Every other article on this site feeds into this one. It is worth assembling the claim from its parts, because stated as a whole it cannot be assessed, and stated as parts it mostly can.
The four supports, each assessed
One: limited dermal penetration
Well supported. Assessment work has consistently found that metal oxide particles remain in the stratum corneum and follicular openings on healthy, intact skin. The qualification about intact skin is in the assessments themselves and matters, as set out in the nanoparticle safety questions. This is the strongest support the claim has.
Two: low allergenic profile
Reasonably supported. The metal oxides are not among the substances that appear frequently in contact allergy series, and photoallergic reactions to certain organic filters are a recognised if uncommon entity. The support is for the filter, not the formula, which is where the difficulty starts.
Three: photostability
Established. Metal oxides do not photodegrade in the molecular sense. As set out in photostability, this is a genuine property and the conclusion usually drawn from it, that mineral products last longer in use, does not follow, because films fail by mechanical removal and redistribution rather than by molecular degradation.
Four: environmental superiority
Not supported, and contradicted in parts. Zinc oxide releases zinc species in water and zinc is aquatically toxic at concentration. Nanoscale titanium dioxide is photocatalytic in sunlit water. Neither biodegrades. This is set out in mineral filters and the environment.
Mineral sunscreen is safer
- What would have to be true
- That a comparison has been made at the level of finished products, with formulation variables controlled, on defined safety endpoints.
- That the individual supports for the claim aggregate into a general one.
- That the costs of the mineral approach have been counted on the same balance.
- What is established
- Metal oxide filters show limited dermal penetration through healthy, intact skin.
- The metal oxides are not frequent contact allergens, and photoallergic reactions to certain organic filters are recognised.
- Metal oxides do not photodegrade in the molecular sense.
- What is not established
- That any comparative assessment of finished products on defined safety endpoints has been made.
- That harm from compliant products using organic filters has been demonstrated in normal use.
- That environmental superiority holds, which the evidence contradicts in places.
- That the costs, including reduced application density from cast and drag, have been counted.
The costs the claim does not count
A safety comparison that only counts one side is not a comparison. Four costs belong on the other side of the balance, and all of them are established elsewhere on this site.
- Reduced application density. Cast and drag push people to apply less, and delivered protection falls steeply with quantity. This is the largest one and it is quantitatively significant, as set out in the quantity problem.
- Exclusion by shade. Products that leave a visible cast on deeper skin are, in practice, not used by the people they do not suit, which is a protection outcome and not a cosmetic one.
- Fragrance and essential oils. Within the natural segment specifically, these are more common, and they are established sensitisers with a photosensitisation dimension. See essential oils and photosensitivity.
- Adjacent claims. The same marketing framework that promotes mineral filters also promotes plant oils as sun protection, which is the most dangerous claim in the category and is dealt with in its own article.
What safer would require to be established
A defined endpoint, a defined population, matched formulations differing only in filter system, and a comparison. Not a mechanism, not a preference, and not the absence of a substance somebody has decided to worry about. Safety in cosmetics is assessed product by product by qualified assessors, which is a different operation from a class comparison, and no regulator produces class comparisons because the framework does not work that way.
This is why the status recorded is unresolved rather than unsupported. It is not that evidence was sought and found wanting. It is that the question, as asked, is not one the evidence base has been arranged to answer, and the supports and costs point in different directions.
Under the status vocabulary published on this site, unsupported means no evidence was located that would support the claim. Unresolved means a real question with incomplete evidence, neither supported nor refuted. This claim has real supports and real counterweights and no comparison. That is what unresolved describes.
Why safer is the wrong shape of word
Safer is comparative, and a comparative term needs three things to mean anything: a comparator, a dimension and a population. Safer than what, in respect of what, for whom.
Once those are supplied, the claim usually becomes either trivially true or clearly false, and in both cases it becomes checkable. Less likely to cause allergic contact dermatitis than a formulation containing a named organic filter, in a sensitised individual, is a claim with a shape. Less likely to be absorbed into viable epidermis on intact skin is a claim with a shape. Better for the sea is a claim with a shape and it fails. Safer, on its own, has no shape at all, and that is precisely why it is the word used.
The same test applies to gentler, cleaner, purer and better for you. A reader encountering any of them can ask the three questions, and a company that cannot answer them has said nothing. This is not a trick reserved for critics: it is the ordinary discipline of making a claim that can be assessed, and it is what the advertising codes ask for.
What a reader can take from it
The category's reasons for existing are better than the argument it usually makes for itself. A person can prefer a mineral formulation because they like the finish, because they have reacted to something in the past, because they want a filter that stays outside the body on principle, or because they want a short and legible ingredient list. Those are all sufficient reasons and none requires the belief that the alternative is dangerous.
What matters more than the choice, on every measure that has been examined, is applying enough, applying it to everywhere that is exposed, reapplying, and treating sunscreen as one measure alongside shade, clothing and timing. The general guidance on that comes from the NHS, from Cancer Research UK and from the British Association of Dermatologists.
If you have had a skin cancer, are on a medicine or a treatment that increases sensitivity to light, have a photosensitivity condition, or are worried about a mole or a change in your skin, the right advice is clinical and specific to you. Contact your GP. The NHS pages on skin cancer symptoms explain what to look for and how to get seen.
The position of this publication
Neither anti mineral nor anti conventional. The metal oxides are good filters with real advantages and real limits. The organic filters are good filters with real advantages and real limits. The category that sells the first by frightening people about the second is doing something that is bad for the reader and, in the long run, bad for itself, because a category whose claims collapse under examination loses the ability to be believed about the things it has right.
Everything examined on this site is collected, with its status, in the register of claims examined, and the vocabulary used to record those statuses is published at claim status. Both exist so that a reader can check whether this publication is applying the same standard to every claim, including the ones it finds sympathetic.