Face balm: getting occlusion onto facial skin without leaving a greasy film

An everyday face balm at 5 percent wax and 25 percent squalane, with the dose per use, a 30 ml jar, a 58 to 64 C pour and the fix for pilling under makeup.

Most face balms fail on feel rather than on function. The occlusion is easy, and any blend of wax, butter and oil will slow water loss from the surface of a cheek. What separates a balm people use nightly from one that sits in a drawer is how fast the film stops being visible, and that is set by three things: the wax level, the choice of light esters, and the quantity you put on. This page fixes all three with numbers.

• Wax 5%
• Squalane 25%
• Dose 0.25-0.4 g
• Pour 58-64 C
• Cure 24 h

What a face balm does, and what it does not

An anhydrous balm has no water in it, so it cannot add water to skin. What it does is sit on the surface and slow transepidermal water loss, the rate at which water already in the tissue leaves. Skin losing less water rehydrates itself from below, and that is the whole of the effect. Nothing in the jar repairs a barrier, and the honest framing of the three functional classes is at occlusive, emollient and humectant.

The size of that effect is worth stating plainly, because balm marketing overstates it. Petrolatum is repeatedly measured as the most occlusive single cosmetic material, cutting water loss by upwards of 90% in standard in vitro comparisons. A blend of plant waxes, butters and oils manages a fraction of that, in the region of 20 to 30% depending on the test and the film thickness. That is a real effect on skin dried by cold or washing, and it is not the same effect. Where those figures come from is at measuring occlusivity.

So a face balm is a finishing step rather than a moisturiser in the full sense. It works best over damp skin or over a humectant layer, where there is water present for it to hold.

Why facial skin takes far less wax than body skin

A body balm runs at 10 to 20% wax and nobody complains. The same blend on a face reads as a mask. Three differences explain it.

The first is mechanical. A wax network has a yield stress: it does not flow until the shear you apply exceeds a threshold, which is why a balm holds its shape in a jar and spreads under a finger. Facial application uses light pressure over a mobile, uneven surface, so a network that needs firm pressure to break deposits a thick patchy film instead of a thin continuous one, as set out at rheology and yield stress. At 5% beeswax the yield stress breaks under the pressure you would use to smooth an eyebrow. At 8% it does not, and the balm you see at two hours is the film you failed to spread.

The second is thermal. Beeswax melts at 62 to 65 C and facial skin sits around 32 to 34 C, so the wax never melts on your face. Whatever the fingertip does not break stays as a network on the surface. Body skin gets more pressure and more friction from clothing, so it hides a stiff film that a face displays.

The third is sebaceous. The face carries the highest density of sebaceous glands on the body, commonly quoted at several hundred per square centimetre on the forehead and nose against under a hundred on the limbs. There is already a lipid film there, so added wax and butter are additive rather than corrective, and the visible result is shine.

The formula, and why each number is there

Percentages are weight percent of the batch and total exactly 100. A 90 g batch fills three 30 ml jars at about 27 g each and leaves roughly 9 g of overage. Multiply any percentage by 0.9 for grams.

Squalane at 25% is the largest single line because it is the one material that gives film without residue. It is a fully saturated C30 hydrocarbon at 25 to 35 mPa s with an iodine value below 1, so it neither oxidises nor needs protecting, and it is close in character to the hydrocarbon fraction of sebum. Its data and the purity question are at squalane. The 20% of coco-caprylate and dicaprylyl carbonate is the dry-touch pair, both far faster spreading than any triglyceride, and both there to shorten the shine rather than to add anything to the film: see cosmetic esters.

The 37.9% of liquid oils is oleic-dominant and low in linoleic, because polyunsaturated oils oxidise and oxidised oil on a face is both a smell problem and an irritation problem. High-oleic sunflower is the workhorse, jojoba is a liquid wax ester rather than a triglyceride and lays a soft film without weight, and meadowfoam is over 95% long-chain C20 and C22 material with unusual oxidative stability. The comparative data is at carrier oils compared.

Butters at 12%, split 8 refined shea and 4 mango, give body and a solid fraction still present at bedtime. Refined shea rather than unrefined, because the unsaponifiable fraction that makes unrefined shea interesting also makes it smell and darken, and mango because an all-shea butter phase grains readily. Tocopherol sits at 0.1% because the measured optimum for alpha-tocopherol in an oil is around 100 parts per million and it can act as a pro-oxidant above that.

Absorption rate, measured as time to lose shine

What people mean by "absorbs well" is not absorption in any strict sense, because very little of a balm crosses the stratum corneum. The film spreads, thins and wicks into the outer corneocyte layers until it stops reflecting light as a continuous sheet, and that is measurable at a kitchen table. Weigh 0.3 g, spread it over a marked 5 by 5 cm patch of inner forearm, and time how long until the specular shine is gone when you turn the arm under a raking light. Run each material on a fresh patch in the same session at the same room temperature, because the number moves with humidity and with how oily your own skin is.

Ten to fifteen minutes is the target for a night balm. Under five and the film is too light to be doing anything at 3 am. Over twenty-five and you will not want to put anything on top of it, which is the pilling problem below. Tune with the ester fraction: 5 points moved from sunflower oil into coco-caprylate takes roughly two to three minutes off, and 5 points the other way puts them back.

Dose per use, which decides the verdict more than the formula does

A face balm at 0.25 g and the same balm at 0.6 g are different products to the person using them. Film thickness roughly doubles, time to lose shine more than doubles because the excess has nowhere to go, and the balm that read as elegant becomes the one that pills under sunscreen. Most people who reformulate a greasy face balm should have halved the dose instead.

Calibrate once. Put a jar on a scale, take out what you would normally use, and read the number. A quantity the size of a small blueberry is close to 0.3 g for this density. Then use that quantity for a fortnight before judging anything.

The table doubles as a diagnostic. A 30 ml jar that empties in three weeks of once-daily use is being used at over 1 g a time, and no reformulation will fix what that feels like. The same reasoning applied to lip products, with the arithmetic behind it, is at how much balm to apply.

Making it: melt, pour at 58 to 64 C, cure 24 hours

1. Weigh everything except the tocopherol into one heatproof jug, tared between additions, on a scale reading to 0.01 g.
2. Melt at 75 to 80 C and hold five minutes. The hold is not optional. Shea and mango carry high-melting stearic fractions that recrystallise as grit if any crystal survives the melt.
3. Cool below 60 C and stir in the tocopherol. Stir, do not whisk. Air in a soft balm shows as a pale foamy crust.
4. Pour at 58 to 64 C. With 5% wax in a mostly liquid phase, the melt starts to cloud at roughly 56 to 60 C, so this window is at or just above onset. Fill in one movement, on a level surface, keeping the set layer under about 20 mm deep so it sets evenly.
5. Cool quickly. Twenty minutes in a refrigerator, then room temperature. Not the freezer. Slow cooling from a partly melted state is what makes butters grainy.
6. Cure 24 hours before assessing feel. Wax crystallisation continues after the balm looks solid, and a balm judged at two hours reads softer and greasier than the same balm the next day.

Pour too hot, above about 70 C, and you get a dished centre and a coarse crystal network that feels waxier at the same 5%. Pour below 56 C and the melt is already thickening, so the surface sets mottled. The general case is at pour temperatures.

Two variants at the same wax level

Both variants keep 5% wax and 12% butters and change only the oil fraction, so the set, the pour window and the cure are identical and you are testing exactly one thing.

The dry variant halves the dry-touch esters and adds avocado, a heavier oleic oil, pushing the shine time out towards twenty to thirty minutes and leaving a film you can still feel at bedtime. The oily-prone variant runs squalane at 30% and the ester pair at 28%, bringing shine time down near six to nine minutes, and swaps three points of shea for mango, the drier of the two on skin. Neither changes the structure, so if the base sets correctly both will.

Layering, sunscreen and pilling

Pilling is mechanical. A film that has not thinned enough is still a discrete layer, and when you rub a second product containing film formers, silicone elastomers or pigment over it, the shear rolls both layers into visible threads. The pills are not the balm alone, they are the balm plus whatever went on top.

Four fixes, in the order worth trying. Cut the dose to 0.25 g. Wait until the shine has gone, ten to fifteen minutes for this formula rather than two. Press the sunscreen on in sections with a flat palm instead of rubbing. If it still pills, stop layering and use the balm at night only.

Comedogenicity, honestly

The 0 to 5 numbers circulating for every oil come from rabbit ear assays on neat materials under occlusion, mostly from the 1970s and 1980s, and the published charts contradict one another because they are compiled from different studies with different methods. A rating tells you what a neat ingredient did to a rabbit, not what 8% of it in a balm does to your jawline, and there is no way to compute a formula's rating from its ingredients. The provenance is picked apart at comedogenic ratings.

What does predict trouble is duller: how much you leave on, how long it stays occluded, how much sebum you produce, and whether you have a history of comedones. That is why the dose section matters more than the ingredient list. A comedo takes weeks to form, so a three-day patch test cannot rule congestion out. Use the balm on one side of the jaw only, at 0.25 g, for four to six weeks, and compare: the staged method is at how to patch test a balm and the evidence at will a balm break me out.

One specific limit: this formula is not Malassezia-safe and should not be described as such. The yeast uses fatty acids in roughly the C11 to C24 range as a carbon source, so most triglycerides and many esters feed it. Squalane does not, but 25% squalane in a balm whose other 62% is metabolisable does not make the product safe for anyone managing that particular problem, and Malassezia-safe balms explains what would.

What goes wrong

Jar contamination deserves its own line. The balm is anhydrous, its water activity sits far below what moulds need, and it requires no preservative. A wet fingertip puts a droplet of water on the surface, and that droplet is where the argument stops holding. A spatula, a jar kept out of a steamy bathroom and a period after opening of six months deal with it: the reasoning is at do balms need preservatives.

The decision rule

Make the base formula, cure it 24 hours, weigh 0.3 g and time the shine on your own forearm. If it clears in ten to fifteen minutes, the formula is right for you and nothing else needs changing. If it is still shiny at twenty-five minutes, or it pills under whatever you wear in the morning, go to the oily-prone column. If it clears in under five minutes and your skin feels tight by the second hour, go to the dry column. In every case change the oil fraction and leave the 5% wax alone, because wax controls the jar and the oils control the face.

Two limits are worth stating. This balm holds water in, it does not put water on, so on dehydrated skin it belongs over something damp rather than on its own. And its occlusivity is a fraction of petrolatum's, a fair trade for the feel and a poor one if what you need is maximum protection on a cracked, weather-beaten patch. That case is a different formula.

Reviewed and updated 6 September 2026. Spotted an error? Tell us and we will fix and log it.