Dreaded orange spots (DOS) and rancid soap

Weeks or months after a batch has cured, small orange spots appear on some of the bars, often with a stale smell. Soapmakers call them dreaded orange spots, or DOS. They are the visible sign of rancidity, the same slow oxidation that spoils a forgotten bottle of cooking oil, and they are one of the few soap faults that has been studied with some care. This guide sets out what is known, what is only believed, and what to do about it.

What DOS looks like

Pale yellow to rust-orange dots or blotches, a few millimetres across, scattered over the bar or gathered towards one end; sometimes a general yellowing instead. They arrive weeks or months after the soap was made, never on the first day, and they don’t wash off. The smell can come first or follow. DeeAnna Rouse of Classic Bells says rancid soap smells musty or “off” before it spots, and that some people liken it to old crayons; Susan Miller Cavitch saw it the other way round, one small yellowish circle first, more over the following weeks, and “an off-odor follows the first circle of rancidity”.

A pale bar of handmade soap on a wooden rack, its cut face scattered with small orange spots and one larger blotch Dreaded orange spots on a cured bar: small, soft-edged and orange to rust, appearing long after the soap was made.

Rancid soap still cleans, and no one has shown it harms the skin; Rouse doubts it would hurt anyone to wash with it. But the smell can linger on the skin, and spotted soap should never be sold or given away.

Other colours can be mistaken for DOS. The timing usually tells them apart.

What you see	What it probably is	The clue
Orange spots or blotches, weeks to months later, with a stale smell	DOS	Late, patchy, smells rancid
An even tan to brown that starts on the outside within days, in a vanilla or bakery scent, or with clove or cassia oil	Browning from the fragrance: Simmons and Appleton list clove, cassia, heliotropin and vanillin as scents that darken soap	Early and even, smells of the fragrance
Tan to orange-brown from the day it was cut, in a milk, honey or sugar soap	Scorched sugars	There from the start
Coloured rings spreading from bits of herb, tea or fresh plant matter	The botanical bleeding	Centred on the plant material

Oxidation, and why some fats are more at risk

A fat goes rancid when oxygen from the air attacks its fatty chains. The attack happens at the double bonds of unsaturated chains, and it runs as a chain reaction: once some chains have been attacked, the products go on to attack others. Along the way the chains break into smaller molecules, among them the aldehydes and ketones that smell rancid, and coloured products that show as yellow and orange.

The unsaturated chains are not all alike. Oleic acid, the main fatty acid of olive oil, has one double bond; linoleic acid has two and linolenic acid three, and the carbon that sits between two double bonds is where oxygen gets in most easily. Ralph Holman and Otto Elmer measured in 1947 that each extra double bond at least doubles the rate of oxidation. In a 2009 storage study at 55 °C, linoleic acid was lost up to twelve times as fast as oleic acid, and linolenic acid up to twenty-six times as fast.

Oleic acid has one double bond, linoleic two and linolenic three. The carbons between two double bonds are the most easily oxidised, which is why oils rich in linoleic and linolenic acids go rancid so much sooner than olive oil or the saturated fats.

The iodine value that soap calculators show is a measure of total unsaturation, and it is a fair rough guide. But it counts a double bond in oleic acid the same as one in linolenic acid, and in 2002 Gerhard Knothe argued that it is too general to predict oxidative stability. The linoleic and linolenic figures for a recipe are the better numbers to watch; the iodine value guide explains the difference.

Where the oxidation happens in a bar

Makers often blame the superfat, the oil left unsaponified on purpose. Kevin Dunn tested that. He aged pure olive oil soap at 60 °C in pure oxygen and in nitrogen, measured its colour from scans, and compared soap made with a 10 % lye discount against soap made with none. The 10 % soap ended up more orange, but both began to turn at the same time, about 75 hours into the test. Soap made with no lye discount at all still went orange.

The simplest explanation is that the soap itself oxidises as well as the leftover oil: sodium linoleate is as unsaturated as the linoleic acid it was made from. Soap boilers knew it; Simmons and Appleton warned in 1908 that “oleates are very liable to go rancid”. The superfat adds to the load, and it is a richer load than the recipe suggests, because the fat that stays unsaponified is not a fair sample of the oils. In a second study Dunn’s students extracted the leftover fat from finished soap and analysed it. In a soap of 90 % palm oil and 10 % grapeseed oil, whose fatty acids were 9 % linoleic, the leftover fat was 17 to 19 % linoleic. The unsaturated oils react more slowly, so they are what remains, and it made no difference whether the grapeseed oil went in at the start or at trace. Dunn called these results preliminary.

The same experiment showed how much oxygen matters. Stored under nitrogen, the olive soap coloured far more slowly than in oxygen. That is why full, closed oil bottles keep better than half-empty ones.

How the orange colour itself forms is less settled than the smell. Lever Brothers’ soap chemists wrote in 1970 that rancid odour was generally accepted to come from oxidation, but that spotting and discolouration were “less clearly understood”; what they could show was that traces of copper and iron made both worse.

What raises the risk

Factor	What the evidence says	What to do
Polyunsaturated oils: sunflower, soybean, cottonseed, grapeseed, hemp, walnut, flax; canola in larger amounts	Strong. The chemistry and the storage studies agree; Cavitch also found cottonseed’s keeping varied with the weather the crop met after ripening	Keep them a small share of the oils, or use high-oleic versions
Old or badly stored oils	Strong. Oxidation starts in the bottle, and Holman and Elmer found free fatty acids oxidise faster than whole fats	Buy fresh, store cool, dark and full, and use within their shelf life
Trace metals: tap water, copper or iron tools, metal racks	Strong. A 1970 Lever Brothers patent reports that as little as 0.2 parts per million of copper had a significant effect on the colour and smell of soap bars, and Dow sells EDTA for bar soap against brown spotting and rancidity. The idea that every spot sits on a speck of metal is Rouse’s observation, not a test	Distilled water, plastic or stainless steel, never cure on bare metal, add a chelator
A high superfat	Measured by Dunn: a deeper colour, not an earlier start	Keep it moderate; Soap Queen sees more DOS at 5 % and above
Heat and light	Strong. Heat is how Dunn sped his test up; light drives its own kind of oxidation, which Hall and Cuppett’s study was about	Cure and store cool and out of the sun
Humid, stuffy curing, or wrapping too soon	Maker experience (Soap Queen, Anne-Marie Faiola)	Cure with airflow and wrap only once cured
Oxidised essential oils	Maker experience: Rouse has seen old lavender oil set off rancidity, and Cavitch found citrus oils “vulnerable after just six months”	Use essential oils within their shelf life
Botanicals rich in copper, such as spirulina	One maker’s photographs, reported by Rouse	Worth knowing; not proven

“Strong” means well established in the chemistry of fats and consistent with what soapmakers see; “maker experience” means consistent reports without a controlled test.

The oils that carry the risk

Oil	Linoleic (%)	Linolenic (%)	Iodine value
Coconut	1–2.5	0–0.2	6.3–10.6
Cocoa butter	1.6–4.8	0–1.5	33–42
Tallow	1–6	up to 1.5	40–53
Shea butter	4.4–7.9	–	30–75
Palm	9–12	0–0.5	50–55
Lard	4–12	up to 1.5	55–65
Olive	3.5–21	up to 1.2	75–94
High-oleic sunflower	2.1–17	0–0.3	78–90
Avocado	6–20	up to 3	75–95
Sweet almond	20–30	0–0.5	85–109
Rice bran	21–42	0.1–2.9	90–115
Canola	15–30	5–14	105–126
Cottonseed	46.7–58.2	0–0.4	100–123
Soybean	48–59	4.5–11	124–139
Sunflower	48.3–74	0–0.3	118–141
Grapeseed	58–78	0–1	128–150
Hemp seed	44.5–65.5	13.5–30.5	145–170

Ranges from the sources behind each oil’s page: the Codex food standards for most, and pharmacopoeia monographs, supplier specifications or published analyses for the rest. A dash means the source gives no figure.

Classic Bells’ rule of thumb is to keep linoleic and linolenic acids together at or below 15 % of a recipe’s fatty acids. Rouse is candid that the rule has no strong scientific basis; it simply seems to work. The chemistry supports its direction, so it is a sensible target, but it is not a line: a bar at 16 % is not doomed, and one at 12 % is not safe from a rusty rack.

The calculator makes the check easy. A bar of olive oil 40 %, coconut oil 25 %, ordinary sunflower oil 20 % and shea butter 15 % comes out at 18.8 % linoleic and linolenic together, with an iodine value of 70. Swap the sunflower oil for high-oleic sunflower oil and the same bar comes out at 8.5 %, with an iodine value of 61, and needs 139.0 g of sodium hydroxide per 1000 g of oils instead of 139.4 g. Our budget bar keeps canola to a quarter of the oils for the same reason, the supermarket oils bar holds it to 15 % and comes out at 12.4 %, and our palm-free classic uses the high-oleic sunflower.

Make the swap. The high-oleic version of this bar needs 139.0 g of sodium hydroxide for 1000 g of oils. The link adds EDTA and rosemary extract at 0.1 % of the oils each, the dose Dunn tested.

Open it in the calculator

What the tests say about additives

Dunn tested the additives soapmakers recommend against DOS, each at 0.1 % of the oil weight, in the same aged olive oil soap.

Additive, 0.1 % of the oils	Result
None	Colour took off at about 75 hours
Grapefruit seed extract, vitamin C, vitamin E or sodium citrate, each alone	No different from no additive
BHT	Held off to about 150 hours, twice as long
Rosemary extract (ROE)	Had not taken off by the end of the test, about 300 hours; stayed below his “noticeably orange” mark
EDTA	The same as rosemary extract
ROE with EDTA, or BHT with EDTA	Past 300 hours, and paler than either alone
BHT with sodium citrate	The best: past 300 hours at about the colour of fresh soap
ROE with sodium citrate	Worse than rosemary extract alone

Dunn, “The Dreaded Orange Spot”, 2005/2006: 1 g samples of soap aged at 60 °C in oxygen, colour measured from daily scans.

How to use them

EDTA. Tetrasodium EDTA is the grade for soap. Dow’s handbook gives 0.1 to 0.2 % for bar soap, Dunn tested 0.1 % of the oils, and Classic Bells gives 0.05 to 0.25 % of the batch against rancidity. It uses no lye. Dow advises adding a chelator early, “during the water phase”, and Dunn adds it to the lye; our calculator puts it in the water before the lye goes in, which suits an ordinary lye solution. Rouse has seen EDTA come out of strong lye as a thick white sludge, though: it stayed dissolved below about 35 % sodium hydroxide in her tests, and another maker saw it come out at 33 %. With a strong solution, dissolve it in a little of the recipe’s water held back for the purpose and stir that into the oils instead. Disodium EDTA is a grade for neutral products; it uses up some lye, which the calculator allows for.

Rosemary extract (ROE). An oil-soluble extract rich in carnosic acid, and a different thing from rosemary essential oil, which does not do the same job. Dunn used 0.1 % of the oils. Food studies suggest more is not better: in soybean oil exposed to light, 0.02 to 0.05 % protected best and 0.5 % acted as a pro-oxidant, and in rapeseed oil stored at room temperature 50 to 200 mg per kg (0.005 to 0.02 %) was the best range. Extracts are sold at different strengths, so follow the supplier’s figure; Classic Bells keeps to 0.05 % of the fats, and adds it to the oils as soon as they are bought rather than waiting for soap day. Our Aleppo-style laurel bar stirs 1 g into its 1000 g of fresh oils, Dunn’s 0.1 %, to carry it through a long cure.

BHT. A synthetic antioxidant that dissolves in warm oil. At 0.1 % of the oils it doubled the time to colour in Dunn’s test, and with 0.1 % sodium citrate it did best of all. EU law allows up to 0.8 % in rinse-off products, so 0.1 % is well inside it. Dunn’s own advice was practical: if your customers prefer natural ingredients, use ROE; if they don’t mind, BHT with sodium citrate is likely to do better.

Sodium citrate. No help against DOS on its own at 0.1 %, but a good partner for BHT. Craft doses are far higher, 1.3–3.9 % of the oils (Classic Bells), and no one has tested those against DOS. In the bar it binds metal traces; it cannot soften the water you rinse with (see soap in hard water).

Vitamin E. It did nothing in Dunn’s test. In food oils, alpha-tocopherol has repeatedly behaved as a pro-oxidant: added to soybean oil at levels from 0.002 to 5 %, it sped oxidation at every level except the lowest. Add it for other reasons if you like, but not to prevent DOS. The antioxidants and chelators guide goes further into all of these.

Storage and handling that help

Oils. Buy them fresh and in amounts you will use. Keep them cool and dark in full, closed containers, and refrigerate fragile ones such as hemp and grapeseed. Smell an oil before you soap it: rancid fat makes rancid soap.
Water. Make the lye with distilled or deionised water. Tap water is often blamed for DOS; the chemistry supports that, and many makers use tap water without trouble, so it is a risk to reduce rather than a certain cause.
Tools. Plastic or stainless steel; never copper, brass or bronze, and not cast iron, which Cavitch found discolours soap.
Curing. On wood, plastic or card, never on bare metal. Rouse warns that metal ions migrate through paint and paper, so cover metal shelving with thick cloth or sturdy plastic. Cure somewhere cool, dry and airy, out of the sun, and wrap bars only once they are cured. Cavitch never stacked curing bars and kept them out of closets and drawers, which she found gave too little air “for maximum protection against rancidity”.
Keep an eye on stock. Store older bars apart from new ones, and use bars made with the less stable oils first.

The curing and storing guide covers racks, airflow and packaging.

If your soap has spots

Rancidity can’t be reversed, and Rouse is clear that rebatching does not cure it. If a bar has only a few spots, cut them out and keep the rest for your own use. Throw away bars that are rancid all through, and any liquid soap that smells rancid, since there are no spots to cut out of a bottle. Don’t sell or give away spotted soap. Keep affected bars away from the rest of the batch, then look for the cause: the age of the oils, the water, the rack, the storage room. The pattern usually points to it. Spots on the faces that touched a rack point to metal; an even yellowing and a rancid smell through every bar point to the oils.