Rod-shaped crystals and the enamel lattice: why nano-hydroxyapatite adheres so well.
Nano-hydroxyapatite adheres to enamel because it is the same mineral arriving at a surface made of itself, and that affinity is real and has been photographed rather than merely asserted. What forms is a deposit on the tooth rather than a join with it: in an in-situ pilot study, particles delivered by a single 30-second rinse were still scattered across enamel slabs worn in two volunteers' mouths two hours later, bridged to the salivary film by connective structures visible under the electron microscope1. The nano-hydroxyapatite in S3 Sensitivity Science™ is under 100 nm, rod-shaped and crystalline; its biomimetic partner is around two microns and less crystalline, one sized to enter and one sized to settle on top. No study has measured how long such a deposit survives a normal day of eating, drinking and brushing, and the authors of the laboratory work that produced the thickest coating anyone has measured on enamel say in their own discussion that its durability in the mouth is unknown2.
What was checked15 peer-reviewed studies, two European safety opinions on hydroxyapatite (nano), product information as published by each brand
- Hydroxyapatite is the mineral most of a tooth is built from, so a hydroxyapatite crystal meeting enamel is not sticking to a foreign surface; it is meeting its own chemistry.
- The crystal in the tube has been measured, not just described: in vitro, three batches of one commercial oral-care grade were rod-like with an average particle length of about 20 to 40 nm, in characterisation work whose authors include the material's manufacturer, and an independent group working on enamel used a nanocrystalline grade of about 35 nm35.
- Shape is not a slogan here, it is a regulatory specification: the European safety opinion on hydroxyapatite (nano) applies only to rod-shaped particles, and states that it is not applicable to needle-shaped ones4.
- In vitro, when a laboratory put nanocrystals on extracted human enamel and read the result by X-ray diffraction, the new layer was hydroxyapatite and its crystals took a preferred orientation relative to the enamel's own; that experiment needed a chemical pretreatment and a 24-hour soak, neither of which a toothbrush provides5.
- S3 carries a rod under 100 nm and a particle around two microns because each does a job the other cannot, one inside the channel and one across the surface.
Why does hydroxyapatite stick to a tooth at all?
Because it is the tooth's own mineral. A 2023 review of nano-hydroxyapatite in dentistry puts enamel at roughly 96% inorganic material by weight and dentine at roughly 65%, and that inorganic fraction is largely one crystalline calcium phosphate: hydroxyapatite, Ca₁₀(PO₄)₆(OH)₂6. Brushing a hydroxyapatite paste against enamel is therefore not gluing an unrelated substance to a tooth. It is presenting a surface with more of what it is already made of.
The same review gives the physical difference between the nano form and coarse hydroxyapatite: higher solubility and higher surface energy, which it attributes to the morphological and structural similarity between nanosized hydroxyapatite particles and the crystals of the tooth itself6. That is as close as the published literature comes to a mechanism for "like binds to like". Note what it is and is not: a statement that the property differs, not a measurement of what the property produces on a tooth, and the review makes no claim of the second kind6.
It helps to know that the mineral in the tube is manufactured. The standard chemist's review of the ingredient, written in 2018 by a university chemist and a senior scientist employed by a German maker of hydroxyapatite toothpastes, treats synthetic hydroxyapatite throughout as an agent built to a specification rather than an extract of anything7. That matters for the rest of this page. A manufactured mineral arrives with a specification, and the next section is about what a regulator wrote into one.
What does a hydroxyapatite crystal actually look like?
Rod-like, and small enough that the honest answer is a distribution rather than a number. In vitro, three batches of one commercial oral-care nano-hydroxyapatite were characterised by dynamic light scattering and transmission electron microscopy and came out rod-shaped, with an average particle length of about 20 to 40 nm; two of that paper's authors are affiliated to the company that makes the material, which is worth carrying alongside the figure3. The nanocrystalline carbonate-substituted hydroxyapatite used in the in-vitro mineralisation work below had a typical particle size of about 35 nm5. Enamel's own apatite sits in the sub-micrometre range, packed into the rods that give enamel its structure5.
Then there is what is actually in a tube, which is not the same question. An independent laboratory extracted and analysed the particulate fraction of four marketed hydroxyapatite toothpastes and a mouthrinse in 2025: the particles it found were amorphous rather than well-formed crystals, and elemental analysis identified silicon and oxygen, with calcium and phosphorus present only in small amounts on the particles8. Silicon and oxygen are silica, the abrasive, and it was silica that dominated the particles that analysis recovered8. That is an in-vitro analysis of formulations and it says nothing about whether any pack misstates its ingredients. It does say that a page about crystal shape should not quietly assume every particle in a toothpaste is a tidy rod.
| Form | Size, as measured | Shape | Crystallinity | Source, with the model named |
|---|---|---|---|---|
| The tooth's own apatite | Crystal lengths in the sub-micrometre range | Elongated crystallites packed into enamel rods | Crystalline, with an ordered direction perpendicular to the enamel-dentine boundary | Narrative review, 2023 (ST-293); X-ray microdiffraction on extracted human enamel (ST-480) |
| Commercial oral-care nano-hydroxyapatite | Average particle length about 20 to 40 nm across three batches | Rod-like | Crystalline | Dynamic light scattering and TEM in vitro; authors include the manufacturer (ST-290) |
| Carbonate-substituted nanocrystalline hydroxyapatite | About 35 nm, forming agglomerates of about 50 to 70 nm on the surface | Nanocrystals | Crystalline; X-ray diffraction found hydroxyapatite and no other phosphate phase | In vitro on extracted human enamel (ST-480) |
| Biomimetic (micron-sized) hydroxyapatite | Around two microns | Not published | Less crystalline than the nano form | S3's declared formulation specification (PO-007) |
| Zinc-carbonate hydroxyapatite | Not stated in the record | Nanocrystals | Not reported | In vivo, teeth extracted after eight weeks of use (ST-031) |
| The particulate extracted from marketed hydroxyapatite toothpastes | Varying sizes | Amorphous particles | Largely amorphous; silicon and oxygen dominant by elemental analysis | In vitro analysis of four marketed toothpastes and a rinse (ST-380) |
| S3's two declared entries | Given as inclusion levels of each ingredient as supplied, 10% nano-hydroxyapatite and 5% biomimetic hydroxyapatite | Rod-shaped for the nano entry; the larger form's shape is not published | One crystalline, one less so | Brand ingredient list, read 2026-09-09 (CR-028) |
Why is the crystal's shape written into the safety specification?
Because a regulator had to decide which particles it was talking about, and it drew the boundary by morphology rather than by concentration alone. Read the operative sentence of the 2023 European opinion on hydroxyapatite (nano), SCCS/1648/22, adopted 21 to 22 March 2023, and the shape comes before everything: the assessment "only applies to the hydroxyapatite (nano) with the following characteristics: composed of rod-shaped particles of which at least 95.8% (in particle number) have an aspect ratio less than 3, and the remaining 4.2% have an aspect ratio not exceeding 4.9; the particles are not coated or surface modified"4. A line further on says what falls outside it: "This Opinion is not applicable to hydroxyapatite (nano) composed of needle-shaped particles"4. The 2025 opinion on submission IV, SCCS/1677/25, adopted 26 June 2025, keeps that architecture with wider tolerances, adds a maximum particle length and widens the exclusion to any material composed of or containing needle-shaped particles9. The full comparison of the two opinions, their concentration ceilings and what they mean for a shopper belongs on the pages about ten per cent and daily use; what matters here is that shape is the first thing the specification pins down.
Aspect ratio is length divided by width, so a particle with an aspect ratio under three is a stubby rod rather than a fibre. The reasoning behind the rod-and-not-needle condition is a safety argument, and this page draws no efficacy conclusion from it. It is worth knowing all the same, because it means the shape a consumer reads as marketing language is the same property a regulator writes into a specification.
One toothpaste has put it on the label. The published ingredient list for the Davids Sensitive + Whitening nano-Hydroxyapatite Toothpaste reads "Nano Hydroxyapatite (rod shaped)", naming the crystal's morphology inside the ingredient declaration itself. This page compares ingredients and stated actions only, not clinical performance, based on what each brand states about its own formula. Prices and formulations may change; always check the pack.
Why does S3 use a rod-shaped crystal and a larger, less crystalline one?
Because there are two places to put mineral and one particle cannot occupy both. The nano-hydroxyapatite in S3 is under 100 nm, rod-shaped and crystalline; the biomimetic hydroxyapatite beside it is around two microns and less crystalline. Two microns is two thousand nanometres, so these are not two grades of the same idea but two different tools, one sized for the inside of a dentine tubule and one sized to lie across the surface above it.
On the pack the two appear as separate entries in the ingredient list, "Hydroxyapatite" and "Hydroxyapatite (nano)", given as inclusion levels of each ingredient as supplied rather than as the weight of mineral in the finished paste. That is a description of a formula, not a claim about how firmly either form holds. No such comparison between brands has been published, and none is cited here, because none exists. What the raw material's own paperwork supports is narrower and duller: the nano-hydroxyapatite in S3, nanoXIM CarePaste, is tested to the SCCS guidance for nanomaterials, and the Committee has separately published an opinion on hydroxyapatite (nano) in oral products. Tested to a guidance is not the same as assessed as a product, and the distinction is the whole of what that sentence says.
What the larger, less crystalline form does once it is on the surface — the mineral layer it forms, and what it releases when the mouth turns acidic — is silo B's subject, and it is set out on the pages about what biomimetic hydroxyapatite is and the protective mineral layer. Where each form ends up is the subject of how nano-hydroxyapatite seals dentine tubules; the size argument itself belongs to what nano-hydroxyapatite is.
What has actually been observed on enamel, and in what model?
More than the sceptics assume and less than the packaging implies. The most informative single experiment for this page is an in-vitro one: it put nanocrystalline carbonate-substituted hydroxyapatite onto segments of sound human enamel and then interrogated the result with X-ray microdiffraction, and it found that the new layer was hydroxyapatite and no other phosphate phase, that it measured 150 to 200 nanometres thick on cross-section, and that its crystals showed a predominant orientation relative to the ordered direction of the apatite already in the enamel5. Crystallising in register with the surface underneath is a real and specific finding, and it is the strongest version of the "same crystal chemistry" story that anyone has published.
Read the method before taking it home. In that in-vitro protocol the enamel was pretreated in calcium hydroxide and then in an amino-acid booster, and the mineral was applied as a 24-hour immersion in a solution at pH 8.55. The arm that came closest to a toothbrush is the in-vitro one that received no pretreatment at all, and it took mineral only in scattered patches of roughly 10 to 100 square micrometres, with no pattern to where the deposit formed5. The layer that came out looking most like enamel needed a chemical pretreatment first. A toothbrush does not provide one.
The thickest measured coating comes from a different setting again. On bovine enamel, a single four-minute professional application of an eggshell-derived nano-apatite together with an acidulated phosphate fluoride solution produced a coating of aggregated particles about five micrometres thick, against one to two micrometres for the fluoride solution alone, and the coated samples lost 0.859 micrometres of enamel to an acid challenge where the fluoride-only samples lost 1.761 and untreated samples lost 6.1842. That is a clinic procedure with a high-concentration fluoride solution, not brushing, and it is quoted here for one principle only: a hydroxyapatite layer can be formed on enamel and can hold up to an acid challenge in a laboratory.
| Study | Model | How the mineral was applied | What was seen on the surface | How long it was watched |
|---|---|---|---|---|
| Seredin 2022 (ST-480) | In vitro, 50 enamel segments from ten extracted human teeth | 24-hour immersion at pH 8.5, after pretreatment with calcium hydroxide and an amino-acid booster | A layer 150 to 200 nanometres thick, confirmed as hydroxyapatite by X-ray diffraction, crystals preferentially oriented to the enamel's own | One reading, no challenge applied |
| Seredin 2022, no-pretreatment arm (ST-480) | As above, enamel left untreated | 24-hour immersion at pH 8.5 | Scattered patches of roughly 10 to 100 square micrometres, no pattern to the microrelief | One reading, no challenge applied |
| Satou 2022 (ST-479) | In vitro, bovine enamel, eight samples per group | One four-minute professional application with an acidulated phosphate fluoride solution | A coating of aggregated particles about five micrometres thick, against one to two micrometres for fluoride alone | One acid challenge; the authors state that durability in the mouth is unknown |
| Nobre 2020 (ST-482) | In situ, bovine enamel slabs on splints worn by two volunteers | One 30-second rinse of a 5% watery suspension | A heterogeneous scatter of particles and clusters, with bridge-like structures to the salivary pellicle at two hours | Immediately after the rinse, at 30 minutes and at two hours |
| Hill 2015 (ST-040) | In vitro, 25 dentine discs from extracted molars | One 30-second application of a nano-hydroxyapatite rinse | The only one of the rinses tested that adequately covered the dentine surface on SEM | Not followed; on fluid flow other rinses did better |
| Lelli 2014 (ST-031) | In vivo, five subjects per group, teeth analysed after extraction | A zinc-carbonate hydroxyapatite toothpaste, used normally for eight weeks | A hydroxyapatite-rich coating on the enamel; the fluoride and potassium nitrate comparators showed no appreciable change | Eight weeks of use, one look at the end |
| Bossù 2019 (ST-029) | In vitro and in vivo, deciduous teeth | A fluoride-free biomimetic hydroxyapatite toothpaste | An apatite coating that resisted brushing | Not stated in the abstract |
Three rows there are worth pausing on. The in-situ study is the only one that watched the mineral on a surface inside a human mouth against a clock, and it saw a deposit surviving a real mouth for two hours after one rinse1. The in-vivo study is the only one that let people brush normally for eight weeks and then looked at their extracted teeth, and it found a hydroxyapatite-rich coating on the enamel where the fluoride and potassium nitrate comparators had left the surface much as they found it11. That study used a zinc-carbonate biomimetic material rather than the nano form, five subjects per group, with no randomisation described, and it measured a surface and not a symptom11. The third is the closest anyone has come to testing whether a surface coating stays put: in a study on deciduous teeth, a fluoride-free biomimetic hydroxyapatite toothpaste deposited an apatite coating that resisted brushing, though the abstract gives no sample size, no duration and no figures12.
Coverage after a single application is easier to demonstrate than coverage that lasts. In vitro, on 25 dentine discs from extracted molars, a nano-hydroxyapatite rinse was the only one of four tested that adequately covered the surface under the electron microscope after one 30-second application; on fluid-flow measurement, in the same study, other rinses did better10. Nothing in that experiment was followed past the single application.
How long does it stay there?
Nobody knows, and that is the most useful sentence on this page. A PubMed search made for this page, looking for a measured retention time or a re-examination after a defined interval for a hydroxyapatite layer left on a tooth by a toothpaste, returned nothing that met either condition.
What exists are three partial answers, none of which reaches a day. The in-situ study followed its particles for two hours and watched them thin out: clusters up to four micrometres immediately after the rinse, fewer and smaller particles at 30 minutes, and at two hours only small particles, with clusters above one micrometre rarely seen1. The authors attribute the decline to continuous adsorption and desorption in the mouth, dissolution in saliva and shearing forces, with some particles readsorbed to the pellicle and others swallowed, and they state plainly that stability over longer periods was not evaluated1. Second, the only test of whether a deposit survives a subsequent week of brushing was done on tubule plugs rather than a surface layer: in vitro, half of each dentine disc was re-brushed with distilled water for a further seven days, and the plugging rate in the 80 nm group fell but stayed above 90%13. Third, the authors of the thick-coating study say in their discussion that strength experiments were not performed, that the durability of the coating in the oral cavity is unknown, and that further work is needed on its resistance to brushing2.
Saliva is not a bystander in any of this. In the in-situ study the salivary pellicle appeared to help: at two hours, higher-magnification electron microscopy showed connective structures between the particles and the pellicle on enamel, and the authors conclude that the pellicle acts as a bridge to the surface rather than a barrier in front of it1. Pull the same question into a permeability model and the direction changes: in vitro, on bovine dentine slices, adding a film of human saliva reversed the ranking of three toothpastes on hydraulic conductance, with the hydroxyapatite arm giving the lowest permeability of the three without saliva and the highest with it, though the authors report that those between-material differences did not reach statistical significance14. Two models, two directions, and no basis for choosing between them. The honest summary is that what a hydroxyapatite crystal meets on a real tooth is not clean mineral but a protein film, and the published work disagrees about what that film does.
There is a reason this evidence is so thin, and the researchers say it themselves. Laboratories build sintered hydroxyapatite pellets as model surfaces because, in their own words, samples of natural teeth do not meet the demands for well-defined, highly reproducible properties15. Teeth vary too much to measure adsorption on reproducibly, so part of what is known about hydroxyapatite adhering to hydroxyapatite has been learnt on a substitute for a tooth.
None of this makes brushing pointless, and none of it makes the deposit durable. On the only in-mouth measurement anyone has published, the layer begins thinning from the moment it is laid down1. It is replaced at the next brushing or it is lost, which is a statement about a routine and not a promise about a layer.
Is "bonds to your enamel" a fair way to describe it?
No, and the accurate words are better anyway. Nothing in the published record shows hydroxyapatite fusing with, integrating into or becoming part of the enamel lattice. What has been shown is narrower and more interesting. In vitro, the mineral crystallises on enamel as a template, in register with the crystals underneath5. In an in-situ study, it could still be found on the surface, held against the salivary film, two hours after a single rinse1. Deposit, adhere, settle, crystallise on. Each of those is supported. "Bonds" is not, and neither is anything that implies the result is fixed in place.
Keeping the weaker word costs less than it looks. The evidence on the outcome people actually care about is separate from the evidence on adhesion, and it is stronger than the mechanism story: a 2023 systematic review and meta-analysis of 44 clinical trials, whose authors include scientists employed by a German maker of hydroxyapatite toothpastes, reported a 39.5% reduction in dentine hypersensitivity against placebo17. That is a claim about the ingredient class, from a conflicted source, and it is weighed properly on the page asking whether nano-hydroxyapatite works for sensitive teeth. The caution around it comes from a source with nothing to gain either way. Reviewing the ingredient in 2021, a group at a university dental school with no declared industry tie wrote in their abstract that clinical data on hydroxyapatite oral care products "is limited with varied results" and that the effectiveness of these compounds against fluoride "has not been established"16.
So the fair description is this. The attraction is real, it has a physical basis in surface energy and crystal chemistry, and the deposit it produces has been photographed on enamel in the laboratory, in a mouth over two hours and on teeth after eight weeks of ordinary brushing. It is a deposit. It is renewed or it is lost. Anyone telling you it bonds to your teeth is describing a picture rather than a measurement. Why enamel is worth protecting in the first place is set out in the Journal's piece on tooth enamel erosion.
Frequently asked questions
Does the hydroxyapatite in S3 bond to my enamel?
It deposits on enamel and adheres to it. In vitro, the mineral has been photographed crystallising on enamel in register with the tooth's own crystals5; in an in-situ pilot, it was still on enamel slabs worn inside the mouth two hours after a rinse1. It does not fuse with the enamel lattice, and nothing in the published record shows that it does. How long the deposit lasts on a tooth in normal use has not been measured, which is why the instruction is to brush twice a day rather than to expect a layer to stay.
Does the layer wash off when you rinse?
Part of it goes, and that is the expected behaviour of a deposit rather than a failure of one. In the in-situ study, particles were rinsed off the specimens with running water before they were even examined, and what remained thinned steadily over two hours through dissolution, shearing and swallowing1. What survives is the fraction held against the salivary film, and it is replaced at the next brushing.
Is nano-hydroxyapatite the same shape as the crystals in my teeth?
Similar, not identical. Enamel's crystallites are elongated apatite packed into rods with lengths in the sub-micrometre range5. In vitro, commercial oral-care nano-hydroxyapatite has been measured as rod-like at about 20 to 40 nm long, in characterisation work whose authors include the material's manufacturer, which is far shorter3. The 2023 review that describes the nano form's higher surface energy attributes it to exactly that morphological and structural similarity to the tooth's own crystals6.
Why do some labels say "rod shaped"?
Because shape is part of the regulatory specification, not because it is a marketing flourish. The European safety opinion on hydroxyapatite (nano) applies only to rod-shaped particles within a stated aspect-ratio distribution, and says it is not applicable to needle-shaped ones4. At least one brand has taken the phrase into its published ingredient list.
Does a micron-sized biomimetic particle stick as well as a nano one?
Nobody has run that comparison in a way that would answer it. The two forms have been studied in different experiments, on different surfaces, with different methods, and no head-to-head measurement of adhesion exists. What the larger form does once it is in place is covered on the page about what biomimetic hydroxyapatite is.
Where S3 sits
S3 carries a rod-shaped crystal under 100 nm and a second, larger and less crystalline particle around two microns, because the two settle in different places. One works inside the channel and one works on the surface above it, and each does a job the other cannot. Neither of those is a statement about how firmly anything holds, and this page does not make one.
S3 Sensitivity Science™ pairs potassium nitrate with two hydroxyapatites and adult-strength fluoride in one daily paste.
See the toothpasteS3 Sensitivity Science™ is one daily toothpaste carrying three actives: potassium nitrate at 5% for the nerve, nano-hydroxyapatite at 10% and biomimetic hydroxyapatite at 5%, both as solution, with full adult-strength fluoride kept in. Calm, strengthen, protect: three actions that sensitive teeth need, from a single tube. Patent-pending S3 Repair Technology™, UK application GB2604755.5. More than 20 practising UK dentists own a stake in S3, and nine founding dentists advise on the formulation. Read more about S3.