Scandium as a Geothermometer: First Experimental Results

Daniel Röhnert (roehnert@em.uni-frankfurt.de)¹, Gerhard Brey¹, Michael Seitz² & Rainer Altherr²

¹ Inst.f.Mineralogie, Senckberganlage 28, 60054 Frankfurt, Germany

² Mineralogisches Inst., Im Neuenheimer Feld 236, 69120 Heidelberg, Germany

The partitioning of Scandium between garnet and clinopyroxene, as well as clinopyroxene and orthopyroxene, in natural rocks has been studied by various authors. A strong temperature dependence led to the formulation of a Sc-thermometer for coexisting CPX and OPX ([1] and earlier work). The empirical formulation as well as the given potential for further geothermobarometric use demand experimental confirmation. As a first step the solubility of Scandium in mantle phases was investigated within the model systems CMS, MAS, NCMS, and NCMAS (+Sc, respectively) with pressures ranging from 3 to 5 GPa and temperatures between 1200 and 1400°C. Charge balance in pyroxenes is mostly attained (NCMS, 4.6-5.4 mol% Sc in CPX at 1300°C, 5 GPa) via monovalent cations (i.e. Na) or to a lesser degree (CMAS, 1.3-1.6 mol% Sc in CPX at 1300°C, 5 GPa) by Tschermak's substitution. A substantial degree of incorporation within CMS-pyroxenes takes place via point defects (0.5 mol% Sc in CPX, 1300°C, 5 GPa). The stability of Scandium-bearing garnet solid solutions was determined within the CMAS+Sc system. At atmospheric pressure the only stable endmember is the CaSc-garnet, while the maximum grt_ss at 5 GPa and 1300°C is Ca_0.4Mg_1.6Sc₂Si₃O₁₂ (CMS) and Mg₃Sc_0.5Al_1.5Si₃O₁₂ (MAS), whereas the MgSc-garnet endmember could not be stabilized. This data is the basis for interpreting experiments within the system NCMAS with Sc added in trace amounts yet still suitable for microprobe analysis. The first results attained show good resemblance to the partitioning data obtained from natural xenolith samples.

Seitz HM, Altherr R & Ludwig T, Geochim. Cosmochim. Ac, in press, (1999).

In situ Study of Olivine-Wadsleyite Transformation Using Impedance Spectroscopy in the Multianvil Apparatus

Claudia Romano (romano@uniroma3.it)¹ & Brent Poe (brent.poe@uni-bayreuth.de)²

¹ Dipartimento di Scienze della Terra, Facoltà di Scienze M.F. N. Università di Roma Tre, Largo San Leonardo Murialdo, 1 00146 Roma, Italia

² Bayerisches Geoinstitut, Universitat Bayreuth, Bayreuth, Germania

Study of the phase transformation of olivine to its high pressure polymorphs wadsleyite and ringwoodite helps our understanding of subduction, mantle convection, and other important processes such as deep focus earthquakes. Due to the high pressure and temperature conditions at which these transformations occur, much of the experimental work on this subject has been accomplished through analysis of samples quenched after various amounts of time under transformation conditions. Recently, however, an in-situ x-ray diffraction study of the transformation of synthetic forsterite to wadsleyite was conducted providing important information about its kinetics). We examine the transformation of a San Carlos olivine in-situ using complex impedance spectroscopy in a multianvil apparatus at pressures and temperatures corresponding to depths of the transition zone. From the complex impedance data we are able to determine electrical conductivity, which differs by a factor of about one hundred between olivine and wadsleyite. This strong contrast provides for a very sensitive means of examining the extent of the transformation in real time. Complex impedance spectroscopy, carried out over a sufficiently wide frequency range, can also provide information about current pathways (e.g. grain interior vs. grain boundary) in polycrystalline samples and aid in the determination of the phase distribution in two-phase aggregates. Therefore, this method should be especially useful at detecting the point at which the higher conductivity wadsleyite phase becomes interconnected on olivine grain boundaries. We complement the study with the examination of quench specimens by micro-Raman and SEM. Early results so far indicate that the magnitude of the conductivity jump from olivine to wadsleyite is lower in these experiments compared to the difference in their conductivities measured separately within their respective stability fields. This might indicate that the pressure dependence on the conductivity of olivine becomes significant above 10 GPa.

Disequilibrium Carbonate-Silicate Immiscibility and Megacryst Growth

Gianluigi Rosatelli (giar@nhm.ac.uk)¹ & Adrian P. Jones (adrian.jones@ucl.ac.uk)²

¹ Dept. Mineralogy Natural History Museum, Cromwell Road, SW7 5BD London, GB

² Dept. Geology University College London, Gower Street, WC1E 6BT, GB

Olivine nephelinite liquids are plausible candidates for parental magmas in many mixed carbonate-silicate volcanic systems. Pargasite amphibole is very close in composition to that of an olivine-nephelinite. Mixtures of natural pargasitic amphibole, synthetic calcite and/or sodium carbonate were used to experimentally investigate the relationship between carbonate and silicate phases at 10-25 kbar in equilibrium conditions. We also explored the transient fields of liquid immiscibility following a complex P-T path, which differs from traditional petrology experiments by simulating disequilibrium conditions. At 10 kbar, from a mixture of pargasite and Na-rich carbonate, we produced a single liquid (T=1200°C), which was then supercooled to 950°C with a cooling rate of 100°C/min. The resulting metastable liquid was than rapidly reheated to 1200°C. Adopting this procedure we obtained, from a single carbonated silicate liquid, two immiscible liquids. The charges were finally cooled at a rate of 1°C/min allowing crystallisation of liquidus phases. In a single capsule we could see (1) a green silicate glass bead surrounded by immiscible Na-rich carbonate quenched liquid; (2) a pyroxene-rich cumulate texture at the capsule base within the carbonate liquid; (3) capsule-long Cpx "megacrysts" surrounded by immiscible carbonate droplet-bearing silicate glass; (4) numerous rounded and curvilinear droplets of immiscible Na-rich carbonate melt and vapour within the silicate-rich glass (5) a large gap at the top of the capsule occupied by a vapour phase evolved during the cooling. We have interpreted these results in comparison with equilibrium experiments and immiscibility data (e.g. Lee and Wyllie, 1997) and with natural occurrences of closely-related alkali silicate-carbonatite rocks. We conclude that immiscibility between silicate and carbonate-rich melts may be produced by decompression in a fast-rising carbonated magma and may contribute to the formation of large megacrysts.

Lee W-J and Wyllie, Contribution to Mineralogy and Petrology, 127, 1-12, (1997).