Phillipsite
Phillipsite is a zeolite series: phillipsite-Ca, -K and -Na on the IMA list. Monoclinic, always twinned, hardness 4 to 4.5, density 2.20; one Handbook sheet.
Phillipsite is a zeolite series, not one species: the IMA list has phillipsite-Ca, phillipsite-K and phillipsite-Na, each (Si10Al6)O32·12H2O with calcium, potassium or sodium dominant. Its one Handbook sheet gives monoclinic crystals, always twinned, to 2 cm, hardness 4 to 4.5, density 2.20. The suffix needs an analysis; the sheet’s Mazé analysis is potassium-dominant by 0.11.
In short
- Three rows, one name. A script over the September 2026 IMA-CNMNC Master List finds phillipsite-Ca, Ca3(Si10Al6)O32·12H2O, USA; phillipsite-K, K6(Si10Al6)O32·12H2O, Italy; and phillipsite-Na, Na6(Si10Al6)O32·12H2O, Italy; all A, 1997 s.p. There is no row for plain phillipsite. The IMA zeolite report (Coombs et al. 1997) treats phillipsite as a series and names members by the most abundant extra-framework cation in atomic proportions.
- One sheet, and it is the potassium one. The Handbook of Mineralogy has a single phillipsite sheet, filed as phillipsite-K and headed with the series formula (K,Na,Ca)1–2(Si,Al)8O16·6H2O, written on half the IMA formula’s 32 oxygens. Phillipsite-Ca, phillipsite-Na and an unsuffixed sheet return 404 (10 November 2026), so the numbers on this page are that sheet’s.
- The Mazé margin. The sheet’s one analysis, from Mazé, Niigata, Japan, recast on 16 oxygens, gives K 0.793, Ca 0.683, Na 0.675 and Fe 0.083, reproducing the printed (K0.79Na0.68Ca0.68Fe0.08)Σ=2.23(Si5.12Al2.81)Σ=7.93O16·6.39H2O. Potassium leads by 0.11 atoms, so the analysed material is phillipsite-K; a shift of that size would change the name.
- Documented failure: charge does not name it. In the same analysis calcium carries 1.37 of the 2.84 positive charges on the extra-framework cations, against potassium’s 0.79. The rule counts atoms, not charge, so the cation carrying most charge is not the one in the name. No hand test, colour or habit gives the suffix; the label honestly reads phillipsite unless an analysis is cited.
- Weight and the cell. D(meas.) 2.20 is 10.91 g in water for a 20.00 g piece. On the sheet’s cell (1011.35 ų) the printed Mazé formula gives 2.197, the measured value, not the printed D(calc.) 2.242; ideal end-members on the same cell give 2.308 (K), 2.149 (Na) and 2.120 (Ca), a 0.19 spread; real phillipsite is mixed, so the balance does not give the suffix.
| Member | IMA formula, status, country | Handbook sheet? | Dominant cation per 32 O | K2O / Na2O / CaO by mass | H2O by mass | Density on the sheet’s cell |
|---|---|---|---|---|---|---|
| Phillipsite-Ca | Ca3(Si10Al6)O32·12H2O; A, 1997 s.p.; USA | No (404) | Ca 3 | 0 / 0 / 13.03% | 16.74% | 2.120 |
| Phillipsite-K | K6(Si10Al6)O32·12H2O; A, 1997 s.p.; Italy | Yes, headed Phillipsite | K 6 | 20.11% / 0 / 0 | 15.38% | 2.308 |
| Phillipsite-Na | Na6(Si10Al6)O32·12H2O; A, 1997 s.p.; Italy | No (404) | Na 6 | 0 / 14.21% / 0 | 16.52% | 2.149 |
| Mazé, Japan (sheet analysis) | Printed: (K0.79Na0.68Ca0.68Fe0.08)Σ=2.23(Si5.12Al2.81)Σ=7.93O16·6.39H2O | On the phillipsite-K sheet | K 1.59, Ca 1.37, Na 1.35: K by 0.22 (0.11 per 16 O) | 5.59 / 3.13 / 5.73% printed | 17.22% printed | 2.197 (printed formula); D(meas.) 2.20, D(calc.) 2.242 |
| Harmotome (IMA list only) | Ba2(Si12Al4)O32·12H2O; A, 1997 s.p.; Germany | Separate sheet | Ba 2 | — | — | — |
Is phillipsite one mineral? Phillipsite-Ca, -K and -Na
What the IMA list prints. Three rows, all status A with a 1997 special-procedure date: phillipsite-Ca, Ca3(Si10Al6)O32·12H2O, country USA; phillipsite-K, K6(Si10Al6)O32·12H2O, Italy; and phillipsite-Na, Na6(Si10Al6)O32·12H2O, Italy. Each has ten silicon to six aluminium, Si:Al 1.667, and twelve waters; only the extra-framework cation differs. harmotome is a separate row, Ba2(Si12Al4)O32·12H2O, A, 1997 s.p., Germany, with Si:Al 3 by its formula.
What the 1997 report adds. The report writes the series as (K,Na,Ca0.5,Ba0.5)x[AlxSi16–xO32]·12H2O, with x from about 4 to about 7, and its rule names each member by attaching to the series name the chemical symbol of the extra-framework cation most abundant in atomic proportions. So “phillipsite” on a label names the series, and the suffix needs an analysis. The same principle on chabazite’s sheets is worked on chabazite; this page does not repeat it. The family as a whole is on which minerals are zeolites, and other zeolites are in the species index.
The Mazé margin: why the sheet’s analysis is phillipsite-K
Method. The sheet’s analysis (1), from Mazé, Niigata Prefecture, Japan, prints SiO2 46.03, Al2O3 21.43, FeO + Fe2O3 0.99, CaO 5.73, Na2O 3.13, K2O 5.59, H2O 17.22, total 100.12. Each oxide is divided by its molar mass and the anhydrous oxygens scaled to 16. With the iron as Fe2O3 the result is K 0.793, Ca 0.683, Na 0.675, Fe 0.083, Si 5.122, Al 2.810 and 6.39 H2O, which reproduces the printed formula to the second decimal; with the iron as FeO it is within 0.01 on every cation.
The margin, and the failure. Of the three alkali and alkaline-earth cations, potassium is 36.9%, calcium 31.7% and sodium 31.4% by atoms: potassium is most abundant by 0.11 per 16 oxygens, so the analysed material is phillipsite-K, consistent with the sheet’s filing. By charge the order reverses: calcium, divalent, supplies 1.37 of 2.84 charges, potassium 0.79. The naming rule counts atoms. A second analysis from the same place could fall on the other side of a 0.11 margin, which is why a locality does not fix the suffix either.
Phillipsite crystals: always twinned
What the sheet prints. Monoclinic, point group 2/m, space group P21/m. Crystals always twinned, pseudosymmetrical, striated parallel to [001] on {010}, to 2 cm; equant to prismatic, or in spherical radiating aggregates. Twinning: cruciform single and double penetration twins on {001}, {021} and {110}, ubiquitous. Cleavage {010} and {100}, distinct; fracture uneven; brittle; hardness 4 to 4.5. Colourless, white, reddish or yellowish, with a white streak and vitreous lustre. The 1997 report adds a pseudo-orthorhombic cell, a ≈ 9.9, b ≈ 14.2, c ≈ 14.2 Å, β ≈ 90°.
Thin section and X-ray. Colourless in thin section; biaxial (+), α 1.483 to 1.504, β 1.484 to 1.509, γ 1.486 to 1.514, 2V 60° to 80°. The sheet’s powder pattern, from the Casal Brunori quarry near Rome, has its strongest lines at 3.206 (100), 7.16 (66), 7.18 (63), 4.12 (41), 3.277 (37), 4.13 (36) and 2.753 (36). A powder pattern identifies the series; the 1997 report gives pseudocell ranges for each member that overlap, so it is the analysis, not the pattern, that gives the suffix.
The cell check: 2.20 measured, 2.242 printed
Worked. The sheet’s cell, a 9.865, b 14.300, c 8.668 Å, β 124.20°, has V = abc·sin β = 1011.35 ų. With Z = 2 and the printed Mazé formula (formula mass 669.0), Z·M·1.66054/V gives 2.197, the measured 2.20, not the printed D(calc.) 2.242; the sheet does not say what composition its 2.242 assumes, and this page does not correct it. A 20.00 g piece reads 10.91 g in water at 2.20 and 11.08 g at 2.242; method on how to measure specific gravity at home.
Documented failure: weighing does not give the suffix. Ideal end-members on the same cell give 2.308 for potassium, 2.149 for sodium and 2.120 for calcium. Real phillipsite is mixed, as the Mazé analysis shows, and its 6.39 H2O is not the formula’s 6, so a density between those values says nothing about which cation leads. Laumontite’s water ladder on laumontite shows the same limit for water.
Where phillipsite is found
Occurrence. Typically in cavities in basalt, of hydrothermal origin; an authigenic mineral in saline lake and hot spring deposits and calcareous deep-sea sediments. Associates: zeolites, apophyllite, calcite, nosean, nepheline, olivine, melilite, celadonite.
Localities, as the Distribution field prints them. Many localities, although uncommon in large crystals. Aci Castello, Sicily, and Capo di Bove and other localities around Rome, Lazio, Italy; Annerod near Giessen, Hesse, and Asbach, Westerwald, Germany; along the Giant’s Causeway, Co. Antrim, Ireland; in Oregon, USA, at Wall Creek near Monument and Mount Vernon, Grant Co., Cape Lookout, Tillamook Co., and Spray, Wheeler Co.; Mazé, Niigata Prefecture, Japan; Richmond, Victoria, and Gads Hill near Liena, Tasmania. The sheet marks no type locality; see what is a type locality. Related zeolites: chabazite, heulandite, stilbite, natrolite. Any hazard question is routed to toxic minerals in a collection; that is not advice. We hold no catalogue and nothing here is offered for sale; the wanted list is the only commercial route.