Oriented quartz + calcic amphibole inclusions in omphacite from the Saualpe and Pohorje Mountain eclogites, Eastern Alps—An assessment of possible formation mechanisms based on IR- and mineral chemical data and water storage in Eastern Alpine eclogites

Lithos 106 (2008) 336–350

Contents lists available at ScienceDirect

Lithos j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / l i t h o s

Oriented quartz + calcic amphibole inclusions in omphacite from the Saualpe and Pohorje Mountain eclogites, Eastern Alps—An assessment of possible formation mechanisms based on IR- and mineral chemical data and water storage in Eastern Alpine eclogites Jürgen Konzett a,⁎, Eugen Libowitzky b, Clivia Hejny a, Christine Miller a, Alberto Zanetti c a b c

Institut für Mineralogie und Petrographie, Universität Innsbruck, Innrain 52, A-6020 Innsbruck, Austria Institut für Mineralogie und Kristallographie, Universität Wien, Althanstrasse 14, A-1090 Wien, Austria CNR-Istituto di Geoscienze e Georisorse, Unità di Pavia, via Ferrata 1, I-27100 Pavia, Italy

a r t i c l e

i n f o

Article history: Received 14 February 2008 Accepted 10 September 2008 Available online 20 September 2008 Keywords: Eastern Alps Eclogite IR spectroscopy Nominally anhydrous phases Exsolution

a b s t r a c t The composition of mineral phases and their modal proportions have been determined for three representative Eoalpine eclogites from the Saualpe type locality/Eastern Austria (sample SKP31) and the Pohorje Massif/Slovenia (CM31/03 and CM15/01) using electron microprobe, laser ICP-MS, IR spectroscopy and modal analysis to evaluate possible mechanisms for the formation of composite oriented calcic amphibole + quartz inclusions (COIs) in omphacite and to assess the relative importance of hydrous and nominally anhydrous phases as H2O carriers in these eclogites. For omphacites in CM31/03 with a zonal distribution of COIs, a comparison of water and trace element concentrations of areas containing COIs and those free of COIs and a comparison with the trace element concentration of calcic amphibole indicate that COIs have formed through an open-system alteration of clinopyroxene and not through a closed system exsolution process. In sample SKP31, both textural and mineral chemical evidence suggests that COIs did not form by exsolution involving a Ca-Eskola component in clinopyroxene but formed by progressive growth under eclogite-facies P-T conditions and prior to the onset of retrogressive symplectite formation analogous to the formation of poikiloblastic quartz–calcic amphibole grains in the matrix. Bulk H2O contents of the eclogites are between ca. 750 and 2150 ppm with 6–25% of the total water contributed by nominally anhydrous minerals (NAMs). Because of high modal amounts of 37–65%, omphacite is the major nominally anhydrous water carrier, containing 145–580 ppm H2O with significant concentration variations on a thin section scale. Due to their very low H2O concentrations of b5–10 ppm (garnet, kyanite) or insignificant modal amounts ≤ 3% (rutile) the remaining NAMs contribute less than 1.5% to the bulk eclogite H2O content. Calcic amphibole forming part of COIs may be a major carrier of H2O as evidenced by CM31/03 containing both COIs and texturally primary calcic amphibole. In this sample calcic amphibole of the COIs contributes 63% whereas primary calcic amphibole only accounts for 13% of the bulk water. The relative order of H2O concentrations in NAMs is H2Orutile N H2Oomphacite ≫ H2Ogarnet for CM31/03 and CM15/01 and H2Orutile ≈ H2Oomphacite ≫ H2Ogarnet in SKP31. © 2008 Elsevier B.V. All rights reserved.

1. Introduction Omphacitic clinopyroxenes containing oriented rods/needles of quartz and more rarely coesite are frequently found in mafic eclogites (Page et al., 2005, and references therein). These oriented inclusions are commonly interpreted as an exsolution texture formed by the retrogressive breakdown of a non-stoichiometric Ca-Eskola component (Ca0.550.5AlSi2O6) in clinopyroxene during exhumation accord⁎ Corresponding author. E-mail addresses: [email protected] (J. Konzett), [email protected] (E. Libowitzky), [email protected] (A. Zanetti). 0024-4937/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.lithos.2008.09.002

ing to a reaction 2 Ca0.550.5AlSi2O6 = CaAl2SiO6 + 3 SiO2, thereby producing a Ca–Tschermak enriched pyroxene while releasing SiO2. Because some of these localities record ultrahigh-pressure (UHP) metamorphic conditions as evidenced by coesite- and/or diamondbearing assemblages (Smith, 1984; Shatsky et al., 1985; Katayama et al., 2000; Song et al., 2003; Zhang et al., 2005) the presence of oriented quartz inclusions in clinopyroxene in combination with thermobarometric calculations have often been used to claim UHP conditions even if only indirect evidence for the former presence of UHP phases could be presented (e.g. Schmädicke and Müller, 2000; Tsai and Liou, 2000; Zhang et al., 2002; Liati et al., 2002; Dobrzhinetskaya et al., 2002; Janák et al., 2004). Using phase analysis

J. Konzett et al. / Lithos 106 (2008) 336–350

in a simplified CaO–MgO–Al2O3–SiO2–TiO2 model eclogite, Day and Mulcahy (2007) showed that formation of SiO2-lamellae need not be related to a CaEs component in omphacite but may result from reactions of minor/accessory phases such as rutile or kyanite with stoichiometric pyroxene phase components. Likewise, these authors showed that vacancies in omphacite may be consumed by SiO2conserving reactions so that a high vacancy content of a pyroxene does not imply a high potential to produce free silica. In some cases, calcic amphibole was identified intergrown with the oriented quartz inclusions, thus forming composite oriented calcic amphibole-quartz inclusions (COIs) (Terry and Robinson, 2001; Liati et al., 2002; Page et al., 2005; Miller and Konzett, 2005). The presence of amphibole requiring a hydrous fluid for its formation shows that mechanisms more complex than the simple CaEs breakdown reaction as defined above must have been responsible for the formation of COIs. Page et al. (2005) pointed out that composite inclusions need not be at all related to a former CaEs component in clinopyroxene but may form instead by secondary breakdown reactions involving a hydrous fluid. Page et al. (2005) also challenged the significance of oriented quartz+calcic amphibole as UHP indicators and reported their extensive growth in clinopyroxene from an eclogite equilibrated at 600–700 °C and approximately 1.5 GPa. Additional evidence for a HP origin of oriented quartz inclusions had already been presented by Gayk et al. (1995) describing an occurrence of oriented quartz needles in clinopyroxene from a felsic high-pressure granulite equilibrated around 1.8 GPa and 1100 °C. This is in agreement with experimental results by Konzett et al. (2008) who stabilized clinopyroxene with a significant CaEs component at 2.5 GPa and 850 °C in a basaltic bulk composition. These authors showed that CaEsss in clinopyroxene is strongly dependent upon bulk composition and temperature but shows very little pressure dependence. In order to assess possible formation mechanisms of COIs in eclogitic clinopyroxene with special emphasis on potential sources of hydrous fluid and to evaluate the water storage capacity of typical Eastern Alpine eclogites we present the results of an IR and mineral chemical investigation of eclogites containing COIs in omphacites from the Koralpe/Austria representing the eclogite type locality and from the Pohorje Mountains/Slovenia, together forming the southernmost portion of the Eoalpine eclogite zone of the Eastern Alps. 2. Geological setting The Saualpe–Koralpe–Pohorje (SKP) complex is part of the southernmost Austroalpine nappe system and represents portions of the former southern margin of the Alpine Thethys immediately north of the Periadriatic lineament (e.g. Stampfli and Mosar, 1997) (Fig. 1). This Austroalpine domain was dissected and accreted to the European continental margin during closure of the Meliata and Vardar oceans, and collision of the Apulian microplate with Europe. The lithology of the SKP complex is dominated by ortho- and paragneisses with intercalations of eclogite bodies ranging in size from a few to several hundred meters. A more detailed description of the tectonic and geological situation is given by Miller et al. (2005) (with references therein) and need not be repeated here. Peak-metamorphic conditions for the SKP complex are controversial: a number of studies (Miller, 1990; Miller and Thöni, 1997; Sassi et al., 2004; Miller et al., 2005; Miller and Konzett, 2005; Miller et al., 2007) derived HP-conditions of 2.2–2.8 GPa at 670–760 °C that were reached around 90 Ma. Janák et al. (2004, 2006) on the other hand, postulated UHP conditions for the Pohorje Massif based on calculated phase equilibria and textural observations involving oriented quartz inclusions in clinoproxene and radial cracks around quartz inclusions in garnet. 3. Sample preparation and analytical techniques The major element composition of minerals was analyzed using a JEOL XJS 8100 electron microprobe with 15 kV acceleration voltage

337

and a 20 nA beam current and acquisition times of 20 s and 10 s on peaks and backgrounds of the X-ray lines. The following standards were used: Si, Al, Ti: synthetic SiO2, Al2O3 and TiO2; Fe, Mn, Mg, Ca, Na, K: natural almandine, tephroite, diopside, bytownite, jadeite and orthoclase. In addition, a diopside standard (USNM-standard 117733; Jarosevich et al., 1980) was analyzed before and after each analytical session in order to increase the reliability of clinopyroxene analysis especially with respect to total cation sums. Raw counts were corrected using the PRZ correction procedure. In-situ trace element analyses of individual mineral phases were carried out using laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) at the CNR-IGG, Unit of Pavia, Italy. The laser probe consists of a Q-switched Nd:YAG laser, model Quantel (Brilliant), whose fundamental emission in the near-IR region (1064 nm) is converted to 266 or 213 nm by harmonic generators. The laser was focused to a spot diameter of 40–60 μm and the ablated material was analyzed by using, alternatively: (i) a double focusing sector–field analyzer Element I (ThermoFinnigan MAT), in which the standard field regulator power stage of the magnet and the ICP torch were upgraded to those of the Element II model; (ii) an Elan DRC-e mass spectrometer. Helium was used as carrier gas with Ar admixed downstream of the ablation cell. NIST SRM 610 was used as external standard and Ca or Si as internal standard. Precision and accuracy of individual analyses were assessed from repeated analyses of the BCR-2 or NIST SRM 612 standards and yielded results usually better than 7% (1σ) and 10% (at ppm level), respectively. Detection limits were typically in the range of 100–500 ppb for Sc, 10–100 ppb for Sr, Zr, Ba, Gd and Pb, 1–10 ppb for Y, Nb, La, Ce, Nd, Sm, Eu, Dy, Er, Yb. Hf and Ta, and usually b1 ppb for Pr, Th and U. A detailed description of instrumental parameters and quantification procedure is given in Tiepolo et al. (2003). X-ray single crystal diffraction data were collected at room temperature using an Oxford Diffraction Gemini R Ultra four-circle diffractometer equipped with graphite monochromated MoKα radiation at 50 kV and 40 mA and a 135 mm diameter Ruby CCD detector. Reflections with intensities smaller than 0.2% of the type h0l: h + l = 2n + 1 violating P2/n symmetry were proven to originate from λ/2 effect and thereafter ignored in the structure refinement. Lattice parameters were found to be very similar to omphacites of comparable chemical composition (cf. McCormick, 1986; Boffa-Ballaran et al., 1998). Structure refinements based on Fo2 were performed using the program SHELXL-97 (Sheldrick, 2008) starting from positional and displacement parameters as in the above mentioned samples. Likewise the site-occupation of the M positions was taken from these samples as staring values for the structure refinement, which eventually converged to an R1-value smaller 0.004 for all studied samples. The density of oriented inclusions in omphacites was determined based on high-quality BSE images (0.25 pixels/µm) using the NIH image software. Only those sections were selected where the long axes of the quartz–calcic amphibole needles are ± parallel to the sample surface. IR absorption measurements of omphacite, garnet and kyanite were performed on a PerkinElmer 1760X FTIR spectrometer connected to a PerkinElmer IR microscope using doubly-polished self-supporting sections with a thickness of 0.145–0.195 mm. In addition selected individual garnet and rutile grains were handpicked from a mineral concentrate (cf. Miller et al., 2005) and embedded in epoxy resin for IR-analysis. Due to their small crystal size, garnet grains were polished to a thickness of 0.125 mm and rutiles to 0.085 mm. A ceramic light source, a KBr beam splitter and a MCT detector were used to collect spectra in the wavenumber range from 600 to 4000 cm− 1. The IR microscope was equipped with three 6×/0.6 N.A. mirror (Cassegrain) lenses in condenser, objective and detector position. A dual aperture system was set up to constrain the measured sample areas and to avoid any stray light. Therefore, selected regions of the polished sample slabs were adjusted on the centre of a flat metal aperture with 100 µm diameter and, in addition, a second

338

J. Konzett et al. / Lithos 106 (2008) 336–350

Fig. 1. Simplified geological map of the Eastern Alps (a) showing the different segments of the Austroalpine basement and the areal distribution of the Cretaceous eclogite-facies metamorphism (modified from Thöni 2002); abbreviations in Fig. 1a are: ÖTZ Ötztal Crystalline Basement; SI Silvretta Crystalline Basement; EW Engadine Window; TE Texel Mountains; TW Tauern Window; SCH Schober Mountains; WÖLZ Wölzer Tauern; GN Gurktal Nappe; SAU and KOR. Saualpe and Koralpe; PO Pohorje Massif; RW Rechnitz Window; SAM southern limit of Austroalpine high-pressure metamorphism; PAL Periadriatic Lineament; (b) showing the geology of the easternmost Austroalpine units including Saualpe, Koralpe and Pohorje Massif with the locations of the samples described in Section 4.

aperture of 600 µm diameter was placed after the objective lens (thus, the effective diameter was also 100 µm). Spectra were averaged from 128 scans at 4 cm− 1 resolution each and compared to a 100% transmittance reference spectrum from an empty aperture. Because of the random orientation of mineral grains in the rock sections, polarized spectra were not obtained, except for one spot, where the pleochroic behavior of elongated amphibole inclusions was tested (see below). Because of the beam shift between the reference and sample acquisition (empty aperture vs. plane parallel platelet with refractive index n N 1), all spectra are influenced by water vapor in the wavenumber region from 3550 to 3900 cm− 1. Therefore, spectra were corrected as follows: all spectra were converted from transmittance to absorbance units to facilitate the various spectral calculations and corrections. The garnet spectra of each sample containing no visible OH absorptions were averaged, background-corrected to give a horizontal line, and shifted to zero absorbance. Thus, the resulting spectra contained only a flat line with the water vapor vibrations/ rotations. These spectra were then subtracted from the other sample spectra to compensate for the water vapor influence. The final spectra were then converted to ASCII format. All these steps were performed

with the Spectrum 2.00 spectroscopy software by PerkinElmer. Plotting of spectra was done with SigmaPlot 4.01 by Jandel Scientific. To calculate water concentrations from the spectra, the areas (Ai) of the absorption bands were analyzed. In case of pyroxenes showing absorption features between 3350 and 3550 cm− 1 (samples CM15 and SKP31) only, the area calculator of the Spectrum 2.0 program was used. In case of additional amphibole bands between 3600 and 3700 cm− 1 (sample CM31/03) the program PeakFit 4 by Jandel Scientific was used. In all cases a straight baseline between approx. 3200 and 3600–3800 cm− 1 was subtracted to ensure relative comparability. Aside from slight variations in the thickness of the doubly-polished rock slabs, the main source of uncertainty in the determination of water contents of clinopyroxenes reported in this study is the usage of unpolarized radiation. Nevertheless, because polarized spectra recorded in one case show that the absorptions are not extremely pleochroic (see below), and because the absorptions are generally very weak (A b 0.3), an approach using Atot = Aunpold 3 was adopted. The water concentrations were then calculated using the calibration devised by Libowitzky and Rossman (1997) which gives integral molar absorption coefficients εi in correlation to the mean wavenumber of the IR

J. Konzett et al. / Lithos 106 (2008) 336–350

absorption. The obtained value εi = 67,000 L·mol− 1·cm− 2 is in excellent agreement with the most recent omphacite-specific calibration by KochMüller et al. (2007), who obtained εi = 65,000 ± 3000 L·mol− 1·cm− 2. The water content is given by c(H2O) [wt.%] = 1.8·Ai,tot [cm−1]/(D[g/cm3]·εi·d [cm]). Densities used were 3.30 g/cm3 for omphacite and amphibole and 3.65 g/cm3 for kyanite. The obtained H2O concentrations are estimated to be accurate with ± 10 rel.%. For IR absorption measurements of rutile (from rock sections and separates) and garnet separates, a Bruker TENSOR27 FTIR spectrometer connected to a Bruker HYPERION IR microscope was used with a globar light source, a KBr beam splitter and a MCT detector to collect spectra in

339

the wavenumber range from 530 to 4000 cm− 1. The IR microscope was equipped with 15×/0.4 N.A. mirror (Cassegrain) lenses in the objective and condenser positions. Dependent upon the crystal size, a rectangular aperture from 30 × 50 to 100 × 100 µm was used to constrain the selected measurement area. Spectra were averaged from 128 scans at 4 cm− 1 resolution each and compared to a 100% transmittance reference spectrum from an empty aperture. Plotting of ASCII-converted spectra was done with SigmaPlot 9.0 by Jandel Scientific. Two polarized spectra could be acquired per rutile crystal section, one containing the direction perpendicular to the c-axis (o) and one parallel to the projection of the c-axis on the section plane (e'). Because

Fig. 2. Photomicrographs of mineral assemblages from samples CM31/03 and SKP31; (a) optical microscopic image (parallel polarizers) of omphacite from CM31/03 showing COIs with a zonal distribution: high COI-density in the center with COI-free area towards coexisting garnet and in the lower right part of the omphacite grain; (b) high-magnification back scattered electron (BSE)-image of COIs showing the two types of COIs (numbers refer to those in Section 4.4.) and the calcic amphibole lamellae without any associated quartz; (c) BSE-image of coarse COIs from sample SKP31; (d) selected area from (c) (black frame) showing a symplectite vein that cuts across a COI (black circle) thus providing information on the relative timing of COI and symplectite formation; (e) BSE-image of a kyanite grain from CM31/03 with a strong Cr zoning; white dots denote locations of microprobe analyses and numbers next to dots are wt.% Cr2O3; (f) BSE-image of K-feldspar–quartz aggregate enclosed in omphacite affected by retrogressive alteration from CM31/03; mineral abbreviations according to Kretz (1983).

340

J. Konzett et al. / Lithos 106 (2008) 336–350

the O-H dipole in rutile is aligned perpendicular to the c-axis, only the opolarization shows maximum absorption whereas the e-polarization (parallel to the c-axis) shows no absorption at all (Beran and Zemann, 1971). The total absorbance, i.e. the summation of the three orthogonal directions is thus obtained by Atot =Ao +Ao +Ae = 2·Ao. (Libowitzky and Rossman,1996). Data acquisition, background correction and calculation of band areas (Ai) were done using the OPUS 5.5 software from Bruker. In all cases a straight baseline between approx. 3130 and 3520 cm− 1 was subtracted. This wide interval was chosen to include the low-intensity bands surrounding the characteristic strong OH stretching band at ∼3280–3300 cm− 1. H2O concentrations were calculated using the rutilespecific calibration of Maldener et al. (2001). This calibration by nuclear reaction analysis (NRA) resulted in an integrated molar absorption coefficient εi = 38,000 L·mol− 1·cm− 2 which is different from and superior to the general calibration trend of Libowitzky and Rossman (1997). The water contents result from the general equation c(H2O) [wt.%] = 1.8·Ai,tot [cm− 1]/ (D[g/cm3]·εi·d[cm]), or recalculated specifically for rutile (Rossman, 2006) from c(H2O) [wt. ppm]= 0.110·αi,tot [cm−2]. The density used for rutile is 4.25 g/cm3, α =A/d. Garnet spectra were acquired with unpolarized radiation, Atot = Aunpol d 3. Because of a curved and tilted background line and interference fringes caused by the thin, doubly-polished platelets, all spectra were treated by (i) background correction with a cubic function and (ii) Fourier smoothing. The areas of the weak absorption bands at ∼3600–3650 cm− 1 were then used for calculation of H2O contents using the calibration of Bell et al. (1995). The detection limit is estimated at

2 ppm H2O. All these steps were performed with PeakFit 4 by Jandel Scientific. 4. Sample petrography and mineral composition Three eclogite samples were selected for this study. Two of them (CM31/03 from the Pohorje Massif and SKP31 from the Saualpe) contain COIs (Fig. 2a-d). A third sample (CM15/01) from the Pohorje Massif without visible quartz inclusions was chosen for comparison and has already been described by Miller et al. (2005, 2007). Representative and averaged major and trace element analyses of mineral phases from these samples are listed in Tables 1a, 1b, 1c and 2. 4.1. Sample CM31/03 (Sample location N46°25.25′E15°31.29 near Jurisna Vas') This is a kyanite-rich granular-textured eclogite with the highpressure assemblage garnet + omphacite + kyanite + quartz + rutile plus accessory Ca-amphibole and phengite. Calcic amphibole appears as matrix grains interstitial between garnet and clinopyroxene and as inclusions in garnet. In addition, trace amounts of zircon, K-feldspar and orthopyroxene could be identified. K-feldspar forms polycrystalline aggregates with quartz, plagioclase and diopside in garnet and omphacite (Fig. 2f. Orthopyroxene was found only once as a small (10 µm) inclusion in garnet intergrown with quartz and associated with kyanite, rutile and quartz inclusions of the same size range.

Table 1a Averaged and representative compositions of minerals from CM31/03 (2) Omp (clear) 10

(3) Omp (reint)

(4) Grt rim

(5) Grt core

# analyses

(1) Omp (exln) 10

(6) Zo

(7) Ky rim

(8) Ky core

SiO2 [wt.%] TiO2 Cr2O3 Al2O3 Fe2O3 FeOtot MnO MgO CaO Na2O K2O H2O⁎ ∑

55.05(25) 0.13(02) 0.07(06) 8.39(07) – 2.47(09) b0.05 12.12(11) 17.97(12) 4.11(10) b0.05 – 100.31(36)

55.09(18) 0.12(03) 0.15(06) 8.20(10) – 2.47(07) b 0.05 12.19(12) 17.92(14) 4.10(04) b 0.05 – 100.25(29)

55.81 0.13 0.06 8.28 – 2.48 b 0.05 12.15 17.32 3.95 b 0.05 0.09 100.32

40.72 b 0.05 0.21 22.69 0.72 14.96 0.28 13.39 7.47 b 0.05 nd – 100.43

Si a.p.f.u. Ti Cr Al Fe3+ Fe2+ Mn Mg Ca Na K OH Cl ∑cat

1.959(03) 0.003(00) 0.002(02) 0.352(03) – 0.073(03) – 0.643(05) 0.685(04) 0.283(07) –

1.962(04) 0.003(01) 0.004(02) 0.344(04) – 0.074(02) – 0.647(06) 0.684(05) 0.283(03) –

1.980 0.003 0.002 0.346 – 0.074 – 0.643 0.658 0.272

4.002(07)

4.002(04)

3.978

CaEs cats jd di hd en Na–Ti-px

0.000 0.040(03) 0.276(08) 0.573(08) 0.074(03) 0.033(05) 0.007(01)

0.000 0.036(04) 0.277(03) 0.576(07) 0.074(03) 0.034(03) 0.006(02)

0.043 0.019 0.266 0.544 0.074 0.048 0.006

(9) Kfs

(10) Opx

(11) Cam

40.68 b 0.05 0.14 22.83 0.98 14.39 0.27 13.90 7.18 b 0.05 nd – 100.37

39.21 b0.05 0.09 32.29 1.42 – b0.05 b0.05 24.29 b0.05 nd 1.96 99.26

37.67 b0.05 1.71 61.37 0.51 – b0.05 nd nd nd nd – 101.26

37.40 b 0.05 0.05 62.52 0.33 – b 0.05 nd nd nd nd – 100.30

65.89 nd nd 18.51 nd nd nd nd b 0.05 0.31 16.43 – 101.15

56.51 b0.05 b0.05 1.68 0.44 9.58 0.11 31.83 0.32 0.16 b0.05 – 100.58

53.01 0.11 0.07 9.64 3.15 1.19 b0.05 18.61 10.38 2.20 0.48 2.21 101.05

2.992 – 0.012 1.965 0.039 0.919 0.018 1.467 0.588 – –

2.982 – 0.008 1.973 0.054 0.882 0.017 1.519 0.564 – –

8.000

8.000

3.003 – 0.005 2.914 0.081 – – – 1.993 – – 1.000 – 7.997

1.010 – 0.036 1.940 0.010 – – – – – –

1.007 – 0.001 1.983 0.007 – – – – – –

3.007 – – 0.995 – – – – – 0.027 0.957

1.965 – – 0.069 0.011 0.279 0.003 1.650 0.012 0.011 –

2.997

2.998

4.987

4.000

7.208 0.011 0.008 1.545 0.322 0.135 – 3.772 1.512 0.580 0.083 2.000 – 15.175

(1) omphacite host in area with COIs; (2) omphacite without detectable COIs; (3) re-integrated omphacite composition using omphacite host (1) and calcic amphibole composition (11); cpx analyses recalculated to 6 ox + Fetot = FeO; Cam analysis recalculated to 13 cat-(Na + K + Ca); mineral abbreviations according to Kretz (1983) except: CaEs Ca-Eskola pyroxene; cats calcium Tschermaks pyroxene; Na–Ti-px Na–Ti-pyroxene; Phe phengite; ⁎wt.% H2O recalculated based on stoichiometric (OH, F, Cl).

J. Konzett et al. / Lithos 106 (2008) 336–350

341

Table 1b Averaged and representative compositions of minerals from SKP31 and CM15/01 (1) Omp (exln) 13

(2) Omp (clear) 11

(3) Grt (rim)

55.08(24) 0.16(10) b0.05 9.51(12) – 3.77(17) b0.05 10.68(08) 16.44(10) 4.91(07) b0.05

55.45(21) 0.14(05) b 0.05 9.67(18) – 3.59(15) b 0.05 10.84(14) 16.58(21) 5.01(12) b 0.05

100.49(39)

Si Ti Cr Al Fe3+ Fe2+ Mn Mg Ca Na K ∑cat CaEs cats jd di hd en Na–Ti-px

# analyses SiO2 TiO2 Cr2O3 Al2O3 Fe2O3 FeOtot MnO MgO CaO Na2O K2O H2O⁎ ∑

(4) Cam

(5) Cam

101.30(45)

40.42 b 0.05 b 0.05 22.65 0.33 19.34 0.25 9.59 9.09 b 0.05 nd nd 101.67

45.02 0.49 0.05 16.19 6.07 1.59 b0.05 15.20 10.60 3.09 b0.05 2.15 100.45

46.59 0.37 0.05 14.15 4.33 4.00 b 0.05 15.12 10.82 2.97 b 0.05 2.14 100.54

1.959(03) 0.004(03) – 0.399(04) – 0.112(05) – 0.567(04) 0.627(04) 0.339(05) – 4.007(04)

1.957(06) 0.004(01) – 0.402(06) – 0.106(04) – 0.570(07) 0.627(08) 0.343(08) – 4.009(03)

3.000 – – 1.981 0.019 1.201 0.016 1.061 0.723 – – 8.000

6.289 0.052 0.006 2.666 0.638 0.186 – 3.165 1.587 0.837 – 15.423

6.532 0.039 0.006 2.338 0.457 0.469 – 3.160 1.625 0.807 – 15.433

0.000 0.040(03) 0.326(06) 0.474(07) 0.111(05) 0.041(03) 0.008(05)

0.000 0.042(06) 0.329(08) 0.479(08) 0.104(04) 0.039(02) 0.007(03)

(6) Cam

(7) Czo

(8) Omp 12

(9) Grt (rim)

(10) Phe

(11) Czo

(12) Cam

44.85 0.39 b0.05 16.46 2.48 6.01 0.07 14.04 11.02 3.54 0.12 2.13 101.11

39.69 0.07 0.06 31.36 2.59 – b 0.05 0.07 24.29 b 0.05 b 0.05 1.97 100.10

54.51(30) 0.12(06) b 0.05 10.03(18) – 4.64(23) b 0.05 9.42(22) 15.42(51 5.64(27) b 0.05 – 100.04(43)

39.49 b0.05 b0.05 22.31 0.10 21.01 0.46 7.78 9.29 nd nd – 100.43

49.38 0.88 0.05 29.15 – 1.67 b 0.05 2.98 0.05 0.83 10.70 4.49 100.10

38.70 b 0.05 b 0.05 29.11 5.92 – b 0.05 b 0.05 24.06 nd nd 1.96 99.82

42.25 b0.05 b0.05 17.62 4.32 8.73 0.08 10.81 10.56 3.57 b0.05 2.07 100.01

6.309 0.041 – 2.729 0.263 0.707 0.008 2.944 1.661 0.966 0.022 15.648

3.024 0.004 0.004 2.816 0.149 – – – 1.983 – – 7.988

1.960(08) 0.003(02) – 0.425(08) – 0.139(07) – 0.505(10) 0.594(20) 0.393(19) – 4.021(07)

2.998 – – 1.997 0.007 1.332 0.030 0.881 0.756 – – 8.000

3.302 0.041 – 2.280 – 0.114 – 0.280 – 0.048 0.950 7.016

5.994 0.005 – 5.313 0.690 – – – 3.992 – – 15.994

6.120 – – 3.008 0.470 1.057 0.010 2.334 1.639 1.003 – 15.642

(1)–(7) SKP31; (8)–(12) CM15/01; (1) omphacite host in area with COIs; (2) omphacite in area without detectable COIs; (4) Ca-amphibole inclusion in garnet; (5) Ca-amphibole with quartz-blebs; (6) Ca-amphibole coexisting with quartz in COI.

Table 1c Averaged analyses of apatite Sample No. # analyses

CM15/01 11

SKP31

P2O5 CaO SrO FeO MnO MgO Na2O Cl F H2O⁎ ∑ −Cl = O −F = O ∑

42.11(10) 55.23(12) 0.22(05) b 0.05 b 0.05 b 0.05 b 0.05 0.31(03) 1.89(16) 0.81(07) 100.64(17) 0.07(01) 0.80(07) 99.77(21)

42.31 55.46 b 0.05 0.14 0.05 b 0.05 b 0.05 1.44 0.50 1.19 101.09 0.32 0.21 100.56

P Ca Sr Fe Mn Mg Mn Mg Na ∑

3.001(02) 4.981(02) 0.011(02) – – – – – – 4.997(05)

3.002 4.980 – 0.010 0.004 – – – – 4.994

Cl F OH

0.044(04) 0.504(04) 0.453(04)

0.205 0.133 0.665

Wt.% H2O recalculated based on stoichiometric (OH, F, Cl).

Cooling and exhumation has led to the formation of minor symplectitic reaction rims around omphacite, phengite and kyanite. Symplectites around omphacite consist of Na-rich plagioclase + diopside-rich clinopyroxene ± secondary Ca-amphibole intergrowths and those around phengite are formed by biotite + Ca-rich plagioclase ± K-feldspar ± albite rims and kyanite. Kyanite is usually surrounded by two-layered symplectites consisting of a spinel-margarite layer towards kyanite and a plagioclase-Ca-amphibole layer towards coexisting clinopyroxene. In a few cases clinoptilolite may be part of the kyanite symplectites (cf. Song et al., 2003). Garnets in CM31/03 are pyrope-rich solid solutions with a narrow compositional range prp49–51grs15–19alm30–31sps0–1 and do not show any significant compositional zoning (Fig. 3a). Clinopyroxene, too, is homogeneous in its major element composition with an observed range di63–65jd26–28cats02–04acm01–03 when normalized to 6 oxygens and 4 cations and possesses a disordered space group symmetry C2/c (Table 3). On individual clinopyroxene grains, no difference in composition was observed between inclusion-free areas and those with COIs recognizable with the petrographic microscope and/or BSE imaging. On the basis of Fetot = FeO, averaged total cation sums were found to be 4.002 ± 0.004 (n = 10) and 4.002 ± 0.007 (n = 10) respectively. To test potential trace element heterogeneities related to the formation of the COIs, LA-ICP-MS analyses were performed on two clinopyroxene grains containing COI-free and COI-bearing areas. For both types of areas, chondrite-normalized REE plots show bell-shaped patterns with a strong enrichment of the MREE (Eu-Tb) and depletion of the LREE, yielding LaN/SmN and TbN/YbN ranging from 0.006–0.018 and 3.5–27.3 respectively. Total REE concentrations are always low (∑REE = 0.58–1.10 ppm) resulting in a maximum enrichment

342

J. Konzett et al. / Lithos 106 (2008) 336–350

Table 2 Trace element concentrations in ppm in minerals from CM31/03 Mineral Li Sc Ti V Cr Co Rb Sr Y Zr Hf Nb Ta Pb Th U La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Omp grain 1 Clear⁎

Exln⁎⁎

Omp grain 2 Clear

Exln

17.9⁎ 20.8 663 189 855 23.7 b 0.04 15.7 0.67 5.5 0.42 b 0.0008 0.031 0.13 0.014 0.30 b 0.0008 0.005 0.001 0.021 0.043 0.044 0.10 0.033 0.21 0.034 0.046 0.005 0.029 0.004

9.8 14.8 671 171 153 22.0 0.05 20.1 0.55 5.6 0.31 0.004 0.001 0.076 0.001 0.001 0.002 0.028 0.011 0.15 0.13 0.10 0.29 0.051 0.24 0.029 0.040 0.004 0.027 0.003

15.1 12.9 657 146 124 21.7 b 0.033 21.5 0.38 5.8 0.29 0.003 b 0.001 0.14 0.001 0.19 0.002 0.019 0.007 0.085 0.12 0.071 0.24 0.036 0.17 0.019 0.014 0.002 0.006 0.002

9.9 18.0 697 178 325 24.1 0.053 19.6 0.84 5.9 0.30 0.002 b 0.001 0.10 0.05 0.61 0.001 0.006 0.004 0.060 0.086 0.070 0.23 0.038 0.28 0.035 0.068 0.008 0.050 0.006

Cam

Grt Rim

Grt Core

Ky

0.75 6.8 1326 102 461 65.3 0.21 11.2 0.84 4.2 0.27 0.005 b 0.002 0.28 0.001 b 0.001 0.0002 0.006 0.003 0.053 0.068 0.055 0.21 0.043 0.28 0.038 0.072 0.007 0.029 0.002

b0.35 35.3 156 24.6 511 66.9 b0.051 0.041 10.0 3.8 0.045 b0.003 b0.003 0.007 0.009 b0.001 b0.002 0.045 b0.0008 0.027 0.14 0.15 0.73 0.24 1.9 0.41 0.79 0.13 0.84 0.13

b 0.32 42.4 233 26.2 595 68.6 b 0.046 b 0.009 13.7 4.4 0.051 b 0.002 0.002 0.025 0.006 b 0.005 b 0.001 0.001 b 0.001 b 0.001 0.021 0.037 0.33 0.19 1.9 0.55 1.37 0.20 1.2 0.14

b0.47 b0.093 56.8 46.2 455 0.055 b0.067 0.50 0.011 b0.021 b0.001 b0.004 b0.001 b0.018 0.001 0.001 b0.002 0.002 b0.002 0.013 b0.001 b0.004 b0.013 0.001 0.038 0.001 b0.002 b0.0006 0.008 b0.003

⁎Analysis in area without detectable COIs; ⁎⁎ analysis in area with COIs.

b2⁎chondritic abundance for the MREE, whereby the inclusion-free areas show slightly lower ∑REE (0.58–0.81 ppm) compared to the areas with visible inclusions (∑REE= 0.81–1.10) (cf. Fig. 4a,b). All analyzed clinopyroxene spots show small positive Eu anomalies with Eu/Eu⁎ in the range 1.27–2.01. Chondrite-normalized LIL-HFS element spectra of COI-free and COI-bearing areas show a strong but variable enrichment of Th and U in both types of areas without significant differences in the absolute values. In this respect Li is a notable exception with a consistent enrichment in the COI-free portions of the omphacites (Fig. 4d,e). According to the classification of Leake et al. (1997) primary calcic amphiboles are magnesio-hornblendes with 10–11 wt.% Al2O3 and 1.6– 2.4 wt.% Na2O whereas secondary calcic amphiboles in kyanite/ omphacite symplectites are tremolitic hornblendes with approximately 5 wt.% Al2O3 and ≤1.0 wt.% Na2O. REE patterns of primary amphiboles are very similar to those of omphacite and with the exception of Cs, Rb, Ba and Li the same is true for the trace element patterns (Fig. 4c,f). An unusual feature of sample CM31/03 is a strong compositional Cr– Al zoning in kyanite ranging from 0.05 to 1.7 wt.% Cr2O3 observed within a single kyanite grain. In most cases the zoning is patchy and irregular but a more regular concentric zoning with Cr-poor cores and Cr-rich rims may also occur with textures very similar to those described by Yang and Rivers (2001) from upper amphibolite facies metapelites (Fig. 2e). Rutile is present as large matrix grains and as inclusions in garnet, clinopyroxene and kyanite. It contains significant amounts of Cr2O3 (0.35–0.86 wt.%) along with minor amounts of Fe2O3 (0.06–0.33 wt.%) and traces of Nb, Al and Zr. The rutile composition is dependent of the coexisting assemblage with the highest Fe and Cr contents found in rutile inclusions in garnet. 4.2. Sample SKP31 (Saualpe; sample location Kuppler Brunn) This is a quartz-rich granular-textured eclogite with the highpressure assemblage garnet + omphacite + zoisite + calcic amphibole+ quartz + rutile. Minor retrogressive alteration has led to narrow

plagioclase-diopside symplectites that surround or crosscut omphacite. As in sample CM31/03 Ca-amphibole appears as a minor additional phase in two generations with texturally early matrix grains or garnetinclusions and texturally late poikiloblastic intergrowths with quartz (cf. Miller et al., 2007). Compared to the suite of Pohorje eclogites examined, SKP31 is unusually rich in apatite that appears as up to 500 µm sized elongated grains in the matrix and as inclusions in garnet. Most apatite grains contain oriented inclusions of CuFe-sulfide. SKP31 garnets are less pyrope-rich than those from sample CM31/03 but with a similarly narrow compositional range prp35–37grs21–24alm39–41sps00–01. A slight compositional zoning can be observed with decreasing Ca and increasing XMg from core to rim, indicating garnet growth during prograde metamorphism (Fig. 3b). SKP31 clinopyroxenes are homogeneous in composition with di56–57jd29–34cats02–05acm00–04 and possess the ordered P2/n space group symmetry (Table 3). By analogy with CM31/03 both inclusion-bearing and inclusion-free areas in SKP31 show nearly ideal total cation sums of 4.007±0.004 and 4.009±0.003 respectively when normalized to 6 oxygens and Fetot =FeO. As opposed to sample CM31/03, amphiboles of the COIs are large enough to be analyzed with the microprobe. Their composition is very similar to those of primary amphiboles that are present as garnetinclusions or as texturally late poikiloblastic grains interstitial between clinopyroxene and garnet. 4.3. Sample CM15/01 (Sample location N46°24.87′E15°27.473 near Kebelj) This sample is a quartz-rich eclogite with the assemblage garnet+ omphacite + quartz + rutile in addition to minor calcic amphibole + clinozoisite and accessory phengite, kyanite, apatite, zircon and pyrite. Omphacite in this sample shows a compositional range di49–50jd31– 34cats02–04acm02–08 and shows space group symmetry P2/n (Table 3). For a more detailed description of mineral textures and compositions of this sample, the reader is referred to Miller et al. (2005, 2007).

J. Konzett et al. / Lithos 106 (2008) 336–350

Fig. 3. Compositional zoning profiles across garnets (a) from sample CM31/03 and (b) from sample SKP31; bars denote the 2σ-error of an individual microprobe analysis for the respective oxide.

4.4. Textures of COIs and re-integrated pyroxene compositions In sample CM31/03 oriented needles/rods in clinopyroxene are immediately recognizable with the petrographic microscope due to their relatively high density and strong relief compared to the host omphacite. While most omphacite grains show a relatively homogeneous distribution of inclusions, some grains exhibit a zonal distribution with a high density of needles in the cores and without any visible inclusions in the outermost portions to 600× magnification. In some cases, the distribution is patchy without core–rim zonal patterns (Fig. 2a). It is with the BSE-imaging mode of the electron microprobe that the composite nature of the needles is best recognized (cf. Page et al., 2005). Individual quartz–calcic amphibole composite needles are 50–100 µm in length and 1–3 µm in diameter perpendicular to their long axes. Two types of composite aggregates can be observed: (1) parallel intergrowths of quartz and calcic amphibole and (2) “sandwiches” with a calcic amphibole lamella stacked between two quartz lamellae (Fig. 2b) very similar to textures observed by Page et al. (2005). Some of the omphacite grains show two sets of lamellae, one with composite aggregates as described above and a second set consisting of very fine (≤1 µm) lamellae of

343

calcic amphibole without any visible quartz oriented perpendicular to the composite needles (Fig. 2b). The term “oriented” is used here to denote a parallel alignment of the long axes of the calcic amphibole or calcic amphibole + quartz lamellae within the host omphacite without any further implications as to possible crystallographic orientation or genesis (e.g. exsolution). An attempt was made to measure the orientation relationship between the omphacite host and COIs using the electron back scatter diffraction (EBSD) technique in a scanning electron microscope. Unfortunately the size of the COIs turned out to be too small to generate an electron diffraction pattern that could be indexed reliably. Nevertheless the term “oriented” is considered appropriate to distinguish this type of texture from a purely random orientation of the needles/rods by analogy with the use of this term in the literature (e.g. Tsai and Liou, 2000; Zhang et al., 2002; Song et al., 2003). When the omphacites are sufficiently idiomorphic it is possible to recognize that the COI long axes are aligned in planes that are parallel to the host's long axes, the latter usually being identical to the c-axis. These textures are very similar to those reported in the literature (compare Fig. 2a with Fig. 4F of Song et al., 2003 or with Fig. 4a of Zhang et al., 2002). No compositional zoning in omphacite in the vicinity of the inclusions as reported by Page et al. (2005) could be observed. Modal determination using image analysis (see above) of areas up to 350×250 µm yields 1.7–2.3% quartz and 3.7–4.5% calcic amphibole with an average of 1.9% quartz and 4.0% calcic amphibole for omphacite grains with the highest density of inclusions. Using a representative composition of amphibole present as inclusions in garnet (analysis 11 in Tables 1a,b,c; the calcic amphibole lamellae were too small to be analyzed with the electron microprobe) and that of clinopyroxene forming the host of the COIs, a re-integrated clinopyroxene composition with ∑cat = 3.979 equivalent to 4.3 mol% CaEsss is obtained (Tables 1a). There is clearly a certain error associated with this method due to variable amounts of quartz and calcic amphibole exposed to the sample surface depending upon the section plane through the composite aggregates but it is still considered the most reliable method available (cf. Page et al., 2005). In sample SKP31, COIs are well visible with the petrographic microscope, too. Compared to CM31/03, they are coarser (up to 90× 20 µm) and show a lower density (Fig. 2c) which precludes a meaningful compositional reintegration. In a few cases, Fe-rich omphacite may be part of the COIs (Tables 1b). All clinopyroxenes containing COIs are rimmed and/or veined by narrow albite–diopside– calcic amphibole symplectites that may crosscut the COIs (Fig. 2d). Some of the COIs also contain calcite and albite in addition to quartz and calcic amphibole indicating the presence of a mixed H2O–CO2 fluid. The significance of these assemblages will be discussed in Section 7. 5. H2O contents of nominally anhydrous eclogitic phases 5.1. Omphacite In general, IR absorption spectra of omphacites analyzed in the present study show a characteristic absorption band which is composed

Table 3 Unit cell parameters of omphacites from the present study compared to selected literature data Sample-No.

a0 [Å]

b0 [Å]

c0 [Å]

β[°]

V [Å3]

ρ [g/cm3]

Space group

CM15/01 CM31/03 SKP31 SBB-45⁎ 74AM39⁎⁎ 78AM12⁎⁎

9.602(1) 9.622(4) 9.622(1) 9.624(5) 9.621(1) 9.619(1)

8.788(1) 8.799(1) 8.801(1) 8.795(4) 8.805(1) 8.807(1)

5.255(1) 5.243(1) 5.252(1) 5.254(2) 5.254(1) 5.256(1)

106.70(1) 106.44(1) 106.59(1) 106.88(4) 106.58(1) 106.69(1)

424.7 424.9 426.2 425.5 426.6 426.5

3.37 3.35 3.35 3.35 3.33 3.35

P2/n C2/c P2/n C2/c C2/c P2/n

⁎ McCormick (1986); ⁎⁎ Boffa-Ballaran et al. (1998).

344

J. Konzett et al. / Lithos 106 (2008) 336–350

Fig. 4. (a)–(c) Chondrite-normalized REE patterns for omphacite (a and b) and primary calcic amphibole (c) from CM31/03; shaded area in (a) represents compositional range of calcic amphibole as shown in (c); the shaded area in (b) represents the compositional range of COI-bearing omphacite as shown in (a) for comparison; black symbol in (c) denotes concentration below the detection limit; (d)–(f) Chondrite-normalized LIL-HFS-element patterns for omphacite and calcic amphibole from CM31/03; shaded area in (d) represents compositional range of calcic amphibole for comparison as shown in (f); filled symbols and open symbols as in (c).

of two component bands at about 3460 and 3525 cm− 1 (cf. Smyth et al., 1991; Katayama and Nakashima, 2003; Koch-Müller et al., 2004) (Figs. 6– 8). These two bands described as group 1 and group 2 bands by KochMüller et al. (2004, 2007) are typical for omphacites from eclogites (e.g. Katayama and Nakashima, 2003; Koch-Müller et al., 2004, 2007). KochMüller et al. (2004) explained the OH bands of groups 1 and 2 by vibrations of O2–H1…O2 and O2–H2…O2 dipoles with a neighboring vacant M2 site and a neighboring Si substituted by Al, respectively. The OH contents (expressed as ppm H2O) calculated from band groups 1 and 2 are in the range 145–580 ppm H2O with significant variations observed within individual grains as well as between grains on a thin section scale. In sample CM31/03, the OH content of five clinopyroxene grains was analyzed, yielding values of 240–580 ppm H2O (Fig. 5a). These values are slightly lower than those observed by Janák et al. (2007), i.e. 580–760 ppm H2O, however, if band group 3 (see below) is included in the total H2O calculation, the obtained values coincide. Three of the

five grains (grains 1, 2 and 5 in Fig. 5a) show a zonal distribution of COIs and on these grains inclusion-bearing and inclusion-free areas were analyzed for comparison. In addition to band groups 1 and 2, all spots analyzed on CM31/03 clinopyroxenes show at least very weak band components in the range 3600–3670 cm− 1 (band group 3). IR bands in the wavenumber region 3600–3624 cm− 1 have been described by Koch-Müller et al. (2004) and attributed to the presence of (sub) microscopic inclusions of sheet silicates in clinopyroxene. In spectra 2-i-1 and 2-i-2 taken from the inclusion-bearing portion of an omphacite, single sharp bands at approximately 3670 cm− 1 are additionally visible. A polarized IR spectrum taken from spot 2-i-2 (E// elongation of inclusions) and shown in Fig. 5b reveals the presence of 4 bands at 3625, 3647, 3657 and 3673 cm− 1 characteristic of amphiboles of the tremolite–pargasite join (Skogby et al., 1990; Skogby and Rossman, 1991; Della Ventura et al., 2003; Jenkins et al., 2003; Libowitzky and Beran, 2004).

J. Konzett et al. / Lithos 106 (2008) 336–350

345

Fig. 6. Unpolarized IR spectra for 6 spots on two omphacite grains from sample SKP31; meaning of numbers as in Fig. 5.

are in the range 145–460 ppm and may strongly vary within single grains as well as among different grains (Fig. 7). As opposed to spectra from CM31/01 and SKP31, bands of band group 1 are dominating the clinopyroxene IR spectra.

Fig. 5. (a) Unpolarized IR spectra for omphacites from sample CM31/03 (10 spots on 5 grains); bold numbers and numbers in italics refer to H2O concentrations in ppm of omphacite derived from band groups 1 and 2 and for sheet silicates/amphibole (band group 3), respectively; labels for analytical spots: 1-c grain 1 clear = spot without detectable COIs; 1-i grain 1 spot with COI inclusions etc.; (b) comparison of unpolarized and polarized spectra of omphacite with COIs from spot 2-i-2. Note that the unpolarized spectrum is approximately the average of both polarized spectra (which is not necessarily true for pleochroic samples and strong absorbance). The four bands between 3600 to 3700 cm− 1 are characteristic of calcic amphibole of the actinolite– ferroactinolite series (see Section 6.1.).

In case of sample SKP31, the OH content of six clinoproxene grains was determined, yielding concentrations in the range 155–475 ppm H2O (Fig. 6). Compared to IR spectra of clinopyroxenes from CM31/03, bands of band group 3 are very faint and amphibole bands are absent which is consistent with results from optical microscopy. In sample CM15/01 no oriented inclusions could be detected in omphacite with optical microscopy or BSE imaging. OH concentrations

Fig. 7. Unpolarized IR spectra for omphacites from sample CM15/01 free of COIs (9 spots on 3 grains); note that band group 1 dominates the spectrum as opposed to spectra from CM31/03 and SKP31; meaning of numbers as in Fig. 5.

346

J. Konzett et al. / Lithos 106 (2008) 336–350

5.2. Kyanite, rutile and garnet Two kyanite grains analyzed from sample CM31/01 show unpolarized IR spectra with a broad dual band at 3260 and 3275 cm− 1 along with a band triplet consisting of sharp bands at 3385, 3410, and 3440 cm− 1 (Fig. 8). This pattern of bands is typical for kyanite from a wide variety of low- and high-pressure environments (e.g. Rossman and Smyth, 1990; Wieczorek et al., 2004; Bell et al., 2004). A quantitative analysis of the kyanite IR spectra yields 5 or 15 ppm H2O using the calibrations of Libowitzky and Rossman (1997) or Bell et al. (2004), respectively. Four rutile grains from samples CM31/01 and SKP31/01 respectively were analyzed in-situ along with omphacite or kyanite. In case of sample 15/01, 16 individual rutile grains embedded in epoxy were analyzed. All IR absorption spectra show a characteristic single band in the wavenumber region 3280–3300 cm− 1 that can be decomposed into two overlapping components at ∼ 3280 and 3300 cm− 1 (Fig. 9). This main absorption feature (plus occasionally weaker side bands) has been described from natural rutiles in a wide variety of geological settings (e.g. Beran and Zemann, 1971; Rossman and Smyth, 1990; Hammer and Beran, 1991; Vlassopoulos et al., 1993; Maldener et al., 2001; Katayama et al., 2006) and also from synthetic rutiles doped with minor elements (e.g. Bromiley and Hilairet, 2005). Water contents calculated using the Maldener et al. (2001) calibration (see above) range from 108 to 323 ppm H2O in SKP31 and from 787 to 1187 ppm H2O in CM31/03. Rutiles from CM15/01 show a range in H2O concentrations from 425 to 650 ppm with an average of 516 ± 80 ppm (Fig. 9). Due to their very low H2O concentrations, no reliable IR analyses of garnets could be obtained from the doubly-polished rock sections. For this reason, individual handpicked grains from sample CM15/01 were used instead. For 19 garnet grains a range in H2O concentrations of b2 to 14 ppm was obtained with an average of 8 ± 4 ppm for 16 grains that show H2O concentrations above the detection limit of ∼2 ppm. The IR spectra show an extremely weak band at approximately 3650 cm− 1 with an additional shoulder at 3612 cm− 1 that are typical for pyroperich garnets (Fig. 10) (cf. Rossman and Smyth, 1990; Bell and Rossman, 1992; Lu and Kepper, 1997; Withers et al., 1998, and references therein; Katayama et al., 2006; Beran and Libowitzky, 2006, and references therein). Experimental evidence suggests that H2O solution mechanisms in garnet are a complex function of fH2O, fO2, aSiO2 and the major/trace element composition of garnet. Lu and Kepper (1997) observed a continuous increase in H2O solubility in natural pyrope-

Fig. 8. Unpolarized IR spectra of kyanite from sample CM31/03 (2 spots on a single grain); the upper spectrum shows some uncompensated water vapor vibrations.

Fig. 9. Polarized IR spectra (only E perpendicular to the c axis, i.e. the direction with maximum absorbance) from rutile crystals of samples CM31/03, CM15/01, and SKP31 (four grains per sample). Water concentrations in ppm H2O were calculated using the calibration of Maldener et al. (2001); spectra have been vertically offset for clarity and normalized to 1 cm sample thickness for direct comparison.

rich garnet from 1.5 to 10 GPa at 1000 °C and identified the incorporation of isolated OH groups as major solution mechanism in accordance with very high concentrations of Li, Na and B available for charge compensation. These authors tentatively assigned absorption bands at 3640–3660 cm− 1 and at 3600 cm− 1 to OH Li and OH B defects respectively. Withers et al. (1998), on the other hand, found a dramatic decrease in H2O solubility in pure synthetic pyrope between 5 and 6 GPa. These authors proposed hydrogrossular substitution as the dominant H2O solution mechanism and explained the drop in H2O solubility by an increase in the partial molar volume of H2O in the hydrogrossular component compared to that in fluid water, leading to

Fig. 10. Unpolarized IR spectrum of 0.125 mm thick garnet from sample CM15/01 showing the maximum H2O content of 14 ppm.

J. Konzett et al. / Lithos 106 (2008) 336–350

a dehydration of garnet at P N 5 GPa even under H2O-saturated conditions. Garnets from sample CM15/01 have uniformly low Li concentrations of b1 ppm [0.53 ± 0.06 ppm (n = 9); Miller et al. (2007) and unpublished data] too low to charge compensate the highest H2O contents of 10–14 ppm observed. Therefore the major IR absorption band in CM15/01 garnets is unlikely to be associated with OH Li defects. A Ti-related OH defect that is characterized by IR absorption at 3512 cm− 1 (Beran and Libowitzky, 2006) can be excluded as well. 6. Discussion While oriented SiO2 inclusions in clinopyroxene may form by the reaction 2 Ca0.550.5AlSi2O6 = CaAl2SiO6 + 3 SiO2 as originally suggested by Smyth (1980), the additional presence of calcic amphibole (e.g. Terry and Robinson, 2001; Liati et al., 2002; Page et al., 2005) indicates that mechanisms more complex than simple CaEs breakdown must be involved. If a CaEs component in clinopyroxene is involved in the formation of COIs, then this process is more likely to be a cooling than a decompression phenomenon. Experimental results by Konzett et al. (2008) have shown that CaEsss in omphacite strongly increases with increasing temperature and bulk Al content but shows very little pressure dependence. Moreover, these authors found ∼ 4 mol% CaEsss in omphacite coexisting with garnet + kyanite + rutile at 2.5 GPa and 850 °C. A very similar P-T behaviour of CaEsss in clinopyroxene from a pelitic bulk composition was observed by Hermann and Spandler (2007). In view of the fact that many rocks with omphacites containing COIs have reached high temperatures ≥ 750 °C (e.g. Schmädicke and Müller, 2000; Terry et al., 2000; Tsai and Liou, 2000; Katayama et al., 2000; Dobrzhinetskaya et al., 2002; Liati et al., 2002; Janák et al., 2004; Page et al., 2005) these experimental results and field observations are not contradictive. Reintegration of omphacites containing COIs from sample CM31/03 yields 4 mol% CaEs component. By comparison with the experimental data mentioned above this would be consistent with the thermobarometric estimates of the SKP rocks suggesting P-T conditions in the quartz stability field. In order to reconcile the SiO2 exsolution hypothesis with the presence of amphibole, Shau et al. (2005) proposed a two-stage growth mechanism for the formation of COIs found in a UHP eclogite from the Sulu Terrane/eastern China, involving a decompressionrelated exsolution of SiO2 from a clinopyroxene containing CaEsss, followed by joint amphibole + quartz growth as a result of clinopyroxene breakdown along quartz–clinopyroxene grain boundaries during retrograde metamorphism. Page et al. (2005) noted that COIs may simply form by retrogressive breakdown of clinopyroxene without any involvement of a CaEs component. Amphibole lamellae in clino- as well as orthopyroxene from crustal and mantle rocks are not uncommon and are probably much more widespread than documented because in many cases these lamellae are only recognizable using TEM imaging (e.g. Smith, 1977; Veblen and Buseck, 1981, and references therein; Ingrin et al., 1989). Of the possible formation mechanisms proposed by Papike et al. (1969), i.e., retrogressive hydration, exsolution, prograde reaction of amphibole and primary epitaxial growth of pyroxene and amphibole, Veblen and Buseck (1981), on the basis of both chemical and textural evidence, considered hydration reactions proceeding in an open-system to be the most likely. In case of CM31/03, a comparison of trace element concentrations in COI-bearing and COI-free portions of omphacites does not reveal significant differences with the exception of Li that is consistently depleted in the COI-bearing portions (cf. Fig. 4d,e; Table 2). A simple mass balance calculation shows that addition of a few percent of calcic amphibole with b1 ppm Li cannot result in the observed Li depletion. Thus, the depletion indicates open-system behavior of the clinopyroxene at least on a limited scale which is more consistent with alteration than with exsolution (cf. Veblen and Buseck, 1981). In fact

347

both experimental and field evidence suggest that Li is easily mobilized in high temperature aqueous solutions (You et al., 1996; Paquin and Altherr, 2002). As opposed to CM31/03, textures in SKP31 provide constraints on the relative timing of COI formation. Because secondary albite–diopside– calcic amphibole symplectite veins may cut across COIs (Fig. 2d), COI formation must have occurred prior to the symplectite formation. A comparison of amphibole compositions shows that calcic amphiboles forming part of the COIs cannot be distinguished from those of the large and late-stage calcic amphiboles forming poikiloblastic intergrowths with quartz (Table 1b; Miller et al., 2007; their Fig. 3c). In addition, calcic amphibole from COIs in SKP31 shows low but well measurable K2O contents (Tables 1b). A closed system transfer of K from clinopyroxene to amphibole indicative of an exsolution origin is highly unlikely because the geodynamic context (even taking into account a possible UHP history) of the Eastern Alpine eclogites is inconsistent with a subduction into the diamond stability field. Pressures in excess of 5 GPa, however, would be required to generate significant K in clinopyroxene in the virtually Cr-free bulk compositions considered here (cf. Harlow, 1997; Tsujimori and Liou, 2005). Thus, based on the combined textural and chemical information we conclude that COI formation in sample SKP31 is entirely unrelated to the breakdown of a CaEs component in omphacite during exhumation. Instead, we assume that COIs formed by progressive growth under eclogite-facies P-T conditions analogous to the formation of poikiloblastic quartz–calcic amphibole grains. This assumption would be consistent with (1) the identical composition of amphiboles from COIs and poikiloblastic quartz–calcic amphibole intergrowths, (2) the mostly bimineralic nature of COIs and poikiloblastic quartz–calcic amphibole intergrowths, (3) chemical evidence for open-system behavior (H, Li, K) and (4) late-stage primary but not retrogressive formation of both COIs and poikiloblastic quartz–calcic amphibole intergrowths as evidenced by symplectites crosscutting COIs and the rare presence of Fe-rich omphacite as part of the COIs. The presence of albite and calcite in addition to quartz and calcic amphibole in some of the COIs can be explained by secondary alteration of the COIs during symplectite formation. 6.1. The source of fluid for the formation of COIs Whatever the mechanism of formation may be, calcic amphibole growth whether present as part of the COIs or as separate calcic amphibole lamellae requires the presence of a hydrous fluid. Such a fluid can be supplied from an external source but can also be generated in-situ from hydrogen stored in the host clinopyroxene (and/or other nominally anhydrous eclogite phases). It has long been known that clinopyroxene from eclogite and granulite facies rocks can incorporate significant amounts of hydrogen with concentrations usually in the range of several hundred ppm (Smyth et al., 1991; Ingrin and Skogby, 2000; Koch-Müller et al., 2004; Katayama et al., 2006) but reaching even concentrations as high as 1600 ppm (Katayama and Nakashima, 2003). The parameters and crystal chemical mechanisms controlling H incorporation into clinopyroxene are still insufficiently understood. While Katayama and Nakashima (2003) and Katayama et al. (2006) observed an increase in H solubility with increasing pressure, Koch-Müller et al. (2004) found an opposite trend in their samples. Likewise, an experimental study by Bromiley and Keppler (2004) showed a decrease in H solubility in pure jadeite from 450 ppm at 2 GPa to approximately 100 ppm at 10 GPa. Based on these findings Bromiley and Keppler (2004) concluded that bulk composition has a more important effect on the H storage capacity of omphacite than pressure and temperature. Image analysis of inclusion-bearing areas of clinopyroxene from sample CM31/01 yields an average of 4.0% calcic amphibole. To form this amount of amphibole, 880 ppm H2O are required assuming a stoichiometric H2O content of 2.2 wt.% in amphibole (X-ray scans did not show any indication for F or Cl). If clinopyroxene is the sole source

348

J. Konzett et al. / Lithos 106 (2008) 336–350

of H2O and considering its current H2O content of 300–600 ppm, then initial H2O contents on the order of 1200–1500 ppm are required. A comparison of the H2O contents of inclusion-free and inclusionbearing clinopyroxene portions from sample CM31/03 does not yield a clear negative correlation between H2O in clinopyroxene (band groups 1 and 2) and in sheet silicates/amphibole (band group 3 and amphibole bands) indicative of a closed system transfer of H2O from clinopyroxene to amphibole that would be expected in case of an exsolution origin of amphibole. It is only grain 2 where inclusionbearing portions show a decreasing OH defect concentration associated with increasing intensity of band group 3/amphibole bands compared to inclusion-free portions which would be consistent with an in-situ H2O transfer. Such a negative correlation was observed by Andrut et al. (2003) in a hydrothermally formed gem-quality diopside but explained by preferential partitioning of H2O into amphibole in the presence of an external fluid rather than by in-situ redistribution of H2O in a closed system. Thus, the formation of COIs in CM31/03 by open-system alteration of clinopyroxene rather than by exsolution in a closed system involving a CaEs component would be consistent with textures and the data on minor, trace element and H2O concentration of the CM31/03 phases as outlined above. A possible explanation for the difference in space group between CM31/03 (disordered C2/c) and SKP31 and CM15/01 (ordered P2/n) is the lower jadeite-content of CM31/03 omphacites that allows a cooling of the pyroxene within the C2/c stability field to lower temperatures (cf. Rossi 1988). 6.2. Water storage capacity of the SKP-eclogites Due to the frequent presence of water in the typical eclogite phases omphacite, garnet and rutile (e.g. Skogby et al., 1990; Smyth et al., 1991; Beran et al., 1993; Bell et al., 1995; Koch-Müller et al., 2004) even eclogites devoid of hydrous phases are capable of storing and transporting significant amounts of H2O into the mantle. Katayama et al. (2006) for example reported bulk-rock water contents of 460 ± 30 and 300 ± 20 ppm for crustal eclogites with garnet + omphacite + coesite/ quartz + rutile with a tendency for increasing OH in both omphacite and garnet with increasing pressure. This is in agreement with data by Xia et al., 2005 and Zhao et al. (2007) who found up to 1735 and 1144 ppm H2O respectively in garnets from Chinese UHP-eclogites. OH-concentrations in rutile and comphacite from the SKP eclogites are well comparable with data reported by Katayama et al. (2006) for eclogites from the Kokchetav massif that reached similar P-T conditions. The OH-content of garnet from CM15/01, however, is at the lowermost end of the concentration range reported for garnets by Katayama et al. (2006). This is thought to be true for samples CM31/03 and SKP31, too, in view of the similarity of IR garnet spectra from the doubly-polished thin sections. Thus, the relative order of H2O contents in NAMs of the SKP eclogites is RtN Cpx≫ Grt (CM15/01); RtN Cpx ≫ Ky ∼ Grt (CM31/03); ∼ 0.03–0.04 (Table 4). This relative Rt∼ Cpx≫ Grt (SKP31) with DGrt/Omp H2O order is similar to that reported by Katayama et al. (2006) for sample A530 but different from their samples F430 and C230 (CpxN RtN Grt), the latter with P-T conditions and assemblage similar to that of SKP31 and CM31/03. Considering the poor present state of knowledge on the interplay between mineral composition and H2O-incorporation into NAMs and on the influence of a reduced water activity in high P-T fluids on H2O partitioning between NAMs, no compelling reason can be given for the unusually low H2O contents in the SKP garnets. Likewise, because the quantitative determination of H2O in garnet from crustal eclogites has been almost entirely restricted to UHP-assemblages (Rossman et al., 1989; Su et al., 2002; Xia et al., 2005; Zhao et al., 2007; ) a comparison of the SKP data with those from more common eclogites from the quartz stability field is not possible. Based on the H2O contents of hydrous and nominally anhydrous minerals, modal analysis of the SKP samples was used to calculate approximate bulk-rock water contents along with the relative proportions of the contributing phases. The results are ∼750 ppm, ∼1000 and

Table 4 H2O contents of mineral phases from SKP eclogites Modal proportion

H2O content

Contribution to bulk H2O

%

ppm

ppm

%

CM31/03 Grt Omp Ky Cam (COI) Cam (prim) Rt ∑

23 65 7 3 0.6 0.6

b10 390 ± 120 10 22000 22000 950 ± 170

b2 254 0.7 660 132 6

b 0.2 24 0.07 63 13 0.6 1053

SKP31 Grt Omp Czo Cam Rt Qtz ∑

40 37 9 1 3 10

b10 370 ± 90 19800 22000 230 ± 90 nd

b4 137 1782 220 7 –

b 0.2 − 6 83 10 0.3 – 2146

CM15/01 Grt Omp Czo Cam Rt Qtz Ap ∑

44 43 0.8 2 2 8 0.2

8±4 281 ± 105 19800 22000 516 ± 80 nd 8100 ± 700

4 121 158 440 10 – 16

0.5 16 21 59 1 – 2 749

Cam (COI): calcic amphibole as part of COIs; Cam (prim) primary calcic amphibole present as large matrix grains.

∼2200 ppm respectively (Table 4) and are well comparable with the data reported by Katayama et al. (2006). It is interesting to note that in CM31/ 03 calcic amphibole lamellae of the COIs contribute 63% to the bulk-rock water content and thus are by far the most important H2O carriers in this sample. By comparison, primary calcic amphibole contributes 13%, whereas the nominally anhydrous phases contribute ∼26% (Table 4). In summary, the variations in H2O concentrations of NAMs observed in the SKP eclogites and a comparison with data from the literature support the conclusion reached by earlier studies (e.g. Andrut et al., 2003; Bromiley and Keppler 2004, Koch-Müller et al., 2004) that the H2O concentration of NAMs and its partitioning in eclogitic assemblages is mainly controlled by fluid and solid phase composition and only to a minor extent by pressure and temperature. 7. Conclusions Textural and mineral chemical data indicate that oriented calcic amphibole-quartz inclusions in omphacite from the Koralpe–Saualpe– Pohorje eclogites of the Eastern Alps did not form by a closed system exsolution process. In case of sample CM31/03, the presence of both COIs and associated sets of calcic amphibole lamellae without any coexisting phases points to a multistage process whereby the involvement of a Ca-Eskola component cannot be ruled out. The calcic amphibole lamellae carry approximately 60% of the bulk water content of CM31/03. Reintegration of COIs and the omphacite host by image analysis yields ∼ 4 mol% Ca-Eskola component—an amount that can be present in omphacite at pressures well below 3 GPa (cf. Konzett et al., 2008), i.e. within the quartz stability field. This is consistent with thermobarometric results and the fact that only quartz has been found so far as inclusions in zircon from the SKP rocks (Sassi et al., 2004; Miller et al., 2005; Miller and Konzett, 2005). The presence of Kfeldspar–quartz intergrowths can be explained by passive enrichment of a K-feldspar component during progressive breakdown of plagioclase (cf. Yang et al., 1998). IR analyses show that omphacite is a

J. Konzett et al. / Lithos 106 (2008) 336–350

potential H2O source for amphibole formation. The lacking negative correlation between H2O contents in clinoproxene (obtained from IR band groups 1 + 2) and in sheet silicates/amphibole (from IR band group 3), however, is inconsistent with a closed system redistribution of H2O from clinopyroxene to amphibole. An open-system growth of amphibole by alteration is further indicated by minor and trace element mineral data. In sample SKP31, COIs are thought to represent late-stage eclogite-facies quartz + calcic amphibole growth analogous to the formation of poikiloblastic quartz + calcic amphibole grains present in the same sample. There is no compelling reason to invoke the former presence of a Ca-Eskola component in SKP31 omphacite. Both omphacite and rutile from the SKP eclogites contain significant but variable concentrations of H2O which are well within the range reported for crustal HP-eclogites (Katayama et al., 2006). Garnets, however, have very low H2O contents close to or below the values of 0.03–0.04. detection limit, resulting in very low DGrt/Omp H2O Image- combined with IR analyses show that between ∼ 6 and ∼ 25% of the bulk H2O content of the three samples investigated is stored in NAMs, however without any significant contribution of garnet and rutile due to a negligible H2O concentration and modal amount respectively. In addition to variable H2O contents of NAMs in individual samples, the relative order of H2O contents is varying, too. This variability demonstrates in line with previous studies that it is water activity and mineral composition and to a much lesser extent pressure and temperature that play a key role in the ability of eclogitic NAMs to incorporate OH defects in their structures and thus to store and transport H2O into the mantle. Acknowledgements We are greatly indebted to Florian Heidelbach from the Bavarian Research Institute of Experimental Geochemistry and Geophysics, University of Bayreuth, for his attempts to provide EBSD-data for the COIs. A. Wagner from the University of Vienna kindly prepared grain mounts of rutile and garnet for IR-measurements. Reviews by Tatsuki Tsujimori and an anonymous reviewer helped to improve the manuscript. Their support is gratefully acknowledged. This study was conducted under the University of Innsbruck, Faculty of Geo and Atmospheric Sciences' research program “geodynamics–geomaterials“. References Andrut, M., Brandstätter, F., Beran, A., 2003. Trace hydrogen zoning in diopside. Mineralogy and Petrology 78, 231–241. Bell, D.R., Rossman, G.R., 1992. The distribution of hydroxyl in garnets from the subcontinental mantle of southern Africa. Contributions to Mineralogy and Petrology 111, 161–178. Bell, D.R., Ihinger, P.D., Rossman, G.R., 1995. Quantitative analysis of trace OH in garnet and pyroxenes. American Mineralogist 80, 465–474. Bell, D.R., Rossman, G.R., Maldener, J., Endisch, D., Rauch, F., 2004. Hydroxide in kyanite: a quantitative determination of the absolute amount and calibration of the IR spectrum. American Mineralogist 89, 998–1003. Beran, A., Zemann, J., 1971. Messung des Ultrarot-Pleochroismus von Mineralen. XI. Der Pleochroismus der OH-Streckfrequenz in Rutil, Anatas, Brookit und Cassiterit. Tschermaks Mineralogische und Petrographische Mitteilungen 15, 71–80. Beran, A., Libowitzky, E., 2006. Water in natural mantle minerals II: Olivine, garnet and accessory minerals. In: Keppler, H., Smyth, J.R. (Eds.), Water in Nominally Anhydrous Minerals. Reviews in Mineralogy and Geochemistry, vol. 62. Geochemical Society, Mineralogical Society of America, pp. 169–191. Beran, A., Langer, K., Andrut, M., 1993. Single crystal infrared spectra in the range of OH fundamentals of paragenetic garnet, omphacite and kyanite in an eclogitic mantle xenolith. Mineralogy ad Petrology 48, 257–268. Boffa-Ballaran, T., Carpenter, M.A., Chiara Domeneghetti, M., Tazzoli, V., 1998. Structural mechanisms of solid solution and cation ordering in augite-jadeite pyroxenes: I. a macroscopic perspective. American Mineralogist 83, 419–433. Bromiley, G.D., Keppler, H., 2004. An experimental investigation of hydroxyl solubility in jadeite and Na–rich clinopyroxenes. Contributions to Mineralogy and Petrology 147, 189–200. Bromiley, G.D., Hilairet, N., 2005. Hydrogen and minor element incorporation in synthetic rutile. Mineralogical Magazine 69, 345–358. Della Ventura, G., Hawthorne, F.C., Robert, J.-L., Iezzi, G., 2003. Synthesis and infrared spectroscopy of amphiboles along the tremolite–pargasite join. European Journal of Mineralogy 15, 341–347.

349

Day, H.W., Mulcahy, S.R., 2007. Excess silica in omphacite and the formation of free silica in eclogite. Journal of Metamorphic Geology 25, 37–50. Dobrzhinetskaya, L.F., Schweinehage, R., Massonne, H.-J., Green, H.W., 2002. Silica precipitates in omphacite from eclogite at Alpe Arami, Switzerland: evidence for deep subduction. Journal of Metamorphic Geology 20, 481–492. Gayk, T., Kleinschrodt, R., Langosch, A., Seidel, E., 1995. Quartz exsolution in clinopyroxene of high-pressure granulite from the Münchberg Massif. European Journal of Mineralogy 7, 1217–1220. Hammer, V.M.F., Beran, A., 1991. Variations in the OH-concentration of rutiles from different geological environments. Mineralogy ad Petrology 45, 1–9. Harlow, G.E., 1997. K in clinopyroxene at high pressure and temperature: an experimental study. American Mineralogist 82, 259–269. Hermann, J., Spandler, K.J., 2007. Sediment melts at sub-arc depths: an experimental study. Journal of Petrology. doi:10.1093/petrology/egm073. Ingrin, J., Skogby, H., 2000. Hydrogen in nominally anhydrous upper-mantle minerals: concentration levels and implications. European Journal of Mineralogy 12, 543–570. Ingrin, J., Latrous, K., Doukhan, J.-C., Doukhan, N., 1989. Water in diopside: an electron microscopy and infrared spectroscopy study. European Journal of Mineralogy 1, 327–341. Janák, M., Froizheim, N., Lupták, B., Vrabec, M., Krogh Ravna, E.J., 2004. First evidence for ultrahigh-pressure metamorphism of eclogites in Pohorje, Slovenia: Tracing deep continental subduction in the Eastern Alps. Tectonics. doi:10.1029/2004TC001641. Janák, M., Froitzheim, N., Vrabec, M., Krogh Ravna, E.J., Hoog, J.C.M., 2006. Ultrahighpressure metamorphism and exhumation of garnet peridotite in Pohorje, Eastern Alps. Journal of Metamorphic Geology 24, 19–31. Janák, M., Skogby, H., Broska, I., 2007. Water content and composition of omphacite in UHP kyanite eclogite from Pohorje, Eastern Alps. Programme with Abstracts of the 6th ECMS, Stockholm, vol. 46. Jarosevich, E., Nelen, J.A., Norberg, J.A., 1980. Reference samples for electron microprobe analysis. Geostandards Newsletters 4, 43–47. Jenkins, D.M., Bozhilov, K.N., Ishida, K., 2003. Infrared and TEM characterization of amphiboles synthesized near the tremolite–pargasite join in the ternary system tremolite–pargasite–cummingtonite. American Mineralogist 88, 1104–1114. Katayama, I., Nakashima, S., 2003. Hydroxyl in clinopyroxene from the deep subducted crust: evidence for H2O transport into the mantle. American Mineralogist 88, 229–234. Katayama, I., Parkinson, C.D., Okamoto, K., Nakajima, Y., Maruyama, S., 2000. Supersilicic clinopyroxene and silica exsolution in UHPM eclogite and pelitic gneiss from the Kokchetav massif, Kazakhstan. American Mineralogist 85, 1368–1374. Katayama, I., Nakashima, S., Yurimoto, H., 2006. Water content in natural eclogite and implication for water transport into the deep upper mantle. Lithos 86, 245–259. Koch-Müller, M., Matsyuk, S.S., Wirth, R., 2004. Hydroxyl in omphacites and omphacitic clinopyroxenes of upper mantle to lower crustal origin beneath the Siberian platform. American Mineralogist 89, 921–931. Koch-Müller, M., Abs-Wurmbach, I., Rhede, D., Kahlenberg, V., Matsyuk, S., 2007. Dehydration experiments on natural omphacites: qualitative and quantitative characterization by various spectroscopic methods. Physics and Chemistry of Minerals 34, 663–678. Konzett, J., Frost, D.J., Proyer, A., Ulmer, P., 2008. The Ca-Eskola component in eclogitic clinopyroxene as a function of pressure, temperature and bulk composition: an experimental study to 15 GPa with possible implications for the formation of oriented SiO2-inclusions in omphacite. Contributions to Mineralogy and Petrology 155, 215–228. Kretz, R., 1983. Symbols for rock-forming minerals. American Mineralogist 68, 277–279. Leake, B.E., Woolley, A.R., Arps, C.E.S., Birch, W.D., Gilbert, M.C., Grice, J.D., Hawthorne, F.C., Kato, A., Kisch, H.J., Krivovichev, V.G., Linthout, K., Laird, J., Mandarino, J.A., Maresch, W.V., Nickel, E.H., Rock, N.M.S., Schumacher, J.C., Smith, D.C., Stephenson, N.C.N., Ungaretti, L., Whittaker, E.J.W., Youzhi, G., 1997. Nomenclature of amphiboles: report of the subcommittee on amphiboles of the International Mineralogical Association, Commission on New Minerals and Mineral Names. American Mineralogist 82, 1019–1038. Liati, A., Gebauer, D., Wysoczanski, R., 2002. U-Pb SHRIMP-dating of zircon domains from UHP garnet-rich mafic rocks and late pegmatoids in the Rhodope zone (N Greece); evidence for Early Cretaceous crystallization and Late Cretaceous metamorphism. Chemical Geology 184, 281–299. Libowitzky, E., Rossman, G.R., 1996. Principles of quantitative absorbance measurements in anisotropic crystals. Physics and Chemistry of Minerals 23, 319–327. Libowitzky, E., Rossman, G.R., 1997. An IR absorption calibration for water in minerals. American Mineralogist 82, 1111–1115. Libowitzky, E., Beran, A., 2004. IR spectroscopic characterisation of hydrous species in minerals. In: Beran, A., Libowitzky, E. (Eds.), Spectroscopic Methods in Mineralogy. EMU Notes in Mineralogy, vol. 6. Eötvös University Press, Budapest, pp. 227–279. Lu, R., Kepper, H., 1997. Water solubility in pyrope to 100 kbar. Contributions to Mineralogy and Petrology 129, 35–42. Maldener, J., Rauch, F., Gavranic, M., Beran, A., 2001. OH absorption coefficients of rutile and cassiterite deduced from nuclear reaction analysis and FTIR spectroscopy. Mineralogy and Petrology 71, 21–29. McCormick, T.C., 1986. Crystal-chemical aspects of non-stoichiometric pyroxenes. American Mineralogist 71, 1434–1440. Miller, C., 1990. Petrology of the type locality eclogites from the Koralpe and Saualpe (Eastern Alps), Austria. Schweizerische Mineralogische und Petrographische Mitteilungen 70, 287–300. Miller, C., Thöni, M., 1997. Eo-Alpine eclogitization of Permian MORB-type gabbros in the Koralpe (Austria): new petrological, geochemical and geochronological data. Chemical Geology 137, 283–310. Miller, C., Konzett, J., 2005. Comment on ”First evidence for ultrahigh-pressure metamorphism of eclogites in Pohorje, Slovenia: tracing deep continental subduction in the eastern Alps”. Tectonics 24. doi:10.1029/2004TC001765.

350

J. Konzett et al. / Lithos 106 (2008) 336–350

Miller, C., Mundil, R., Thöni, M., Konzett, J., 2005. Refining the timing of eclogite metamorphism: a geochemical, petrological, Sm–Nd and U–Pb case study from the Pohorje Mountains, Slovenia (Eastern Alps). Contributions to Mineralogy and Petrology 150, 70–84. Miller, C., Zanetti, A., Thöni, M., Konzett, J., 2007. Eclogitization of gabbroic rocks: redistribution of trace elements and Zr in rutile thermometry in an Eo-Alpine subduction zone (Eastern Alps). Chemical Geology 239, 96–123. Page, F.Z., Essene, E.J., Mukasa, S.B., 2005. Quartz exsolution in clinopyroxene is not proof of ultrahigh pressures: evidence from eclogites from the Eastern Blue Ridge, Southern Appalachians, U.S.A. American Mineralogist 90, 1092–1099. Papike, J.J., Ross, M., Clark, J.R., 1969. Crystal-chemical characterization of clinoamphiboles based on five new structure refinements. Mineralogical Society of America Special Paper 2, 117–136. Paquin, J., Altherr, R., 2002. Subduction-related lithium metasomatism during exhumation of the Alpe Arami ultrahigh-pressure garnet peridotite (Central Alps, Switzerland). Contributions to Mineralogy and Petrology 143, 623–640. Rossi, G., 1988. A review of the crystal-chemistry of clinopyroxenes in eclogites and other high pressure rocks. In: Smith, D.C. (Ed.), Eclogites and eclogite-facies rocks. Elsevier, Amsterdam, pp. 237–270. Rossman, G.R., 2006. Analytical methods for measuring water in nominally anhydrous minerals. In: Keppler, H., Smyth, J.R. (Eds.), Water in Nominally Anhydrous Minerals Reviews in Mineralogy and Geochemistry, vol. 62. Geochemical Society, Mineralogical Society of America, pp. 1–28. Rossman, G.R., Smyth, J.R., 1990. Hydroxyl contents of accessory minerals in mantle eclogites and related rocks. American Mineralogist 75, 775–780. Rossman, G.R., Beran, A., Langer, K., 1989. The hydrous component of pyrope from the Dora Maira Massif, Western Alps. European Journal of Mineralogy 1, 151–154. Sassi, R., Mazzoli, C., Miller, C., Konzett, J., 2004. Geochemistry and metamorphic evolution of the Pohorje Mountain eclogites from the easternmost Austroalpine basement, Eastern Alps, Slovenia. Lithos 78, 235–261. Schmädicke, E., Müller, W.F., 2000. Unusual exsolution phenomena in omphacite and partial replacement of phengite by phlogopite + kyanite in an eclogite from the Erzgebirge. Contributions to Mineralogy and Petrology 139, 629–642. Shatsky, V.S., Sobolev, N.V., Stenina, N.G., 1985. Structural peculiarities of pyroxenes from eclogites. Terra Cognita 5, 436–437. Shau, Y.-H., Tsai, H.–C., Liu, Y.-H., Yu, S.-C., Yang, J., Xu, C., 2005. Transmission electron microscopic study of quartz rods with intergrown amphibole within omphacite in eclogites from the Sulu ultrahigh-pressure metamorphic terrane, eastern China. 7th International Eclogite Conference Abstracts. Mitteilungen der Österreichischen Mineralogischen Gesellschaft, vol. 150, p. 139. Sheldrick, G.M., 2008. A short history of SHELX. Acta Crystallographica A64, 112–122. Skogby, H., Rossman, G.R., 1991. The intensity of amphibole OH bands in the infrared absorption spectrum. Physics and Chemistry of Minerals 18, 64–68. Skogby, H., Bell, D.R., Rossman, G.R., 1990. Hydroxide in pyroxene: variations in the natural environment. American Mineralogist 75, 764–774. Smith, P.P.K., 1977. An electron microscopic study of amphibole lamellae in augite. Contributions to Mineralogy and Petrology 59, 317–322. Smith, D.C., 1984. Coesite in clinopyroxene in the Caledonides and its implications for geodynamics. Nature 310, 641–644. Smyth, J.R., 1980. Cation vacancies and the crystal chemistry of breakdown reactions in kimberlitic omphacites. American Mineralogist 65, 1185–1191. Smyth, J.R., Bell, D.R., Rossman, G.R., 1991. Incorporation of hydroxyl in upper-mantle clinopyroxenes. Nature 351, 732–735. Song, S.G., Yang, J.S., Xu, Z.Q., Liou, J.G., Shi, R.D., 2003. Metamorphic evolution of the coesite-bearing ultrahigh-pressure terrane in the North Qaidam, Northern Tibet, NW China. Journal of Metamorphic Geology 21, 631–644.

Stampfli, G.M., Mosar, J., 1997. The plate tectonics of the western Tethyan regions. IGCP 369, comparative evolution of peri-Tethyan rift basins, Barcelona, pp. 1–6. Su, W., You, Z.D., Cong, B.L., Ye, K., Zhong, Z.Q., 2002. Cluster of water molecules in garnet from ultrahigh-pressure eclogite. Geology 30, 611–614. Terry, M.P., Robinson, P., 2001. Evidence for supersilicic pyroxene in an UHP kyanite eclogite, Western Gneiss Region, Norway. Eleventh Annual V.M. Goldschmidt Conference, abstract 3842. Terry, M.P., Robinson, P., Krogh Ravna, E.J., 2000. Kyanite eclogite thermobarometry and evidence for thrusting of UHP over HP metamorphic rocks, Nordøyane, Western Gneiss Region, Norway. American Mineralogist 85, 1637–1650. Thöni, M., 2002. Sm–Nd isotope systematics in garnet from different lithologies (Eastern Alps): age results, and an evaluation of potential problems for Sm–Nd garnet chronometry. Chemical Geology 185, 255–281. Tiepolo, M., Bottazzi, P., Palenzona, M., Vannucci, R., 2003. A laser probe coupled with ICP-double-focusing sector–field mass spectrometer for in situ analysis of geological samples and U–Pb dating of zircon. Canadian Mineralogist 41, 259–272. Tsai, C.-H., Liou, J.G., 2000. Eclogite-facies relics and inferred ultrahigh-pressure metamorphism in the North Dabie Complex, central China. American Mineralogist 85, 1–8. Tsujimori, T., Liou, J.G., 2005. Low-pressure and low-temperature K-bearing kosmochloric diopside from the Osayama serpentinite mélange, SW Japan. American Mineralogist 90, 1629–1635. Veblen, D.R., Buseck, P.R., 1981. Hydrous pyriboles and sheet silicates in pyroxenes and uralites: intergrowth microstructures and reaction mechanisms. American Mineralogist 66, 1107–1134. Vlassopoulos, D., Rossman, G.R., Haggerty, S.E., 1993. Coupled substitution of H and minor elements in rutile and the implications of high OH contents in Nb- and Crrich rutile from the upper mantle. American Mineralogist 78, 1181–1191. Wieczorek, A., Libowitzky, E., Beran, A., 2004. A model for the OH defect incorporation in kyanite based on polarized IR spectroscopic investigations. Schweizerische Mineralogische und Petrographische Mitteilungen 84, 333–343. Withers, A.C., Wood, B.J., Carroll, M.R., 1998. The OH content of pyrope at high pressure. Chemical Geology 147, 161–171. Xia, Q.-K., Sheng, Y.-M., Yang, X.-Z., Yu, H.-M., 2005. Heterogeneity of water in garnets from UHP eclogites, eastern Dabieshan, China. Chemical Geology 224, 237–246. Yang, P., Rivers, T., 2001. Chromium and manganese zoning in pelitic garnet and kyanite: Spiral, overprint, and oscillatory (?) zoning patterns and the role of growth rate. Journal of Metamorphic Geology 19, 455–474. Yang, J., Godard, G., Smith, D.C., 1998. K-feldspar-bearing coesite pseudomorphs in an eclogite from Lanshantou (Eastern China). European Journal of Mineralogy 10, 969–985. You, C.-F., Castillo, P.R., Gieskes, J.M., Chan, L.H., Spivack, A.J., 1996. Trace element behavior in hydrothermal experiments: implications for fluid processes at shallow depths in subduction zones. Earth and Planetary Science Letters 140, 41–52. Zhang, L., Ellis, D.F., Jiang, W., 2002. Ultrahigh-pressure metamorphism in western Tianshan, China: Part I. Evidence from inclusions of coesite pseudomorphs in garnet and from quartz exsolution lamellae in omphacite in eclogites. American Mineralogist 87, 853–860. Zhang, L., Song, S., Liou, J.G., Ai, Y., Li, X., 2005. Relict coesite exsolution in omphacite from Western Tianshan eclogites, China. American Mineralogist 90, 181–186. Zhao, Z-F., Chen, B., Zheng, Y-F., Chen, R-X.., Wu, Y-B., 2007. Mineral oxygen isotope and hydroxyl content changes in ultrahigh-pressure eclogite-gneiss contacts from Chinese Continental Scientific Drilling Project cores. Journal of Metamorphic Geology 25, 165–186.