"Pomegranate seed oil: is it worth the price?"

Pomegranate seed oil: is it worth the price?

The bottle is 30 ml. It is dark glass, it is priced in pounds like a decent bottle of gin, and the reason you have not pressed buy yet is that everything you have read about it says the same three words: rich in punicic acid. Not what punicic acid does. Not how long the oil lasts once you have opened it. Just that it is rich in it, and expensive, and both of those are presented as though they were the same fact.

So here is the honest answer, and it comes in two halves.

Pomegranate seed oil is a genuinely unusual material, and what makes it unusual is a piece of chemistry rather than a benefit. That chemistry is real, measurable and interesting — and it is the same chemistry that makes a second oil, tung, a wood finish rather than a cosmetic ingredient. Once you understand what a conjugated triene does, the price, the short shelf life, the film on the neck of the bottle and the contradictory research all follow from one structural fact.

This is the exotics corner of our guide to plant oils by fatty-acid group. One boundary before we start: nothing here is a claim about what any oil does to skin in a finished product. The skin-directed research on this oil is entirely in cell culture, and we will be specific about that when we get to it.

What you are actually paying for: double bonds with no spacer

Nearly every polyunsaturated fatty acid you handle is methylene-interrupted, which means each pair of double bonds is separated by one CH₂ group — a single carbon acting as a spacer. Linoleic acid (18:2, double bonds at positions 9 and 12) and alpha-linolenic acid (18:3, at 9, 12 and 15) are both built that way. It is the ordinary arrangement, and the whole linoleic group shares it.

A conjugated system removes the spacers. In punicic acid the double bonds sit at 9, 11 and 13 — alternating with single bonds along five carbons, nothing in between. Manzari Tavakoli and colleagues in Scientific Reports (2025) give the full geometry as C18:3 9c, 11t, 13c: cis at 9, trans at 11, cis at 13. Alpha-eleostearic acid, the one in tung oil, is the same skeleton with one more trans bond — 9c, 11t, 13t.

Three consequences come out of that arrangement, and they are the three things you are buying:

How much punicic acid is in it holds up under scrutiny, which is not always the case with exotic oils. Zielińska and colleagues in Molecules (2022) analysed cold-pressed pomegranate seed oil by gas chromatography with flame-ionisation detection coupled with infrared spectroscopy, and report the oil consisting mainly of 9,11,13-octadecatrienoic acid at 73.19 weight per cent of total fatty acids, with nine components identified in all, ranging from 0.58 to 73.19 wt%. Manzari Tavakoli and colleagues, profiling thirty Iranian cultivars, likewise found punicic acid the dominant fatty acid across the set, with stearic acid the least prevalent.

What those two papers support is "punicic acid dominates". They do not establish a general percentage band you can hold a supplier to, and there is no Codex fallback to check against: the Codex Standard for Named Vegetable Oils lists per-oil ranges for arachis, coconut, cottonseed, grapeseed, maize, palm, rapeseed, sesame, soya, sunflower and the rest — and pomegranate is not among them, nor is tung. There is no international reference range for a pomegranate gas chromatography trace. The supplier's own specification is all you have.

At the bench: ask for the punicic acid percentage by gas chromatography for your lot, and write the supplier's own stated range next to it. This is one of the few oils where you cannot fall back on a standard to tell you whether a number is plausible.

The same chemistry, sold in a hardware shop

Tung oil comes from the seed of Vernicia fordii and is the purest expression of everything above. Zhang and colleagues in PLoS ONE (2014) profiled 41 tung tree accessions and found alpha-eleostearic acid at 77 per cent of total fatty acids; Suzuki and colleagues in the Journal of Wood Science (2021) describe the oil as approximately 80 per cent alpha-eleostearic acid.

What that produces is a drying oil, and the mechanism is stated plainly by Wang, Wang, Artz and Padua in the Journal of Agricultural and Food Chemistry (2008): oil drying is a chemical reaction in which polyunsaturated fatty acids undergo autoxidation, and during drying the oils polymerise and form water-resistant films. They studied linseed and tung specifically, tracking the disappearance of the 3010 cm⁻¹ infrared band as an index of oxidation rate, and note that the paint industry accelerates the reaction with metal driers that are often not safe even for food contact.

In Britain you will meet tung oil on a shelf in a hardware shop rather than at a cosmetic supplier, usually as a floor or worktop finish and frequently blended with solvents and driers that have nothing to do with the oil. We are not making a regulatory statement about it here — we looked for one and did not find anything worth quoting — but we do not need to. An oil whose documented behaviour is to polymerise into a film is not a material you want distributed through a cream, and the blend in the tin is not the oil anyway.

In our formulators' experience the useful demonstration for any oil with a conjugated triene costs nothing: leave a thin smear on a glass slide overnight, somewhere you will see it in the morning. Pomegranate seed oil goes tacky and yellows. Tung sets.

At the bench: run that smear test on any new lot of pomegranate seed oil before it goes anywhere near a formula, and again on the same lot three months later. You are watching for the speed of the change, and it is the cheapest stability information you will ever collect.

Why it goes off so fast — and the one thing that helps

The measured picture is more specific, and more useful, than "conjugated oils go off quickly".

Tsuzuki and colleagues in Lipids (2004) auto-oxidised thin layers of linoleic acid, two conjugated linoleic acid isomers, alpha-linolenic acid and alpha-eleostearic acid at 37 °C, following the fatty acid remaining, the oxygen absorbed, the lipid hydroperoxide content and TBARS (a common measure of the later breakdown products of oxidation). Alpha-eleostearic acid oxidised faster than the unconjugated acids and faster than both conjugated linoleic acid isomers. But the hydroperoxide and TBARS contents after its oxidation were low, which the authors read as production of only small amounts of rapid-reacting secondary products.

That pairing is worth holding onto, because it is a trap: with these acids, a reassuring peroxide figure does not necessarily mean slow oxidation.

Two further results in the same paper are the practical ones. Conjugated fatty acids esterified into triacylglycerol — that is, sitting in an intact oil rather than floating as free acids — were more stable than the free acids. And adding alpha-tocopherol raised the stability of the conjugated acids to a level similar to the unconjugated ones.

Commercial reality tracks all of this. Siol and colleagues in Foods (2024) took pomegranate seed oils from four different producers and stored them for one month in the dark at refrigerator temperature — that is, under better conditions than yours. Acid values for most of the oils stayed within the permissible Codex level. Peroxide values exceeded the permissible level in all of them after that month. The authors state it flatly: the oils did not exhibit good oxidative stability, with very short oxidation induction times for all of them. They also found linoleic acid predominating in the sn-2 position of the triacylglycerols — the middle of the three attachment points on the glycerol backbone — which places punicic acid largely on the two outer positions of the molecule.

Light adds a second, separate route of loss that has nothing to do with oxygen. Manzari Tavakoli and colleagues confirmed by nuclear magnetic resonance that punicic acid isomerises to beta-eleostearic acid (C18:3 9t, 11t, 13t) under direct sunlight, cis bonds converting to trans. That is geometric isomerisation, not oxidation; the study demonstrates it under direct sun and does not quantify what ordinary shop lighting does. Their own storage recommendation is room temperature, protected from sunlight, with oxygen exposure limited.

At the bench: buy small and buy often, decant into small opaque containers filled to the neck rather than opening the big bottle repeatedly, keep a working stock separate from the reserve, and ask whether the tocopherol on the specification is native or added — either is fine, but you need to know which, because it changes what happens when you dilute the oil into a formula.

Does it do anything for skin? Two 2025 papers, pointing opposite ways

Here is where honesty costs something. There is skin-directed research on pomegranate seed oil, it is recent, and it is entirely in cell culture. It also does not agree with itself.

Costa, Machado, Pintado and Silva in Inflammopharmacology (2025) tested the oil on HaCaT keratinocytes and HDFa dermal fibroblasts — two standard cultured human skin cell lines. They report no cytotoxic effects up to 500 µg/mL, significantly suppressed pro-inflammatory signalling molecules (IL-6, IL-8 and TNF-α) under an inflammatory challenge, and wound closure in fibroblasts accelerated by approximately 20 per cent over 24 hours.

Poljšak, Kočevar Glavač, Ravnikar and Jeras in Scientific Reports (2025) compared coconut, olive, linden, poppy, pomegranate, marigold and linseed oils on human keratinocytes and fibroblasts in vitro, using proliferation and gap-closure assays. Their finding on this oil points the other way: oils rich in linoleic acid significantly promoted proliferation, unsaponifiable compounds showed no effect, and punicic acid and pomegranate seed oil had a significant negative impact. On gap closure the oils as a group produced no notable effect.

Two papers, same year, both in cells: one reporting accelerated fibroblast gap closure with the whole oil, one reporting a significant negative effect on proliferation from both the whole oil and the isolated acid. They used different cell lines, different assays and different concentrations, so they are not directly comparable — and neither examined intact human skin or a finished product.

The correct reading of a pair like that is not to pick the one you prefer. It is that the question is open.

At the bench: if you use this oil, use it because of what it is and what it costs you to keep, not because of a cell-culture result. And keep the percentage modest: we have found no published study establishing a use level for pomegranate seed oil in a topical product, and we would rather say that than invent a number. Treat it as a specialist addition to an oil phase and run your own stability testing rather than borrowing someone else's percentage.

The iodine value on the certificate is lying to you

This one deserves its own heading, because it is a specific, documented trap rather than a general caution.

Iodine value is the classic laboratory measure of how unsaturated an oil is: how much iodine the double bonds will take up, by the Wijs method. Formulators read it as a rough proxy for how prone an oil is to oxidation, which is a shaky habit at the best of times — we take it apart in oxidative stability of oils — and with a conjugated triene it goes further wrong than usual.

Suzuki and colleagues measured iodine values by the Wijs method for several biodiesels and found them almost identical to the values calculated from the composition — except for tung, which read significantly lower by Wijs than by calculation. Their NMR before and after halogenation showed that the conjugated double-bond signals had not disappeared, which they attribute to iodine's large atomic radius preventing complete reaction with the three conjugated bonds.

So a conjugated-triene oil can present a modest iodine value on a certificate while being one of the least stable materials on your shelf. The number is not wrong about what it measured; it is measuring something that stops corresponding to unsaturation in this particular structure.

At the bench: for pomegranate seed oil, ignore iodine value as a stability signal entirely. Ask instead for peroxide value with the date it was measured, and for the pressing date rather than only a best-before. Given the induction times Siol and colleagues report, a six-month-old lot is a different material from a six-week-old one.

What to ask for on the certificate

At the bench: if the seller cannot give you a peroxide value with a date on it, treat the price as buying you an unknown. For this oil that is not pedantry — it is the single measurement that separates a fresh lot from an expensive one that has already turned.

The British part: you are the one who prints the date

There is a consequence of all this that catches small British brands, and it has nothing to do with chemistry.

If you sell a product in Great Britain, Article 19(1)(c) of the UK Cosmetics Regulation requires the label to carry the date of minimum durability — the date until which the product, stored under appropriate conditions, will continue to fulfil its initial function — preceded by the hourglass symbol or the words "best used before the end of". Where that durability exceeds 30 months, the date is replaced by the period-after-opening symbol, the open jar, followed by the time in months or years.

That number is yours. Nobody hands it to you; it goes in your product information file with the reasoning behind it, and it is part of what the safety assessment covers.

Now put that beside Siol's result. Four commercial pomegranate seed oils, stored dark and cold, all past the Codex peroxide limit inside a month. The open jar on your label says twelve months, or six, because the template said so — and the oil inside it has an induction period measured in weeks. That is not a labelling technicality, it is the thing your customer will notice first, in the form of a smell.

The same applies to what you write on the front of the pack. Cell-culture findings are not evidence for a claim about a finished cosmetic; Article 20(2) puts the duty on a named responsible person to ensure claim wording complies with the common criteria, and what that means for an ingredient-led sentence is set out in what a UK label can say about an oil.

At the bench: decide the durability date after you have watched your own oil, in your own packaging, at the temperatures your product will actually see — not from the supplier's best-before and not from the template. Start the stability sample on the day the lot arrives, not on the day the formula is finished.

So — is it worth it?

Only you can price that, but at least price the right thing.

You are not buying a benefit; you are buying a plant oil dominated by a single conjugated triene, at 73.19 weight per cent in the cold-pressed sample Zielińska and colleagues characterised, with no compositional standard behind it, an induction period short enough that four commercial lots failed a peroxide limit in a month in the fridge, a documented isomerisation in sunlight, and an iodine value that under-reads. What it does to human skin remains, on the published evidence, an open question with two 2025 cell papers pointing in opposite directions.

Tung shares the chemistry and takes it to its conclusion at 77 to 80 per cent alpha-eleostearic acid, which is a virtue in a wood finish and disqualifying in a cosmetic.

Both are caveats before they are ingredients. If you buy pomegranate seed oil anyway — and there is nothing wrong with buying it — buy it small, keep it dark, date it, and let the stability test rather than the marketing decide how much of it goes into the formula.

Sources

• Zielińska A. et al. Cold-pressed pomegranate seed oil: study of punicic acid properties by coupling of GC/FID and FTIR. Molecules 27(18), 5863, 2022. PubMed 36144599
• Manzari Tavakoli M., Nikjoo Tavabi N., Rezadoost H., Fathi F., Nejad Ebrahimi S. Comprehensive fatty acids profiling of thirty Iranian pomegranate seed oil cultivars; oil stability and quality through various storage conditions. Scientific Reports 15(1), 21833, 2025. PubMed 40596481
• Siol M. et al. Chromatographic and thermal characteristics, and hydrolytic and oxidative stability of commercial pomegranate seed oil. Foods 13(9), 1370, 2024. PubMed 38731741
• Tsuzuki T. et al. Oxidation rate of conjugated linoleic acid and conjugated linolenic acid is slowed by triacylglycerol esterification and alpha-tocopherol. Lipids 39(5), 475–480, 2004. PubMed 15506243
• Costa E. M., Machado M., Pintado M., Silva S. Evaluating pomegranate seed oil for potential topical applications: safety, anti-inflammatory activity and wound healing in skin cell models. Inflammopharmacology 33(11), 7045–7055, 2025. PubMed 41065959
• Poljšak N., Kočevar Glavač N., Ravnikar M., Jeras M. Influence of vegetable oils and their constituents on in vitro human keratinocyte and fibroblast proliferation and migration. Scientific Reports 15(1), 26898, 2025. PubMed 40707519
• Zhang L. et al. Fatty acid profile and unigene-derived simple sequence repeat markers in tung tree (Vernicia fordii). PLoS ONE 9(8), e105298, 2014. PubMed 25167054
• Suzuki T., Sumimoto K., Fukada K., Katayama T. Iodine value of tung biodiesel fuel using Wijs method is significantly lower than calculated value. Journal of Wood Science 67(1), 55, 2021. doi:10.1186/s10086-021-01987-3
• Wang Y., Wang Q., Artz W. E., Padua G. W. Fourier transform infrared spectra of drying oils treated by irradiation. Journal of Agricultural and Food Chemistry 56(9), 3043–3048, 2008. PubMed 18410115
• FAO/WHO Codex Alimentarius Commission. Standard for Named Vegetable Oils, CXS 210-1999 (2015 amendment) — product definitions and Appendix quality limits. CXS_210e_2015 (PDF) — consulted 22 September 2026.
• Regulation (EC) No 1223/2009, Article 19 (Labelling), as it has effect in Great Britain. legislation.gov.uk — consulted 22 September 2026.
• Regulation (EC) No 1223/2009, Article 20 (Product claims), as retained GB law. legislation.gov.uk — consulted 22 September 2026.