Oxygen Diffusion in Diopside

Jannick Ingrin (ingrin@cict.fr), Olivier Jaoul (jaoul@cict.fr) & Laure Pacaud

LMTG-CNRS (UMR 5563) Minéralogie, Université Paul-Sabatier, 39 allées Jules Gues, 31000 Toulousede, France

Till now, oxygen diffusion data in diopside were scattered and difficult to use in particular for geospeedometric purposes based on ¹⁸O/¹⁶O exchange in relatively dry metamorphic facies, because data came from hydrothermal conditions (Farver, 1989; Elphick & Graham, 1990) or polycrystalline material (Connolly & Muehlenbachs, 1988) or natural single crystals, but only along c (Ryerson & McKeegan, 1994).

We have measured ¹⁸O/¹⁶O exchange rate in diopside single crystals (one is synthetic Ca_0.95Mg_1.05Si₂O₆, the other is natural with Fe/Si = 0.018 and Cr/Si = 0.005) at room pressure, under dry atmosphere (as opposed to hydrothermal conditions). The temperature range was 1050-1370°C. Oxygen partial pressure p¹⁸O₂ was controlled between 10^-2 and 10^-12 atm. Diffusion profiles were analysed using the ¹⁸O(p(alpha)) nuclear reaction allowing determination of diffusion characteristic lengths between 100 nm and a few µm.

Results for pure synthetic diopside show an important anisotropy with a diffusion coefficient D (=D_oexp(-E/RT)pO₂^m) along b one order of magnitude lower than along a* and c. (a*=bxc). LogD_o(m²/s): b=-8.4±1.0; a*=-9.31±0.8; c=-10.9±0.7; E (kJ/mol): b=333±28; a*=281±21; c=228±20. Values of E and D_o along c for the natural specimen are almost identical to those of the synthetic, and give similar values of D as Ryerson and McKeegan (1994) who had however different E and D_o). For the synthetic, D is almost independent of pO₂ along the three directions.

Anisotropy is easily explainable just by structural considerations on the possible oxygen sites used for oxygen migration. Our data can be extrapolated to lower temperatures without too large error bars and therefore can be used with success for geospeedometry based on oxygen isotopic exchange between diopside and calcite such as done by Edwards and Valley (1998) for the Adirondack Mts with cooling rate between 2 and 4°C/Ma.

Farver JR, Earth Planet. Sci. Lett, 92, 386-396, (1989).

Elphick SC & Graham CM, Terra Abstr, 7, 72-72, (1990).

Connolly C & Muehlenbachs K, Geochim. Cosmochim. Acta, 52, 1585-1591, (1988).

Ryerson FJ & McKeegan KD, Geochim. Cosmochim. Acta, 58, 3713-3734, (1994).

Edwards KJ & Valley JW, Geochim. Cosmochim. Acta, 62, 2265-2277, (1998).

Assignment of Infrared (OH)-Stretching Bands of Richteritic Amphiboles Through Heat-treatment

Kiyotaka Ishida (kiyota@rc.kyushu-u.ac.jp)

Graduate School of Social and Cultural Studies, 4-2-1 Ropponmatsu, Chuo-ku, Fukuoka 810-8560, Japan, Japan

Assignment of infrared (OH)-stretching bands of some A-site occupied and ^[‚S]Al (tetrahedral Al)-free or nearly free richteritic amphiboles have been made through a heat-treatment in air. Two (OH)-stretching bands of (MgMgMg)-OH-A-O^2-/F^-/Cl^- and (MgMgMg)-(OH)-V (V=vacancy) configurations persist up to high temperatures and disappear at similar temperature: with increasing temperatures the (OH)-stretching band A* of (MgMgMg)-(OH)-A-(OH) (A=A site cation) configuration near 3730 cm^-1 shifts downward to near 3700 cm^-1 resulted in the band A** of (MgMgMg)-(OH)-A-O^2- configuration; a repulsive force from proton has absented through dehydrogenation of O(3)H^- to O^2- and then the position of it's A-site's alkali ions move to the dehydrogenated side because an attractive force is in operative (cf. Robert et al., 1999). Some natural richteritic amphiboles in which the hydroxyl ions are substituted with not only O^2- but also F^- and Cl^- these two kinds of (MgMgMg)-OH-A stretching bands are observed at near 3730 and 3700 cm^-1, in which A-site's alkali ions move also to F^- /Cl^- substitute side because an attractive force is operative between them. In these manner, (MgMgMg)-(OH)-K and (MgMgMg)-(OH)-Na bands shift downward -20 cm^-1 and -26~-29 cm^-1 by heat-treatment, respectively, reflecting the ionic radius difference of A-site occupying ions.

Robert JL, Della Ventura G & Hawthorne FC, Am.Mineralogist, 84, 86-91, (1999).

Structural Characterization at Variable Temperature of Phengitic Crystals with Chemical and Polytypic Zoning

Gabriella Ivaldi (ivaldi@dsmp.unito.it), Nadia Curetti, Giovanni Ferraris (ferraris@dsmp.unito.it) & Roberto Compagnoni

Dip. Sci. Mineral. Petrol., Univ. Torino, Via Valperga Caluso 35, 10125 Torino, Italy

Pluricentimetric crystals of phengitic mica occur in a metamorphic dyke crosscutting an eclogitic body of the Sesia Zone at Cima Pal, Val Savenca. The crystals show sharp colour zoning oscillating between concentric green and white bands; white bands show lower Fe content. The colour zoning is overlapped by a structural zoning of 3T and 2 M1 polytypes. Independently from colour, 3T shows a lower degree of celadonite substitution (Si about 3.4 apfu) than 2 M1 (Si about 3.5 apfu). The two polytypes differ also in their interlayer content which is more deficient in 3T (about 0.86K + 0.05Na) than in 2 M1 (about 0.96K + 0.02Na).

Fragments of the two polytypes with different Fe content have been investigated by X-ray single-crystal diffractometry between 25 and 700°C in steps of 200°C. The crystal structures have been refined using sets of diffracted intensities collected at 25 and 700°C. Octahedral cation order appears clearly; the detection of tetrahedral order is more problematic because Si and Al are almost isoelectronic and Al is low. Cation order agrees with neutron results from 3T of Dora-Maira (Pavese et al., 1997) and petrologic hypotheses (Sassi et al., 1994). The cell parameters increase linearly with temperature and their slope is a function of both chemical and structural status, as are other structural parameters, like trigonal rotation.

Since it cannot be imagined that P-T conditions were significantly oscillating during the dyke formation, the oscillatory switching between the polytypes [which can coexist at the same P-T conditions (Frey et al., 1983)] must be correlated with the celadonite substitution (Stöckhert, 1985). In principle, the amount of this substitution could either depend on fluid composition and start the switch or be consequent to the switch of polytype; in the latter case, the starter could be a stacking fault.

Frey M, Hunziker JC, Jäger E & Stern WB, Contrib. Mineral. Petrol., 83, 185-197, (1983).

Pvese A, Ferraris G, Prencipe M & Ibberson R, Eur. J. Mineral, 9, 1183-1190, (1997).

Sassi FP, Guidotti C, Rieder M & De Pieri R, Eur. J. Mineral, 6, 151-160, (1994).

Stöckhert B, Contrib. Mineral. Petrol, 89, 52-58, (1985).