Juliette Imbach (imbach@scm.saclay.cea.fr)1, Fabrice Brunet (brunet@mailhost.geologie.ens.fr)2, Thibault Charpentier (charpent@spec.saclay.cea.fr)1 & Joseph Virlet (virlet@scm.saclay.cea.fr)1
1 Service de Chimie, Moléculaire, Gif Sur Yvette 91191
2 Laboratoire de Géologie, UMR 8538-CNRS, Ecole Normale Supérieure, Paris 75005
We have synthesized hydrothermally at 650°C and 0.15 GPa, yttrium-bearing silico-phosphate apatites,Ca5-xYx(PO4)3-x (SiO4)xOH with x = [0, 3]. In the apatite structure, Ca and Y atoms are 9-fold (site I) and 7-fold (site II) coordinated, P and Si atoms occupy a single tetrahedral site. The oxygen of the hydroxyl groups is surrounded by three atoms of the site II. The hexagonal unit-cell volume determined by XRPD on our synthetic products decreases from the phosphate to the silicate pole. Inspection of the diffraction-peak intensities indicates that Y atoms are preferentially partitioned into the site II. 1H MAS NMR spectra show a single resonance at 0.2 ppm for hydroxylapatite. This line still appears in the silicate end-member where hydroxylapatite-like environments are therefore preserved for some OH groups. As expected, the Y incorporation gives rise to new OH environments evidenced by two additional resonances centered around 1.5 and 4.9 ppm. 31P MAS NMR spectrum of hydroxylapatite shows a single line at 2.8 ppm. For intermediate compositions, a broader line occurs at 1.7 ppm presumably due to the proximity of yttrium. In the heteronuclear 1H-31P correlation maps, the proton lines at 0.2 ppm and 1.5 ppm are correlated to the 31P lines at 2.8 ppm and 1.7 ppm, respectively. These resonances correspond to protons which have phosphate anions in their neighbourhood. Adversely, the proton line at 4.9 ppm is not correlated to any 31P lines. It may correspond to protons surrounded by silicate anions. Assuming that the charge balance (YSi = CaP) is locally achieved, these protons should also be close to yttrium atoms. FTIR shows the usual hydroxylapatite band at 3572 cm-1 and a broad band at around 3544 cm-1. Proton spectra (FTIR and 1H NMR) display a global intensity decrease from the phosphate pole to the silicate pole. This result suggests that protons may charge-balance part of the Ca-Y replacement.
Jannick Ingrin (ingrin@cict.fr)
LMTG UMR5563, Minéralogie, 39 allées Jules Guesde, 31000 Toulouse, France
Quantitative measurement of hydrogen content in single crystals can be determine by polarised infrared spectroscopy using calibrated extinction coefficients (Bell et al., 1995; Libowitzky and Rossman, 1997). However, such techniques necessitate to record polarised infrared spectra in at least 2 perpendicular crystallographic orientations and three different polarised directions. Such procedure is difficult to apply to polycrystalline mantle xenoliths with small grain size. An alternative possibility is to perform a single infrared analysis on each individual crystal of the xenolith and to build an averaged infrared spectra representative of each mineral phase present in the xenolith.
We report results from infrared analyses performed on a xenolith from Kilbourne Hole (New Mexico) recovered from about 50 km depth (1030°C, 1.7 GPa; Glucklich, 1992). Infrared profiles realised within single crystals and across grain boundaries show no hydrogen concentration gradient even at distances less than 10 microns from grain boundaries. In these conditions, the hydrogen content of each mineral phase can be assumed homogeneous and each phase is in equilibrium with the others. More than 100 analyses were recorded and averaged, allowing to measure the hydrogen content in olivine, enstatite and diopside and the partition coefficients of hydrogen F between these phases:
Fenstatite/olivine = 170 and Fdiopside/olivine = 360.
The total amount of hydrogen transported by the xenolith is estimated to an equivalent of 100 ppm H2O. These results are compared to hydrogen solubility data obtained from high-pressure experiments performed on olivine, enstatite and diopside.
Xenoliths from different origins and the same origin but recovered from different depths are currently investigated.
Bell DR, Ihinger PD & Rossman GR, Amer. Mineral., 80, 465-474, (1995).
Glücklich, Thesis Univ. Paris 7, pp 270, (1992).
Libowitzky E & Rossman GR, Amer. Mineral., 82, 1111-1115, (1997).
Jannick Ingrin (ingrin@cict.fr) & Marc Blanchard
LMTG, UMR5563, Minéralogie, 39 Allées Jules Guesde, 31000 Toulouse, France
Experimental data on the kinetic of H-D exchange in amphiboles have been widely used to evaluate the rate of oxidation of mantle-derived amphibole xenoliths during their ascent towards the surface and to decipher on the possible variation in the oxidation state of mantle metasomatic fluids (Dyar et al., 1993; King et al., 1999). Unfortunately, previous experimental kinetic data have been measured on powder amphibole. This technique is highly dependent on the shape model used for individual grains and how the average grain size is determined (Graham et al., 1984). It generally leads to large uncertainties on the diffusion coefficients determination. It also precludes any possibility to collect data on the anisotropy of hydrogen diffusion in single crystals.
H-D exchange experiments were performed between a gas and thin plates of kaersutite (40 to 80 microns thick) cut perpendicular to [010] and [001] directions. Data were collected in the temperature range of 600 and 900°C. Diffusion coefficients along [010] fitted to an Arrhenius function:
D (m2/s) = 2 10-14 exp(-(104 kJ/mol)/RT)
give results slightly above, with comparable activation energy, to the average values deduced by Graham et al. (1984) for a model of grains with plate shape. However, preliminary data collected along [001] direction suggest that diffusion of hydrogen in amphibole is anisotrope. The rate of diffusion along [001] is around one order of magnitude faster than along [010] direction.
These data are used to simulate the dehydrogenation rate of single crystals kaersutite sampled from the mantle within the time scale of few hours to few days.
Dyar MD, Mackwell SJ, Mc Guire AV, Cross, LR & Roberton JD, Amer. Mineral., 78, 968-979, (1993).
Graham CM, Harmon RM & Sheppard SMF, Amer. Mineral., 69, 128-138, (1984).
King PL, Hervig RL, Holloway JR, Vennemann TW & Righter K, Geochim. Cosmochim. Acta, 63, 3635-3651, (1999).
Index of EMPG VIII Volume
Further EMPG VIII Information
Index of the Journal of Conference Abstracts
Cambridge Publications Home Page