59KyaniteKyanite: a blade with two kinds of hardnessKyanite often grows as long, flattened blades, sometimes bent or twisted.+
The places listed are examples from historical records. They do not guarantee a gem’s origin or describe current mine output. Treatment, natural or laboratory growth, mixed composition and condition are separate questions. The cautious care advice is not a laboratory cleaning method.
Composition
Al2SiO5
Crystal system
Triclinic
Mohs hardness along the crystal
5.5 Mohs (Along the crystal)
Mohs hardness across the crystal
7 Mohs (Across the crystal)
A blade with two kinds of hardness
Kyanite often forms long, flat blades. Some bend or twist. Blue is its best-known colour, but the species need not be blue. Nor does blue alone identify it. White, green, grey and other colours occur. A key feature is directional hardness. A scratch meets more resistance in one direction than another. This does not mean the stone has two layers. The difference comes from its crystal structure. Perfect cleavage is a separate weak point: the crystal can split along set planes. Resistance to scratches is not resistance to splitting.
Kyanite, andalusite and sillimanite share the formula Al₂SiO₅. Their atoms form different structures. Kyanite often forms when aluminium-rich rocks change under relatively high pressure. This process is called metamorphism. It occurs in schists and gneisses, and in some nearby veins and pegmatites. Examples come from the Austrian Alps, Switzerland, Minas Gerais in Brazil, Kenya, and Virginia and North Carolina in the United States. These places are examples. They are not an origin fingerprint for a finished gem.
Kyanite has more to its colour than one blue swatch. A 1982 gemmological study examined East African material. It linked different blues to combinations of iron, titanium and chromium. It also recorded one colour-change specimen. The evidence concerns those stones. It does not mean all kyanite changes colour. The study also showed why faceting can be hard. Cleavage and uneven directional hardness both matter. A cutter must consider how the crystal behaves at the wheel, as well as how it will look face-up.
In 2009, GIA examined orange faceted kyanites from Loliondo, Tanzania. It also studied fragments from Loliondo and Mautia Hill. Spectroscopy and chemical tests linked the orange colour to trivalent manganese, a form of manganese in the crystal. The report compared these natural samples with earlier work. In the 1970s, researchers had made small orange-yellow kyanite crystals containing manganese. That work shows the mineral can be made in a laboratory. It does not tell us what a particular retailer is selling. Those are different claims and need different evidence.
A photo cannot tell us whether a blue faceted stone is kyanite, sapphire, glass or something else. Good colour does not prove that kyanite is untreated. Ask separately about identity and treatment. For an important purchase, seek a suitable examination. Do not rely on an online scratch test. GIA describes identification as a mix of observations and analytical methods. These can help distinguish treatments and natural from laboratory-grown material.
Choose protected, low-impact wear. Store the stone separately and remove it before rough work. Avoid ultrasonic or steam cleaning. A sound, identified stone may suit a brief, gentle clean with mild soap and lukewarm water. Fractures, possible filling or a fragile setting need individual advice. This care puts cleavage and condition first. It does not rely on the highest hardness number.
Draw a long crystal outline with two arrows at right angles. Label their directions using the property notes above. Put cleavage in a separate box. Which notes concern scratches? Which concern splitting? What do they leave unknown about a finished jewel? Use the drawing to organise questions. It is not a test to perform on a stone.
Typical mineral properties and representative historical occurrences; no proof of individual-stone origin, current mine operation or a pure composition for a mixed ornamental object.
1982 observations concern the investigated East African specimens, including a particular color-change example; not every kyanite or a current market survey.
2009 orange kyanite investigation and discussion of earlier experimental synthesis. Experimental capability is distinct from any retailer-specific origin claim.
General treatment/fracture-aware jewellery care supports conservative guidance only; not a tested cleaning protocol for every listed specimen. Individual condition and the material-specific precautions in this profile still matter.
Blue kyanite blades with quartz, recorded in GeoDIL #940 from Minas Gerais, Brazil.Twyla Baker · CC0 1.0 · Photo record. Original GeoDIL item JPEG bytes retained; no project crop/recolour/reencode/enlargement.Responsive WebP copy · full frame retained
Responsive display copies: proportionally downsampled and encoded as WebP by Goutte de Rosée; full frame retained, with no crop, retouching, recolouring or enlargement. Display orientation follows the original EXIF; any colour profile and attribution metadata are retained. The original-file description above remains applicable to the preserved original.
A blue blade is an observation, not an identification
The blade-like shape can help describe a kyanite specimen. Blue is not its only recorded colour. The Handbook also lists white, grey, green and black. Describe what you see before giving it a mineral name. A photo cannot supply every property needed to identify it.
Kyanite, andalusite and sillimanite have the same composition but different structures. Read about them together, while keeping their names distinct. Kyanite’s hardness depends on direction, and it has cleavage. One scratch-resistance number is therefore not enough to describe it. These facts are not a reason to damage a finished stone with a home test.
The mineral handbook gives different scratch hardness values for different directions in kyanite. It also records clear cleavage. One number would hide that difference. These facts help explain the care advice above. They do not justify scratching or striking a finished stone to identify it.
GIA’s guide to coloured-stone analysis explains how measurements and light work together. Diffused light can show colour zoning. Reflected light can show the surface. An angled fibre-optic light can help reveal inclusions. A microscope image is made using a chosen method. It is more than a larger version of a listing photo.
This is general laboratory context. It does not mean every technique or origin-report service is offered for kyanite. Ask which tests and observations support a particular stone’s identity. Its stated place of origin may rest on other evidence. Keep those claims apart. A polished blue appearance cannot prove origin or treatment.
Read the andalusite profile and the discussion of sillimanite on this page. The mineral-species guide keeps their identities distinct. Start with the names and structures, then compare looks. Shared chemistry does not make names interchangeable. Shared blue colour does not prove the same material. For a specimen or jewel, consider its actual identity, condition and planned use together. A colour photo cannot answer all three.
