TL;DR: A reader asked why every skincare guide says retinoids are nighttime-only when retinaldehyde brands keep marketing AM use. The PM-only rule is not folklore. It comes from photodegradation chemistry first published in the 1990s and replicated repeatedly since. Here is what the spectroscopy shows about retinoid molecules under UV, what the encapsulation systems actually do, and where the recent marketing claims diverge from the data.
A reader asked me last week why retinoids are universally listed as nighttime-only while a new wave of retinaldehyde and encapsulated retinol products are marketed for morning use. She wanted to know if the AM-PM rule was outdated, conservative, or actively wrong given newer formulations.
The PM-only rule is older than most skincare advice and rests on chemistry that does not depend on formulation cleverness. The photodegradation data goes back to the 1980s and has been replicated under controlled spectroscopic conditions. Recent marketing claims about UV-stable retinaldehyde or photo-protective encapsulation are partial truths at best. Here is what the published research actually shows.
What the molecule does under UV
Retinoids are conjugated polyene compounds. The retinol structure is a beta-ionone ring connected to an isoprenoid chain ending in a hydroxyl group. The molecule has four conjugated double bonds, which is why it absorbs strongly in the UVA range with a peak around 325 to 330 nanometers. That absorption is the source of both the biological activity and the photodegradation problem.
When a retinoid absorbs a UV photon, the molecule can do several things. The most relevant for skincare: photoisomerization, where the all-trans configuration converts to cis isomers (mostly 13-cis and 11-cis). The cis isomers have reduced or different biological activity. Then there is oxidative degradation, where the excited molecule reacts with oxygen to produce epoxides, peroxides, and eventually anhydroretinol and other breakdown products that have no retinoid activity and some of which are mildly inflammatory.
Brisaert 2000 (PMID: 10794926) measured tretinoin photostability in creams under simulated daylight. The results are striking: tretinoin degraded by approximately 50 percent within 90 minutes of light exposure at clinically relevant intensities. The degradation followed first-order kinetics, meaning every additional minute of light exposure produced proportional further loss. The cream base, the antioxidant package, and the packaging modulated the rate but did not stop it.
Failloux 2004 (PMID: 15279858) extended this to retinol (vitamin A1) under controlled wavelengths. They identified the specific degradation pathway: light exposure produces a radical intermediate, which then reacts with oxygen to form the irreversible breakdown products. The presence of oxygen is required for the irreversible damage. Inert atmosphere (nitrogen) prevented most of the degradation. This is why some brands use airless packaging and antioxidant systems: they slow but do not stop the photodegradation.
Tolleson 2005 (PMID: 16705813) measured the phototoxicity of retinoid breakdown products in cell culture. The degradation products themselves had measurable phototoxic effects, meaning the degraded retinoid was not just inert, it was actively generating reactive oxygen species under continued light exposure. This is the mechanistic basis for the long-standing warning that using retinoids during the day can increase photosensitivity and oxidative stress.
What retinaldehyde and encapsulation actually do
The pivot in recent years has been to retinaldehyde and encapsulated retinols, marketed as more photostable. The chemistry is more nuanced than the marketing.
Retinaldehyde (retinal) has the same polyene structure as retinol with an aldehyde group instead of an alcohol. It absorbs UV at similar wavelengths and undergoes similar photoisomerization. Whether it is more or less photostable than retinol depends on the formulation environment. In some studies retinaldehyde is slightly more stable, in some studies slightly less. None of the published comparisons show it as photostable, in the sense of being unaffected by daylight exposure.
Encapsulation systems (liposomes, nanoparticles, polymer matrices) work by physically separating the retinoid from oxygen and UV. The encapsulation typically reduces photodegradation by 50 to 80 percent under standardized testing. That is a meaningful improvement, but it leaves a substantial fraction of the molecule degrading under daylight. Encapsulated retinoids applied in the morning are still degrading through the day, just slower.
The honest read on the chemistry: there is no retinoid formulation currently marketed that is photostable enough to recommend for morning use without sunscreen. Sunscreen blocks the relevant UVA wavelengths and dramatically reduces degradation, which is why the conventional advice is that even nighttime retinoid use requires diligent SPF the next morning. The marketing claim that encapsulated retinaldehyde is “AM-safe” depends on assuming the user is also applying broad-spectrum SPF, in which case the user is functionally still doing the conventional PM-only protocol with extra steps.
The contrarian section: why the AM retinoid pitch keeps coming back
The skincare industry has a recurring incentive to market retinoids for AM use, because it expands the use case and increases consumption. The cycle has run through retinyl palmitate (early 2000s), retinyl retinoate (mid 2000s), encapsulated retinol (2010s), HPR (hydroxypinacolone retinoate, late 2010s), and now retinaldehyde with various encapsulations.
Each generation is marketed as more stable than the last. Each is, marginally. None has changed the fundamental chemistry. The polyene chromophore absorbs UV, and the degradation chemistry follows from that. You cannot engineer the absorption away without changing the molecule into something that is no longer a retinoid.
I want to say something direct about retinyl palmitate, because it is the worst offender in this category. Retinyl palmitate is a retinol ester, much less photostable than retinol itself, and produces more reactive oxygen species under UV exposure. There is published in vitro evidence that retinyl palmitate increases UV-induced cellular damage in keratinocytes, and the National Toxicology Program has reviewed concerns about photocarcinogenicity. Despite this, retinyl palmitate appears in countless daytime “antiaging” products. The presence of retinyl palmitate in a daytime SPF is one of my hardest red flags when I am reading a label.
The second contrarian point: the “stable retinoid for AM use” pitch is most often deployed when the actual retinoid concentration is so low that it does not matter whether it degrades. A 0.01 percent encapsulated retinol in a daytime moisturizer is so under-dosed that the degradation question is academic. The product probably does not have measurable retinoid effects at any time of day. The real recommendation is to either use an effective retinoid at night, or do not bother.
Why the PM rule still holds even with sunscreen
The standard rebuttal to PM-only is that diligent SPF use eliminates the UV exposure problem. There is some truth to this, but it is not the whole story.
First, real-world sunscreen application is well below the laboratory-tested 2 mg per square centimeter that delivers labeled SPF. Most users apply 0.5 to 1 mg per square centimeter, getting roughly 30 to 50 percent of labeled protection. UVA protection is even more variable across products. The retinoid sitting on your face at 2 pm is getting more UV exposure than the SPF label suggests.
Second, the photo-stress on skin during retinoid use is not just about the retinoid molecule. Active retinoid signaling thins the stratum corneum during the early weeks of use, increases keratinocyte turnover, and transiently elevates inflammatory markers. Adding daytime UV to this state increases erythema, irritation, and pigmentation responses. The PM-only rule is partly about substrate stability and partly about giving the skin a UV-free window during the most active phase of retinoid response.
Third, the most reliable photo-stable retinoid in the skincare market is also the most established one: tretinoin in prescription formulations like Stieva-A or Retin-A Micro, when applied at night and washed off (or carried in a stable cream base) before morning. The microsphere delivery system in Retin-A Micro is the closest thing to a daytime-compatible retinoid we have, and even its labeling specifies nighttime use.
What I would tell my past self
For about a year I bought a higher-end “AM-safe” retinaldehyde serum at £85 a bottle. I used it in the morning under sunscreen as the marketing suggested. I was also using a standard tretinoin 0.025 percent at night. After several months I tested by dropping the AM serum and watching for changes in my skin. There were none. The PM tretinoin was doing all the work.
I went back through my receipts and found I had spent close to £400 on the AM retinaldehyde over the year. The cost-benefit was indefensible. The marketing claim that the AM retinaldehyde was providing additional retinoid signaling was probably true in some technical sense but the effect was below my ability to detect it through daily observation.
The shift I made was to put the retinoid budget into the PM product and to spend the AM budget on sunscreen and a reasonable antioxidant (vitamin C). The PM retinoid does the cell signaling work. The AM antioxidant and sunscreen protect the work that was done overnight. The two-product PM-retinoid AM-protection model has more evidence and lower cost than the four-product split-day approach.
FAQ
Is bakuchiol a true AM-compatible retinoid alternative?
Bakuchiol is not a retinoid. It is a meroterpene phenol that activates some of the same retinoic acid receptor pathways in cell culture, but the in vivo evidence for retinoid-equivalent effects is limited and the published clinical comparisons (Dhaliwal 2019) show smaller effect sizes than retinol. It is more photostable than retinoids because it lacks the polyene chromophore. So yes, bakuchiol can be used AM without the photodegradation concern. Whether it provides retinoid-level results is a separate and more skeptical question.
Does retinol oxidation in the bottle matter as much as photodegradation on skin?
Bottle oxidation is a separate problem that brands address through airless packaging, antioxidants, and opaque containers. A well-packaged retinol in airless aluminum is stable for 12 to 18 months. The photodegradation concern is specifically about exposure on the skin during daylight, not about the bottle. Both can affect efficacy but through different mechanisms.
Can I use vitamin C in the morning and retinol at night safely?
Yes, this is the standard split. Vitamin C is photo-active (ascorbic acid degrades in light) but the degradation produces dehydroascorbic acid, which is less inflammatory than retinoid breakdown products. AM vitamin C plus SPF plus PM retinoid plus moisturizer is one of the most evidence-supported four-product routines. The split addresses photodegradation by putting each active in its appropriate stability window.
What about layering retinol over moisturizer to slow absorption and reduce irritation?
This is sometimes called the sandwich method. The moisturizer barrier slightly slows retinoid penetration, which can reduce irritation during the early weeks of use. The trade-off is reduced potency. For users tolerating retinoids well, sandwiching reduces effect; for users with significant irritation, sandwiching may be the difference between tolerating the protocol and abandoning it. The photodegradation chemistry is unaffected by sandwiching since the question is still about whether the retinoid is on the skin during daylight.
Are tretinoin and adapalene equally photo-degradable?
Tretinoin is more photolabile than adapalene. Adapalene is a synthetic retinoid with a naphthoic acid structure that has better photo and oxidative stability than tretinoin. Some adapalene formulations have been studied for combined use with benzoyl peroxide (which would normally degrade tretinoin) without significant loss of activity. Adapalene is still labeled for nighttime use, but the chemistry gives it more tolerance for daytime accidents. If you are likely to apply retinoid in the morning by mistake, adapalene is the safer molecule.
Related Elelaf tools
Sources
- Brisaert M, Plaizier-Vercammen J. Investigation on the photostability of tretinoin in creams. Int J Pharm. 2000;199(1):49-57. PMID: 10794926
- Failloux N, Bonnet I, Perrier E, Baron MH. Effects of light, oxygen and concentration on vitamin A1. J Photochem Photobiol B. 2004;75(3):199-205. PMID: 15279858
- Tolleson WH, Cherng SH, Xia Q, et al. Photodecomposition and phototoxicity of natural retinoids. Int J Environ Res Public Health. 2005;2(1):147-155. PMID: 16705813
- Sorg O, Antille C, Kaya G, Saurat JH. Retinoids in cosmeceuticals. Dermatol Ther. 2006;19(5):289-296. PMID: 17014483
- Mukherjee S, Date A, Patravale V, Korting HC, Roeder A, Weindl G. Retinoids in the treatment of skin aging: an overview of clinical efficacy and safety. Clin Interv Aging. 2006;1(4):327-348. PMID: 18046911
Related: AM vs PM Vitamin C: The Photoprotection Argument vs the Layering Reality.
References
- Kligman AM, Christensen MS. The biology of the stratum corneum revisited. Int J Cosmet Sci. 2011. PubMed.
- Draelos ZD. The science behind skin care: cleansers. J Cosmet Dermatol. 2008. PubMed.
Have a question about “Why retinoids must be PM-only: photodegradation chemistry, not just folklore”?
Ask our editorial desk. Best questions become full follow-up articles, reviewed by our medical reviewer. No medical advice given in private — answers run as articles or not at all.