Herbal infused oils: methods, ratios and the water problem

Three infusion methods with times and temperatures, the drying step that decides whether the oil is safe, what transfers into it, and a herb table with cautions.

An infused oil is the oldest technology in a salve maker's cupboard, and the one where folklore and microbiology part company most sharply. The method is simple. The failure mode is not, and it decides whether a batch of calendula oil is a good ingredient or a genuine hazard, on one choice made before the oil was poured. This page covers the three working methods with times and temperatures, the ratios, what ends up in the oil, and where the real risk sits.

Infusion, macerate, tincture, essential oil

Four words used interchangeably that should not be.

• Infused oil or macerate: plant material steeped in a fatty carrier oil, then strained out. The two words mean the same thing, with "macerate" the more formal.
• Tincture: plant material steeped in ethanol, or ethanol and water. A different set of compounds, and it cannot go into an anhydrous balm because the solvent is not oil soluble.
• Essential oil: the volatile fraction, separated by distillation, with no fatty oil present. Hundreds of times more concentrated than an infusion and dosed accordingly, as covered in essential oils in balms.
• CO2 extract: supercritical carbon dioxide extraction, concentrated and lipophilic, no solvent residue.

The consequence is scale. An infused oil is a bulk ingredient, used at 20 to 90% of a salve, delivering a small quantity of plant constituents in a lot of oil. It is mild by construction.

What actually transfers into the oil

Oil is a non-polar solvent, so it takes up non-polar things: carotenoids and chlorophyll (why calendula oil is orange and plantain oil green), triterpenes and sterols, fat-soluble resins, some flavonoid aglycones, cuticle waxes, and part of the plant's own volatile oils.

It does not take up water-soluble constituents in any meaningful quantity: tannins, mucilage, most glycosides, polysaccharides, vitamin C, most alkaloid salts. If a plant's traditional use rests on a water-soluble constituent, an oil infusion is the wrong preparation, and steeping longer will not change it. That is the most useful thing to understand about infused oils and almost never stated.

The drying step, and why it is the whole safety story

A finished anhydrous balm is microbiologically low risk because it has no aqueous phase. ISO 29621 lists water activity below 0.75 as one of six qualifying factors, though 0.75 is a pragmatic industry line rather than a biological one: the same standard cites 0.6 as the approximate floor for growth, and some moulds work between the two. Oil is not at zero either. Measured on peanut oil, 722 ppm of water gives 0.75 and 923 ppm gives 0.93, and vegetable oils carry 300 to 1000 ppm as supplied. There is still no bulk water for a cell to live in, which is why an oil needs no preservative, as set out in do balms need preservatives, but the margin is not the comfortable 0.2 usually quoted.

Fresh plant material breaks it outright. A leaf is 80 to 90% water. Submerging it in oil does not remove that water, it seals it in behind a barrier that keeps air out. Those are the conditions Clostridium botulinum requires: FDA's minima put proteolytic types at 0.935 and non-proteolytic at 0.97, both comfortably exceeded inside a fresh leaf.

Drying targets: brittle, not leathery. Leaves crumble between the fingers, stems snap. A dehydrator at 35 to 40 C, or a warm airy room out of sunlight for a week to ten days, gets most material below the 10% residual moisture where it keeps. Wilting for a day is not drying. Mould in finished product is covered in mould in balm, workspace controls in workshop hygiene.

The garlic-in-oil precedent, and what it actually proves

This is not a theoretical hazard, and the case history is specific enough to be worth knowing.

• Vancouver, 1985. A restaurant outbreak traced to chopped garlic in soybean oil produced 34 cases of type B botulism in two clusters, across Canada, the United States and the Netherlands. Symptom onset ran to ten days, delaying diagnosis.
• Kingston, New York, February 1989. Three cases from garlic bread made with a garlic-in-oil product. The leftover product had a pH of 5.7 and high concentrations of organism and toxin. FDA responded by requiring microbial inhibitors or acidifying agents in these products.
• The controlled study (Solomon and Kautter, Journal of Food Protection, volume 51, 1988, page 862). In bottled chopped garlic in soybean oil, one spore per gram produced toxin in 15 days at 35 C, and five spores per gram produced toxin at room temperature over 75 days. Five type A strains were isolated from 115 bulbs of fresh garlic. The finding that matters most to a maker: even when highly toxic, the garlic looked and smelled acceptable.

Garlic-in-oil is now named in the US Food Code. Read the mechanism before generalising it to every infusion. C. botulinum needs three things at once: anaerobic conditions, a water activity of at least 0.935, and a pH above 4.6. The oil supplied only the first. The garlic supplied the water, the near-neutral pH and the spores. An infusion made with fully dried botanicals sits far below the water threshold and cannot support growth or toxin production however anaerobic it is. Drying does not remove spores, it removes the conditions they need.

Ratios and a worked example

Work by weight. A cup of calendula petals and a cup of chopped root differ fourfold or more in mass, and volume ratios are how salve recipes end up ambiguous.

Worked example. For 200 g of finished calendula oil, start with 40 g of dried petals and 250 g of olive or high-oleic sunflower oil (1:6.25, allowing for absorption). After pressing you should recover 195 to 210 g. Add 0.4 g of mixed tocopherols, 0.2%. Fluffy petals float, so use a narrow jar or a glass weight.

For a stronger oil, do a double infusion: strain, then pour the recovered oil over a second charge of dried herb. That beats extending the time on one charge, because the oil approaches saturation for the more soluble constituents quickly.

Method 1: cold or solar infusion

1. Fill and cover. Pack dried herb loosely into a clean dry jar and cover with oil by at least 2 cm. Anything above the oil line will mould.
2. Stand warm and dark. Two to six weeks at room temperature. The traditional windowsill gives warmth but also light, which oxidises the oil and bleaches carotenoids. A warm cupboard is better.
3. Shake daily. Not ceremonial: it moves saturated oil away from the plant surface and restarts the concentration gradient.
4. Check the seal. Condensation under the lid means moisture is present. Wipe it and watch the batch.
5. Strain at four weeks for most leaf and flower material. Longer adds oxidation more than extraction.

Solar infusion suits heat-sensitive material and makers with time. Its weaknesses are light, weather-dependent temperature, and a long window in which a not-quite-dry herb spoils the batch.

Method 2: low heat infusion

Put the jar or a stainless bowl in a water bath and hold the oil at 50 to 70 C for four to eight hours, stirring occasionally. Do not let it simmer: bubbles mean water is boiling off plant material that was not dry enough, and above about 85 C you degrade the carrier and any tocopherol in it. Put the thermometer in the oil, not the water.

This is the repeatable method. It finishes in a day and the temperature is a number you can write down. The cost is some oxidation of the carrier, so a stable oil matters: olive, high-oleic sunflower, jojoba and fractionated coconut cope, grapeseed, hemp and rosehip do not. The ranking is in carrier oils compared, and jojoba's durability in jojoba oil.

Method 3: controlled slow cooker or oven

A slow cooker on "keep warm" typically sits at 70 to 80 C, the top of the useful range, so stand the jars in water rather than putting oil on the ceramic. Six to eight hours is plenty. An oven at its lowest setting is usually 70 to 90 C and often overshoots: verify with a thermometer and prop the door if needed. The long warm method suits tough material, the roots, barks, resins and seeds a cold infusion barely touches. Chop or grind first, since extraction is limited by surface area more than time.

Straining, and getting a clear oil

Strain through two layers of muslin or fine nylon mesh, then press to recover retained oil. The pressed fraction is cloudier and richer in fines, so keep it separate for a clean product.

Let the strained oil stand undisturbed for 24 to 48 hours in a tall jar. Fines and trace water drop to the bottom. Decant off the top and discard the last centimetre. This step separates an oil that keeps a year from one that smells off in three months, because the fines carry moisture and enzyme activity. Droplets at the bottom, rather than dusty sediment, are water: do not use that batch in a shelf-stable balm.

Shelf life of an infused oil

An infusion inherits the oxidative stability of its carrier and usually loses a little. Chlorophyll is a photosensitiser, plant fines carry trace metals that catalyse oxidation, and the handling adds oxygen. Store in full, tightly capped, dark bottles in a cool dry cupboard, and date them. The mechanism, and the difference between an antioxidant and a preservative, are in rancidity and oxidation and vitamin E and antioxidants. Tocopherol at straining slows what has not yet happened and does nothing for a stale oil.

Working rule: remaining life is the carrier's stated life, minus the time it already spent in your cupboard, minus a further month for the process. Confirm with a retained sample rather than trusting the arithmetic.

Herbs, traditional uses and cautions

Where the infused oil goes next

An infused oil is an ingredient, not a product. It replaces some or all of the plain carrier in a salve's oil phase, typically 50 to 90% alongside 8 to 15% wax, the structure set out in herbal salve and worked through in gardeners hand salve. Keep the antioxidant, and label with the botanical name and part used rather than a folk name. Where the material carries a caution above, that caution belongs on the label, not only in your notes.

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