Among the thermochemical data for rock-forming minerals in the system Mg0-Si02-H20, those for orthoenstatite and for talc are highly uncertain. The tightly-bracketed reactions (1) talc = 3 enstatite + quartz + steam, and (2) talc + forsterite = 5 enstatite + steam permit the simultaneous calculation of the standard Gibbs free energies of formation for both orthoenstatite and talc: −1457.27 ± 4 kJ mol−1 and −5516.37 ± 10 kJ mol−1, respectively. Using the calculated free energies of talc and enstatite and six reversed brackets for four reactions involving anthophyllite (Greenwood, 1963), the standard Gibbs free energy of formation of anthophyllite was recomputed to be −11,331.8 kJ mol −1 with an estimated uncertainty of ±17 kJ mol−1; this value is more positive than previous estimates because of the revised value for enstatite.
The reversed brackets of Johannes (1968) for the reaction clinochrysotile + brucite = 2 forsterite + 3 steam were used to calculate the free energy of formation of clinochrysotile. The value obtained (−4,030.84 ± 4.0 kJ mol−1) agrees well with published values. As an independent check, the free energy of talc was calculated, using the calculated free energy of clinochrysotile and a single reversed bracket for the reaction 5 clinochrysotile = 6 forsterite + talc + steam (Chernosky, 1973). The result, −5,522.91 ± 15.62 kJ mol−1, is in satisfactory agreement with the above-cited value.
Our results confirm the geometry of the P−T diagram for anthophyllite, enstatite, talc, forsterite, quartz, and steam suggested by Greenwood, through inconsistencies in the slopes of Chernosky's experimental results and those of Greenwood remain to be resolved. Nevertheless, our results indicate that an internally consistent P−T net, using the experimental data of Chernosky (1976) and Greenwood (1963) for anthophyllite, enstatite, talc, forsterite, and quartz, should be possible.