Ratanjot (Alkanet Root): Rogan Josh Red, Hair Claims, Liver Warning

Ratanjot (alkanet root) is a dark, papery root that bleeds deep red into hot oil, and it is the real reason a Kashmiri rogan josh glows crimson without being fiery. The same root is sold as a hair oil ingredient and a skin remedy. Two things most sellers leave out: the colour only works in fat, never in water, and the plant family it comes from carries liver-damaging alkaloids, so the root is for colouring oil, not for eating, drinking or giving to children.

Prefer to listen? This post as a 2 min video · Watch the video

What plant is ratanjot, and what does the root look like?

Ratanjot is the dried root of plants in the borage family that store a red pigment in their bark. In the Mediterranean and the Middle East the source is Alkanna tinctoria, dyer’s alkanet, used since Roman times to tint wine, wax and cosmetics. In Kashmir, Ladakh and the Pakistani side of the Himalaya the roots come from Arnebia species, especially Arnebia benthamii, which also goes into the Unani compound sold as gaozaban. Herbalists treat all of them as one commodity, “alkanet”, because the pigment is the same.

The dried root is a finger-thick, twisted stick. The outer bark is thin, papery and purple-black, flaking off in layers; the wood inside is pale and dull. Rub the bark between wet fingers and they stain purple. That bark is where the colour lives, which is why thin roots with lots of bark colour better than thick woody ones.

The pigment is a pair of near-identical naphthoquinone molecules, alkannin and shikonin, together known as shikalkin, and they are lipophilic, which means they dissolve in fat and alcohol and barely at all in water. That one fact explains everything. It explains the cooking technique. It explains why home experiments fail.

Why rogan josh and tandoori chicken are red

A proper Kashmiri rogan josh is deep red and only medium hot. Part of that is Kashmiri chilli, which is mild and bright, and part is dried cockscomb flower (mawal) steeped for colour. The traditional third source is ratanjot, dissolved in the hot fat at the start or stirred in as coloured ghee at the end. The root itself is never eaten; it is lifted out. The red oil on old-style tandoori chicken worked the same way before synthetic food colours took over.

Today much of the red in restaurant food is synthetic colour, not ratanjot, and in home cooking a good Kashmiri chilli or paprika does most of the work without any root at all. Ratanjot is optional. Tradition, not necessity. Our paprika vs Kashmiri chilli vs cayenne page compares those. If you want ratanjot’s particular wine-red, use the fat method above and keep the amount to a piece of root per pot. We are deliberately not giving a rogan josh recipe here; this page is about the colouring step only.

Ratanjot for hair, skin, lips and burns: the claims

What people say

Reels show ratanjot soaked in coconut or sesame oil to “reverse grey hair”, “stop hair fall”, heal burns, darken lips and lighten scars. Some sellers call it a natural hair dye.

Our honest take A pretty red oil; not a grey-hair or growth fix

• Grey hair: the pigment coats the hair shaft faintly and rinses out. It does not restore melanin and cannot make grey hair grow in dark. No evidence supports the reversal claim.
• Hair growth and hair fall: no human trials. Any benefit people notice is the oiling and scalp massage, which any plain oil provides. See our hair oil comparison for what oils can and cannot do.
• Wounds and burns: shikonin and alkannin show anti-inflammatory and wound-healing activity in laboratory and animal work, and shikonin-based ointments have been used in East Asian medicine for centuries. That is a reason to take the tradition seriously, not a licence to put a homemade root oil on a fresh burn. A burn needs about 20 minutes of cool running water, a clean covering and a doctor if it blisters widely, is on the face, hands or genitals, or is on a child. Never butter, toothpaste, ice or herbal oil on a fresh burn.
• Lips and skin lightening: ratanjot oil tints lips red; it does not lighten anything.

If you use ratanjot in a hair or lip oil, patch test 48 hours before on the inner forearm, keep it off broken skin and out of the eyes, and expect it to stain pillowcases, collars and light-coloured skin for a day or two.

The liver warning: pyrrolizidine alkaloids

This is the part of the ratanjot story almost nobody tells. The borage family is one of the main plant families that make pyrrolizidine alkaloids (PAs), the same compounds that make comfrey tea dangerous, and in the liver these alkaloids are converted into reactive molecules that attack the lining of small blood vessels. The characteristic injury is hepatic veno-occlusive disease. The liver swells. Fluid collects in the belly. In bad cases the liver fails. Europe’s food-safety authority concluded that these alkaloids may act as genotoxic carcinogens in people, and that regular intake from herbal products can come uncomfortably close to doses that cause acute harm. Suckling infants are more sensitive than adults.

Both ratanjot sources are affected. Researchers analysing Arnebia benthamii from the Neelum valley of Azad Kashmir identified heliotrine, echimidine, lycopsamine and europine, all hepatotoxic PAs, and advised that its use as an internal medicine be stopped until proper quality controls exist. Alkanna tinctoria is also recorded as containing PAs, with levels poorly defined. This is why alkanet is not an authorised food colour in the European Union, and why we describe it as a colouring step in fat, where only traces pass into the dish, rather than as something to eat or drink.

Is ratanjot safe to eat?

No. Traditional cooking never ate it. It coloured the fat and was removed, and the amount of pigment and alkaloid that crosses into a whole pot of curry from a 3 cm piece steeped for half a minute is tiny, which is why this one use seems reasonable to us for healthy adults who want the traditional colour. Everything beyond that is off: no ratanjot tea, decoction or “kadha”, no chewing the root, no daily “ratanjot water”, no ratanjot in a child’s food, and nothing at all during pregnancy or breastfeeding, because PAs pass into milk and across the placenta.

Real ratanjot vs the poisonous namesake

In Hindi-speaking India “ratanjot” is also a common name for Jatropha curcas, the physic nut or jungli arandi, a hedge shrub with black oval seeds about the size of a peanut. Those seeds contain curcin, a toxalbumin, and a purgative oil, and they are a recognised cause of accidental poisoning in children, who eat them thinking they are nuts. Indian paediatric case series and French poison-centre reports describe vomiting, abdominal pain, diarrhoea and lethargy within hours. Nobody colours food with Jatropha, but the shared name means an online order or a confused pansari can hand you the wrong thing, and a child in the house can find it. Check before you buy. Check again before you store it.

1. Look at the form. True ratanjot is a twisted root with flaking purple-black bark and a pale woody core. A hard black seed, kernel or nut is Jatropha. Do not use it for anything.
2. Do the oil test. A pinch of real root in a teaspoon of warm oil turns it red in seconds. Jatropha seed does not colour oil. A root that gives no colour is stale, woody or something else.
3. Check for dye. Rub the root on a wet white tissue: real bark leaves a purple-red streak but the inner wood stays pale. A root that bleeds bright red from the wood too, or that colours cold water strongly, has probably been dyed to look fresh, since alkannin barely dissolves in water.
4. Ask the botanical name. A seller who says Alkanna or Arnebia knows the product. Anything described as “ratanjot beej” (seed) is not alkanet.
5. Prefer thin bark-rich pieces over thick woody chunks; the colour is in the bark.

The same logic that catches dyed ratanjot also catches dyed chilli and turmeric; see lal mirch adulteration tests.

Who should avoid ratanjot, and what means stop

What to use instead for red colour

For most dishes you do not need ratanjot. Really. Kashmiri chilli powder gives red with mild heat; paprika gives red with none; a tablespoon of either bloomed in hot oil for ten seconds gives much of the glow. Dried cockscomb flower (mawal) steeped in hot water is the other traditional Kashmiri colour and is water-soluble, so it works in the gravy itself. Beetroot gives a purple-red that fades with long cooking. Tomato paste cooked down adds depth. Synthetic red food colours are legal in small amounts in Pakistan and India but check that the powder is a labelled food-grade colour, not an industrial dye; lead chromate in turmeric is the cautionary tale. More kitchen chemistry is in Kitchen Science.

Bottom line

Ratanjot is alkanet root, a fat-soluble red dye from the borage family. Used the traditional way, a small piece steeped in hot ghee and then removed, it gives rogan josh its colour with only traces reaching the plate. It does not reverse grey hair or grow new hair, and its wound tradition is no reason to put homemade oil on a burn. Because its alkaloids can injure the liver, nobody should drink it, eat the root, or give it to children or pregnant women, and anyone sold a black seed under the name ratanjot has been sold a poison, not a spice.

Ratanjot: questions people ask

Sources

• EFSA CONTAM Panel. Risks for human health related to the presence of pyrrolizidine alkaloids in honey, tea, herbal infusions and food supplements. EFSA Journal, 2017
• Toxic pyrrolizidine alkaloids provide a warning sign to overuse of the ethnomedicine Arnebia benthamii. Journal of Ethnopharmacology, 2018
• In vitro antioxidant and cytotoxic activities of Arnebia benthamii: a critically endangered medicinal plant of Kashmir Valley. Oxidative Medicine and Cellular Longevity, 2014 (shikonin/alkannin as the lipophilic red pigment sold as ratanjot)
• Alkannin: source, water solubility, E103 status
• Neeli A et al. Jatropha curcas poisoning in six children: a case series. Indian Journal of Forensic and Community Medicine, 2025
• Langrand J et al. Poisoning with Jatropha curcas: 24 cases reported to Paris and Marseille poison centres. Bulletin de la Société de Pathologie Exotique, 2015