Is Lonsdaleite Harder Than Diamond?
The Handbook prints lonsdaleite at hardness 3 and diamond at 10. A 2009 calculation said 58% stronger; 2025–26 lab tests found slightly to 50% harder.
Not as an established fact. The Handbook sheet prints lonsdaleite at hardness 3 against diamond’s 10, with no Vickers figure for either. A 2009 calculation gave 58% higher indentation strength. Four 2025–2026 papers on laboratory-made hexagonal diamond report: two, slightly higher hardness than cubic diamond; one, 165 GPa, about 50% harder; one, 155 GPa.
In short
- The sheets. The Handbook of Mineralogy lonsdaleite sheet prints Hardness = 3 and VHN = n.d.; the diamond sheet prints Hardness = 10 and VHN = n.d. Lonsdaleite’s natural material is very fine-grained aggregates to 0.7 mm, mixed with diamond, first found in the residue from dissolving 5 kg of the Canyon Diablo meteorite.
- The calculation. Pan et al. 2009, Physical Review Letters 102, 055503 used first-principles calculations and reported that under an indenter lonsdaleite produces indentation strength 58% higher than the corresponding figure for diamond. That is a simulation, not a measurement.
- The laboratory samples. All four report on laboratory-made hexagonal diamond: Yuan et al. 2025, Science Bulletin 70, 1257, 165 ± 4 GPa on (100) planes, about 50% harder than single-crystal cubic diamond; Chen et al. 2025, Nature Materials 24, 513, 155 GPa; Yang et al. 2025, Nature 644, 370, only slightly higher than cubic diamond; Lai et al. 2026, Nature 651, 621, slightly higher than cubic diamond.
- Documented failure: the natural grain. Németh et al. 2014, Nature Communications 5, 5447 concluded that lonsdaleite is faulted and twinned cubic diamond and does not exist as a discrete material, and Németh et al. 2023, Philosophical Transactions of the Royal Society A 381, 20220344 that the Canyon Diablo grains are a nanocomposite. None of the four hardness reports concerns natural lonsdaleite.
- The three-question rule. Before believing “harder than diamond”, ask: measured or calculated; natural grain or laboratory block; and against which diamond, on which plane. On the sources here, laboratory hexagonal diamond is reported at least slightly harder in all three abstracts that compare; natural lonsdaleite is unmeasured.
| Source | Material | Method | Figure as printed | Comparison with diamond, as printed |
|---|---|---|---|---|
| Handbook lonsdaleite sheet | Natural, Canyon Diablo [TL] and other meteorites and impact sites | Reference sheet | Hardness = 3; VHN = n.d. | None stated |
| Handbook diamond sheet | Natural diamond | Reference sheet | Hardness = 10; VHN = n.d. | — |
| Pan et al. 2009, Phys. Rev. Lett. | Lonsdaleite, modelled | Calculation (first-principles) | Indentation strength | 58% higher than diamond |
| Yuan et al. 2025, Sci. Bull. | Laboratory hexagonal diamond, mm-sized, from graphite | Experiment on the synthesized sample | 165 ± 4 GPa on (100) planes | About 50% harder than single-crystal cubic diamond |
| Chen et al. 2025, Nat. Mater. | Laboratory hexagonal diamond, mm-sized block | Experiment on the synthesized block | 155 GPa | None stated in the abstract |
| Yang et al. 2025, Nature | Laboratory bulk hexagonal diamond, 100 µm to mm-sized | Synthesis and extensive characterization; hardness method not stated in the abstract | Not given in the abstract | Only slightly higher than cubic diamond |
| Lai et al. 2026, Nature | Laboratory phase-pure hexagonal diamond, mm-sized | Characterization plus theoretical simulation; the abstract does not say which gave the hardness | Not given in the abstract | Slightly higher than cubic diamond |
| Németh et al. 2014, Nat. Commun. | Samples attributed to lonsdaleite | Electron microscopy | No hardness | Calls it faulted and twinned cubic diamond |
| Németh et al. 2023, Phil. Trans. R. Soc. A | Canyon Diablo grains | X-ray diffraction and microscopy | No hardness | Nanocomposite of stacking-disordered diamond and diaphite |
What hardness does the Handbook give lonsdaleite?
The Handbook of Mineralogy lonsdaleite sheet prints Hardness = 3, with VHN = n.d., and this page quotes it exactly as printed. The rendered sheet was checked: the figure is 3. The diamond sheet prints Hardness = 10, also with VHN = n.d., so neither sheet carries a measured Vickers number. Read side by side, the two sheets put lonsdaleite seven Mohs steps below diamond. None of the five hardness papers on this page reports lonsdaleite or hexagonal diamond as softer than diamond, so this page does not use the 3 to rank them.
What else the sheet records. Lonsdaleite is hexagonal, space group P63/mmc, a = 2.51 and c = 4.12 Å; grayish in crystals, pale yellowish or brown in broken fragments, adamantine, n slightly above 2.404, density over 3.20 measured and 3.51 calculated, against diamond’s 3.511 and 3.515. It was discovered in about 200 mg of residue from dissolving 5 kg of the Canyon Diablo meteorite, the type locality; the sheet also names the Goalpara and Allan Hills 77283 meteorites, placers in northern Sakha, the Tunguska site, the Ries crater and the Sudbury structure. The sheet says it may be formed by impact shock, or be a product of direct crystallization in the parent body.
What the IMA list records. Lonsdaleite, C, status A, 1966-044, USA, first reference Nature 214 (1967), 587, second reference Physics and Chemistry of Minerals 52 (2025), 30. Its place among the carbon rows, and why natural diamond is a mineral, is on is diamond a mineral; graphite has the soft end of the carbon family.
Was lonsdaleite calculated or measured to be harder than diamond?
The famous figure is a calculation; the experimental reports came later, on laboratory material, and they disagree about how much harder. Pan et al. 2009, Physical Review Letters 102, 055503 showed by first-principles calculations that large compressive pressures under an indenter can drive a bond-flipping transformation, and reported that the same mechanism produces indentation strength in lonsdaleite (hexagonal diamond) 58% higher than the corresponding figure for diamond. Its title opens “Harder than diamond”. No sample was measured.
Four laboratory reports, 2025–2026. Yuan et al. 2025, Science Bulletin 70, 1257 made millimetre-sized, nearly pure hexagonal diamond from graphite and reported an ultrahard hardness of 165 ± 4 GPa on (100) planes, about 50% harder than single-crystal cubic diamond. Chen et al. 2025, Nature Materials 24, 513 reported a millimetre-sized, highly oriented block with a very high hardness of 155 GPa. Yang et al. 2025, Nature 644, 370 reported bulk hexagonal diamond whose hardness is only slightly higher than cubic diamond. Lai et al. 2026, Nature 651, 621 reported phase-pure material with a slightly higher hardness than cubic diamond.
Reading them together. Of the 4 laboratory abstracts, 3 compare with cubic diamond and all 3 say harder; 2 of those 3 say only slightly. The 165 and 155 GPa figures differ by 10 GPa. This page does not convert GPa to Mohs, and does not average the papers. How scratch hardness differs from indentation hardness is on how to test mineral hardness properly.
Is lonsdaleite a real mineral? The natural-grain problem
Documented failure: the natural grain. Every sample with a reported hardness above was made in a laboratory, and the natural material has been argued not to be a discrete phase at all. Németh et al. 2014, Nature Communications 5, 5447 reported that samples displaying features attributed to lonsdaleite consist of cubic diamond dominated by extensive {113} twins and {111} stacking faults, and titled the paper “Lonsdaleite is faulted and twinned cubic diamond and does not exist as a discrete material”. Németh et al. 2023, Philosophical Transactions of the Royal Society A 381, 20220344 reported that the Canyon Diablo diamond-like grains are a nanocomposite dominated by cubic/hexagonal stacking-disordered diamond and diaphite domains.
The other side, attributed. The September 2026 IMA list still carries lonsdaleite as an approved species, A, 1966-044. The 2025 and 2026 synthesis papers report nearly pure or phase-pure hexagonal diamond; Lai et al. say their findings resolve the long-standing controversy on the existence of hexagonal diamond as a discrete carbon phase. Those are synthetic samples. This page does not decide whether the Canyon Diablo grains are lonsdaleite in that sense.
The names. Pan et al. write “lonsdaleite (hexagonal diamond)”; Németh et al. 2014 write lonsdaleite, also called hexagonal diamond; Németh et al. 2023 say that since 1967 the name lonsdaleite was used interchangeably with hexagonal diamond. The 2025–2026 papers say hexagonal diamond throughout. Meteorite questions start on how do I know if my rock is a meteorite.
How to read a harder-than-diamond claim: the three-question rule
The three-question rule: before accepting “harder than diamond”, ask whether the figure was measured or calculated, on a natural grain or a laboratory block, and against which diamond. Applied to the sources here: the 58% figure is calculated; the 165 GPa and 155 GPa figures are measured on laboratory blocks; the 50% comparison is against single-crystal cubic diamond on the (100) planes of the hexagonal sample; and the two “slightly higher” results are against cubic diamond without a figure in the abstract.
The verdict, qualified. Laboratory-made hexagonal diamond is reported at least slightly harder than cubic diamond in every abstract that makes the comparison, and Yuan et al. give a figure with an uncertainty, 165 ± 4 GPa. Natural lonsdaleite from meteorites has no measured hardness in any source here, and its own sheet prints 3. So “lonsdaleite is harder than diamond” is a fair summary of the 2025–2026 laboratory work and not a statement about a meteorite grain. Diamond simulants are a separate question, on is zircon a diamond.
No lonsdaleite specimen is discussed or offered here; this is a reading of published work. Other guides are under the collecting guides. Nothing on this site is offered for sale; the wanted list is the only commercial route.