Kyanite
Kyanite is Al2SiO5, hardness 5.5 along the blade and 7 across it, density 3.53 to 3.65. Blades reach 0.5 m and are typically bent or twisted, not straight.
Kyanite is aluminium silicate, Al2SiO5, triclinic, with two different hardnesses on one crystal: 5.5 along the blade and 7 across it. Density is 3.53 to 3.65, cleavage is perfect on {100}, and the habit is bladed. Blades reach 0.5 m in quartz veins and are typically bent, and the locality is always a metamorphic one.
In short
- A steel knife scratches kyanite one way and not the other. The Handbook of Mineralogy gives hardness as 5.5 parallel to [001] and 7 parallel to [100]. Scratch along the length of a blade and it marks; scratch across the blade and it does not. No other common collection species behaves like this.
- The old name records exactly that. Kyanite was long called disthene, from the Greek for two strengths, for the directional hardness. The modern name is from the Greek for blue, for the colour — so the two names between them describe the two diagnostic properties.
- It is one of three minerals with the same formula. Kyanite, andalusite and sillimanite are trimorphs of Al2SiO5, and which one forms records the pressure and temperature of metamorphism. Kyanite is the high-pressure form.
- Perfect cleavage on {100} and good on {010}, meeting at 79 degrees. That measured angle is a genuine identification tool on a cleaved fragment, and it is why kyanite blades split lengthwise so readily.
- Blades reach half a metre and are typically bent or twisted, which is normal rather than damage. Fine blue crystals come from the Zillertal in Tirol, Alpe Sponda in Switzerland, Minas Gerais in Brazil, Machakos in Kenya, and Willis Mountain in Virginia.
| Species | Formula | Hardness | Density | How to recognise it |
|---|---|---|---|---|
| Kyanite | Al2SiO5 | 5.5 along, 7 across | 3.53–3.65 | Blue bladed crystals, bent; two hardnesses |
| Andalusite | Al2SiO5 | 6.5–7.5 | 3.13–3.21 | Square prisms; chiastolite shows a dark cross |
| Sillimanite | Al2SiO5 | 6.5–7.5 | 3.23–3.27 | Fibrous to acicular; the fibrolite habit |
| Staurolite | Fe2Al9Si4O23(OH) | 7–7.5 | 3.7–3.8 | Cruciform twins; brown, not blue |
| Corundum | Al2O3 | 9 | 4.0 | Far harder; barrel-shaped hexagonal prisms |
| Glaucophane | Na2(Mg,Fe)3Al2Si8O22(OH)2 | 6 | 3.0–3.2 | Blue amphibole; fibrous, two cleavages at 56° |
The two-hardness test, step by step
This is the only field test on this site that produces two different answers on the same crystal, and it is completely reliable.
Take a bladed crystal and a steel point — a knife blade, a scriber, or the steel of a pocket file, all of which are about hardness 5.5. Work on the flat of the blade, on an inconspicuous area near an edge rather than in the middle of a display face.
First, draw the point along the length of the blade. It will leave a mark. The hardness in that direction is 5.5, which is the same as the steel, and steel of that hardness will scratch it.
Then draw the point across the blade, at right angles to the length. It will not mark. The hardness in that direction is 7, harder than steel and harder than quartz.
Wipe the surface between the two attempts, because steel leaves a grey metallic smear that looks like a scratch and rubs off. If the along-the-blade direction scratches and the across-the-blade direction does not, the specimen is kyanite and the test is finished. Nothing else blue, bladed and common in collections does this. The general sequence for unknowns is on how to identify a mineral specimen.
Why the same formula makes three different minerals
Kyanite, andalusite and sillimanite are trimorphous: identical in composition, Al2SiO5, and different in structure. Which one crystallises depends on the pressure and temperature at the time, and this makes the three of them one of the most-used indicators in metamorphic petrology. Kyanite is the dense, high-pressure form, which is why its density of 3.53 to 3.65 is noticeably higher than andalusite's 3.13 to 3.21.
The Handbook of Mineralogy gives kyanite's occurrence as gneisses, schists, included pegmatites, and quartz veins, from moderately high-pressure regional metamorphism of principally pelitic rocks, with staurolite, andalusite, sillimanite, talc, gedrite, mullite and corundum as associates. In collecting terms that means kyanite comes from continental collision zones — the Alps, the Appalachians, the Brazilian shield — and not from basalt cavities or from oxidised ore.
For a collector this is the useful part: the species carries its own provenance story. A blue blade in a quartz vein from a schist belt is a coherent object; the same blade said to come from a hydrothermal copper deposit is not. Where a stated locality does not fit the geology, it is the locality that should be doubted — which is the argument on specimens without a locality.
Bent blades, cleavage and what counts as damage
The Handbook records kyanite crystals as bladed and tabular on {100}, elongated parallel to [001], reaching 0.5 m, and typically bent or twisted. Lamellar twinning on {100} is common.
A bent kyanite blade is not a damaged kyanite blade. It is the normal growth habit of the species, recording deformation while the rock was still hot. Collectors new to the species sometimes reject curved crystals as distorted; on kyanite, a gentle curve is characteristic and a perfectly straight blade is the less usual object.
What is damage is a clean break. Cleavage is perfect on {100} and good on {010}, and the Handbook gives the angle between the two faces as 79 degrees. Parting occurs on {001}. Fracture is splintery. Together that means a blade will split lengthwise along its flat with very little provocation and will snap across only under real force.
In storage, support the whole blade. A long kyanite laid across two high points in a drawer has its own weight working on a perfect cleavage. Large blades should lie on a flat bed — a cut Plastazote tray rather than a nest of tissue. The general principles are on handling and storage, and the cleavage-against-fracture distinction is worked through on cleavage, parting or fracture.
Colour, and what a good specimen is
The name means blue, and the best material is blue — the Handbook lists colour as blue, white, rarely green, grey, yellow, pink or black, and notes it can be zoned, with weak pleochroism running from colourless through violet-blue to cobalt blue in thick sections.
Colour zoning is the norm rather than a fault. Much kyanite is deeper blue along the core of the blade and paler at the edges, and Brazilian material in particular shows sharp zoning. A uniformly saturated blade is unusual and is what the better specimens are selected for.
What makes a good specimen, in order: colour saturation, then whether the blade is terminated, then whether it stands free of the matrix, then size. Size comes last because the species reaches half a metre and large kyanite is not scarce; a 4 cm sharply terminated deep blue blade standing proud of white quartz is a better object than a 20 cm pale blade lying in schist. Nepalese and Brazilian material set the current standard for colour.
Type material is held at the Mining Academy in Freiberg, Germany, number 22491, and the IMA-CNMNC list gives Austria as the country of first description. Specific silicate material we are looking for is listed on the wanted list.