Making lipstick: a wax skeleton, 10 percent pigment and a bullet that releases

A lipstick formula at 24 percent total wax and 8 to 12 percent pigment, with the three-wax skeleton, an 82 C pour into a 12.1 mm mould, chilling and flame polish.

A lipstick is not a tinted balm poured into a different shape. It carries five to ten times the pigment, stands up unsupported at 40 C, comes out of a mould without tearing, and still lays a film down under the light pressure of a hand. Those four requirements pull against each other, and this formula is an attempt to hold all of them at once.

• Wax 14 + 6 + 2 percent
• Pigment 8 to 12 percent
• Paste 40 percent in castor
• Pour 78 to 84 C
• Bullet 12.1 mm, about 3.8 g

The problem a lipstick sets that a balm does not

A twist-up balm is extended four to six millimetres at a time inside an 11.4 mm barrel that supports it almost to the tip. A lipstick is drawn out ten to fifteen millimetres and is judged on whether the tip keeps its edge after a fortnight in a handbag. Treat the extended bullet as a cantilever and the consequence falls out: bending stress at the base is tip load multiplied by extension, so the same finger pressure at three times the extension is three times the stress at the root. That is why a stick that works as a tinted lip balm snaps at the case rim when poured as a lipstick.

The opposing requirement is deposit, and nobody presses hard on a lipstick. Rest a bullet on a kitchen scale and draw it across at the angle you would use on a lip: the reading lands around 80 to 150 g for most people, near 100 g for the light-handed. That is a workshop measurement rather than a standard, but it sets the order of magnitude. The stick has to be self-supporting at 40 C and give up a film under about the weight of a bar of chocolate.

Pigment makes both problems worse. Ten percent of hard mineral particulate is inert filler: it thickens the melt and interrupts the continuity of the wax crystal lattice as the stick sets, so a coloured stick is measurably more brittle than the same base uncoloured. A lipstick therefore needs more wax than a balm of the same firmness, and a blend chosen for toughness rather than raw hardness.

The three-wax skeleton, and what each one is for

Three waxes are not showing off. Each covers a failure the other two cause.

Beeswax at 14 percent is the toughness. Its mixed ester, hydrocarbon and free acid fractions crystallise into a coarse, plastic network with a needle penetration around 22 dmm, soft as waxes go, and that plasticity is what stops a heavily pigmented stick shattering. At 62 to 65 C it melts low enough for the bullet to yield under a light draw. Beeswax is here on mechanical grounds, not tradition.

Candelilla at 6 percent is the snap and the release. Roughly three quarters long-chain alkane, principally hentriacontane, it forms a finer network than beeswax and gels an oil at 1 to 2 percent against beeswax's 2. That shows up as a clean shear edge on the lip and as a bullet that lets go of a silicone mould in one piece, and its NF range of 68.5 to 72.5 C lifts the system's set point. The substitution arithmetic is on the candelilla wax page.

Carnauba at 2 percent is the drop point and the gloss. It is the hardest solid on the bench at 0 to 1 dmm and the weakest oil structurant of the three, needing 4 to 5 percent to gel an oil at all, so using it for hardness is a mistake. The 80 to 86 C drop point is the real reason it is here: two points lift the temperature at which the bullet gives up its shape by several degrees at almost no cost in brittleness. Above 3 percent it turns gritty and matte, as carnauba wax sets out.

The last row explains why two numbers get quoted for one formula. Lipstick is built as a colourless base which then has pigment paste stirred into it, and formulators quote wax against that base: 22 g of wax in an uncoloured 90 g base is 24 percent, and the same wax in the finished 100 g stick is 22 percent. Use the base figure when comparing skeletons across shades, or a matte at 12 percent pigment and a sheer at 8 will look like different formulas when they are not.

Pigment: 10 percent, ground into castor at 40 percent loading

Dry pigment stirred into a hot melt does not disperse. It wets poorly, agglomerates, and settles out of the mould in the twenty minutes before the stick sets, which is the commonest way a first lipstick goes wrong. Grind it into castor oil first. Forty percent loading means 40 parts pigment to 60 parts castor by weight, so a 100 g batch carrying 10 g of pigment commits 15 g of castor and a 25 g paste, a quarter of the formula, before a single wax is weighed. That is close to the practical ceiling: at 50 percent the mix behaves like putty, and below 30 percent you dilute the base with more castor than the shade needs.

Castor oil is the vehicle because nothing else in the plant oil cupboard wets mineral pigment as well: its ricinoleic acid carries a free hydroxyl group on the chain, which makes it far more polar than an ordinary triglyceride and lets it displace air from the pigment surface. That same polarity gives lipstick its gloss and cling. Grind on something flat and hard. A glass board and a stiff palette knife, working the paste into a film and scraping it back a dozen times, applies real shear, whereas a pestle and mortar mostly stirs. Draw the paste down a white card to check: at 40 percent it should be a smooth band with no specks against the light, and specks are what a customer will feel.

The shade below is a mid warm red: 4 percent red iron oxide, 1.5 percent yellow, 1.5 percent black and 3 percent titanium dioxide. Yellow warms, black deepens and greys, and titanium dioxide is the opacity lever. Every point of it lightens the shade toward a pastel and pushes the finish toward matte, and above 4 percent the stick reads chalky. Three percent is roughly where the film stops being a stain and becomes a coat of paint. The rest of the pigment family, including what mica does and does not do, is in colourants and tints.

The formula

Weight percent of the total batch, adding to exactly 100. The gram columns cover a 100 g batch, roughly 22 bullets, and a 250 g batch, roughly 55. Build the paste first, then the base, then combine.

The esters do specific work. C12-15 alkyl benzoate wets iron oxides better than any aliphatic ester and has a refractive index near 1.48 against roughly 1.45 for a plain triglyceride, so it adds gloss where more castor would only add tack, while caprylic/capric triglyceride shortens the greasy afterfeel 30 percent castor leaves. Both are covered in cosmetic esters. With neither to hand, put the 8 points into jojoba and accept a heavier stick.

Melting, and the compromise the mould forces on you

Melt phase A in a water bath and hold at 85 to 88 C (185 to 190 F) until it is water-clear. It has to be that high because carnauba does not go fully into solution below roughly 82 C, and under-melted carnauba is the commonest cause of gritty specks appearing on a finished bullet a day later. Hold, do not rush. Stir the paste in at 85 C for a full two minutes, lift the jug out of the bath, stir in the tocopherol and go straight to the pour.

Pouring the 12.1 mm mould

Twelve point one millimetres is the standard bullet diameter, and mould and case have to be a matched pair: a 12.1 mm bullet rattles in a 12.7 mm cup and will not enter an 11.8 mm one. Buy them together and check the specification before the first pour, not after twenty-two bullets have set. Wax shrinks a few percent by volume on solidification, so a bullet from a nominal 12.1 mm cavity comes out a couple of tenths under nominal, and that shrinkage is exactly the clearance the cup needs.

1. Warm the mould to 35 to 40 C. A cold mould chills the leading edge of the pour before the cavity is full, giving horizontal drag lines up the side of the bullet. Ten minutes on a radiator is enough, and above 45 C release takes far longer.
2. Pour at 78 to 84 C, in one pass. Fill each cavity in a single continuous movement, because stopping halfway leaves a seam that survives polishing. Work fast: you have a couple of minutes across a full plate before the jug goes back in the bath.
3. Overfill and scrape. Let each cavity dome slightly proud, wait ninety seconds for the surface to skin, then draw a metal scraper flat across the mould face. Too early and it smears, too late and you are chiselling.
4. Chill 20 minutes at 4 to 6 C. Fast cooling gives the fine, uniform crystal network that makes a bullet strong and glossy; a slow bench cool gives a coarser network, a duller surface and a weaker stick. Not a freezer: the thermal shock cracks bullets down the axis.
5. Release. Flex a silicone mould from the base and let each bullet drop into a gloved hand. If they stick they are not cold enough, and another five minutes usually fixes it.

The reasoning behind those temperatures is at pour temperatures. Let the bullets rest 24 hours before you judge hardness, because wax crystallisation continues well past the point at which a stick is visibly solid.

Flame polish and fitting the bullet

A released bullet carries a faintly textured surface that mirrors the mould rather than the material. Flame polishing melts the outermost few tens of microns and lets surface tension pull it smooth. Hold the bullet by its base with tweezers or in the case and pass it through the edge of a small flame, or under a hot air gun on low, in one rotation lasting under a second per face. Stop as soon as the surface goes wet-looking; if the tip rounds, the bullet needs remoulding. A hot air gun is slower and far harder to ruin a bullet with.

Fitting is the step people rush. The bullet base pushes into the cup of the swivel mechanism, which grips on friction alone, so push straight down with a flat tool rather than on the sides of the taper and seat it until the cup lip is flush. A bullet left half a millimetre proud extends too far when wound out, lengthening the lever, and that is a common reason a stick snaps at the rim in normal use.

Two tests worth running before you sell anything

The stand test at 40 C. Wind a bullet out 12 mm, stand it upright at 40 C (104 F) for an hour and look. A pass is no slump, no bend and no beads of oil. It is a proxy for drop point, which for a lipstick should sit near 60 to 65 C against roughly 45 to 55 C for a soft tube balm; drop point proper needs a standard method such as ASTM D127 and equipment most makers do not have.

Breaking strength. Wind a bullet out 12 mm, clamp the case horizontally at a table edge, hang a small container from the tip on a thread and add water until the bullet snaps. Record the mass. This is the workshop version of the bullet breakpoint test the industry runs on a texture analyser.

Run both on three bullets from one batch. The spread tells you as much as the average, and a wide spread usually means uneven chilling across the mould plate. Firmness methods generally are collected in measuring balm hardness.

What goes wrong

Adjusting the formula

• Firmer bullet: raise candelilla from 6 to 7 percent and drop high-oleic sunflower by the same point. Candelilla buys hardness with the least loss of payoff. Raising beeswax instead makes the stick waxier on the lip.
• More heat resistance: carnauba from 2 to 3 percent out of the jojoba. Watch for drag and a matte finish, and stop at 3.
• Deeper coverage: pigment from 10 to 12 percent, which means an 18 g castor commitment to the paste. Add a point of wax for the extra brittleness.
• Sheerer: pigment to 8 percent with titanium dioxide cut to 1.5. Sheer shades need the white reduced, not just the total.
• Glossier: C12-15 alkyl benzoate from 5 to 8 percent out of the sunflower. More castor instead adds tack as well as gloss.
• Vegan: replace 14 percent beeswax with roughly 11 percent candelilla plus 2 percent rice bran wax and return the balance to castor. Expect a drier, snappier, more brittle bullet, and test breaking strength before committing. The conversion is in vegan beeswax substitutes.
• Scaling: above about 500 g the melt cools too fast between cavities to pour a full plate by hand. Use two jugs, or size sub-batches to one plate each with the batch calculator.

What this formula will not do

It is not long-wear and cannot be made long-wear. Transfer-resistant lipstick works on a different principle: a volatile isoparaffin carrier that flashes off in under a minute, leaving a film-forming polymer that locks pigment against the lip. Wax plays almost no part, and those carriers have flash points around 40 to 45 C and are not sold in hobby quantities. A wax bullet transfers, and the honest target is four to five hours.

It offers no sun protection. The 3 percent titanium dioxide is there for opacity, gives no meaningful protection at that level, and must not be described as if it did; the reasons are in SPF in lip balm.

The regulatory work is yours rather than the formula's. Every colourant has to be confirmed as permitted for lip use in the market you sell into, with the evidence on file, and the two markets do not carry identical lists. Titanium dioxide is the live case: the EU withdrew it as a food additive in 2022 after EFSA concluded genotoxicity could not be ruled out, and its position in oral and lip cosmetics has been examined since, so read the current Annex IV entry rather than a two year old blog post. The duties are in selling balms in the UK and EU and labelling cosmetics in the US. If you cannot show which listing covers each pigment in your stick, make it for yourself and stop there.

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