Abstract

Relations among Fe-Ti oxides, calcium magnesium iron pyroxenes and olivines, and quartz are governed by the reaction QUIIF:

 
SiO2quartz+2Fe2TiO4ulvöspinel=2FeTiO4ilmenite+Fe2SiO4fayalite

and numerous derivative equilibria in the system Fe-O-CaO-MgO-SiO2-TiO2. By combining internally consistent thermodynamic solution models for iron magnesium titanium oxides and calcium magnesium iron pyroxenes and olivines, we have calibrated these equilibria (Lindsley and Frost, 1992). In this paper we show how they are applied to the interpretation of a wide range of volcanic and plutonic rocks. Depending on the assemblage, the pyroxene QUIIF equilibria permit calculation of temperature, pressure, fO2, and silica activity for numerous rocks.

Pyroxene QUIIF equilibria place constraints on the intrinsic parameters that controlled the evolution of three well-studied igneous bodies: the Bishop Tuff, Thingmuli Volcano, and the Skaergaard Intrusion. The Fe-Ti oxides of the Bishop Tuffare not in equilibrium with ttre coexisting pyroxenes and quartz. Compositions of argite, Opx, and titaniferous magnetite are essentially constant throughout the pyroxene-bearing portion of Bishop Tuff and could have been in equilibrium with quartz at 824 ± 15 °C, 2700 ± 2000 bars, and fO2, 1.39 ± 0.05 log units above that of the FMQ buffer. The compositions of the ilmenites have been modified in response to a thermal event not recorded in the pyroxenes that could have occurred either within the magma chamber immediately before the tuff was ejected or possibly within the tuffafter eruption.

Thingmuli Volcano in Iceland contains several rock types that have assemblages suitable for applying pyroxene QUilF. Rocks that contain pigeonite and augite together with ilmenite and single-phase titaniferous magrretite generally have similar pyroxene and oxide temperatures. Pyroxene-QUIIF relations greatly restrict uncertainties in T and fO2 rrelative to those for oxide data alone.

Crystallization of Skaergaard magma followed a relatively reducing trend from fO2 near FMQ, where oxides first appear, to nearly 2 log units below FMQ in the Sandwich horizon, values that are all higher than those indicated by measurements of intrinsic fO2. QUIIF relations show unequivocally that the silicates could not have been in equilibrium with any titaniferous magnetite and ilmenite atthe fO2, indicated by the intrinsic measurements.

We have also applied the pyroxene QUIIF equilibria to many other rock suites. Calc-alkalic volcanic rocks that contain low-Ca pyroxenes tend to have equilibrated at the same relative fO2 (about 1–2 log units above FMQ) even though they represent a range of composition from andesite to rhyolite. In contrast, tholeiitic volcanic rocks, as typified by Thingmuli, crystallized at much lower fO2 (generally about I log unit below FMQ) and, like calc-alkalic suites, show little or no change in relative fO2 with temperature. Plutonic suites show similar ranges in fO2 with, layered gabbros of calc-alkalic affinity having fO2 above that of FMQ and those of tholeiitic affinity having fO2 below FMQ.

This content is PDF only. Please click on the PDF icon to access.

First Page Preview

First page PDF preview
You do not currently have access to this article.