Sweet almond oil: the default balm oil, measured properly
Sweet almond oil carries 62 to 73 percent oleic acid, iodine value 92 to 106 and a 12 month shelf life. Use levels, refining, and the nut allergen question.
Sweet almond oil is the oil most people start with and then keep buying out of habit. It earns the place on feel and price rather than on stability, because it carries enough linoleic acid to give it a genuinely limited life, and it is a declared tree nut in every allergen list that matters. This page gives the composition, the shelf life you can defend, the use levels that hold up in a stick, and the two questions suppliers are asked most: what to put on the label, and whether the bottle is really almond.
Sweet almond oil (INCI Prunus Amygdalus Dulcis Oil) runs 62 to 73 percent oleic, 20 to 30 percent linoleic and 6 to 8 percent palmitic. That gives an iodine value of about 92 to 106, a bis-allylic index near 20 to 31, and a realistic 12 months unopened at 20 C from the pressing date. Use 10 to 60 percent in balms, but keep it under about 40 percent in a twist-up stick.
What is actually in the bottle
The oil is pressed from the kernel of Prunus dulcis, a Rosaceae drupe. Botanically it is a stone fruit seed rather than a true nut, which matters not at all to a regulator: almond is a listed tree nut in the EU, the UK and the US alike. Kernels carry roughly half their weight as oil, and sweet cultivars are selected for a very low amygdalin content, which is the whole difference between sweet and bitter almonds.
Composition is a triglyceride story with almost nothing exotic in it. Two fatty acids account for around nine tenths of the oil, and the ratio between them is what predicts both the feel and the shelf life.
| Component or value | Typical | What it does |
|---|---|---|
| Oleic acid (C18:1) | 62-73% | The bulk of the oil. One double bond, no bis-allylic hydrogen, so it barely oxidises on a balm's timescale. |
| Linoleic acid (C18:2) | 20-30% | The whole of the stability problem. One bis-allylic position per molecule. |
| Palmitic acid (C16:0) | 6-8% | Saturated, contributes a little body without making the oil cloud in a cold room. |
| Stearic acid (C18:0) | 1-2% | Minor. Almond is not a source of hard fat. |
| Palmitoleic acid (C16:1) | 0.4-0.8% | Trace, useful as an identity marker. |
| Alpha-linolenic acid (C18:3) | under 0.5% | Effectively absent, which is why almond outlasts oils of similar iodine value. |
| Total tocopherols, virgin | 350-500 mg/kg | Overwhelmingly alpha-tocopherol. Refining removes a large part of it. |
| Iodine value (Wijs, ISO 3961) | 92-106 | Published spans run wider, roughly 85 to 109, depending on origin. |
| Bis-allylic index | 20-31 | Linoleic plus twice linolenic. The number that ranks oxidation rate. |
| Relative density at 20 C | about 0.910-0.915 | Useful when you convert a formula from weight to volume. |
Sweet almond oil and bitter almond oil are unrelated products. The first is a pressed fixed oil used at tens of percent. The second is a steam-distilled essential oil that is largely benzaldehyde, dosed in fractions of a percent, and restricted for its hydrocyanic acid content. If a bottle labelled almond oil smells strongly of marzipan and comes in 10 mL, it is not an emollient.
Iodine value, and what it really predicts
Iodine value is the mass of iodine, in grams, that reacts with 100 g of fat under the Wijs method (ISO 3961, or AOCS Cd 1d-92). Sweet almond oil sits at 92 to 106 in most commercial data, which places it below standard sunflower at 118 to 141 and above high-oleic sunflower at 78 to 90. As an identity check that is worth having: a bottle testing at 130 is not almond oil.
As a stability figure it is blunt, because it counts double bonds and is blind to how they are arranged. The reactive site in an oil is the bis-allylic methylene, a CH2 with a double bond on either side, and only polyunsaturated chains have one. Linoleic acid has a single bis-allylic position, alpha-linolenic has two, and oleic acid, which is most of this oil, has none at all. Weighting linoleic once and linolenic twice puts sweet almond at an index of 20 to 31, which is the useful number. The full argument is in rancidity and oxidation, and the same index across every oil in the cupboard is tabulated in the oil comparison table.
What that index buys you in practice is a middle position. Almond oxidises several times faster than macadamia or a high-oleic cultivar, and several times slower than grapeseed or standard sunflower. In a blend the highest-index component sets the pace, so 10 percent rosehip in an almond base makes an almond-and-rosehip product with a rosehip shelf life, not an almond one. The weighted arithmetic is what the oil blend calculator exists for.
The shelf life you can defend
Twelve months unopened at 20 C, in a full container, dark, is the figure to work with. It is a trade convention rather than a measurement, and it is the number this site uses because it matches both supplier practice and the position of almond on the bis-allylic ranking. Nobody has published a real-time study on cosmetic almond oil that would let you defend a longer one.
Three corrections turn that convention into something usable. Count from the pressing date and not from the day the bottle arrived, because an oil that spent four months in a distributor's warehouse gives you eight. Halve your expectations for a part-used bottle standing in a warm workshop with 400 mL of air above 100 mL of oil, since headspace oxygen and temperature both matter more than the oil choice did. And check the peroxide value on the certificate: the general Codex standard for edible fats and oils allows up to 10 milliequivalents of oxygen per kg for a refined oil and 15 for a cold pressed one, while genuinely fresh oil tests at 0.1 to 1, so a compliant certificate is not the same thing as a fresh delivery. Ask for the pressing date and the peroxide value before the drum ships.
Adding antioxidant is the last move, not the first. Virgin almond oil already carries 350 to 500 mg/kg of tocopherols, mostly alpha, which is several times the measured optimum for alpha-tocopherol in a bulk oil, so dosing more alpha into it is close to pointless. Mixed tocopherols or rosemary extract at the levels given in vitamin E and antioxidants are a better use of the money, and the colourless, odourless alternative is discussed under synthetic antioxidants. None of them reverse anything: once the oil smells of crayons or old paint the aldehydes already exist, which is the subject of rancid balm.
Refined against cold pressed
The two grades are different products sold under one name. Cold pressed oil is expressed mechanically at low temperature and filtered; refined oil is usually expelled or solvent extracted, then neutralised, bleached and deodorised. The refining steps strip colour, odour, free fatty acids, phospholipids and a good part of the native tocopherol, and they also remove most of what is left of the almond protein.
| Property | Cold pressed (virgin) | Refined |
|---|---|---|
| Colour | Pale gold to amber | Near colourless to very pale straw |
| Odour and taste | Distinctly nutty, faintly marzipan | Bland, close to neutral |
| Total tocopherols | 350-500 mg/kg | Roughly a third to a half lower after deodorisation |
| Residual protein | Higher, not systematically measured | Very low. Comparable refined oils measure 0.2 to 2.2 micrograms per mL |
| Batch consistency | Varies with harvest and cultivar | Consistent, which matters for a repeatable line |
| Relative cost | Roughly one and a half to two and a half times refined | The cheaper of the two by a wide margin |
| Choose it when | The nutty note suits the product, or a marketing position requires virgin oil | Fragrance is delicate, colour must stay pale, or allergen exposure is being minimised |
The stability comparison surprises people. Refining removes tocopherol, so a refined oil can be less stable than the virgin grade unless the refiner added tocopherol back, which most do. Ask. It is one of the few questions where a supplier's answer changes what you should write on the best-before date.
How much to use, and where it thins a stick
Almond oil contributes nothing to structure. Every percentage point of it is a point that is not wax or butter, so the usable window is set by what else is in the formula rather than by the oil itself.
| Product | Sweet almond | Constraint |
|---|---|---|
| Lip balm, twist tube | 10-35% | Above about 40 percent the stick loses its edge and smears rather than draws a line. |
| Lip balm, tin | 15-45% | A tin tolerates far more liquid oil, because a fingertip does the work the stick body would. |
| Body balm or salve | 20-60% | The classic one-part-wax-to-eight-parts-oil salve puts almond near the top of this band. |
| Massage balm | 30-60% | Slip is the point. Almond is the traditional choice for exactly this. |
| Cleansing balm | 10-30% | Usually a supporting oil, since the solubilising work is done elsewhere. |
| Whipped butter | 10-25% | Too much liquid oil and the whip collapses on standing. |
The thinning point in a stick is worth stating precisely, because it is where most first formulas fail. In a 4.25 g oval tube with 12 to 14 percent beeswax, total liquid oils above roughly 55 percent give a stick that survives the mould and then bends against a warm lip. If you want almond above 40 percent, you are choosing to add 1 to 2 points of wax for every 10 points of liquid oil you add, or to move some of that liquid into a butter. The alternative is a tin. Do the arithmetic in the wax ratio calculator rather than by eye, follow the method in how to make lip balm, and if a batch lands wrong the correction ladder is in balm too hard or too soft.
Swap in a stage. Take a working formula, replace 10 percentage points of its liquid oil with almond, pour three tubes and leave them 24 hours before judging. Balms harden for a day after pouring, so a stick assessed ten minutes out of the mould always reads too soft and you will over-correct the wax.
How it feels, and what to pair it with
Almond oil applies with medium slip, sinks in at a moderate rate, and leaves a light residual film that reads as conditioned rather than greasy. It is thin, with a viscosity in the same region as sunflower and an order of magnitude below castor, which is why it is the standard base for anything that has to be spread over a large area. It is the reason a massage product built on almond glides for minutes rather than dragging after thirty seconds.
It is not remarkable at anything, which is a virtue in a base oil and a limitation in a hero one. If you want faster absorption, apricot kernel and camellia go in quicker. If you want cushion and a fast sink together, macadamia does that. If you want gloss and cling on a lip, you need castor oil at 8 to 15 percent alongside it, because almond alone gives a matte, thin film. And if you want the same job done with a shelf life measured in years rather than months, the honest comparison is in jojoba oil against sweet almond oil. A blend of 60 percent almond and 40 percent of a stable partner keeps most of the feel and buys real time, which is the general argument in carrier oils compared.
The comedogenic rating of 2
Sweet almond oil is listed at 2 on the 0 to 5 comedogenicity scale in almost every circulating table. It is worth knowing what that number rests on. The scale comes from the rabbit ear assay developed in dermatology in the 1970s and 1980s: neat material was applied to the external ear canal of a rabbit, repeatedly, for around two weeks, and the resulting follicular hyperkeratosis was graded. The rabbit ear follicle is far more reactive than human facial skin, the test used undiluted ingredients rather than in-use concentrations, and much of the compiled table circulating online has no published primary source at all.
So the 2 tells you almond oil is not inert and not among the worst offenders, and nothing more precise. A 30 percent almond content in a balm is not the same exposure as neat oil under occlusion, and individual response varies more than the ratings do. The full critique is in comedogenic ratings. One genuine constraint does apply: almond oil is a triglyceride of chain lengths that Malassezia yeasts can use, so it is not suitable for a formula built to starve them, which is the subject of malassezia safe balms.
The nut allergen question, and the label
Almond is a listed tree nut allergen in both the EU and the US food systems, and the tree nut oil that turns up most often in balms. Two facts have to be held together. Refined seed oils carry very little protein, in the region of 0.2 to 2.2 micrograms per mL for the oils where this has been measured, and reported reactions to cosmetic nut oils are uncommon. But no controlled feeding trial has ever been run on almond oil, cold pressed grades carry substantially more protein than refined ones, and lip products are partly ingested.
The labelling position is stated wrongly by a great many suppliers. There is no declarable allergen class for nut oils in cosmetics anywhere: not in the EU, the UK, the US, Canada or Australia. What cosmetics law requires instead is the full ingredient list in INCI nomenclature, and because INCI names are binomial the botanical origin is already disclosed. Prunus Amygdalus Dulcis Oil is sweet almond, and a reader who knows that has everything the law provides. Adding "contains sweet almond oil" in plain English alongside a complete INCI list is permitted and genuinely helpful. Claiming "nut free" while using it is not. The detail, including the cross-reactivity picture and the studio contamination routes, is in nut oil allergy, and a formula that avoids the problem entirely is in nut free balm.
If you sell at markets, expect the question in person. The defensible answer is that the product contains sweet almond oil, that it is on the ingredient list under its botanical name, and that anyone with a diagnosed nut allergy should follow their clinician's advice rather than yours. Do not offer reassurance about refining grades over a stall table. The ingredient list is the honest answer, and it is a complete one.
Adulteration with apricot kernel oil
Sweet almond oil is one of the more frequently adulterated cosmetic oils, and the usual diluent is apricot or peach kernel oil. The motive is straightforward: apricot kernel is cheaper, comes from the same genus, feels almost identical on skin, and overlaps almond so heavily on fatty acid profile that a routine gas chromatogram of the methyl esters cannot separate a modest substitution. Apricot kernel runs about 58 to 66 percent oleic and 25 to 32 percent linoleic against almond's 62 to 73 and 20 to 30. A 20 percent addition disappears inside the natural variation of the cultivars. The practice is old enough to have its own name in historical pharmacy, where peach and apricot kernel oil was sold as persic oil and used deliberately as an almond substitute.
The analysis that does separate them is the unsaponifiable fraction, specifically the sterols. Almond oil sterols are dominated by beta-sitosterol, and pharmacopoeial monographs for almond oil include a sterol composition test with a cap on campesterol as a proportion of total sterols, precisely because the kernel oils used to dilute it carry proportionally more. Triacylglycerol profiling by high performance liquid chromatography does the same job from a different angle. Neither is a bench test, which is the point: you cannot detect this by smell, colour or feel, and neither can your customer.
What you can do is buy on documentation. Ask for a certificate of analysis that names the method and reports the fatty acid profile and the sterol composition, not just a peroxide value, and treat an unusually cheap cold pressed almond oil as a question rather than a bargain. The case for using it on purpose, allergen declaration and all, is in apricot and peach kernel oil.
When not to use it
The decision rule is short. Use sweet almond oil when you want a competent, cheap, medium-slip base oil, when the product will be used within a year, and when a tree nut on the label is not a problem for the people buying it. Those conditions are met more often than not, which is why it is the default.
Drop it when any one of them fails. For a lip balm aimed at children, a product sold into a school or hospital setting, or anything marketed as allergen conscious, use apricot kernel, camellia or high-oleic sunflower instead and say so plainly; lip balm for children covers the wider question. For a product that has to survive two summers in a warm van, almond is the wrong substrate and a high-oleic oil costs no more. For anything where the residual nutty note fights the fragrance, buy the refined grade or change oil. And do not reach for it to fix a texture problem: almond oil thins, and thinning is rarely what a disappointing balm actually needs.
The honest limit on everything above is variability. Cultivar, region, season and pressing move the oleic to linoleic ratio by more than any formulation choice you will make, and the numbers on this page are bands rather than constants. If a formula only works when your almond oil comes in at the top of its oleic range, the formula is fragile and the next drum will prove it. Buy on a certificate, record the lot against the batch, and treat published composition tables as a way to compare oils with each other rather than as a fact about the bottle on your bench.
Frequently asked questions
Is sweet almond oil comedogenic?
It is usually rated 2 on a 0 to 5 scale, which places it in the middle. That rating comes from a rabbit ear assay using undiluted material, not from human testing at the concentrations used in a balm, and much of the circulating table has no published source. Treat it as a rough flag rather than a prediction, and patch test if you are pore-prone.
How long does sweet almond oil last?
About 12 months unopened at 20 C, counted from the pressing date rather than from delivery. A part-used bottle in a warm workshop, with a large air space above the oil, will do considerably less. Refrigeration extends it. The limit comes from 20 to 30 percent linoleic acid, which oxidises far faster than the oleic acid making up most of the oil.
Should I buy refined or cold pressed sweet almond oil?
Refined for pale, delicately fragranced or allergen-conscious products, and for batch-to-batch consistency. Cold pressed if the nutty note suits the product or the marketing position requires a virgin oil. Cold pressed typically costs one and a half to two and a half times as much, keeps more of its 350 to 500 mg/kg of tocopherols, and carries more residual almond protein.
Can someone with a nut allergy use a balm containing sweet almond oil?
That is a question for their allergy clinician, not for a maker. Refined oils carry very little protein and reactions to cosmetic nut oils are uncommon, but almond is a listed tree nut, no controlled trial has been run on almond oil specifically, and cold pressed grades carry more protein. Declare it plainly and let the customer decide.
How much sweet almond oil goes in a lip balm?
Ten to 35 percent by weight in a twist-up tube and 15 to 45 percent in a tin. Above roughly 40 percent in a stick, with a normal 12 to 14 percent beeswax load, the balm smears instead of drawing a clean line. If you want more almond than that, add 1 to 2 points of wax for every 10 points of liquid oil, or fill tins instead.
Is apricot kernel oil a substitute for sweet almond oil?
Yes, and a good one. It feels slightly lighter and faster, has a similar fatty acid profile, and is not a declarable tree nut in either the EU or the US lists. It is also the oil most often used to adulterate almond, because the two are so hard to separate without sterol analysis. Choosing it deliberately is a sound move for a nut-conscious range.
Sources and further reading
- European Directorate for the Quality of Medicines and HealthCare, European Pharmacopoeia monographs: Almond oil, virgin and Almond oil, refined, Strasbourg.
- United States Pharmacopeial Convention, USP-NF monograph: Almond Oil, Rockville MD.
- Cosmetic Ingredient Review, Safety assessment of Prunus-derived ingredients as used in cosmetics, Washington DC.
- European Commission, CosIng cosmetic ingredient database, entry for Prunus Amygdalus Dulcis Oil.
- Codex Alimentarius Commission, CXS 19-1981, General standard for edible fats and oils not covered by individual standards, FAO and WHO, Rome.
- International Organization for Standardization, ISO 3961, Animal and vegetable fats and oils: determination of iodine value, Geneva.
- Yada, Lapsley and Huang, A review of composition studies of cultivated almonds, Journal of Food Composition and Analysis, 2011.
Reviewed and updated 6 September 2026. Spotted an error? Tell us and we will fix and log it.