Soap problems, from soda ash to seizing: causes and fixes

This page covers the problems makers meet most often in cold process soap, grouped by when you notice them. Each row says what the fault looks like, what usually causes it, what to do with the batch in front of you, what to change next time, and whether the soap is safe to use. The sections after the tables deal with the judgement calls, and five guides go deeper into the commonest problems.

Is it unsafe, or only ugly?

Soap is unsafe when it carries free lye, and that happens in two ways. Either the batch had more lye than its oils could use, from a weighing slip, a forgotten oil or a recipe nobody ran through a calculator, or the lye never mixed evenly into the oils because the batter was poured at false trace, separated in the mould or held undissolved lye. Nearly everything else on this page is about looks, feel or shelf life.

The signs of free lye are:

• pockets of clear liquid in the cut bars, or liquid weeping from fresh soap;
• hard, gritty or glassy white crystals inside the bar;
• an oily layer over a hard, chalky one;
• a bar that is harsh and crumbly all through, or that stings.

To check a suspect batch:

1. Wait. DeeAnna Rouse of Classic Bells waits two to three days before testing, to let saponification finish.
2. Cut it and look, with gloves on. Pockets, crystals and layers show on a cut face, and they are the plainest evidence there is.
3. Try a drop of phenolphthalein on a fresh cut. It turns pink from about pH 8.2, so a faint pink that creeps in slowly shows on well-made soap too; an instant, vivid magenta, above all in spots or streaks, is the look of free lye.
4. Measure pH properly, or not at all. On a calibrated meter, a 1 % solution of well-made soap reads about 9 to 10, a little higher if it is mostly olive oil, and no pure soap in Kevin Dunn’s figures reaches 12, so 12 or more means free alkali. Rouse finds cheap strips read 2 to 3 units low, and 1 part per thousand of excess lye raises the pH by only about 0.3, so a strip can flag a grossly lye-heavy bar but cannot clear a doubtful one.

The definitive check is a titration of the free alkali; Dunn treats less than 1 part per thousand of free sodium hydroxide as safe. A small excess fades as the soap cures: in one of Dunn’s bars, the free alkali at the bottom fell from 4.9 to 0.9 parts per thousand over 60 days. Pockets of lye don’t fade. The traditional zap test, a fingertip touched to the tongue, puts soap in your mouth at the moment it is most likely to be caustic, and our pH guide explains why we don’t recommend it.

The soap pH and testing guide goes through each test in detail.

The problems at a glance

In the pot

“Test first” means handle with gloves and check for free lye, as above, before anyone uses the soap.

In the mould

Most of the heat problems in this table are cosmetic; the overheating guide explains when they are not.

When you cut

A cut face with a smooth knife tells air marks from true white spots; see below.

During cure and storage

“Own use only”: rancid soap is not known to harm skin, but it smells, and it should not be sold or given away.

In the pot

How fast batter moves is decided by the fragrance, the heat, the water and the fats. Eugenol, the main aroma chemical in clove and cinnamon leaf oils, is the best-measured accelerator: in Dunn’s measurements, batter thickened faster with more of it, as it did with more heat and less water. The seizing guide covers what to do mid-batch and how to test a fragrance before trusting it with a full loaf. False trace, which is cool fat setting rather than soap forming, is covered in the trace guide. Separation is the one pot problem that becomes a safety problem, and the soft soap guide covers it in full.

In the mould

A loaf heats itself. Too much heat cracks it or pushes it out of the mould, heat that reaches the middle but not the edges leaves a partial gel, and a cool loaf with its top open to the air grows soda ash. The overheating guide and the soda ash guide cover these, and the gel phase guide explains how to steer a loaf one way or the other.

Partial gel is cosmetic, and there may be no undoing it: Kelly Cable finds nothing changes the look once the soap has set. Rouse does try, warming the soap, cut or uncut, at the oven’s lowest setting, usually 63 to 77 °C (145 to 170 °F), for about an hour and letting it cool slowly; it may help soften the ring, and she can’t promise it will go, though some makers who have tried it say it becomes less obvious. Cable prevents it by putting the fresh loaf in a 77 °C (170 °F) oven until it has gelled evenly. Next time, either insulate the whole mould or chill it: the halfway house is what causes the ring.

Soap stuck in a mould is still too soft, and small silicone cavities, which lose their heat and rarely gel, are the usual culprits. Soap Queen freezes a stubborn mould for 24 hours and Anne-Marie Faiola freezes soap solid if it tears; frozen soap no longer flexes, so it pops out. Sodium lactate helps too: Corbion, which makes it, sells it for bar soap to harden bars and help them release from moulds. The unmoulding guide and the moulds guide cover the rest.

When you cut

Holes. Round holes are air: whipped in by a stick blender switched on before it was tipped to let its trapped air out, splashed in with the lye, or caught when thick batter was poured. Soap Queen burps the blender, pours the lye down its shaft and taps the mould firmly once it is filled. A long, smooth tunnel down the middle of a loaf is a heat tunnel, left by overheating.

White spots. Most white flecks inside a bar are high-melting fat that set before it could mix in. Amanda Aaron of Lovin Soap got them with her lye at 16 °C (60 °F) and oils at 27 °C (80 °F), and soaps butter-rich recipes at 38 to 49 °C (100 to 120 °F) to avoid them; palm oil scooped from a part-melted tub leaves its stearin in clumps the same way, so melt and stir the whole container before weighing. She also shows that a wire cutter can drag tiny air bubbles into round marks that look like spots: cut a slice with a smooth knife, and if the marks vanish, they were air. Hard, glassy or damp crystals are a different matter; treat them as lye until tested. White streaks can go either way. Alicia Grosso finds that streaks which neither leak nor sparkle usually mean fragrance or colour not fully stirred in, a matter of looks, while sparkly, crusty white means too much lye; Susan Miller Cavitch’s rule is to make sure white swirls are not shiny chunks of lye before shrugging them off.

Lye pockets. Cavitch warns that watery or oily puddles in the pot “signal a poorly mixed solution that will result in pockets of solid lye in the final bars”. If the last of the batter at the bottom of the pot is watery and uneven, the mixing wasn’t finished, and she leaves it out of the mould.

Crumbly or brittle bars. Test first, because one cause is too much lye. If the soap passes, the usual cause is timing. Rouse cuts when a loaf feels like fridge-cold cheddar; if it has gone too hard, she warms it on the oven’s lowest setting, checking every 10 to 15 minutes, and cuts it with a flat blade or a wire, because a wedge-shaped knife splits firm soap. Salt bars go hard fast and need cutting within hours.

Oily patches. A thin film of oil on top of a new loaf, less than about 3 mm (1/8 in), can soak back in over one to three days, Faiola finds. If it doesn’t, or the oil sits in pockets, treat the batch as separated. Oil weeping from cured bars is often fragrance that never mixed in: wipe it off, and next time use less of that fragrance and stir it in more thoroughly. Grosso draws the line between the two kinds of leak: oil seeping from cut bars is usually too much essential oil and seldom serious, while a watery leak from the block is caustic lye solution from a weighing mistake, and the batch goes in the bin.

During the cure

Two of the problems that appear after cutting come from the air. Sweating is glycerin pulling water from humid air and does no harm; orange spots are oxygen turning the fats rancid, and are permanent.

Colour. Fresh soap is strongly alkaline, and many colourants change in alkali. Oxides and ultramarines hold their colour. Faiola finds some micas shift, blues towards purple and greens towards coral, while others fade entirely, and many food dyes fade to nothing in fresh cold process soap. Natural colours move most: alkannin, the pigment in alkanet root, goes from red through purple to blue as the pH rises, and curcumin, in turmeric, turns red in strong alkali. Many natural colours also fade in light. Vanilla-type fragrances turn soap tan to brown over days or weeks, starting at the surface; that is the vanillin, not a fault in the soap. The colouring guide and the natural colourants guide go further.

Fragrance. Light top notes, citrus above all, fade first, and soap’s alkali breaks some aroma chemicals down. Faiola finds ordinary orange oil vanishes from cold process soap, while five-fold or ten-fold orange lasts better. The dose matters too, within the supplier’s safe limit, and so does storage: cured soap keeps its scent longer wrapped, cool and dark. The scenting guide covers doses and limits.

Rescue, rebatch or bin

The rescue cook is ordinary hot process on a failed batch; the hot process guide covers the method and the rebatching guide how to add a missing oil by the numbers.

To bin a batch safely, let it harden, then wrap it in two or three bags, wearing gloves and goggles, and put it out with the household rubbish. Never pour soap batter or lye down a drain.