Is andradite a garnet?
Yes: andradite is one of 14 species in the IMA garnet group, hardness 6.5 to 7. A three-site check on a real specimen’s analysis, and when the name changes.
Yes. Andradite, Ca3Fe3+2(SiO4)3, is one of 14 species in the garnet group of the IMA-approved garnet supergroup (Grew et al. 2013): cubic, hardness 6.5 to 7, refractive index 1.887. Worked on the Handbook’s Serbian analysis, calcium holds 97.7% of the X site, ferric iron 99.0% of Y and silicon 98.3% of Z.
In short
- The definition. Grew et al., American Mineralogist 98 (2013) define the garnet supergroup as all minerals isostructural with garnet: 32 approved species with the general formula {X3}[Y2](Z3)φ12, X, Y and Z being dodecahedral, octahedral and tetrahedral sites. Groups are split by total charge at Z; the garnet group, charge 12, holds 14 species, andradite among them. GIA lists andradite in its garnet group facts.
- The three-site check. Grew et al. name a garnet by finding the dominant valence, then the dominant cation, at each site. The Handbook of Mineralogy sheet’s Reškovic stream, Serbia analysis prints (Ca3.01Mg0.07)(Fe3+1.94Al0.02)(Si2.95Al0.05)O12: worked, calcium is 97.7% of X, ferric iron 99.0% of Y and silicon 98.3% of Z, a Z charge of 11.95. Each site names andradite.
- When the answer changes. With aluminium dominant at Y the name is grossular, with chromium uvarovite, with Mn2+ dominant at X calderite, and with the schorlomite component dominant schorlomite. All four have IMA rows and all are in the garnet supergroup, so the answer to the question stays yes; only the species changes.
- Documented failure: the optics. The andradite sheet says isotropic, typically weakly anisotropic. GIA gives garnet birefringence none; Grew et al. report birefringent grossular–andradite garnets at up to 0.015, normally biaxial. And two garnets, henritermierite and holtstamite, are tetragonal. Neither cubic shape nor a dark polariscope defines a garnet; the structure does.
- Is andradite magnetic? USGS Open-File Report 99-529 pulled 9 andradite samples on a laboratory separator at 0.10 to 0.80 amp, best 0.30 to 0.50: the range the report calls paramagnetic, against magnetite’s best of 0.01. The sheet prints no magnetic property, and neither source tests a hand magnet.
| Site (Grew et al.) | Andradite end-member | Reškovic analysis, worked share | The name changes when | It becomes (IMA row) | Still in the garnet supergroup? |
|---|---|---|---|---|---|
| X, dodecahedral | Ca3 | Ca 3.01, Mg 0.07: 97.7% Ca | Mn2+ dominates X | Calderite, Mn2+3Fe3+2(SiO4)3: G, 1909 | Yes: garnet group |
| Y, octahedral | Fe3+2 | Fe3+ 1.94, Al 0.02: 99.0% Fe3+ | Al dominates the trivalent cations | Grossular, Ca3Al2(SiO4)3: A, 1962 s.p. | Yes: garnet group |
| Y, octahedral | Fe3+2 | No Cr reported | Cr dominates | Uvarovite, Ca3Cr2(SiO4)3: A, 1967 s.p. | Yes: garnet group |
| Y and Z together | Fe3+ at Y, Si at Z | TiO2 trace | The schorlomite component, Ti at Y with Fe3+ at Z, is the most abundant | Schorlomite, Ca3Ti2(SiFe3+2)O12: G, 1846 | Yes: schorlomite group, Z charge 10 |
| Z, tetrahedral | Si3, charge 12 | Si 2.95, Al 0.05: 98.3% Si; charge 11.95 | Vacancies with OH outnumber Si | Fe3+ analogue of katoite: reported to 35 mol% in andradite, not yet dominant | No species exists |
Is andradite a garnet? What the IMA garnet report says
Andradite is a garnet: a species of the garnet group in the IMA-approved garnet supergroup. Grew et al., American Mineralogist 98 (2013) say the supergroup includes all minerals isostructural with garnet, regardless of what elements occupy the four atomic sites, and count 32 approved species. Twenty-nine of them fall into five groups by total charge at the Z site: 8, 9, 10, 12 and 15. Their Table 1 lists 14 in the garnet group, charge 12, among them andradite, {Ca3}[Fe3+2](Si3)O12. The live how to tell if garnet is real page uses the same count.
What is not a garnet. The IMA list has rows for species, not for garnet itself: a case-insensitive scan finds 0 lines containing garnet. Grew et al. exclude look-alikes on structure: griphite is not isostructural with garnet, wadalite is too distant, and the tetragonal almandine–pyrope phase has garnet stoichiometry but not its structure. Vesuvianite, which grows with andradite, is not a garnet either: see is vesuvianite a garnet.
The species page, andradite, carries the sheet data, the variety names and the titanium boundary with schorlomite; those are not repeated here.
The three-site check, worked on a real andradite
The three-site check: a garnet is named by the dominant valence and then the dominant cation at X, Y and Z, and on the Handbook’s analysed andradite every site points the same way. Grew et al.’s procedure is to calculate a formula on 12 anions and 8 cations, allocate the cations to sites, then identify the dominant valence at each site and the dominant cation for that valence, which names both group and species.
Worked. The sheet’s Reškovic stream, Serbia analysis gives 34.91% SiO2, 0.69% Al2O3, 30.40% Fe2O3, 0.58% MgO and 33.20% CaO. Recalculated on 12 oxygens, with water and trace titanium left out, that is Si 2.954, Al 0.069, Fe3+ 1.936, Mg 0.073 and Ca 3.011, matching the printed (Ca3.01Mg0.07)(Fe3+1.94Al0.02)(Si2.95Al0.05)O12. At X calcium is 97.7%; at Y ferric iron is 99.0%; at Z silicon is 98.3%, with 0.05 aluminium making a Z charge of 11.95 against the garnet group’s 12.
What the check needs. An analysis with ferric and ferrous iron resolved, or a recalculation that assumes them; the sheet’s formula treats all iron as Fe3+. Without an analysis the honest label is the species the locality and properties support, written as such; labelling practice is on how to label and catalogue a mineral collection.
When does andradite stop being andradite?
Andradite changes species, never supergroup, when a different cation takes over a site. At Y, aluminium dominant makes it grossular, and the sheet names that series; chromium dominant makes it uvarovite. At X, Mn2+ dominant makes it calderite, {Mn2+3}[Fe3+2](Si3)O12 in Grew et al.’s Table 1; the calderite sheet gives n = 1.875 to 1.934 and density 4.05. Titanium at Y balanced by ferric iron at Z leads to schorlomite, in the schorlomite group; the worked 15 percent line is on the andradite species page.
A boundary not yet crossed. Grew et al.’s Table 6 lists components reported but not yet found to be dominant in natural garnet. The Fe3+ analogue of katoite, Ca3Fe3+2(OH)12, is there, at up to 35 mol% in andradite. Andradite rich in hydroxyl is still andradite, and the IMA list has no species for the hydrous end-member, by script.
Mixed stones. GIA says many garnets are chemical mixtures of two or more garnet species. The grossular–andradite blend sold as Mali garnet is discussed on the grossular page; green grossular is on is tsavorite a garnet and orange on is hessonite a garnet.
Is andradite isotropic? Why optics cannot settle it
A documented failure: the optical shortcut for a garnet, one refractive index and no birefringence, does not hold for andradite. The andradite sheet prints isotropic, typically weakly anisotropic, n = 1.887. The International Gem Society says some demantoids may show anomalous birefringence and that a polariscope can help decide whether a stone is truly birefringent. Grew et al., American Mineralogist 98 (2013) report birefringent grossular garnets as normally biaxial, at 0.001 to 0.010, and up to 0.015 for compositions with considerable andradite component.
Symmetry fails too. Grew et al. say most garnets are cubic but that henritermierite and holtstamite, both OH-bearing, are tetragonal, and still garnets. Crystal shape is evidence for a specimen; it is not the definition of the supergroup.
What still works. At 1.887 andradite sits far above glass and above the other common garnets; the reading ladder is on how to tell if garnet is real. In water a 20.00 g piece reads 14.74 to 14.87 g; the method is on how to measure specific gravity at home.
Is andradite magnetic?
On a laboratory separator andradite is pulled in the range USGS calls paramagnetic, and no source here tests it with a hand magnet. Rosenblum and Brownfield, USGS Open-File Report 99-529, Table 1, passed 9 andradite samples through a Frantz separator: total range 0.10 to 0.80 amp, best 0.30 to 0.50. The report defines paramagnetic minerals as less magnetic than ferromagnetic ones and separable at up to 1.70 amp; magnetite’s best range is 0.01 amp. Andradite is not flagged as often locked with other species.
The other garnets on the same table. Almandite, the report’s name for almandine, 0.10 to 0.60, best 0.20 to 0.40, 13 samples; grossularite 0.30 to 1.20, best 0.60 to 0.90; pyrope 0.20 to 0.80, best 0.50 to 0.60; spessartite 0.10 to 0.40, best 0.30. Uvarovite has no row. The ranges overlap, so the separator does not identify a garnet species. The andradite sheet prints no magnetic property, and lists magnetite among its associates.
On this site, the magnetic questions are gathered on which minerals are magnetic and is chromite magnetic; the classification questions are indexed at the collecting guides. Nothing on this site is offered for sale; the wanted list is the only commercial route.