AENIGMATITE-ILMENITE-CLINOPYROXENE EQUILBIRA AND APPLICATIONS TO GEOTHERMOMETRY AND OXYGEN BAROMETRY IN PANTELLERITIC MELTS: EXAMPLES FROM PANTELLERIA (ITALY) AND EBURRU (KENYA)

            Recent studies have demonstrated the utility and reliability of QUIlF equilibria in constraining intensive crystallization parameters (T-P-fO₂-a_SiO2) in peralkaline igneous rocks with assemblages of fayalite-clinopyroxene-ilmenite.  Briefly: temperature is constrained by clinopyroxene-fayalite equilibria; the addition of silica activity (or presence of quartz) can constrain pressure, or vice versa; and the addition of ilmenite allows the calculation of oxygen fugacity.  However, many highly-evolved pantellerites have assemblages of clinopyroxene-ilmenite-aenigmatite or clinopyroxene-aenigmatite; the lack of fayalite precludes the use of QUIlF equilibria to determine T- fO₂ in these rocks.  We demonstrate that temperature and oxygen fugacity in highly-evolved pantelleritic melts can be reasonably constrained, given a set of assumptions (fixed P, a_SiO2 = 1.0 relative to quartz, and a_Na2Si2O5 = 1.0), by the following equilibria:

            1) Aen + O₂ = Ilm + 2 Hem + 2 SiO₂ + Na₂Si₂O₅,

            2) 2 SiO₂ + 2 Na₂Si₂O₅ + 2 Fa + O₂ = 4 Aeg,

            3) 2 Aeg = Hem + 2 SiO₂ + Na₂Si₂O₅,

            4) Aen + Na₂Si₂O₅ + O₂ = 4 Aeg + Ilm, and

            5) 2 Fs (cpx) + O₂ = 4 SiO₂ + 2 Hem.

Reaction 1 defines a univariant curve in T-fO₂ space for assemblages with aenigmatite-ilmenite.  Reaction 2 defines the fayalite-out reaction, and the intersection of this curve with 1 provides a maximum T.  Reaction 3 defines the rhombohedral oxide-out reaction, and provides a minimum T for assemblages with ilmenite and a maximum T for assemblages without ilmenite.  Reaction 4 is the upper boundary of the “no-oxide” field, and provides a minimum T for assemblages without oxides.  Reaction 5 is the displaced AHQ buffer, one of the QUIlF reactions.  The intersections of curves defined by reactions 1 and 5 may be able to provide a reasonable estimate of T- fO₂ for assemblages of aenigmatite-ilmenite-clinopyroxene.  Since the free energy data for aenigmatite and aegirine used in these calculations are from estimates, rather than experiment, curves for equilibria 1-4 should be drawn with a very broad brush!  However, results from these techniques are consistent with both experimental data and results from other techniques applied to suites from Pantelleria, Italy, and Eburru, Kenya.