To quote Logan R. Kearsley's answer to this question regarding stronger-than-normal bones for making big but proportionate humanoids:

And if you wanna go a little further out there... Sapphire

Sapphire is aluminum oxide, Al2O3. It has a compressive strength of 2 gigapascals, so even if you accept some losses in incorporating it into a biological composite, you're still starting out way ahead of natural bone. Aluminum is not currently know to play any significant role in biology, but it is bioavailable in ion form (e.g., as aluminum citrate) and accumulates in the biosphere, so it should be available in normal food supplies--and if biology can handle laying down oxidized iron crystals, I'm sure something can be worked out for depositing oxidized aluminum!

Now, it's true that "biology can handle laying down oxidized iron crystals"; specifically, the teeth of limpets use monocrystaline whiskers of goethite suspended in a matrix of collagen.

However, sapphires are a bit different from this. Sapphires are made out of corundum with traces of other metals in it, which, according to the Wikipedia section on synthetic versions of it, seems to require very high temperatures - in the thousands of degrees - in order to form; specifically, temperatures that seem incompatible with life as we know it. Moreover, some of those reactions require bases (which sometimes do doubleplusungood things to terrestrial biochemistry) or fluorines. This is different from goethite, for which I cannot find anything similar; goethite apparently forms when other iron-containing minerals are weathered, and does not involve high temperatures, bases, or the like. This is why limpets can use goethite to make teeth.

Now, I have no idea whether or not there's a way to make sapphires in a way that doesn't involve high temperatures, overly-acidic or overly-basic chemicals, or other conditions inimical to life as we know it, which is where you come in.

Is there, in fact, some kind of biological method - capable of existing within Earth-based biology - of producing sapphire, corundum or aluminum oxide bones?

Good answers will either explain how this is undoable or will cite some form of chemical reaction capable of doing this that can exist within a Earthly creature's body without melting them, burning them, corroding them, or otherwise doing something unpleasant.

The BEST answers will cite a chemical reaction that DOES involve high temperatures, basic chemicals, and the like, but said best answers will ALSO provide a way for it to work with Earthly life.

  • $\begingroup$ note sapphire is just Corundum, unless you are purposely adding specific trace impurities. Is there a reason you are calling is sapphire? $\endgroup$
    – John
    Commented Dec 11, 2021 at 4:46
  • $\begingroup$ @John It's what Logan R. Kearsley described it as, so I'm sticking to accuracy; I do know that it's corundum, and mentioned corundum in my answer. Also, it's clickbait. $\endgroup$
    Commented Dec 11, 2021 at 4:50
  • $\begingroup$ Just checking, have you considered nacre? Human femur clocks in at 205MPa vs nacre's 300-500MPa. Tensile strength of nacre isn't that great though so I don't know how well it'll ultimately perform for a whole skeleton. $\endgroup$
    – Lemming
    Commented Dec 11, 2021 at 7:02
  • $\begingroup$ @Lemming I know, but I want crystal bones. Also, these bones are for something VERY large, where every megapascal of compressive strength counts - sure, technically it'll be crystalline fibers of sapphire/corundum suspended in a collagen matrix, but it's what those crystals are made of that counts. $\endgroup$
    Commented Dec 11, 2021 at 7:04
  • $\begingroup$ I think your only option for aluminum processing as far as earth-life goes is via aluminum hydroxide, whose compressive strengths of all of its polymorphs from doyleite through nordstrandite aren't anything to write home about. Incorporation of already-formed sapphires would be your best bet if you want sapphires as part of your bones. $\endgroup$
    – Lemming
    Commented Dec 11, 2021 at 8:40

3 Answers 3


Nothing on a biology based on water will get you aluminium oxide at ambient temperature, all you'll get is aluminium hydroxide or compounds that has the OH- bound to the aluminium atom.

  • Aluminium hydroxide - Std enthalpy of formation (ΔfH298) is −1277 kJ·mol−1 - this one is a quite deep

  • Aluminium oxide - Std enthalpy of formation (ΔfH298) is −1675.7 kJ·mol−1

Aluminium oxide is more stable than the hydroxide, and it's nice that's so, otherwise we wouldn't have aluminium in metallic form - the compact oxide layer that forms on the surface of metallic aluminium protects it from further corrosion.
But it doesn't happen for the hydroxide to spontaneously eliminate the water and fall into a deeper potential energy well - one needs external application of energy to activate the decomposition reaction.

I couldn't find the the activation energy value for the decomposition of aluminium hydroxide to its oxide and water. But, since it has to do with strong heating and elimination the water as vapors, it cannot be lower than the vaporization enthalpy of water - were it to be so, you'd see the "dehydration" of aluminium hydroxide happening under the boiling point of water (see also Hess's law).
So, the activation energy can only be greater than 40.66kJ·mol−1 - the sad reality that the organism trying to get aluminium oxide from its hydroxide having to calcinate (i.e. more than boil) that water away.

The consequence of having aluminium hydroxide that deep into the "potential energy well of formation" means that all the other salts (or ionic compounds of aluminium - like ) will be either:

  • hydrated in the presence of water - thus getting those basic salts of aluminium or aluminates; or
  • decompose into a weak acid and aluminium hydroxide - the case of aluminium alkoxide where the alcohol is forced into a weak acid role - see aluminium triethoxide or diethylaluminum ethoxide or aluminium isopropoxide - because of their reaction to water, those substances will be strongly corrosive (in the GHS05 category); or
  • is insoluble in water - like the AlPO4 (which itself requires quite drastic conditions to prepare - hydrothermal synthesis usually happens at 200c+ and high pressure).
  • $\begingroup$ Given that the whole point of enzymes is to reduce or eliminate activation energy, I'm not convinced that having a huge activation energy for dehydration is, in itself, actually a barrier to biological use. $\endgroup$ Commented Dec 13, 2021 at 17:14
  • 1
    $\begingroup$ @LoganR.Kearsley the enzymes will catalyze reactions with an activation energy at about the energy of an UV photon. Going over this energy, you'd ask the biology to be resistant to oxidative stresses beyond the normal capability of repair. And you are asking the elimination of multiple OH- chemically bonded to an Al atom, in an aqueous medium that is just happy to provide them back. $\endgroup$ Commented Dec 13, 2021 at 19:17
  • $\begingroup$ @LoganR.Kearsley Here pubmed.ncbi.nlm.nih.gov/11833776 "strong interaction of Al3+, the main Al toxic form, with oxygen donor ligands (proteins, nucleic acids, polysaccharides) results in the inhibition of cell division, cell extension, and transport. Although the identification of Al tolerance genes is under way, the mechanism of their expression remains obscure." And the entire "similar articles" is full of "aluminium toxicity for plants" - with the highest degree of adaptation being "tolerance". (ctnd) $\endgroup$ Commented Dec 13, 2021 at 19:31
  • $\begingroup$ (ctnd) Since plants can't move, one would think some would have evolved to use Al if there's be an advantage to use it (after all, many plants use silicon, even if there's no energetic metabolic advantage to use it, i.e. can't use it for food). Yet it doesn't seem to have happened for the case of Aluminium. $\endgroup$ Commented Dec 13, 2021 at 19:33
  • $\begingroup$ Decent points, thanks. $\endgroup$ Commented Dec 13, 2021 at 19:42

To quote Wikipedia:

The Verneuil process allows the production of flawless single-crystal sapphire and ruby gems of much larger size than normally found in nature. It is also possible to grow gem-quality synthetic corundum by flux-growth and hydrothermal synthesis. Because of the simplicity of the methods involved in corundum synthesis, large quantities of these crystals have become available on the market at a fraction of the cost of natural stones.

This tells me that people are no longer interested in doing much research in new methods of forming sapphires synthetically. Like you said aluminium is not used by organisms. So biologists aren't interested in it either. That just leaves speculation.

So far as i understand it, assumes all older models of crystal growth that the crystals grow monomer by monomer (unit by unit). In this theory no biological catalysts have been theorised. At least not for mineral crystals like sapphire. I think this is due to crystals themselves not being bio-available. Which indicates that enzymes would have a hard time catalysing the addition of new monomers to the crystal. Without this catalysis you need the conditions they use in synthetic growth.

This article in science talks about a new theory on how crystals form. Where they think that crystals also grow by incorporating larger particles than just the monomers. But it only came out in 2015 and there has only been limited work done using this newer model.

In conclusion biological synthesis of sapphire is probably not feasible. Though i don't see a problem with salvaging the sapphire from the environment.


You may have better luck looking for information on alumina, as "sapphire" is generally bulk crystalline forms. Here's a paper on alumina nanofiber production by electrospinning aluminum-containing polymer fibers using aluminum isopropoxide as an aluminum-containing organic chemical...but the last step in the process is calcination at 1100 °C.

Even if you might be able to biologically handle compounds like aluminum isopropoxide, biologically producing such compounds from ambient aluminum compounds is a stretch. Calcium and magnesium are commonly present in both soluble and insoluble compounds that take little energy to move it between, but aluminum is mostly present as oxides and aluminosilicates that would be biologically very difficult to work with.

And even if biologically feasible, why would such a system evolve? Bony tissues evolved in large part for storage of calcium and phosphorus, which both have a huge range of biological roles. Aluminum doesn't really have any useful chemical properties that would favor organisms keeping it around, and formation of alumina would be too hard to reverse anyway.

  • 1
    $\begingroup$ It would be best if the biology of that organism was to be totally not based on water - aluminum isopropoxide decomposes in the presence of water into isopropyl alcohol and Al(OH)3 $\endgroup$ Commented Dec 13, 2021 at 1:04
  • $\begingroup$ Some other organometallic might work better, aluminum isopropoxide is just what they used in that paper. It's not inconceivable for it to isolate such compounds to dry and unreactive environments. It'd be very difficult, lossy, and energy-intensive, but not as much so as biologically replicating the calcination step. $\endgroup$ Commented Dec 13, 2021 at 13:25
  • $\begingroup$ "Some other organometallic might work better" - I looked for such, couldn't find any. $\endgroup$ Commented Dec 13, 2021 at 19:18

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .