Why Foundation Turns Orange or Darker

Foundation genuinely does change colour over a day. That part is measurable, and it has been measured.

The word everybody uses for it — oxidation — is almost certainly the wrong one. And the mechanism that best fits the evidence is not what you would guess: the foundation does not turn into a darker paint. It turns into a more transparent veil, and more of your own skin shows through it.

The honest headline, though, is that this is a thinly studied corner of cosmetic science and the researchers working on it say so themselves.

Why "oxidation" is the wrong word

Oxidation has a narrow chemical meaning — loss of electrons, or an increase in oxidation number. The everyday cosmetic use of the word does not describe that.

The clearest statement of the problem comes from the researchers themselves. A 2025 study in Scientific Reports puts it directly:

"While this phenomenon is often colloquially described as 'oxidation,' the underlying mechanisms remain insufficiently explained in cosmetic science. Since the pigments themselves (e.g., iron oxides and titanium dioxide) are already in stable oxidized forms, oth[er factors]…"

Peer-reviewed cosmetic science describing the internet's favourite explanation as a colloquialism whose mechanism is unexplained is a stronger statement than "experts disagree".

Someone actually ran the experiment

An industry study presented through the Society of Cosmetic Chemists did the obvious thing nobody on the internet had bothered to do: coat a substrate with foundation and expose it separately to sebum, to sweat, and to sunlight, then measure the brightness change.

The result was lopsided. Artificial sebum produced a brightness drop of roughly −1 to −6. Sweat stayed within −1. Sunlight — half the sample in light, half in the dark for 24 hours — also stayed within −1.

Sebum dominated by something like a factor of six, and the "light makes it oxidise" hypothesis was tested head-on and came out near noise.

What sebum actually does — and it isn't chemistry

Here is the reframe that makes the whole thing click.

A 2024 study in the International Journal of Cosmetic Science added artificial sebum to liquid foundation and measured the colour against a standard opacity chart. The finding:

"The influences of sebum addition on liquid foundation darkening had a significant positive correlation with the increase in transmissivities (R² = 0.852, p < 0.01)."

Darkening tracked transmissivity — how much light passes through. The film was not becoming a different colour. It was becoming more see-through.

That is ordinary physics, and it is the same reason a wet stone looks darker than a dry one. Filling the air gaps in a rough, light-scattering layer with a liquid reduces how much light bounces back out. Sebum soaking into a foundation film does exactly that.

And it has a consequence that matches who complains about this. If the mechanism is loss of opacity, then the more covering your foundation was doing, the more there is to lose — which fits the Scientific Reports finding that darker shades drifted less in lightness than lighter shades within each brand.

But it also happens on plastic

This is the single most interesting result in the literature, and it stops the sebum story from being the whole answer.

The Scientific Reports team measured 32 shades from four brands over eight hours — applied to an opacity chart. No skin. No sebum. No sweat. No body heat.

The colour still drifted. And it drifted in the direction people complain about: hue angle decreased for most shades in three of the four brands, and as the paper explains, "the decrease in hue angle suggests that the foundation appeared redder than its initial color."

So "my foundation turned orange" is a real, instrumented observation — even though the usual explanation for it is wrong. Whatever is happening, part of it happens to the film on its own, as it dries and consolidates. It cannot all be your skin chemistry, and it cannot all be sebum.

The authors are candid about the limitation cutting the other way too: discoloration "may be more accurately assessed when foundation shades are applied to human skin rather than inert substrates."

The other thing people mean by "it looked different"

There is a second, completely separate failure that gets folded into this one, and keeping them apart is most of the practical value here.

Metamerism is when two things that match under one light stop matching under another. The CIE defines it as spectrally different colour stimuli having the same tristimulus values in a specified system — different materials, matching under one illuminant, diverging under another.

That explains a shade you matched under shop lighting looking wrong the moment you step outside. It is instantaneous, and it is about two different lights.

It cannot explain a shade that looked right at 8am and looks orange at 2pm under the same office lighting. That is a change over time in one condition, which is a different problem with a different cause. Almost every article on this topic conflates the two.

Working out which one you have

What we could not establish

We want to be clear about how thin the evidence base here is. We located four relevant primary studies. Two are paywalled, one is an industry conference paper. Both research groups that address the topic directly say the area is understudied, and the 2025 paper says the mechanism is "insufficiently explained".

Specifically unresolved:

• How much of the shift is sebum and how much is the film consolidating on its own, on a real face. The two effects have only been measured separately, on artificial substrates.
• Whether black iron oxide actually oxidises during wear. It is the one pigment that chemically could. We found no measurement either way.
• Any quantitative role for metamerism in the shop-versus-daylight mismatch. The phenomenon is well defined; we found no fetchable study measuring it between a foundation and a wearer's skin under retail versus daylight illuminants.
• The refractive-index explanation that circulates in beauty media. It is attributed to the 2024 sebum paper's discussion, which is behind a paywall. We read only the abstract and are not going to repeat an argument we have not seen.
• The US colour-additive regulation text. The eCFR route to 21 CFR 73.2250 redirected to a bot wall during research. We have relied on the FAO/WHO specification and the EU colorant annex for pigment identity instead.

Sources consulted, retrieved 28 July 2026:

• Yolanda S, Kim B, Kim M, Suk H-J. Discoloration of liquid foundation across the shade color diversity. Sci Rep 2025. Quoted for the "colloquially described as oxidation" statement, the opacity-chart method, and the hue-angle direction.
• Zhang Q, Chen G, Hou Y, et al. Int J Cosmet Sci 2024;46(1):142–152. Abstract only — the full text is paywalled. Quoted for the transmissivity correlation.
• FAO/WHO JECFA, "Iron Oxides", FAO JECFA Monographs 5 (2008). Cited for the chemical identity of CI 77491, 77492 and 77499.
• Regulation (EC) No 1223/2009, Annex IV — list of colorants allowed in cosmetic products.
• CIE International Lighting Vocabulary, term 17-23-006 — definition of metamerism.
• Lekner J, Dorf MC. Why some things are darker when wet. Appl Opt 1988;27(7):1278–80.
• Huang H, Wang Y, Zhang J, et al. Study on the factors affecting the darkening of liquid foundations. Hosted by the Society of Cosmetic Chemists. An industry conference paper, described rather than quoted, and used as supporting evidence only.

This article names no product and characterises none. GlowSteal has not tested any product for it. General information, not medical advice.