Quantitative 3D measurement of ilmenite abundance in Alpe Arami olivine by confocal microscopy: Confirmation of high-pressure origin

A critical aspect of the debate about the origin and conditions of metamorphism of the Alpe Arami (AA) peridotite is the disagreement over how much ilmenite is contained in the older generation of olivine and therefore how much TiO2 might have been dissolved at high pressure and temperature. We have now determined quantitatively the 3-dimensional distribution of ilmenite in AA olivine by confocal laser scanning microscopy (CLSM). The CLSM measurements show an average concentration of 0.31 vol% of ilmenite in olivine, with individual grains containing up to 1.2 vol%. This translates into average and maximum concentrations of 0.23 and 0.9 wt% TiO2 in olivine, respectively, and confirms the original estimation of maximum concentration of ~1 vol% TiO2. The vast majority of ilmenite in AA olivine is distributed randomly (although topotactically oriented) and, in all cases, is accompanied by chromite in a ratio of ~4:1. These observations are consistent with an origin of the ilmenite (and chromite) by exsolution from an olivine solid solution at P = 9– 12 GPa and temperatures above the stability field of titanian clinohumite, but are not consistent with suggested breakdown of titanian clinohumite. Combining these results with other recent findings suggests that exsolution followed deformation under relatively high fugacity of H2O, and that the high solubility of TiO2 is probably explained by pressure-induced accommodation of Ti in the tetrahedral site of silicates. degrees to a finer-grained aggregate (second generation) containing interstitial oxides. The abundance of rod-shaped inclusions of (Fe,Mg)TiO3 contained within first-generation AA olivine was reported by Dobrzhinetskaya et al. (1996) to be >1 vol% (implying >0.6 wt% TiO2), based upon digital image analysis of polished sections viewed in reflected light. Using a broad-beam microprobe technique, Hacker et al. (1997; see also response by Green et al. 1997a) analyzed a crystal of this material provided to them by the original authors and concluded that the integrated amount of TiO2 in the olivine was only 0.02– 0.03 wt%, even though Hacker confirmed the volume-fraction measurements of the original authors using backscattered electron imaging with a scanning electron microscope (Green et al. 1997a). This inconsistency might have been resolved had the inclusions been determined to have a much lower TiO2 content than originally thought, but subsequent work has shown that all of the rods are ilmenite (Hacker et al. 1997; Green et al. 1997a; Risold et al. 2001), with no evidence of substitution of any element for Ti. Creating even greater confusion, the microprobe method of Hacker et al. (1997) returned the same low concentration of TiO2 in a specimen of their own material for which the original authors measured a volume fraction of 0.02 vol% titanate, a value consistent with the microprobe measurements (Green et al. 1997a). It is clear, therefore, that the amount of TiO2 present in olivine of the specimens of AA peridotite originally examined by Dobrzhinetskaya et al. (1996) remains to be characterized definitively. We have revisited this subject by directly measuring the volume fraction of ilmenite in the cores of the oldest generation of olivine in selected specimens of the AA garnet lherzolite by applying confocal laser scanning microscopy (CLSM), as BOZHILOV ET AL.: 3D MEASUREMENT OF ILMENITE IN ALPE ARAMI OLIVINE 597 well as large-area chemical analysis with a scanning electron microscope (SEM) and an electron probe microanalyzer (EPMA). We find that both methods of analysis confirm the original abundances of ilmenite and TiO2 reported by Dobrzhinetskaya et al. (1996). We then apply a stringent test of the models of origin of these oxides proposed by Dobrzhinetskaya et al. (1996) and by Risold et al. (2001), and find that the observations are consistent with the former model but inconsistent with the latter one.