Propolis: the hive material behind most beeswax allergy reports

Propolis extract at 0.1 to 2 percent, flavonoid content, alcohol removal, and the contact allergy data that makes it the first suspect in a beeswax reaction.

Propolis is the resin bees gather from tree buds and bark, mix with wax and enzymes, and use to seal, varnish and disinfect the hive. It is also the most likely reason a beeswax balm made somebody itch. This page covers what is in it, which extract form actually works in an anhydrous base, how much to use, what the patch test data really says, and what you may and may not claim for it in the UK and EU.

• INCI: Propolis Extract
• 0.1 to 2% resin solids
• 40-55% resin, 25-35% wax
• Frequent contact allergen
• Not vegan

What propolis is made of

Raw propolis is not one substance. It is plant resin, bee wax, volatile oils and hive debris in proportions that shift with the forage available, the season and how the beekeeper collected it. Scraped propolis carries more wood and paint fragments than propolis harvested from a plastic collection mat, and the difference shows up in the extract yield.

Within the resin fraction, the compounds that get quoted are flavonoid aglycones (pinocembrin, galangin, chrysin, pinobanksin, apigenin), free phenolic acids (caffeic, ferulic, p-coumaric, cinnamic) and, importantly for the allergy section below, the caffeic acid esters. Caffeic acid phenethyl ester, usually written CAPE, and 3-methyl-2-butenyl caffeate are the two names worth remembering, because they are the identified sensitisers rather than incidental markers. Total flavonoid content in raw temperate propolis commonly runs 2 to 6 percent by mass; standardised dry extracts are sold at 8 to 15 percent flavonoids, which is why a percentage on your formula sheet means nothing until you say which input it refers to.

Poplar, green and red are three materials with one name

Propolis chemistry follows the trees, not the bees. A European propolis and a Brazilian green propolis share a name, a smell family and almost nothing else on a chromatogram. If a supplier will not tell you the botanical and geographic origin, you do not know what you have bought.

Two other types turn up in trade: Mediterranean propolis from Cupressaceae, rich in diterpenes rather than flavonoids, and Pacific propolis from Macaranga species, characterised by prenylated flavanones. Both behave differently again. Practically, the chemotype decides the colour you will get in a pale balm and, as the allergy section sets out, may change the sensitisation risk. It does not much change the usage level.

Which extract to buy

The form matters more than the percentage, because most of the forms sold are simply not compatible with an anhydrous product.

The glycol and glycerin forms are the trap. They are the standard propolis inputs for serums and toners, they are widely stocked, and they will not go into a balm. Adding a glycerin extract to a molten butter and wax base gives you visible beads and, once cooled, a mottled surface with sticky spots. The physical chemistry behind that is set out in solubility in anhydrous formulas, and the wider problem of polar ingredients in a fat base is the same one described under glycerin and humectants. Any water riding in with them is also a contamination route worth taking seriously, for the reasons in moisture control.

Getting the alcohol out of a tincture

A tincture is 70 to 96 percent ethanol by volume. Poured into a hot oil phase it flashes off unevenly, spits, carries the aroma compounds away with it and leaves you with no idea how much resin actually stayed behind. Reduce it separately and weigh the endpoint.

1. Work out the target solids. A tincture labelled 20 percent leaves 0.2 g of resin per gram. For 0.5 percent resin in a 100 g batch you need 0.5 g of solids, so weigh 2.5 g of tincture.
2. Weigh the empty dish, then the tincture into it. Record both. Use a shallow glass or stainless dish, not plastic, and give it as much surface area as you can.
3. Reduce over a water bath at 50 to 60 C. Away from any flame or hot plate element. Ethanol vapour is flammable and this is the step where people get careless. Ventilate the room.
4. Stop when the mass hits the calculated endpoint. Dish plus 0.5 g in the example. It will be a dark, tacky, brittle residue that will not scrape out cleanly.
5. Dissolve the residue in part of the oil phase. Add 5 to 10 g of the batch oil to the same dish and warm to 60 to 70 C with stirring until the residue is fully in solution, then add that to the main batch. Dissolving in the dish is what recovers the material you would otherwise leave behind.

Add the dissolved resin at the same point you would add any heat sensitive material, at the low end of your working window rather than into the boiling wax. The general rules for that window are in pour temperatures. Propolis resin itself softens somewhere around 60 to 70 C and is brittle when cold, so it will not seize in a normal balm; the concern is aroma loss and darkening, not solubility.

How much to use, and what you get at each level

All figures below are resin solids as a percentage of the finished balm, not tincture weight. Propolis is a colouring, scenting, tackifying ingredient at every level that does anything, so the ceiling is usually set by the sensory result long before it is set by safety.

Zero point three to 0.5 percent is where most balms should land. Below 0.1 percent you are paying for an ingredient list entry. Above 2 percent you are making a resin salve, not a balm, and the closest useful comparison is the handling of tree resin described in pine resin salve, which is tacky by design. If the smell is fighting your fragrance rather than adding to it, the diagnosis is in balm smells too strong, and the answer is usually less propolis rather than more fragrance.

The allergy data, without the exaggeration

Propolis is a well characterised contact allergen, and the sensitisers are the caffeic acid esters in the resin. It is patch tested at 10 percent in petrolatum in the standard preparations, and it cross reacts with balsam of Peru (Myroxylon pereirae), colophony, benzoin and poplar bud, so a person positive to any of those has a raised chance of reacting to a propolis balm.

Three things follow from that table. First, propolis sits among the more frequent cosmetic allergens: a 2 to 4 percent positivity rate in a dermatitis clinic puts it in the same territory as fragrance mix components rather than with the inert structural materials. Second, the test material matters enormously. The Brazilian preparation returned 14.7 percent in one season against 2.2 percent for a Chinese preparation two years later in the same testing programme, a spread that says as much about the preparation as about propolis itself. Third, the beekeeper story is more moderate than it is usually told. In a survey of 1,051 German beekeepers, 3.6 percent reported allergic reactions to propolis, and only about a quarter of those had recognised the allergy before being asked. Occupational exposure raises risk and, more interestingly, it hides it.

For a maker, the practical consequences are ordinary. Propolis is not an ingredient to hide in a formula sold as gentle. It should be declared clearly on the pack, for which the conventions are in reading a balm label. If a customer reports burning or itching, propolis is the first thing to look at, ahead of essential oils, and the elimination approach is set out in balm stings or burns. Anyone with a known reaction to bee products should be pointed at how to patch test a balm before they use a whole tin, and the same logic that applies to declared fragrance allergens applies here even though propolis is not on the EU allergen list.

Why propolis travels with beeswax, and what filtering will not do

Bees do not keep propolis and wax in separate compartments. Propolis is smeared over comb, frames and cappings, so rendered wax carries it in, and the caffeate esters are lipophilic, which means they dissolve into molten wax rather than sitting on it as particles. That distinction is the whole point. Straining molten wax through muslin or a fine sieve removes pollen, wood, bee parts and undissolved propolis lumps. It does not remove anything already in solution.

Bleaching goes further. Filtration through activated carbon, the usual route to a white grade, adsorbs coloured and aromatic material and will take out a good deal of the dissolved propolis with it. That is why white beeswax smells of almost nothing and why reactions are reported more often with raw and yellow grades. But no pharmacopoeia monograph sets a propolis limit for either grade, no supplier certificate reports one, and there is no test you can run to declare a batch propolis free. The honest position is on the page for beeswax: bleached wax carries less, neither grade is guaranteed clear, and a person with a confirmed propolis allergy should treat all beeswax as suspect rather than trusting the grade. The full comparison of the two allergens is in beeswax and propolis allergy.

Standardisation and what a certificate should show

Because the plant source varies, propolis is one of the few balm ingredients where a supplier certificate carries real information rather than confirming the obvious. Ask for it before you order, not after.

Two colorimetric numbers alone are weak evidence. Total polyphenol and total flavonoid assays both respond to compounds that are not the ones you are buying, and neither distinguishes a poplar extract from a green one. It is the fingerprint that tells you what the material is. The general approach to supplier documentation, including what to do when a supplier will not produce it, is in sourcing ingredients.

What you may claim in the UK and EU

Propolis has a genuine antimicrobial literature behind it, which is exactly why the claim rules bite. Under the EU Cosmetic Products Regulation 1223/2009, and the retained GB version of it, a cosmetic is defined by function: cleaning, perfuming, changing appearance, correcting body odours, protecting and keeping in good condition. Say that a balm kills bacteria, is antiseptic, heals a wound or treats a cold sore and you have described a medicinal product, which needs a marketing authorisation, and the MHRA applies a "presentation and function" test that looks at your whole marketing, not just the pack.

Claims that stay inside the cosmetic definition still have to satisfy the common criteria in Regulation (EU) 655/2013, retained in GB: legal compliance, truthfulness, evidential support, honesty, fairness and informed decision making. "Contains propolis, traditionally used in hive materials" is a statement about the ingredient. "Propolis balm for cuts" is not a cosmetic claim at all. The line, claim by claim, is worked through in cosmetic vs drug claims, and the notification and responsible person duties that sit behind it are in selling balms in the UK and EU.

One specific trap. Propolis is not listed in Annex V, the list of preservatives permitted in EU and GB cosmetics, so you cannot use it as a preservative or say that it preserves the product. A claim that the product resists microbial growth because of propolis also drifts towards biocidal product territory. In an anhydrous balm the question is moot for a different reason, set out in preservatives for anhydrous products: there is no water for organisms to grow in. Your safety assessor will want the propolis inclusion level, the extract documentation and a sensitisation rationale in the report, which is covered in safety assessment and the CPSR.

When propolis is the wrong ingredient

Propolis is a poor choice in four situations, and none of them is about efficacy. It does not belong in a product marketed as suitable for sensitive or reactive skin, because a 2 to 4 percent clinic positivity rate is not compatible with that positioning. It does not belong in a pale or delicately fragranced product, because at any level where it is doing something it is also colouring and scenting the batch. It rules the product out of a vegan range, along with beeswax, lanolin and honey, for the reasons in vegan and cruelty free labels. And it does not belong in anything you intend to describe as antibacterial, because the moment you say that, the product is no longer a cosmetic.

The decision rule is short. If you want a hive character in a salve, want the colour and smell to be part of the product, and are willing to declare a known allergen prominently, buy a standardised dry extract with a named botanical origin and run it at 0.3 to 0.5 percent resin solids. If you want the antimicrobial story, buy nothing, because you cannot sell it. And if you want an occlusive that clings and repairs chapped skin without an allergy conversation, lanolin does more work, though it brings its own sensitisation debate. Propolis is a character ingredient. Formulate it as one.

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