Cordierite
Cordierite, Mg2Al3(Si5AlO18), is the mineral cut as iolite: orthorhombic, hardness 7 to 7.5, density 2.60 to 2.66, with yellow, violet and blue pleochroism.
Cordierite is a magnesium aluminium silicate, Mg2Al3(Si5AlO18) on the IMA list: orthorhombic, hardness 7 to 7.5, density 2.60 to 2.66, blue to violet and strongly pleochroic. Cut as a gem it is called iolite, a trade name GIA uses for the same mineral. Sekaninaite is its iron analogue; indialite has the same formula but is hexagonal.
In short
- One species, one gem name. The September 2026 IMA-CNMNC Master List lists cordierite as Mg2Al3(Si5AlO18), status G, 1813, with Germany queried; the Handbook of Mineralogy sheet writes (Mg,Fe2+)2Al4Si5O18. GIA says cordierite is better known by its trade name, iolite. The full name question is on is cordierite the same as iolite.
- The two-by-two. By script, four IMA rows have the formula X2Al3(Si5AlO18): cordierite (Mg) and sekaninaite (Fe2+), both orthorhombic on their sheets, and indialite (Mg) and ferroindialite (Fe2+, approved 2013), both hexagonal. Magnesium or iron, and orthorhombic or hexagonal, are the two questions that name the specimen.
- The iron share, worked. From the sheets’ formula lines, Fe2+/(Fe2+ + Mg) is 4.0% and 9.0% in the two cordierite analyses, 85.3% in the Dolní Bory sekaninaite and 58.9% in the Bellerberg ferroindialite. The indialite sheet’s Unazuki analysis is 54.9%, iron over magnesium; that sheet is dated 2001, before ferroindialite was approved.
- Three colours. The sheet’s pleochroism is X pale yellow or green, Y violet or blue-violet, Z pale blue, with absorption Z > Y > X. GIA says bluish iolites show colourless to yellow, blue-grey and dark violet, and that from some angles a bluish iolite can look completely colourless or yellow.
- The scale does not separate it from quartz. A 20.00 g piece reads 12.31 to 12.48 g in water as cordierite, and quartz reads 12.45 g, inside the window. Sekaninaite reads 12.75 to 12.78 g, only 0.27 g above. Pleochroism and the refractive index do the work.
| Species, IMA formula | IMA status | Crystal system | Hardness | Density; 20.00 g piece in water | Indices | Pleochroism | Fe share in the sheet analyses |
|---|---|---|---|---|---|---|---|
| Cordierite, Mg2Al3(Si5AlO18) | G, 1813, Germany ? | Orthorhombic | 7 to 7.5 | 2.60 to 2.66; 12.31 to 12.48 g | α 1.527–1.560; γ 1.537–1.578 | X pale yellow, green; Y violet, blue-violet; Z pale blue | 4.0% (White Well); 9.0% (Smith Ridge) |
| Sekaninaite, Fe2+2Al3(Si5AlO18) | A, 1967-047, Czech Republic | Orthorhombic | 7 to 7.5 | 2.76 to 2.77; 12.75 to 12.78 g | α 1.561; γ 1.576 | X colourless; Y blue; Z pale blue | 85.3% (Dolní Bory) |
| Indialite, Mg2Al3(Si5AlO18) | G, 1954, India | Hexagonal | 7 to 7.5 | 2.512 (synthetic); 12.04 g | ε 1.532–1.535; ω 1.537–1.539 | None printed | 32.0% (Bokaro); 54.9% (Unazuki) |
| Ferroindialite, Fe2+2Al3(Si5AlO18) | A, 2013-016, Germany | Hexagonal | 7 | 2.66; 12.48 g | α 1.539; γ 1.554, anomalous | Weak: X colourless, Z pale violet | 58.9% (Bellerberg) |
| Quartz, SiO2 (for the scale) | A, 1967 s.p. | Hexagonal, point group 32 | 7 | 2.65; 12.45 g | Not compared here | Not compared here | No iron in the formula |
What is cordierite? A magnesium aluminium silicate
The names. The September 2026 IMA-CNMNC Master List gives cordierite as Mg2Al3(Si5AlO18), status G, with the year 1813 and the country printed as Germany followed by a question mark. The Handbook sheet writes (Mg,Fe2+)2Al4Si5O18 and says the name honours Pierre Louis A. Cordier (1777–1861), the French mining engineer and geologist who first studied the species. Both total Al4Si5O18; the IMA form writes the magnesium end and groups one aluminium with the silicon, and the Handbook form allows iron beside the magnesium.
The composition, worked. For the magnesium end, Mg2Al4Si5O18, the formula mass is 584.94 and the mineral is 13.78% MgO, 34.86% Al2O3 and 51.36% SiO2. The sheet’s White Well, Western Australia analysis gives 12.8 MgO, 33.5 Al2O3 and 50.2 SiO2, with 0.84 FeO. Both of its analyses also print H2O+ of 1.69 and 1.71, which the ideal formula does not carry; one of the sheet’s references is a 1977 study of low hydrous cordierite.
What it looks like, per the sheet. Orthorhombic and pseudohexagonal; short prismatic crystals striated parallel to [001], to 18 cm, but typically granular to compact and massive. Twinning is common on {110} and {130}, simple, lamellar or cyclical. Cleavage is fair on {100}, poor on {001} and {010}; fracture subconchoidal; brittle. Hardness 7 to 7.5; density 2.60 to 2.66 measured, 2.505 calculated, as printed. Colour blue, smoky blue, bluish violet, also greenish, yellowish brown or grey; vitreous.
Cordierite, sekaninaite, indialite: the two-by-two
The two-by-two: the IMA list has four rows with the formula X2Al3(Si5AlO18), and two questions sort them: is magnesium or iron dominant, and is the crystal orthorhombic or hexagonal? Magnesium and orthorhombic is cordierite; iron and orthorhombic, sekaninaite (A, 1967-047, Czech Republic); magnesium and hexagonal, indialite (G, 1954, India); iron and hexagonal, ferroindialite (A, 2013-016, Germany). Method: the list as text, every row whose formula contains (Si5AlO18); the script finds six, and the other two, adachiite and alumino-oxy-rossmanite, have longer formulas that also carry (BO3)3.
The iron share, worked from the sheets. As Fe2+/(Fe2+ + Mg) from each formula line: cordierite, White Well 4.0% and Smith Ridge, Idaho 9.0%; sekaninaite, Dolní Bory, Czech Republic 85.3%; ferroindialite, Bellerberg, Eifel 58.9%; indialite, Bokaro coalfield, India 32.0%, and Unazuki, Japan 54.9%. The Unazuki analysis has more iron than magnesium, Fe 1.07 against Mg 0.88. The indialite sheet is dated 2001 and ferroindialite was approved in 2013, so the sheet could not have used the newer name; we do not rename the Unazuki material.
What the sheets say about the boundaries. The cordierite sheet says it is dimorphous with indialite and forms a series with sekaninaite; the sekaninaite sheet says it forms a series with cordierite, and its colour is blue to blue-violet. The ferroindialite sheet puts it in the beryl group. No source read for this page states where along the iron share one name gives way to the other, so this page gives the measured shares and not a cut-off.
Why does cordierite change colour when turned? Pleochroism
Cordierite shows a different colour along each of its three optical directions. The Handbook sheet gives X pale yellow or green, Y violet or blue-violet, and Z pale blue, with absorption Z > Y > X and orientation X = c, Y = a, Z = b. GIA’s iolite page says that when you turn iolite you see three distinct colours in three crystal directions, and that strongly pleochroic iolite has been incorrectly called water sapphire.
GIA’s descriptions on its iolite description page: iolites that appear violet show light violet, dark violet and yellow-brown; bluish iolites show colourless to yellow, blue-grey and dark violet, and from some angles a bluish iolite can look completely colourless or yellow. GIA attributes the violet-to-blue hue to iron, and says the deep blues of some iolites are thought to share sapphire’s cause, an iron-titanium charge transfer.
The relatives differ. The sekaninaite sheet gives X colourless, Y blue, Z pale blue, with no yellow; the ferroindialite sheet gives weak pleochroism, X colourless, Z pale violet; the indialite sheet prints none. A yellow direction is therefore a cordierite pointer among the four, on their sheets. Iolite is compared with benitoite and sapphire on benitoite.
Weighing cordierite: where the scale fails
A 20.00 g piece of cordierite reads 12.31 to 12.48 g in water, and quartz reads 12.45 g, inside that window. Worked as 20 − 20/D from the sheet densities, 2.60 to 2.66, and quartz’s 2.65, which is the figure the live quartz page uses. The scale cannot separate the two; the method is on how to measure specific gravity at home.
Against its own relatives. Sekaninaite, at 2.76 to 2.77, reads 12.75 to 12.78 g, 0.27 g above cordierite’s top. Ferroindialite, at 2.66, reads 12.48 g, the same as cordierite’s top. Indialite’s only measured density on its sheet is for synthetic material, 2.512, which reads 12.04 g. GIA gives iolite a specific gravity of 2.61, 12.34 g.
What does separate them. On the sheets, cordierite’s indices run from 1.527 to 1.578; and cordierite’s three-colour pleochroism includes the yellow direction its iron-rich relatives lack. Birefringence, worked from the matched ends of the sheet ranges, is 0.010 to 0.018, the same figures the benitoite page prints.
Where cordierite is found: Bodenmais, Orijärvi, Madagascar and Sri Lanka
The Occurrence field: in thermally metamorphosed argillaceous sediments and high-grade regionally metamorphosed schists, gneisses and granulites; in mafic igneous rocks and granites; detrital. Its associates are sillimanite, potassic feldspar, muscovite, biotite, corundum, spinel, garnet and andalusite; the almandine sheet lists cordierite in turn.
The Distribution field calls it widely distributed and names localities for abundant material: Bodenmais, Bavaria, Germany; Orijärvi and Leppävirta, Finland; Kragerö, near Tvedestrand, and Akland, Söndeled, near Risör, Norway, as the sheet spells them; the Mt Bity district, south of Antsirabe, Madagascar; the Treasure Casket claims, Fungure Reserve, Zimbabwe; around Namib, Namibia; crystals from the gem gravels of Sri Lanka; around Tiruchchirappalli and Coimbatore, Tamil Nadu, India; the Harts Range, Northern Territory, Australia; Richmond, New Hampshire and Haddam, Connecticut; and fine crystals from Thompson, Manitoba.
No British or Irish locality is on the cordierite sheet, by script. The sekaninaite sheet names Brockley, Rathlin Island, Co. Antrim, which it lists under Ireland. GIA adds Kenya, central Tanzania, India, Brazil and Norway as iolite sources, and a Madagascar deposit discovered in 1994. The rest of the reference section is at the species index. We hold no catalogue and nothing here is offered for sale; the wanted list is the only commercial route.