Is kyanite a feldspar?
No. Kyanite, Al2SiO5, is an island silicate with no alkali and 62.92% alumina; feldspars are framework silicates. A 20 g kyanite reads 14.33 g or more in water.
No. Kyanite is aluminium silicate, Al2OSiO4 on the IMA list, an island silicate with isolated SiO4 groups; feldspars are framework silicates of sodium, potassium or calcium. Kyanite carries no alkali: the sheet’s analysis has 0.02% CaO, Na2O and K2O and 63.28% alumina. A 20.00 g piece reads 14.33 g or more in water; feldspars read 12.75 g or less.
In short
- The verdict. The Tulane University mineralogy notes (Nelson), silicate structures define nesosilicates as minerals of isolated SiO4 tetrahedra and name the Al2SiO5 minerals among them; the tectosilicate notes say the feldspars have a framework of SiO4 tetrahedra sharing all their corner oxygens. Kyanite is Al2OSiO4 on the September 2026 IMA-CNMNC Master List, trimorphous with andalusite and sillimanite on its Handbook of Mineralogy sheet.
- The alkali–alumina check. Every feldspar end member carries a large cation: albite 11.82% Na2O, orthoclase 16.92% K2O, anorthite 20.16% CaO, worked from the IMA formulas. Alumina tops out at anorthite’s 36.65%. The kyanite sheet’s Elovyi, Karelia analysis has 0.02% CaO, Na2O and K2O together and 63.28% Al2O3; the ideal is 62.92%. No alkali and nearly two-thirds alumina is not a feldspar.
- The reference shelf agrees. A script over the 13 sheets named here finds every one of the 10 feldspar sheets citing Deer, Howie and Zussman, Rock-forming minerals, volume 4, framework silicates, and the kyanite, andalusite and sillimanite sheets citing volume 1A, orthosilicates.
- Choose the test by the specimen. With an analysis, the alkali–alumina check. A clean blade, the balance: 14.33 to 14.52 g in water for 20.00 g against the feldspars’ 12.13 to 12.75 g. A cut stone, the refractometer: kyanite 1.710 to 1.734, anorthite, the highest feldspar sheet, 1.573 to 1.590. Hardness chooses nothing.
- Where the check fails, documented. It works one way only. Nepheline and petalite carry alkalis, sit on the same volume 4 shelf, and are not feldspars. A kyanite blade on quartz weighs as a mixture: quartz reads 12.45 g, inside the feldspar band.
| Test | Kyanite | Feldspar (10 Handbook sheets) | Does it choose? | Wrong for (documented failure) |
|---|---|---|---|---|
| Alkali and calcium oxides | 0.02% (Elovyi analysis); 0% in Al2SiO5 | End members: Na2O 11.82%, K2O 16.92%, CaO 20.16% | Yes, with an analysis | Nepheline and petalite carry alkalis and are not feldspars |
| Alumina, Al2O3 | 62.92% ideal; 63.28% (Elovyi) | 18.32% (K) to 36.65% (anorthite) | Yes, with an analysis | Andalusite and sillimanite are 62.92% too |
| Structure class | Nesosilicate: isolated SiO4 (Tulane); volume 1A, orthosilicates (sheet) | Framework, all corners shared (Tulane); volume 4, framework silicates (sheets) | Yes, from the literature | Nepheline and petalite are on volume 4 too |
| 20.00 g piece in water | 14.33 to 14.52 g (D 3.53 to 3.65) | 12.13 g (microcline) to 12.75 g (anorthite) | Yes, on a clean single-mineral piece | Kyanite on quartz (12.45 g) or in schist matrix |
| Refractive index | α 1.710 to 1.718, γ 1.724 to 1.734 | Highest sheet: anorthite α 1.573, γ 1.590 | Yes, on a polished face | Instrument test; needs a flat face |
| Hardness | 5.5 ∥ [001], 7 ∥ [100] | 6 to 6.5 (sanidine, anorthoclase 6) | No | Every specimen: one direction is softer, the other harder |
| Cleavage | Perfect {100}, good {010}, at 79°; parting {001} | Two cleavages near 90° where an angle is printed | Partly, on a clean cleaved fragment | Matrix pieces and bent blades |
| Colour | Blue, white, rarely green, grey, yellow, pink, black | Microcline may be blue; albite bluish | No | Blue specimens of either |
Is kyanite a feldspar? Choose by chemistry, not by look
Kyanite is not a feldspar: it has no alkali and it is not a framework silicate. The September 2026 IMA list gives kyanite as Al2OSiO4, A, 1967 s.p., Austria, and the Handbook of Mineralogy kyanite sheet as Al2SiO5, trimorphous with andalusite and sillimanite. The IMA way of writing it, Al2O(SiO4), keeps one oxygen outside the silicate group. The feldspar rows read albite Na(AlSi3O8), anorthite Ca(Al2Si2O8) and microcline, orthoclase and sanidine K(AlSi3O8): a large cation beside a group of four tetrahedral cations and eight oxygens.
The structure, sourced. The Tulane notes say nesosilicates, or island silicates, are built of SiO4 tetrahedra that share no corner oxygens with each other, and name olivines, garnets, the Al2SiO5 minerals and staurolite among the common ones. Their tectosilicate notes give the feldspars three end members, KAlSi3O8, NaAlSi3O8 and CaAl2Si2O8, and a framework structure of SiO4 tetrahedra sharing all of the corner oxygens, with a Si to O ratio of 1:2. A whole-line scan of the IMA list finds 0 lines containing feldspar: it lists species, not structure classes.
The reference shelf. Each Handbook sheet cites the standard reference work, Deer, Howie and Zussman, Rock-forming minerals. By script over the 13 sheets named on this page, all 10 feldspar sheets cite volume 4, framework silicates, and the kyanite, andalusite and sillimanite sheets cite volume 1A (2nd edition), orthosilicates. The three Al2SiO5 species are on kyanite, andalusite and sillimanite; the weight rule that separates them from each other is on the sillimanite page.
The alkali–alumina check, worked
The check: a feldspar carries several per cent of Na2O, K2O or CaO and no more than about 37% Al2O3; kyanite carries almost none and about 63%. On standard atomic weights the IMA end members give albite, NaAlSi3O8, 11.82% Na2O and 19.44% Al2O3; orthoclase, microcline and sanidine, KAlSi3O8, 16.92% K2O and 18.32% Al2O3; anorthite, CaAl2Si2O8, 20.16% CaO and 36.65% Al2O3. Kyanite, Al2SiO5, is 62.92% Al2O3 and 37.08% SiO2 with nothing else, the ideal column the sheet prints.
Against a real analysis. The sheet’s Elovyi, Karelia microprobe analysis reads SiO2 36.68, Al2O3 63.28, FeO 0.37, Cr2O3 0.06, MgO 0.02, CaO 0.01, Na2O 0.00 and K2O 0.01. Calcium, sodium and potassium oxides together are 0.02%, less than a five-hundredth of albite’s 11.82% Na2O. The alumina is 26.63 points above anorthite’s. Worked, any mixture of the three end members carries at least 11.82% of the three oxides and at most 36.65% Al2O3, so the gap holds across the series; the plagioclase arithmetic is on is labradorite a feldspar.
Where it fails. The check is one-way. No alkali rules out feldspar, but alkali does not prove it: nepheline and petalite carry sodium, potassium or lithium and are not feldspars, and the tests for those are on is nepheline a feldspar and is petalite a feldspar. And a specimen needs an analysis before the check can be used.
Can you tell kyanite from feldspar without an analysis?
Yes, on a clean piece, by weight. Worked as 20 − 20/D, a 20.00 g kyanite reads 14.33 to 14.52 g in water on its sheet’s 3.53 to 3.65. Across the 10 feldspar sheets the range is 12.13 g (microcline, 2.54) to 12.75 g (anorthite, 2.76). The gap is at least 1.58 g, wide enough for a balance reading to 0.01 g; the method is on how to measure specific gravity at home.
The refractometer agrees. Kyanite’s indices run α 1.710 to 1.718 and γ 1.724 to 1.734. The highest feldspar sheet, anorthite, stops at γ 1.590, 0.120 below kyanite’s lowest index. That is an instrument test on a polished face, not a field one.
Cleavage helps on a clean fragment. Kyanite cleaves perfectly on {100} and well on {010}, meeting at 79°, with a parting on {001}, read from the rendered sheet. The feldspar sheets that print an angle give about 90°, 90° and about 94°. The cleavage-angle survey across many species is on how to identify feldspar and is not repeated here.
Why hardness and colour do not choose
Kyanite’s hardness depends on direction. The sheet gives 5.5 parallel to [001], along the blade, and 7 parallel to [100]; the feldspar sheets give 6 to 6.5, or 6 for sanidine and anorthoclase. A scratch along a kyanite blade marks it more easily than feldspar, and one across it less easily, so a single scratch can point either way. The two-hardness test is the subject of the kyanite page; method on how to test mineral hardness properly.
Colour overlaps. Kyanite’s sheet gives blue, white, rarely green, grey, yellow, pink and black. Microcline’s gives red, green and blue among its colours, and albite’s bluish and greenish; blue-green microcline sold as amazonite is on is amazonite a feldspar. A blue blade is consistent with kyanite, not proof of it.
Who each test is wrong for
Each test fits some specimens and fails others. The alkali–alumina check is decisive with an analysis and useless without one. The balance is right for a clean, loose blade and wrong for kyanite on quartz or in schist, because quartz reads 12.45 g, inside the feldspar band. The refractometer needs a polished face. Cleavage needs a clean fragment. Hardness and colour are wrong for every specimen.
The same structure question for kyanite’s companion. Staurolite, first on kyanite’s association line, is an island silicate too, and is often asked about as a garnet: see is staurolite a garnet. The feldspars themselves are on microcline, orthoclase and the collecting guides. We hold no catalogue and nothing here is offered for sale; the wanted list is the only commercial route.