problem is wax or butter. An ester will not fix greasiness caused by too much total oil either: moving weight between oils changes the character of the residue, not its quantity, which is the distinction greasy, heavy afterfeel turns on.
Esters also give nothing back on the ingredient list. There are no tocopherols, no sterols, no unsaponifiable fraction, nothing that supports a story about the material, and a buyer scanning a label reads them as chemicals. That is a positioning cost, not a performance one, but it is real and it is why some makers work around them with squalane and light seed oils instead.
The decision rule is short. Buy caprylic/capric triglyceride first, because it does 80 percent of the work at the lowest price. Add C12-15 alkyl benzoate second if you make anything coloured or glossy, because nothing else combines pigment wetting, gloss and a dry finish. Add coco-caprylate third if you need a genuinely fast, dry break in a face or body product. Everything after that is a refinement, and if you cannot state which sensory complaint the fourth ester is fixing, you do not need it.
Frequently asked questions
Is caprylic/capric triglyceride the same as fractionated coconut oil?
They are the same material under two names. "Fractionated coconut oil" is a supplier description; Caprylic/Capric Triglyceride is the INCI name that has to appear on the label. It is not coconut oil. The whole oil has been split and the lauric acid and other long-chain saturates removed, leaving a thin, colourless liquid that stays liquid in a fridge.
How much ester can I put in a balm?
Between 5 and 40 percent by weight. Eight to fifteen percent is the corrective band that fixes drag without changing the character of the product. Above about 20 percent you are rebuilding the formula rather than adjusting it, the balm softens, and you should expect to add roughly a percentage point of wax back to hold the same firmness.
Which ester is best for a tinted lip balm?
C12-15 alkyl benzoate. It is polar enough to wet iron oxides so the colour disperses evenly, its aromatic ring gives it a refractive index near 1.48 which adds gloss, and it dries down without tack. Use it at 3 to 15 percent alongside castor oil rather than instead of it, since castor still contributes the cling.
Do cosmetic esters go rancid?
Almost none of them do. Seven of the eight common balm esters are fully saturated, so they have no bis-allylic hydrogens for autoxidation to attack and suppliers typically quote two to three years unopened. Heptyl undecylenate has one terminal double bond, which raises its calculated iodine value to about 90 but leaves it roughly as stable as oleic acid.
Are cosmetic esters natural?
It depends entirely on the feedstock, not the chemistry. Coco-caprylate, isoamyl laurate and heptyl undecylenate are made wholly from plant sources and have COSMOS-approved grades. Isononyl isononanoate is petrochemical with no plant route. C12-15 alkyl benzoate is part renewable but its benzoic acid is usually petrochemical, so it generally fails the standard.
Will esters clog pores?
Some are more likely to than others, and the ranking is not the one usually quoted. Isopropyl myristate and isopropyl palmitate have the strongest comedogenic record and are worth avoiding on the face. Caprylic/capric triglyceride, coco-caprylate and dicaprylyl carbonate have much less against them. The limits of these ratings are set out in comedogenic ratings.
What replaces dimethicone in an anhydrous formula?
Coco-caprylate is the usual answer for the fast, dry break of a volatile silicone, and dicaprylyl carbonate or isoamyl laurate for the slip. None of them reproduce the persistence of a high-molecular-weight dimethicone film, so a formula that relies on long wear or water resistance will feel different rather than equivalent.
Sources and further reading
- Personal Care Products Council, International Cosmetic Ingredient Dictionary and Handbook, Washington DC (entries for Caprylic/Capric Triglyceride, Coco-Caprylate, Dicaprylyl Carbonate, C12-15 Alkyl Benzoate and Cetearyl Ethylhexanoate).
- European Commission, CosIng cosmetic ingredient database, function and INCI entries for the esters listed.
- Cosmetic Ingredient Review, Safety assessment of alkyl benzoates as used in cosmetics, Washington DC.
- Cosmetic Ingredient Review, Safety assessment of triglycerides as used in cosmetics, Washington DC (covers caprylic/capric triglyceride).
- International Organization for Standardization, ISO 16128-1:2016 and ISO 16128-2:2017, Guidelines on technical definitions and criteria for natural and organic cosmetic ingredients and products.
- COSMOS-standard AISBL, COSMOS-standard: cosmetics organic and natural standard, Brussels, including the permitted processes and petrochemical moiety limits.
- International Organization for Standardization, ISO 3104, Petroleum products: transparent and opaque liquids, determination of kinematic viscosity, the method behind most published emollient viscosities.
Reviewed and updated 6 September 2026. Spotted an error? Tell us and we will fix and log it.