How skin works
Sunscreen expires. What that means is unknown SPF, not zero.
Yes: in the US a date on the box, in the EU an open-jar symbol with the months it is good for once opened. Past that date nobody can tell you the number on the tube, because the things that fail do not fail together: the emulsion first, then two of the filters, and the minerals never. Here is the order, from our own data.
The short answer
Expired sunscreen is a sunscreen whose SPF nobody has measured any more. Not zero: unknown.
- Does sunscreen expire?
- Yes. In the US it is an over-the-counter drug and carries an expiration date set by stability testing, or none if the maker showed three years of stability. In the EU it is a cosmetic: a "best used before" date if it lasts 30 months or less, otherwise the open-jar symbol with the months after opening.
- Does expired sunscreen still work?
- Nobody can say what its SPF is. The number on the box was measured on a fresh, even film; what an aged tube forms is not that film, and the two filters that break down in light may have broken down. Unknown is the honest answer, and unknown is not a number you should stand behind in the sun.
- What breaks down first?
- The emulsion (it separates, and the film it spreads is no longer the one the SPF was measured on), then the photolabile filters (avobenzone and octinoxate, unless a stabiliser stands beside them on the label). Zinc oxide and titanium dioxide do not photodegrade.
- Can I use it if unopened?
- In the US the date is the date, opened or not. In the EU the open-jar months start when you open it; before that the maker’s own durability applies, and the label does not print it.
- How can you tell?
- Separation, a smell, a texture that changed, a colour that changed - and none of them is a measurement; the date is.
What fails first: the ladder
A sunscreen is four things that age at four speeds: a base that holds oil and water together, the molecules that absorb the light, the preservative that keeps the water phase clean, and, in a mineral one, two pigments that do not care about light at all. The number on the box belongs to all four at once, which is why nobody can tell you what is left when one of them goes.
The emulsion
The SPF on the box was measured on a smooth film of set thickness. The film sets the number; a formula that has separated no longer forms the film it was measured on.
The SPF on the box was measured on a film of a set thickness; the thickness the film actually forms sets the protection.
What the source says
- This result establishes the relationship between sun protection and the film thickness distribution actually formed by the applied sunscreen and demonstrates that variation in SPF between formulations is primarily due to their film forming properties.
A thinner, less even film measured a lower SPF with the same filters.
What the source says
- The vehicle had a significant impact, low vehicle viscosity resulting in a smaller [mean thickness], larger [mean] to median ratio, and lower SPF in vitro than high viscosity; continuous oil phase produced the largest [mean thickness] and SPF values.
Not known: Nobody has measured what a shelf-aged, separated sunscreen does to the film: the thickness-to-SPF link is measured on fresh vehicles, and the step from "separated" to "thinner film" is the reader’s own eye.
The photolabile filters
Two organic filters break down in light on their own: avobenzone and octinoxate. Whether they hold depends on what else is on the label - bemotrizinol or octocrylene beside avobenzone - so the question to ask of a bottle is "is there bemotrizinol or octocrylene on the label".
Avobenzone, the UVA filter in most US and many EU sunscreens, is not photostable on its own.
What the source says
- Avobenzone is an ultraviolet (UV) filter that is often included in sunscreen formulations despite its lack of photostability.
Beside bemotrizinol it holds: the SPF and the UVA ratio were sustained to 30 MED.
What the source says
- With spectrophotometrical methods to assess the sun protection factor (SPF) and UVA ratio and chromatographical methods to determine the amount of UV filters recovered after irradiation we showed that Tinosorb S prevented the photodegradation of AVB in a concentration-dependent way, leading to a sustained SPF and UVA ratio even after irradiation with doses of up to 30 MED.
Beside octocrylene it is improved.
What the source says
- In addition, OC improved the photostability of OMC, AVB and BP-3.
Octinoxate is the second labile filter: half of it went in an emulsion under UV, a third when encapsulated.
What the source says
- The photodegradation of the sunscreen agent in emulsion vehicles was reduced by encapsulation into the PLGA nanoparticles (the extent of degradation was 35.3% for the sunscreen-loaded nanoparticles compared to 52.3% for free trans-EHMC) whereas the EC nanoparticle system had no significant effect.
And avobenzone makes octinoxate worse when the two share a bottle.
What the source says
- Since AVB was shown to destabilize ethylhexyl methoxycinnamate (EHM) we tested the effect of Tinosorb S in sunscreens containing this UV filter combination.
The preservative
"Expired sunscreen rash" is a real search, and nobody has measured a rash from an expired sunscreen: what is measured is contamination of cosmetics while in use, and the closure that lets it in. A preservative system is proven at release, in a challenge test, not at the date; whether a sunscreen past its date grows anything, no study has measured.
What is measured is contamination of cosmetics while in use: in one survey of 91 products, about one in ten rinse-off products carried bacteria, and the authors call contaminated products relatively uncommon and some of them a potential hazard.
What the source says
- Contaminated cosmetic products are relatively uncommon, but some products, unable to suppress the growth of several micro-organisms, represent a potential health hazard.
And the closure decides how much gets in: an unpreserved lotion used for two weeks was contaminated in 71% of screw-cap bottles and 10% of pump-tops.
What the source says
- The greatest amount of protection (i.e., lowest contamination incidence) was provided by the flip-cap closure for the shampoo container (0%) and the pump-top closure for the skin lotion container (10%).
Not known: PubMed, 2026-09-21: (sunscreen OR cosmetic) AND (expired OR "expiry date") AND (preservative OR microbial OR contamination OR "challenge test") returned no measurement of an expired sun product; (sunscreen OR "sun protection product") AND (expired OR "expiration date") AND (rash OR dermatitis OR "skin reaction" OR irritation) returned 16 records, none on an expired product. The rung is a question, not a finding.
The minerals
Zinc oxide and titanium dioxide do not photodegrade. What still fails in a mineral sunscreen is the emulsion above.
Zinc Oxide: measured stable.
What the source says of Zinc Oxide
- It was concluded by the applicant that all three model O/W formulations with uncoated (Z-COTE) and coated (Z-COTE HP1, Zano 10 Plus) ZnO were photostable over the whole UV range between 290 and 400 nm.
SCCS/1489/12 (EU Scientific Committee opinion, read in full)
Titanium Dioxide: stated photostable, not measured.
What the source says of Titanium Dioxide
- All materials are stated to be photostable.
SCCS/1516/13 (EU Scientific Committee opinion, read in full)
The 16 filters, one line each
The same 16 glossary rows the sunscreen pages use, with the light answer each carries: 1 break down in light, 1 is labile alone and holds beside the right partner, 5 were measured stable, 3 are stated stable by an opinion or a review and not measured, and 6 have no measurement in the open literature. Where the answer is a statement, the sentence is printed so you can see it is one. The table does not borrow the makers’ sheets.
| Filter | In light | What the source says |
|---|---|---|
| Avobenzone | labile alone; holds beside bemotrizinol | With spectrophotometrical methods to assess the sun protection factor (SPF) and UVA ratio and chromatographical methods to determine the amount of UV filters recovered after irradiation we showed that Tinosorb S prevented the photodegradation of AVB in a concentration-dependent way, leading to a sustained SPF and UVA ratio even after irradiation with doses of up to 30 MED.PMID 11594052, abstract read, not the full text |
| Octinoxate | breaks down in light | The photodegradation of the sunscreen agent in emulsion vehicles was reduced by encapsulation into the PLGA nanoparticles (the extent of degradation was 35.3% for the sunscreen-loaded nanoparticles compared to 52.3% for free trans-EHMC) whereas the EC nanoparticle system had no significant effect.PMID 12270607, abstract read, not the full text |
| Octocrylene | measured stable | The formulations that were investigated contained or not (vehicle: formulation 1) a combination of 0.6 % (w/w) vitamin A palmitate (1,700,000 UI/g), 2 % (w/w) vitamin E acetate and 2% (w/w) ascorbyl tetraisopalmitate (formulation 2) supplemented with a photounstable UV filter combination octyl methoxycinnamate (OMC), avobenzone (AVB) and 4-methylbenzilidene camphor (MBC) (formulation 3) or with a photostable UV filter combination OMC, benzophenone-3 (BP-3) and octocrylene (OC) (formulation 4).PMID 17614223, abstract read, not the full text |
| Homosalate | measured stable | The decrease in homosalate content ranged between 0 - 2.7% and thus, homosalate was shown to be stable under these conditions.SCCS/1622/20 (EU Scientific Committee opinion, read in full) |
| Tinosorb S | measured stable | Considering HPLC, spectrophotometric and GC-MS results, it is suggested that formulations containing BMZ were considered the most photostable.PMID 26159738, abstract read, not the full text |
| Tinosorb M | measured stable | Excellent UVA1 photostability showed sunscreen S8 (EHMC, EHT and methylene bis-benzotriazolyl tetramethylbutylphenol); Auc-UVA1 Index was of 1.00.PMID 21277959, abstract read, not the full text |
| Uvinul A Plus | stated photostable, not measured | DHHB is highly valued for its exceptional UV-absorbing properties (especially UVA radiation), excellent photostability, and safety profile.SCCS/1678/25 (EU Scientific Committee opinion, read in full) |
| Mexoryl SX | stated photostable, not measured | Mexoryl, a novel ultraviolet A filter, provides efficient ultraviolet A coverage, better photostability, and enhanced water resistance.PMID 17805138, abstract read, not the full text |
| Ethylhexyl Triazone | not measured in the open literature | Nothing measured alone. The one paper on it is photophysics: We propose that this involves ultrafast internal conversion of the initially photoexcited n¹ππ* state that couples to the ground state via a 1¹ππ*/S₀ conical intersection, enabling multiple absorption and recovery cycles, as one would anticipate from a highly efficient filter.PMID 28437110, abstract read, not the full text |
| Octisalate | not measured in the open literature | Nothing measured in the open literature (PubMed, 21 September 2026); the sheets say photostable, and we do not borrow them. |
| Mexoryl XL | not measured in the open literature | Nothing measured in the open literature (PubMed, 21 September 2026); the sheets say photostable, and we do not borrow them. |
| Iscotrizinol | not measured in the open literature | Nothing measured in the open literature (PubMed, 21 September 2026); the sheets say photostable, and we do not borrow them. |
| Zinc Oxide | measured stable | It was concluded by the applicant that all three model O/W formulations with uncoated (Z-COTE) and coated (Z-COTE HP1, Zano 10 Plus) ZnO were photostable over the whole UV range between 290 and 400 nm.SCCS/1489/12 (EU Scientific Committee opinion, read in full) |
| Titanium Dioxide | stated photostable, not measured | All materials are stated to be photostable.SCCS/1516/13 (EU Scientific Committee opinion, read in full) |
| Ensulizole | not measured in the open literature | Nothing measured in the open literature (PubMed, 21 September 2026); the sheets say photostable, and we do not borrow them. |
| Polysilicone-15 | not measured in the open literature | Nothing measured in the open literature (PubMed, 21 September 2026); the sheets say photostable, and we do not borrow them. |
The two labels: a US box and an EU tube
The reason two people argue about whether sunscreen expires is that they are holding two different objects. In the US a sunscreen is a drug and its box carries a drug’s date. In the EU it is a cosmetic and its tube carries a cosmetic’s symbol. Both are readable once you know which you have.
The US box
- an expiration date, or none when the maker has shown three years of stability
- no date on the box: expired three years after purchase, by the FDA’s own line
- "if unopened" changes nothing: the date is the date
In the US a sunscreen is an over-the-counter drug, and a drug product carries an expiration date set by stability testing.
What the instrument says
- To assure that a drug product meets applicable standards of identity, strength, quality, and purity at the time of use, it shall bear an expiration date determined by appropriate stability testing described in § 211.166.
The exception is a product shown stable for at least three years: it may carry no date, and the FDA tells you to treat it as expired three years after purchase.
What the instrument says
- Pending consideration of a proposed exemption, published in the Federal Register of September 29, 1978, the requirements in this section shall not be enforced for human OTC drug products if their labeling does not bear dosage limitations and they are stable for at least 3 years as supported by appropriate stability data.
The FDA’s own consumer line: a sunscreen without a date should be treated as expired three years after purchase; keep the container out of direct sun.
What the instrument says
- A sunscreen product that doesn’t have an expiration date should be considered expired three years after purchase.
FDA, "Sunscreen: How to Help Protect Your Skin from the Sun": read on fda.gov, 21 September 2026
Read in the instrument’s own text on 21 September 2026: 21 CFR 211.137; the FDA’s consumer page.
The EU tube
- lasts 30 months or less: a "best used before the end of" date, or the hourglass symbol beside it
- lasts longer than 30 months: the open-jar symbol with the months after opening - "12M"
- "if unopened": the months start when you open it; before that, the maker’s own durability, which the label does not print
In the EU a sunscreen is a cosmetic, and a cosmetic that lasts 30 months or less carries a date of minimum durability.
What the instrument says
- Indication of the date of minimum durability shall not be mandatory for cosmetic products with a minimum durability of more than 30 months.
A cosmetic that lasts longer than 30 months carries instead the open-jar symbol with the months it is safe for after opening.
What the instrument says
- For such products, there shall be an indication of the period of time after opening for which the product is safe and can be used without any harm to the consumer.
Read in the instrument’s own text on 21 September 2026: Regulation (EC) 1223/2009, Article 19(1)(c); the symbols are its Annex VII.
Heat, the car, the beach bag
The industry ages a formula on purpose: a sample sits at 40 °C and 75% humidity for six months, the accelerated condition of the ICH stability guideline, and a formula that holds there is read as holding on the shelf. That is what the number on the box rests on. It is a test of the fresh, sealed product in a cabinet, not of a tube left on a dashboard, and it implies exactly this much: a well-made sunscreen tolerates months of warmth in its closed pack. It does not measure your car.
The industry ages a formula faster at 40 °C and 75% humidity, the accelerated condition of the ICH stability guideline; a formula that holds six months there is read as holding on the shelf.
What the source says
- Results obtained from the stability study indicated that the formulation possesses 50+ in vitro SPF value and remained stable for 6 months and 12 months under storage at 40 ± 2 °C and 75 ± 5% relative humidity; and -20 °C ± 5 °C respectively.
In the one six-month study that heated a finished sunscreen to 50 °C, the UV filters and the SPF held while the plant antioxidant in it did not.
What the source says
- The contents of catechin (С) and epicatechin (EC) in sunscreen with GSE were reduced to 37% and 50% at 50°C, respectively, whereas UV filters appeared to be stable.
What the FDA does say about heat is a storage line, not a number.
What the source says
- FDA recommends you not expose sunscreen containers to direct sun.
FDA, the same consumer page, on storage: read on fda.gov, 21 September 2026
Folklore, not a finding: “a summer in the car halves your SPF”. Nobody has measured a sunscreen through a summer in a car. What is measured is above: six months at 40 °C in a cabinet, and in one study 50 °C, with the filters holding. The honest line is the FDA’s: keep the container out of direct sun, and if it has separated, it is not the product that was tested.
How to tell, and why none of it is a measurement
Four things people read off an old tube. Each says something true about the base, and none of them says anything about the SPF, because the SPF cannot be seen. The only measurement is the date.
It has separated
the emulsion has broken; the film it forms is no longer the one the SPF was measured on
It smells different
an oil or a fragrance component has oxidised; it says nothing about the filters
The texture has changed
the same as separation, earlier
The colour has changed
something has oxidised; which thing, the eye cannot tell
Our 8 sunscreens, and where each one is weak
The 8 sunscreens we decode, with the regime their pack falls under, the period after opening as read off the maker’s page on 21 September 2026 (never a reseller’s, never guessed: “not printed” means the maker prints none), the filters each carries, and therefore which rung of the ladder is its weak point.
| Sunscreen | Regime | After opening | Filters | Weak rung |
|---|---|---|---|---|
| Neutrogena Hydro Boost Water Gel SPF 50a tube | EUthe EU pack (Hydrating Fluid SPF 50): date or open-jar symbol | not printed | the photolabile filter: avobenzone, with octocrylene beside it | |
| Beauty of Joseon Relief Sun: Rice + Probiotics SPF50+a tube | Koreaa Korean pack: an expiry date is printed; the maker’s FAQ gives 12 months after opening | 12M | the emulsion: none of its filters is a labile one | |
| Round Lab Birch Juice Moisturizing Sun Creama tube | Koreaa Korean pack: an expiry date is printed; the maker prints no months after opening | not printed | the emulsion: none of its filters is a labile one | |
| Supergoop! Unseen Sunscreen SPF 50a tube | USan OTC drug: the expiration date sits on the crimped edge (supergoop.com Q&A) | not printed | the photolabile filter: avobenzone, with octocrylene beside it | |
| La Roche-Posay Anthelios UVMune 400 Invisible Fluid SPF50+a tube | EUa cosmetic under 1223/2009: date or open-jar symbol on the pack | not printed | the photolabile filter: avobenzone, with Tinosorb S beside it | |
| CeraVe Hydrating Mineral Sunscreen Face Lotion SPF 30a tube | USan OTC drug: the expiration date is on the label (cerave.com) | not printed | the emulsion: its minerals do not photodegrade, so the film is what ages | |
| ISDIN Fotoprotector Fusion Water SPF 50a tube | EUa cosmetic under 1223/2009: date or open-jar symbol on the pack | not printed | the photolabile filter: avobenzone, with octocrylene and Tinosorb S beside it | |
| Bioré UV Aqua Rich Watery Essence SPF50+a tube | Japana Japanese pack: no period after opening is required; Kao says three years unopened and "use as soon as possible" after | not printed | the photolabile filter: octinoxate, with Tinosorb S beside it |
Once a bottle is open, the months are yours to count. My Shelf remembers the day you opened it and the number on its open-jar symbol, and says when it runs out.
Sources
Every claim above rests on one sentence quoted from its source, inside its fold. The identifiers below were resolved on 21 September 2026: the papers through PubMed, the two instruments in their own text. Where an abstract stopped short of the claim, so did we.
- 21 CFR 211.137(a), expiration dating. 21 CFR 211.137(a) (US federal regulation), read in full
- 21 CFR 211.137(h), the three-year exemption. 21 CFR 211.137(h) (US federal regulation), read in full
- Regulation (EC) 1223/2009, Article 19(1)(c). EU Regulation 1223/2009 Article 19(1)(c), read in full
- Sohn 2016, film thickness distribution and the SPF in vitro. PMID 27045276, abstract read, not the full text
- Sohn 2014, film thickness of different vehicles. PMID 25408957, abstract read, not the full text
- Holt 2021, avobenzone in formulation models. PMID 34694312, abstract read, not the full text
- Chatelain & Gabard 2001, photostabilisation by bemotrizinol. PMID 11594052, abstract read, not the full text
- Gaspar & Maia Campos 2006, four filter combinations. PMID 16289662, abstract read, not the full text
- Perugini 2002, octinoxate in emulsion vehicles. PMID 12270607, abstract read, not the full text
- Campana 2006, cosmetics during their use by consumers. PMID 16910936, abstract read, not the full text
- Brannan & Dille 1990, closure type and in-use contamination. PMID 2339896, abstract read, not the full text
- Sharma Bora 2019, a sunscreen cream through ICH accelerated storage. PMID 31116617, abstract read, not the full text
- Yarovaya 2022, a sunscreen at 30 and 50 °C for six months. PMID 35971885, abstract read, not the full text
- FDA, “Sunscreen: How to Help Protect Your Skin from the Sun”. fda.gov, read 21 September 2026
- Avobenzone: the glossary row’s light source. PMID 11594052, abstract read, not the full text
- Octinoxate: the glossary row’s light source. PMID 12270607, abstract read, not the full text
- Octocrylene: the glossary row’s light source. PMID 17614223, abstract read, not the full text
- Tinosorb S: the glossary row’s light source. PMID 26159738, abstract read, not the full text
- Tinosorb M: the glossary row’s light source. PMID 21277959, abstract read, not the full text
- Mexoryl SX: the glossary row’s light source. PMID 17805138, abstract read, not the full text
- Ethylhexyl Triazone: the glossary row’s light source. PMID 28437110, abstract read, not the full text
- Homosalate: the glossary row’s light source. SCCS/1622/20 (EU Scientific Committee opinion, read in full)
- Uvinul A Plus: the glossary row’s light source. SCCS/1678/25 (EU Scientific Committee opinion, read in full)
- Zinc Oxide: the glossary row’s light source. SCCS/1489/12 (EU Scientific Committee opinion, read in full)
- Titanium Dioxide: the glossary row’s light source. SCCS/1516/13 (EU Scientific Committee opinion, read in full)
The filters’ answers come from our own knowledge base, whose rows carry their own citations on their own pages; the eight bottles’ packs and periods after opening are read as the catalogue records them, with the maker’s page and the date in each row.
Corrections and updates
Nothing has been corrected on this page since it was published. If that changes, the change will be listed here with its date rather than quietly edited in.
Last re-read against its sources on . If you find an error, tell us and we will fix it and say so.
Get the next one
We read the primary papers, write to the researchers, and print what they say. If that is useful, the free plain-words routine guide comes with it. No spam, and nothing sold here.