Pigment Formulation FAQ

Sometimes, but if the product has to be acidic, ultramarine is the wrong blue.

Ultramarine (CI 77007) is acid-sensitive and can decompose with release of hydrogen sulphide, which shows up as odour and fading. The acid sources in a real formula are broader than people expect: AHAs, salicylic acid, vitamin C and its derivatives, some organic-acid preservatives, and any emulsion running at low pH.

Three routes, in order of reliability. Keep the system neutral to mildly alkaline and confirm the actual window by compatibility test rather than by guesswork. Keep the acid actives in a separate phase where the format allows. And run an accelerated sample — 40°C for a month plus light — checking for shade shift, gas and odour.

If the concept is inherently acidic, switch to an acid-stable blue: certain organic blue lakes, or iron blue, each of which needs its own permitted-use check for the market and the application area.

Three things on top of permitted use: migration resistance, odour, and the narrower lip list itself.

Permitted use is the gate. Lip products are partly ingested, and every market runs a shorter list for lips than for face. A CI number that is fine in a blusher can be prohibited on lips, so check the lip list for the target market instead of reusing a face formula.

Migration is the experience. A lipstick carries a lot of wax and oil, and a pigment with poor migration resistance bleeds past the lip line, transfers to cups and stains. Lakes resist migration better than dyes; inorganic pigments such as the iron oxides resist it very well but read muted.

Odour is the one people forget. Some pigments carry a faint smell that is amplified at the lips, and impurities can react with fragrance to produce an off note. Assess it on its own during sampling rather than through the fragrance.

A long-wear, non-transfer claim is not a pigment decision alone — the film former and the treatment have to work with it.

It delays the fade; it does not fix it. If light fastness is the weakness, treat it as a selection criterion rather than something to patch later.

Assess first. Draw the pigment down in the target base, expose it under xenon or UV for 24–72 hours, read ΔE and compare visually. Above ΔE 2–3 the change is visible, though the acceptable threshold is the product's to set.

Then the mitigation, in three layers: choose an inherently light-fast pigment, since inorganics — iron oxides, chromium oxide green — generally outperform organic lakes; add a UV absorber to cut the energy reaching the pigment; and use a coating or treatment to lower photocatalytic activity at the surface.

Be clear about the limit. A UV absorber slows a fade that is inherent to the chromophore; it does not stop it. For products that live in light — a compact carried in a bag, anything left in a car — light fastness belongs in the selection brief from the start. We hold a range of light-stable water-soluble colours for exactly this need, with test data available on request.

There is no general number — it depends on the pigment's strength, the base colour and the target depth. Typically 0.5–10%, higher for deep shades.

Strength varies widely. Organic lakes are strong, so 1–3% gives clear colour. Inorganic pigments such as the iron oxides are weak and often need 5–10%. A white base containing TiO2 dilutes everything and pushes the number up; a transparent base reads more directly.

Overshooting costs three things. Skin feel degrades, because pigment absorbs oil, so the formula turns dry and draggy. Pressed strength falls, because pigment particles do not bond to the platelet fillers as well. And cost rises, especially with lakes.

There is a quieter problem: at high loading with poor dispersion, the surplus floats and you get speckling and rub-off on top of the wear problem.

Sample three levels — 1%, 3%, 6% — read L/a/b, and take the lowest point that hits the target depth. Do not fix it by eye in a single pass.

Usually all three, in that order. Work through them in sequence instead of adding dispersant and hoping.

Wetting comes first. Pigment has a large surface area and adsorbs air, so tipping it dry into a liquid gives powder-wrapped air pockets that no amount of later shear will open. Wet it into a paste with a small amount of compatible oil or carrier first, then dilute. A treated grade — AS or SI — wets markedly better and often solves this step on its own.

Dispersion next. Agglomerates need real shear to break, but too much shear fractures particles and shifts hue. Liquids go through a three-roll mill or bead mill; powders through high-shear mixing and screening. Judge by fineness of grind or a microscope slide, not by whether the colour looks even.

Stabilisation last: keep it from re-agglomerating with a dispersant or a higher yield value, and include freeze-thaw in the stability work because temperature cycling accelerates it. For a pressed powder with speckles, check blend uniformity as well — a pigment with a very different density from the base stratifies in the blender, so build the colour up by serial dilution.

Not by rule, but in a silicone system carrying a water-resistance or long-wear claim it is close to unavoidable.

The treated grades on our list carry the same pigment with different layers: SI for slip, water resistance and dispersion in silicone; AS for dispersion in oils and silicones with lower oil absorption and more loading headroom.

Where it is effectively required: silicone-containing formulas claiming water resistance or wear; high pigment loads such as liquid foundations and slurries; systems whose viscosity is sensitive to oil absorption; and anything needing long-term stability.

Where it can be skipped: pressed powders and eyeshadows with low oil content, and cost-driven filler formulas where the treated premium buys nothing.

Remember that changing the treatment changes the formula. Re-test dispersion, viscosity, stability and shade — the layer alters optical behaviour, and pale and bright shades show it first.

Four groups: CI number and permitted-use statement, heavy-metal report, microbiological report, and whatever the target market adds on top.

The CI statement is the core document. It has to name the CI number, the permitted application area — face, eye, lip — and the basis for the maximum use level. Our specification sheets carry the CI number per grade, and the COA should match it exactly.

Heavy metals — lead, arsenic, mercury, cadmium — matter because pigments are mineral or synthetic and the impurity profile follows the route and the process. Microbiological data matters most for naturally derived colours.

The market-specific layer is what gets missed: extra labelling for certain CI numbers, notification requirements for nano-scale material, tighter purity limits for lip products. Check each market separately instead of trying to cover all of them with one dossier.

Build a raw-material compliance file: lock the grade, take the COA per batch, and refresh the regulatory statement annually. The cost of doing that is small; the cost of hunting documents during an urgent shipment is not.

It is adhesion and compatibility, not a pigment that is inherently loose.

Three causes. The pigment does not bond to the platelet fillers — talc, mica — so it sits at the surface and wipes away. The formula is low in oil, so the powder has nothing to grip with. Or the loading is above what the base can carry and the surplus becomes loose dust.

The fixes follow the cause. A treated pigment improves compatibility with the oil phase and lifts adhesion. Adjusting the base — more platelet filler, or a filler with better adhesion — gives the pigment something to hold on to. And keeping the loading inside the base's capacity helps, using a high-strength lake so you need less.

If the problem is a liquid foundation wearing off rather than rubbing off, look at the film former and the adhesion promoter instead; the pigment is usually innocent. Quantify it with a rub test at fixed pressure and stroke count, plus a wear test, so comparisons are numbers rather than impressions.

Fix the grade, control incoming batches, and build a little tolerance into pale shades. The formula cannot eliminate it.

Pigments are mineral or synthetic and natural variation exists — inorganics especially, where the ore source matters. Three layers of control: hold one grade and one supplier instead of switching on price; require a coloristic value or ΔE on the COA, which most suppliers can provide; and inspect on receipt against a retained standard, rejecting or re-matching outside the agreed window.

Inside the formula, pale shades are the sensitive ones. Keeping a small headroom in the pigment loading lets you trim rather than reformulate. Deep shades absorb more drift but still need a fixed standard.

The habit that beats data: retain a drawdown panel from every batch — same base, same process — and compare panels on the next delivery. It is faster than reconciling L/a/b across suppliers.

If the same SKU is made at more than one site, tighten the standard and single-source the pigment. Two sites plus two pigment sources is where drift becomes unfixable.