Abstract
Mineral incompatibilities (based on a complete review of natural occurrences), relevant experimental data, and a computer program ("REACTION") have been used to model the probable stabilities of beryllium-bearing minerals in the system CaO-BeO-SiO 2 -P 2 O 5 -F 2 O (sub -1) , in terms of the chemical potentials of the Lewis acid components P 2 O 5 and F 2 O (sub -1) . The somewhat complicated phase relations are greatly simplified if it is assumed that the gangue minerals (in this case, mainly quartz plus fluorite, CaF 2 , or fluorapatite, Ca 5 (PO 4 ) 3 F) are present "in excess," so that only one Be-ore mineral is stable in a given region of the 3mu F 2 O (sub -1) -- mu P 2 O 5 diagram. The resulting diagram shows that gugiaite, Ca 2 BeSi 2 O 7 , and other Be-bearing calc-silicates in skarns, are stable only under relatively high temperature or basic conditions. With increasing activities of the acid components F 2 O (sub -1) and P 2 O 5 , phenakite, Be 2 SiO 4 , becomes stable with fluorite or fluorapatite in typical greisen and vein assemblages (especially in replaced carbonate rocks). Under the very high P 2 O 5 , low F 2 O (sub -1) conditions characteristic of the late stages of certain pegmatites, hurlbutite, CaBe 2 (PO 4 ) 2 , becomes stable, whereas under high activities of both P 2 O 5 and F 2 O (sub -1) , herderite, CaBePO 4 F, becomes stable in very late-stage pegmatites and in greisens. Under conditions more F 2 O (sub -1) -rich than those normally encountered in nature, the phase CaBeF 4 might occur (especially in fluid inclusions).When Al 2 O 3 is added to the model system, the resulting diagrams show the breakdown reactions of beryl, Be 3 Al 2 Si 6 O 18 , to phenakite plus topaz in the fluorite stability field, or to herderite plus topaz in the fluorapatite stability field.
