Journal of Petrology Advance Access published online on January 7, 2009
Journal of Petrology, doi:10.1093/petrology/egn075
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Chloritoid-Bearing Mineral Assemblages in High-Pressure Metapelites from the Bughea Complex, Leaota Massif (South Carpathians)
bu1
1Geological Institute of Romania, 1 Caransebe
St., RO-012271 Bucharest 32, Romania
2Institut Für Mineralogie und Kristallchemie, Azenbergstr. 18, D-70174 Stuttgart, Germany
Received January 11, 2008; Revised typescript accepted December 3, 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL BACKGROUNDBULK-ROCK COMPOSITIONAL DATAMINERAL ASSEMBLAGES AND TEXTURAL...MINERAL COMPOSITIONS AND ZONINGP-T CONDITIONSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
High-Mg chloritoid (XMg = 0·40–0·47) and relatively high-Mg staurolite (XMg = 0·25–0·28) coexisting with kyanite and garnet were identified in a mica–garnet-rich rock associated with very high-pressure eclogites in the Bughea Complex of the Leaota Massif (South Carpathians). Major and trace element geochemical data for both fresh eclogites and associated rocks which represent a metasomatic or retrograde alteration rind of the eclogites, indicate a pelitic precursor. Magnesian chloritoid was found as inclusions in garnet as part of a chloritoid–kyanite–garnet assemblage which is indicative of high-pressure conditions. The host garnet shows a typically prograde chemical zoning pattern. The chloritoid-bearing assemblage is confined to the inner part of the garnet porphyroblasts, whereas the matrix assemblage in equilibrium with Mg-rich garnet rims has exceeded the thermal stability limit of chloritoid. Pressure–temperature pseudosections for simplified compositions approaching the rock bulk-chemistry show a high-pressure field for the identified chloritoid-bearing assemblage in good agreement with pressure–temperature estimates in the CFMASH and KCFMASH chemical subsystems using analysed mineral compositions. The derived pressure–temperature path is clockwise, indicating overprinting during exhumation from 1·8 GPa and 580°C to 1·15 GPa and 620°C, at a water activity approaching aH2O = 1. These conditions were attained in a subduction mélange indicating transient thermal perturbations of a subduction channel.
KEY WORDS: high-pressure metapelite; Mg-rich chloritoid; P–T path; P–T pseudosection; very high-pressure eclogite
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDBULK-ROCK COMPOSITIONAL DATAMINERAL ASSEMBLAGES AND TEXTURAL...MINERAL COMPOSITIONS AND ZONINGP-T CONDITIONSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
Following the studies by Abraham & Schreyer (1976
), Schreyer & Baller (1977), Chopin (1981), and Chopin & Schreyer (1983), new pathways opened in the hitherto uncharted territories of high-pressure (HP) mineral assemblages in metamorphosed pelitic rocks. As a direct consequence, the pressure–temperature (P–T) field of interest for rocks with pelitic bulk compositions expanded significantly, allowing the full geodynamic significance of previously inconceivable high-pressure histories recorded in rocks of crustal origin to be appreciated (Schreyer, 1988a, 1988b).
A wealth of index mineral species or mineral assemblages, critical in documenting HP metamorphic conditions, as well as their phase relationships in P–T space resulted from new field and experimental data acquired at a sustained rate. Experimentally determined stability fields and phase relationships of the Mg end-members of common rock-forming silicates such as magnesiocarpholite (Chopin & Schreyer, 1983), magnesiochloritoid (Chopin & Schreyer, 1983; Fockenberg, 1995), and magnesiostaurolite (Fockenberg, 1998), or assemblages such as talc–kyanite, talc–phengite (Massonne & Schreyer, 1989), and pyrope–coesite (Chopin, 1984) provided important insights in improving understanding and quantification of the HP conditions sustained by Mg-rich pelites. Nevertheless, the incidence of the relevant phases and assemblages mentioned above is often restricted to particular compositional and P–T conditions, such as low metamorphic grade for carpholite, silica-undersaturated compositions for magnesiochloritoid and magnesiostaurolite (Fockenberg, 1998), and/or unusually Cr-rich compositions for Mg-rich staurolite (Gil Ibarguchi et al., 1991; Costin & Luffi, 2001), or high oxygen fugacity for Fe3+-bearing phases such as yoderite (Fockenberg & Schreyer, 1991).
In normal iron-rich metapelites, the low- to medium-temperature formation of almandine-rich garnet (often containing substantial amounts of Ca and Mn) exerts a powerful screening effect on the appearance of other Fe–Mg phases with more restricted stability fields, leading to a wide P–T stability field of monotonous assemblages such as phengitic mica–garnet–kyanite at high pressures (see Massonne et al., 2007; Caddick & Thompson, 2008). The presence of the unusual chloritoid–garnet–kyanite assemblage in the Saxothuringian of the Bohemian Massif (Konopásek, 2001), the Alps (Pognante et al., 1980; Stöckhert et al., 1997; Tropper et al., 1999), and the Kokchetav Massif in northern Kazakhstan (Massonne & Schreyer, 1989), as well as their own identification of this assemblage in the Raspas Complex, Ecuadorian Andes, led Gabriele et al. (2003) to claim it as a key eclogite-facies assemblage. Calculations in the FMASH system indicate a minimum pressure of 1·2 GPa for this assemblage (Konopásek, 2001); however, the presence of additional components such as CaO and MnO fractionating into garnet and widening its stability field could possibly lower the above-mentioned minimum pressure limit. It follows therefore that the composition of the coexisting phases rather than the mere presence of the assemblage would offer a pathway to constraining the HP evolution of the rocks in question. A key role is apparently played by pressure-sensitive Mg incorporation in chloritoid, as is apparent from the HP stability field of Mg-chloritoid (Chopin & Schreyer, 1983) and experimental Fe–Mg partitioning data for the garnet–chloritoid pair (Koch-Müller & Wirth, 2001). The equally HP character of the chloritoid–garnet–kyanite assemblage in Na-bearing metapelites has been documented by Coggon & Powell (2002) and Wei & Powell (2004).
So far only a restricted number of natural occurrences of chloritoid with XMg exceeding 0·25 have been identified. In addition to those listed by Chopin & Schreyer (1983) and mentioned above, which coexist with kyanite and garnet and are mostly located in the Alps, other such chloritoids have been found in blueschist terrains, for instance, in New Caledonia (Ghent et al., 1987), Oman (El-Shazly & Liou, 1991), and Brittany (Bosse et al., 2002). Here we report the presence of Mg-rich chloritoid and (Fe, Mg)-staurolite associated with kyanite and garnet in a retrogressed very high-pressure (VHP, as defined by Liou et al., 1998) eclogite formed from a pelitic precursor rock from the Leaota Massif and its P–T–aH2O estimates in the KCFMASH system.
|
GEOLOGICAL BACKGROUND |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDBULK-ROCK COMPOSITIONAL DATAMINERAL ASSEMBLAGES AND TEXTURAL...MINERAL COMPOSITIONS AND ZONINGP-T CONDITIONSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The metamorphic basement of the Leaota Massif, South Carpathians (Fig. 1), consists of a flat-lying sequence of structurally concordant units displaying internal and/or mutual lithological and metamorphic contrasts. One of these units, the Bughea Complex (BC) is a mélange containing HP lenses embedded in a medium-pressure semipelitic matrix. It is emplaced between a medium-grade gneissic unit probably originating in an arc setting (Negulescu, 2006
) and structurally higher units of pelitic or meta-igneous composition. These latter units display an upward decrease in apparent metamorphic grade from low to very low grade.
|
The matrix of the BC consists of amphibole mica-schists, garnet–chloritoid mica-schists, amphibolites and minor marbles and quartzites (Gheuca & Dinic
, 1986). This matrix includes HP lenses of common eclogite (different occurrences were described by Gherasi et al., 1971; Mrun
iu et al., 1997; Medaris et al., 2003), VHP eclogite (Sbu, 2000) and metagabbronorite (Negulescu, 2006). Unlike the large number of common eclogites known from the BC, there is only a single occurrence of VHP eclogite (Fig. 1), which can be distinguished from the other eclogites by its unusual mineralogy and chemical composition and calculated peak pressure, which consistently place in the coesite stability field. The peak metamorphic conditions for the matrix of the BC are 557–585°C, 0·85–0·95 GPa, aH2O = 0·7 ± 0·15 (Sbu, 2000), whereas the common eclogites and metagabbronorites equilibrated in the range 550–680°C, 1·7–2·6 GPa (Negulescu, 2006). The VHP eclogite records maximum conditions of 2·8–3·2 GPa, 780–840°C, along a near-peak heating–decompression path (Sbu, 2000). The VHP-eclogite body is composed of coarse garnet (1–3 cm) embedded in a matrix of clinopyroxene, kyanite and retrograde white mica and amphibole (Fig. 2b). Leucocratic domains of white mica (phengite, paragonite) and quartz can be observed around the garnets and in the rock matrix. The VHP body grades laterally into distinct rock-types that do not appear in other settings inside the BC; namely, relatively finer-grained (up to 5 mm) massive mica- and garnet-rich rocks (Fig. 2c) and subordinate garnet- and mica-bearing pelitic gneisses. This typical association in the field and the gradual transition of the VHP eclogite towards the rock types mentioned (Fig. 2) support the interpretation that these massive rocks and gneisses originated from the marginal alteration of the VHP eclogite, representing metasomatic or retrograde rinds, as also indicated by chemical data presented below. The massive finer-grained rock displays remarkably homogeneous features in both hand-specimen and thin sections based on observations of the abundant blocks found both in situ and downstream in the river alluvia. The garnet–kyanite–chloritoid assemblage discussed at length below was identified in these isotropic mica- and garnet-bearing rocks bordering the VHP eclogite. Given the homogeneity of the studied rock types, two samples were selected for bulk-rock and mineral chemical analysis. One of these samples, displaying the most typical and complete mineral assemblage and microtexture, as well as the clearest and most extensive zoning pattern of garnet porphyroblasts, was used for thermodynamic calculations of P–T conditions.
|
|
BULK-ROCK COMPOSITIONAL DATA |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDBULK-ROCK COMPOSITIONAL DATAMINERAL ASSEMBLAGES AND TEXTURAL...MINERAL COMPOSITIONS AND ZONINGP-T CONDITIONSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
As noted by S
bu (2000), the VHP eclogite has a Si-rich, alkaline character (see Table 1) and a highly fractionated normalized multi-element pattern unlike all the other eclogites, which display mainly mid-ocean ridge basalt (MORB)-like characteristics. The strong trace element fractionation was tentatively ascribed by Sbu (2000) to an ocean-island magmatic precursor.
|
Additional major element (X-ray fluorescence; XRF) and trace element analyses (inductively coupled plasma; ICP-MS) performed during this study on both the unaltered VHP eclogite and its retrogressed rind [for details of analytical techniques and conditions see Massonne & Czambor (2007
)] confirmed this fractionated trace element characteristic. Chondrite-normalized rare earth element (REE) patterns are strikingly similar between the VHP eclogite (sample ELG) and its outer rind (samples RU007 and VH6) (Fig. 3). The REE characteristics of these rocks perfectly match the North American Shale Composite (NASC) of Taylor & McLennan (1985), suggesting that they represent metasediments of pelitic composition (Fig. 4). The slight downward shift of the REE pattern of the rind may be a dilution effect related to the silica enrichment of the rind (Table 1). There is also some depletion of the relatively immobile major elements (Al2O3, FeO, MgO), high field strength elements (HFSE) and immobile trace elements (Cr, Ni) (Fig. 5). A marked enrichment in large ion lithophile elements (LILE) and chalcophile elements is also recorded, whereas halogens display contrasting behaviour. Fluorine, which tends to fractionate into the solid phases (hydrous minerals such as chlorite, chloritoid, and especially mica), is enriched whereas chlorine, which fractionates into fluids, is depleted (see also Wood & Walther, 1986). The REE patterns (Fig. 3) are ‘spoon-shaped’ and enriched in the middle REE (MREE) and light REE (LREE). Strontium is depleted to the same degree as the LREE. Manganese appears enriched in the alteration rind; although Mn behaves as a highly compatible element, as it is readily incorporated in garnet, it would be expected to undergo a slight apparent depletion considering the Si-dilution effect. The actual relationship may most probably be caused by inter-sample variation, potentially giving an amplified effect at concentrations as low as those recorded in the analysed rocks. One of the most salient features in Fig. 5 is the strong Ca and Na depletion. The most plausible explanation appears to be connected to the decompressional breakdown of clinopyroxene (impure jadeite and omphacite), which is abundant in the pristine VHP eclogite. Godard & Mabit (1998) found that symplectitic breakdown of omphacite occurs with Na and Ca release. Sbu et al. (2002) advocated, on the other hand, relatively long-range diffusion of these components by fluid-mediated mass transfer to explain kyanite dissolution following pyroxene breakdown in eclogites. In these circumstances, fluid exchange between the rock (containing Na and Ca-enriched fluids) and its surroundings (containing Si and K-rich fluids) is likely to occur.
|
|
|
Chemical alteration caused a strong modal increase of quartz and micas in the rind rocks. This is most probably related to subduction-zone metasomatism, as also indicated by the LILE enrichment of the BC matrix rocks themselves, demonstrated by the high Ba contents of K white micas (Negulescu & S
bu, 1999). |
MINERAL ASSEMBLAGES AND TEXTURAL RELATIONSHIPS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDBULK-ROCK COMPOSITIONAL DATAMINERAL ASSEMBLAGES AND TEXTURAL...MINERAL COMPOSITIONS AND ZONINGP-T CONDITIONSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The mineral assemblages of the fresh VHP eclogites were comprehensively described by S
bu (2000). These assemblages and textural relationships reveal several stages of a burial and exhumation evolution. The well-preserved eclogitic assemblage consists of jadeite-rich clinopyroxene (Jd76–50), kyanite, almandine-rich garnet, quartz, allanite, and phengite. The retrograde assemblages are rich in white mica (phengite, paragonite), Na-amphiboles, Na-Ca-amphiboles and albite.
The eclogite rind zone is mainly composed of a grey, fine-grained, massive, mica-rich rock. The mineral assemblage of the matrix is phengite + paragonite + quartz + rutile + kyanite + epidote. Blasts of garnet up to 5 mm in size, often displaying a web-like structure, are embedded in a matrix rich in quartz and white mica. Kyanite with quartz and rutile inclusions is frequent in the matrix. Mostly, kyanite represents recognizable relics from the unaltered parent rock and is corroded by coarse bunches of white mica (Figs 6 and 7). Uncorroded kyanite intimately associated with spongy garnet occurs rarely in the matrix (Fig. 8). No trace of jadeite-rich clinopyroxene is preserved but pseudomorphic lumps of finely intergrown albite + quartz including randomly oriented chlorite and biotite laths imply the original existence of such pyroxene porphyroblasts (Figs 8–10). Potassic white-mica is partly replaced along grain boundaries by a thin selvage of biotite, sometimes accompanied by chlorite.
|
|
|
|
|
Poikiloblastic or web garnets commonly include quartz, rutile and white-mica, with subordinate amounts of epidote, chloritoid, staurolite and kyanite. Regularities in the distribution of the inclusions are conspicuous, correlating with chemical zonation (described below): (1) the garnet cores are almost free of inclusions (except rare epidote); (2) an intermediate garnet zone contains abundant quartz and epidote inclusions, also containing chloritoid and staurolite in its external part and often displaying a web structure; (3) the garnet rims contain less abundant, large-sized inclusions, consisting of white mica (phengite, paragonite) and quartz. Chloritoid and staurolite occur as single-phase or composite inclusions in garnet, sometimes replaced by chlorite and biotite (Fig. 11), together with kyanite and rare Na–Ca mica (Fig. 12).
|
|
These textural relationships suggest the presence of three successive mineral assemblages, of which the first is confined to the intermediate zone of the garnet and its inclusions:
I, garnet (intermediate zone) + chloritoid + kyanite + epidote + quartz ± (Fe, Mg)-staurolite;
II, matrix assemblage: garnet (rim) + phengite + paragonite + quartz + epidote + kyanite;
III, retrograde assemblage: albite + chlorite + biotite ( + phengite, paragonite).
|
MINERAL COMPOSITIONS AND ZONING |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDBULK-ROCK COMPOSITIONAL DATAMINERAL ASSEMBLAGES AND TEXTURAL...MINERAL COMPOSITIONS AND ZONINGP-T CONDITIONSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
Mineral compositions were analyzed using a Cameca SX100 electron microprobe (EMP) with five wavelength-dispersive spectrometers. Operating conditions were 15 kV accelerating voltage, 15 nA beam current and focused beam for anhydrous minerals, and 10 nA and beam spot of 6 µm for hydrous or volatile-bearing minerals. The counting time was 20 s for Ba, Na, K, Fe, Ti, Ca, Cr, Cl, Si, Al, Mg and Mn, and 30 s for Zn and F. A set of standards composed of natural minerals and synthetic glasses was used for calibration in connection with an on-line PAP-correction routine.
Mineral zoning was investigated using both compositional maps and line-profiles. The conditions for obtaining compositional maps were 15 kV, 30 nA and 100 ms counting time per step. Ka radiations of Fe, Mn, Ca and Mg were simultaneously detected using large LIF, large PET, PET and TAP crystals, respectively. Back-scattered electron (BSE) images were also recorded during the stage scan.
Electron microprobe analyses were recalculated, using an automated routine, as structural formulae and ideal end-member concentrations by iteratively modifying Fe (and when necessary Mn) oxidation state to simultaneously fulfil stoichiometric and charge balance constraints as outlined below for each mineral group. The 1
standard deviation and detection limits for each oxide component were recalculated from the standard deviation of the measured element concentrations determined by the microprobe. The corresponding standard deviations of the calculated quantities were estimated (Tables 2–8) from error propagation. Elements below the detection limit are not included in the tables.
|
|
|
|
|
|
|
Garnet was recalculated on the basis of 24 balanced charges and eight cations p.f.u. (per formula unit) in three tetrahedral, two octahedral and three higher-coordination sites, using the site-filling sequence listed in Table 2. Calculations resulted in small quantities of Fe3+ in the highly coordinated site. Although this result might represent an artefact arising from analytical error, it was preferred over non-stoichiometry/charge imbalance. Garnet compositions vary significantly from Prp8·4Alm56·2Grs19·9Sps13 in the core to Prp33·7Alm55·2Grs7·1Sps0·5 at the rims of single grains (Table 2). The Fe/Mg value decreases accordingly from 6·7 to 1· 6. The cation distribution diagrams and BSE images (Fig. 13) reveal a sharp concentric zonality corresponding to the three zones indicated by the distribution and nature of the mineral inclusions. Multiple idiomorphic cores displaying high compositional gradients are agglutinated by a compositionally more homogeneous intermediate zone. The outer part of the garnets is represented by a relatively Mg-rich homogeneous mantle.
|
The chemical zonality in relationship with the distribution of the inclusions may provide a qualitative estimate of both the position of a given section in the garnet porphyroblast, essential in analyzing irregularly zoned, often coalescent crystals as in the case discussed, and the growth history of the garnet porphyroblasts. Compositional gradients during porphyroblastic growth are primarily a reflection of changing P–T conditions and coexisting assemblages, as well as chemical fractionation, with no time implication, yet specific compositions inside the porphyroblasts (especially when a uniform trend can be documented) and particular inclusion patterns relate to a definite relative time sequence during crystal growth. Thereby in random sections through porphyroblasts, those with steeper gradients (and larger chemical ranges) within the same compositional range are closer to the geometrical centre of a zone. On the other hand, the abundance of the inclusions is an inverse measure of the growth rate of the porphyroblasts as compared with the contemporaneous dissolution rate of the overgrown matrix minerals. Relatively fast porphyroblastic growth favours the preservation of abundant inclusions, with web textures resulting from extreme growth rates, whereas slow growth produces inclusion-poor crystals. According to these criteria, the morphology of the garnets is consistent with slow growth of the idiomorphic cores followed by fast agglutinating growth of the intermediate zone, and a decrease in growth rate coincident with further modal increase during mantle formation.
The garnet porphyroblasts represented in Fig. 13 display almost the entire measured compositional range of the garnets in sample RU007. Figure 13a shows a section cutting into three agglutinated cores of which one nearly exposes its centre, whereas Fig. 13b shows a section cutting into the intermediate zone (close to a core located below the section plane) and the crystal mantle. By virtue of the position of the section, Fig. 13a allows an estimate of the typical volume ratio between the three zones: 62% mantle and 38% intermediate zone plus core.
Because of the complex structure of the porphyroblasts, we chose to summarize the compositional range in the chemical profiles in Fig. 13c–e rather than attempting to illustrate it on a single profile. Manganese zonation in the garnet core displays a typically bell-shaped pattern (Fig. 13c). The intermediate garnet zone is characterized by a continuous decrease of Mn and Fe/Mg and a uniform Ca content (Fig. 13d and e). The mantles show a significant jump in Mg at the inner boundary, followed by continuous increase towards the outermost rim (Table 2, Fig. 13e). Iron is almost homogeneously distributed through the entire crystal. Thus, the element concentration maps (Fig. 14) reveal the detailed growth history of the garnet porphyroblasts, with concentric growth, agglutination and overgrowth stages. The outer intermediate zone and especially its boundary towards the mantle is the location in which chloritoid and subordinate staurolite inclusions almost exclusively occur (Fig. 14c and f).
|
The abrupt rise in the Mg content of the garnet and the disappearance of chloritoid towards the rim document massive garnet growth at the expense of chloritoid with a higher XMg than the coexisting garnet. The absence of chloritoid in the matrix indicates that all chloritoid was consumed during the growth of the garnet mantles, being preserved only in armoured inclusions lacking the necessary reactants (quartz ± phengite). Calcium depletion in the mantles could be related to plagioclase growth in the matrix. A narrow fringe at the garnet rims displays lower Mg, and especially higher Mn and Ca contents (Fig. 14), pointing to minor retrograde corrosion of the garnet porphyroblasts. Interestingly, the Na contents in garnet are frequently above detection limit (especially in cores) (Table 2).
The Ca–Fe–Mg and Mn–Fe–Mg diagrams (Fig. 15) display the chemical trend of garnet growth in the rinds in comparison with the unaltered VHP-eclogite core (Sbu, 2000). The high Mn contents in the cores of the studied garnets suggest complete resorption of the homogeneous and low-Mn garnet pre-existing in the fresh VHP eclogites. No measured compositions corresponding to relics of these low-Mn garnets have been identified.
|
Chloritoid and staurolite occur exclusively as inclusions in the outer part of the intermediate zone of garnet (Fig. 14c and f), near the high-Mg overgrowth boundary, and are occasionally corroded by chlorite and biotite. Chloritoid was recalculated on a basis of 12 charges, four cation sites and two OH + F sites. Compositions (Table 3) are remarkably magnesian (XMg = 0·40–0·47).
Staurolite was recalculated on basis of 96 charges and Fe3+ was considered to amount to 3% of the total Fe following Hawthorne et al. (1993, and references therein). A regression of 62 complete staurolite analyses from the same paper and Holdaway et al. (1986) was also used to estimate Li and H contents, respectively, from the R3+-normalized Mg vs Li, and cationic sum vs Si correlations. Complete filling of the T(1) tetrahedral and octahedral sites was assumed as in Table 4, the T(2) tetrahedral occupancy and OH + F p.f.u. resulting from the calculations. The above-mentioned correlations consistently indicate no Li in any of the analysed staurolites. Their compositions show XMg values slightly but constantly exceeding 0·25 (Table 4).
Epidote occurs as inclusions in all garnet zones as well as in the rock matrix. Analyses were recalculated to 25 charges, one OH + F and site-filling according to Table 5. Fe and Mn oxidation states were iteratively adjusted for the closest match of the ideal crystal-chemical formula.
Phyllosilicates, mainly white micas (phengite, paragonite), are very abundant in the matrix, around large kyanite relics and as inclusions in garnets. The structural formula of white mica was recalculated on the basis of 22 balanced charges, two OH + F + Cl p.f.u. and iterative increase of Fe3+ starting from Fe3+ = 0 in case of over-occupancy of one of the cationic sites (usually the interlayer site). As all white micas analysed yielded incomplete occupancies of the interlayer position and oxide totals close to 100% upon recalculation, all iron was left in the ferrous state. The compositions of phengite grains (Table 6), which are more common than paragonite, are fairly homogeneous. However, a slight decrease of Si p.f.u. towards the outermost rim could be noticed in a profile line (analyses 1pr–10pr in Table 6), from 3·38–3·39 p.f.u. in the crystal core to 3·34–3·36 Si p.f.u. in the rim. The paragonite compositions (Table 6) are homogeneous (85–88 mol % paragonite end-member).
An unusual (Ca, Na)-mica was identified as chemically homogeneous inclusions (both optically and in BSE images) in staurolite (Fig. 12, Table 7). This is identical in composition to the Na–(Mg,Fe)-margarite mica reported by Godard & Smith (1999). Distinctive characteristics are the extensive Na substitution and intermediate filling of the octahedral layer with significant Mg and Fe in addition to Al. Expressed in ideal end-members, the analysed mica contains (in molar quantities) 55·1% margarite, 13·3% paragonite, 10·1% preiswerkite–eastonite, 11·2% interlayer-vacant terms, 9% of the theoretical ‘Mica L’ of Smith (1988) and minor muscovite (1· 4%). According to Godard & Smith (1999), a critical factor for the occurrence of such micas is the retrogression of eclogitic rocks containing chemical domains low in Si and high in Na and Al, such as kyanite surrounded by jadeite-rich clinopyroxene.
Biotite was recalculated according to the same procedure as for white mica, but somewhat arbitrarily choosing an Fe3+/Fetot value of 0·25, which was found to consistently yield good analysis totals. Significant Fe3+ in biotite is also indicated by its deep brown pleochroism and the reasoning that biotite most probably incorporates the bulk of the ferric iron in the rock, in addition to the by far less abundant epidote. Biotite is a secondary phase. Its composition differs according to the textural setting: biotite corroding staurolite (Fig. 16a) is less magnesian than biotite in the albite–oligoclase lumps and that formed at the expense of phengite (Fig. 16b).
|
Chlorite is an abundant secondary phase formed at the expense of chloritoid and in albite lumps. In other instances it appears in ambiguous textural settings, in both the matrix and in composite inclusions; neither setting excludes a primary genesis. Chlorite was recalculated to 28 charges and eight (OH + F + Cl) groups. Cations were ascribed in the order indicated in Table 6 to four tetrahedrally and six octahedrally coordinated positions. Large cations were assumed to occur in interlayer positions. Chlorite is chemically relatively homogeneous and close to clinochlore end-member composition (Table 8).
|
P–T CONDITIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDBULK-ROCK COMPOSITIONAL DATAMINERAL ASSEMBLAGES AND TEXTURAL...MINERAL COMPOSITIONS AND ZONINGP-T CONDITIONSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The above-described textural relationships allow the identification of two pervasive metamorphic assemblages. Poikiloblastic garnets are the host of the older assemblage [namely, the critical garnet–chloritoid–kyanite association, documented by the included phases (chloritoid, staurolite, kyanite, epidote ± chlorite)], whereas the matrix contains a chloritoid-free, kyanite–garnet-bearing assemblage. The compositions of the ferromagnesian phases were plotted in an AFM diagram (Fig. 17) showing that in this simplified chemography staurolite cannot be in equilibrium with the above-mentioned critical assemblages. The occurrence of staurolite, however, may be the result of the stabilizing effect of additional components not considered in the projection (e.g. ZnO; see Table 4). Considerable amounts of Li have been reported from metapelitic staurolites (e.g. Holdaway et al., 1986
), but in our case no indication of the presence of Li could be found (see above).
|
A P–T pseudosection was calculated with the PERPLE_X computer package (Connolly, 1990
, Perp07_Windows version, available at http://www.perplex.ethz.ch/perplex/ibm_and_mac_archives) using the ‘hp02ver.dat’ thermodynamic dataset updated from Holland & Powell (1998). The solid-solution models used are those identified by the following abbreviations in the ‘solut_08.dat’ file of the above-mentioned package, modified from Holland & Powell (1998) according to the references therein: Chl(HP) for chlorite, Gt(HP) for garnet, Ctd(HP) for chloritoid, cOmph(HP) for omphacite, Pheng(HP) for phengite, Carp for carpholite, T for talc. The last two models assume ideal solution properties, as well as the model adopted for epidote, where ideal Al–Fe3+ mixing on one octahedral site was assumed. For phengite solid solutions we preferred a simpler and older model over the more recent ones available [Mica(CH1), Mica(CH2), using the parameters derived by Coggon & Holland (2002)]. All three models available assume ideal celadonite–muscovite interactions and do not consider trioctahedral substitutions, whereas both non-ideal behaviour of the celadonite components and octahedral occupancy typically exceeding two cations p.f.u. are well documented (Massonne & Spurka, 1997). The Pheng(HP) model, however, delivers Si isopleths substantially closer to the values actually found in the studied rock. The input rock composition resulted from the projection of the actual rock composition (RU007, Table 1) into the KFMASH system through all other components. The resulting pseudosection (Fig. 18a) predicts that garnet, chloritoid and kyanite indeed coexist as a HP assemblage along with phengite with Si contents between 3·3 and 3·4 p.f.u. in a P–T field around 600°C and pressures between 1·7 and 2·6 GPa. Resulting XMg values of garnet and coexisting chloritoid are 0·24–0·35 and 0·37–0·49, respectively (Fig. 18a). Thermal breakdown of chloritoid is predicted to produce assemblages containing talc or chlorite before all available magnesium can be incorporated in garnet with increasing temperature.
|
The actual rock composition in the K2O–Na2O–CaO–FeO–MgO–MnO–Al2O3–Fe2O3– SiO2–H2O system was considered in association with additional solution properties for epidote and acmite-bearing ordered omphacitic clinopyroxene (see above), under the following assumptions: (1) the iron speciation in modelled sample RU007 containing this assemblage was considered identical to that in the companion sample VH6 (Table 1), titrimetrically analyzed for iron oxidation state, but in which no chloritoid was identified in the available thin sections; (2) TiO2 and P2O5 were assigned to rutile and apatite, respectively, a corresponding quantity of CaO being subtracted from the bulk composition; (3) quartz and water were considered as excess phases and O2 (and, thus, Fe3+) content fixed. The resulting P–T pseudosection failed to reproduce the assemblages observed in the studied rock because instead of epidote hematite was predicted to coexist with the chloritoid–kyanite–garnet assemblage. Therefore, we operated a further simplifying assumption by projecting the rock composition through a constrained quantity of epidote with 0·75 Fe3+ p.f.u. (approximating the average composition of epidote ubiquitous in all mineral assemblages). A projection through epidote is equivalent to the assumption that this phase stably coexists with all other phases involved in the mineral reactions. If this assumption was incorrect, epidote would represent a metastable relic separated from the reacting assemblage, and it still appears justified to subtract it from the rock composition to obtain the effective bulk composition involved in metamorphic reactions. The quantity of epidote subtracted was chosen in such a way as to allow enough Ca left in the system for a grossular content in garnet matching the Mn/Ca ratio of garnet rims. No correction for Mn in epidote was considered, permitted by the small concentration of this cation in both epidote and whole-rock, and the subordinate modal amount of epidote. After the subtraction of epidote, Fe3+ was still present in the system considered. Most probably the remaining Fe3+ is at present hosted in biotite, but at higher pressures was more likely to enter acmite-bearing clinopyroxene. Although this projection through epidote with a fixed composition has a limited thermodynamic legitimacy, it is expected reasonably to predict phase relationships around the P–T conditions under which the assemblage equilibrated.
The calculated pseudosection (Fig. 18b) predicts a stability field for the chloritoid–garnet–kyanite assemblage together with phengite, paragonite and clinopyroxene ( + quartz, rutile, apatite, epidote) at pressures between 2 and 2·2 GPa in a narrow thermal domain around 600°C. Considering the probable origin of the biotite and chlorite-bearing albite lumps from former clinopyroxene, this seems to be a reasonable reproduction of the assemblage actually observed in the studied rock. An additional narrow field, containing chlorite in addition to the other phases mentioned above, would be stable at pressures down to 1·8 GPa. Hematite, on the other hand, appears to be stable only at lower pressures than those corresponding to the garnet–kyanite–chloritoid assemblage.
An independent assessment of the equilibrium conditions of relevant assemblages was also attempted with the PTGIBBS program package of Brandelik & Massonne (2004), employing the thermodynamic dataset of Berman (1988, 1990) augmented by compatible data for pyrope, Mg-chloritoid, Mg–Al-celadonite and muscovite (Massonne, 1995) and Fe–Al-celadonite (Massonne & Spurka, 1997). Garnet was modelled as an ionic solid solution with mixing on two sites: non-ideal asymmetric Ca–Fe–Mg-Mn solution on the eight-fold coordinated site and ideal solution on the octahedral site (Massonne, 1992) whereas phengite was considered a non-ideal asymmetric six-component molecular solution (Massonne, 1995). Chloritoid was treated as an ideal solid solution (Vidal et al., 2001). Clinozoisite was considered also as an ideal solution involving mixing of Al and Fe3+ on a single site (aCzo = Al – 2).
The two key assemblages were investigated by selecting phases in apparent microtextural equilibrium: (1) the HP chloritoid-bearing assemblage, represented by garnet with intermediate zone compositions, Mg-rich chloritoid and epidote included in it (aCzo = 0·21), kyanite and quartz; (2) the matrix assemblage, consisting of garnet and phengite of rim composition, matrix epidote (aCzo = 0·26), kyanite and quartz. For each assemblage, the whole range of garnet compositions corresponding to the appropriate growth zone was investigated to identify the composition yielding the closest intersection of calculated univariant lines. Typically, best matches were found for garnet compositions still belonging to the appropriate chemical zone, but appearing texturally older than the other mineral phases considered. Garnet compositions (Table 2) with a texturally more favourable position yielded unrealistically high temperatures ( > 700°C) and poor intersections. In the first case, it seems likely that the actual inclusion of a phase postdates its formation, in such a way that the inclusions are in equilibrium with compositions older than those of the immediately adjacent host (gt4pr1/5, Table 2), and consequently the derived conditions may truly characterize the equilibrium assemblage. In the second case, the outermost rims of garnet are most likely to stably coexist with the matrix minerals. The tighter intersections obtained with garnet compositions from the inner mantle (grt4pr2/4, Table 2) would imply that phengite ceased growing or was even consumed as the modal abundance of garnet continued to increase. In this case the derived P–T point does not exactly define the peak temperature conditions, but still approximates them close enough to give a useful image of the heating–decompressional path.
The equilibrium conditions for the HP chloritoid-bearing assemblage were calculated by equilibria in the CFMASH system. Because of compositional degeneracy, the number of distinct reactions is only eight, intersecting in three invariant points, of which one is water-conservative. To determine the aH2O at specific P–T conditions, the water-present invariant points were iteratively modified by changing the water activity to finally match the water-absent invariant point (geohygrometry according to Massonne, 1992).
The eight above-mentioned equilibria for the CFMASH system (Fig. 19a) intersect at 1·78 GPa and 580°C for a water activity aH2O = 1. These P–T conditions for the garnet–chloritoid–kyanite assemblage are compatible with the estimated conditions for other similar occurrences elsewhere: 1·7–2·1 GPa and 550–650°C for the Sesia Zone, Italian Alps (e.g. Tropper & Essene, 2002), 1·9–2·3 GPa, 590 ± 25°C for the Tauern Window, Austrian Alps (Holland, 1979; Stöckhert et al., 1997) and 2 GPa, 550–600°C for the Raspas Complex, Ecuadorian Andes (Gabriele et al., 2003).
|
To derive the P–T conditions recorded by the matrix, the eight distinct equilibria in the KCFMASH system were considered, characterizing the assemblage garnet–phengite–quartz–kyanite–epidote–excess water and also yielding three invariant points, of which one is water-conservative. These equilibria (Fig. 19b) intersect at 1·15 GPa and 620°C for the mineral compositions of the matrix assemblage. Water activity aH2O was again found to be close to unity.
The P–T conditions thus derived indicate a heating–decompression path that implies a change from omphacitic clinopyroxene stable conditions to those of plagioclase. This P–T array is supported by garnet mantle compositions lower in Ca and the presence of the albite-rich bundles in the matrix, which most probably originate from the replacement of Na-rich clinopyroxene.
The P–T evolution of the VHP-eclogitic precursor (stages 1–5, Fig. 19) described by Sbu (2000) documents an intricate exhumation path recording a heating episode (stages 4 and 5) during decompression. According to the estimated P–T conditions of the VHP-eclogite rind (between a and b in Fig. 19) the rock experienced even a more significant heating during its exhumation than originally assumed.
|
DISCUSSION AND CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDBULK-ROCK COMPOSITIONAL DATAMINERAL ASSEMBLAGES AND TEXTURAL...MINERAL COMPOSITIONS AND ZONINGP-T CONDITIONSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The garnet- and mica-rich rocks surrounding the VHP eclogite in the Leaota Massif represent its metasomatic–retrograde marginal alteration zone, as proven by close spatial association and geochemical features. During the alteration, pervasive growth of the critical chloritoid–garnet–kyanite HP assemblage is discernible from inclusions in garnet porphyroblasts. This assemblage was, in turn, replaced by a chloritoid-free, kyanite-bearing assemblage prevailing in the matrix. The resulting P–T array records a decompression episode during which a temperature increase of about 40°C occurred. Considering the complete resorption of the pre-existing garnet in the unaltered VHP eclogite, as well as the advanced obliteration of the pre-existing textures, the chloritoid-bearing stage may have also resulted from a reheating episode acting on rocks that had already experienced cooling along the exhumation path (point 4 in Fig. 18). It should be noted that the chemical zonation of garnet also indicates continuous temperature increase during its entire growth history.
The chloritoid-bearing assemblage most probably originated from decomposition of garnet and kyanite originally present in the parental eclogite at VHP conditions, as a result of the influx of an external fluid percolating the rock. Subordinate staurolite growth occurred either because of the stabilizing effect of additional cations such as Zn or in micro-domains that failed to reach bulk equilibrium (high-Al–Mg, low-Si environments such as former jadeitic pyroxene sites).
The enrichment of Mg in the garnet rims can be related to resorption of Mg-rich chloritoid forming Mg-richer garnet and kyanite (± chlorite) at P–T conditions recorded by the matrix (1·15 GPa and 600–620°C). Kyanite also reacted with omphacitic clinopyroxene to form paragonite at this stage. The final retrograde stage of the evolution is recorded by chlorite and biotite formed at the expense of chloritoid armoured relics, phengite, and staurolite (but never garnet). The albite lumps probably represent pseudomorphs resulting from the breakdown of former clinopyroxene at this stage.
The assemblage chloritoid–garnet–kyanite indicates a low thermal gradient (largely between 5 and 10°C/km) of the type usually prevailing during the early stages of metamorphism. This assemblage, appearing exclusively in rocks of pelitic composition, is a key indicator for the involvement of continental crust or continent-derived sediments in subduction zones, or extensive crustal thickening during a collisional process. The high value of this indicator is also due to the rarity of Mg-rich chloritoid, in direct relationship with the tight chemical constraints it provides.
The mineral assemblages preserved in the marginal alteration products of the VHP eclogite thus reveal a part of a complex retrograde path during exhumation of metapelitic rocks subject to VHP metamorphic conditions. In addition, considering also the corresponding near-peak metamorphic history (Sbu, 2000) and the relationships with the adjacent rocks, a geotectonic scenario can be constrained that is typical for a subduction-channel environment. Deep subduction of pelitic rocks of crustal derivation was followed by detachment from the subducting slab by an upward-directed mass flow and ascent along a low-gradient P–T path indicating syn-subduction exhumation. The marginal chemical alteration was due to interactions with subduction-zone fluids, also consistent with the invoked subduction-channel environment. This feature is shared by the HP metapelites with their present host, a semipelitic variegated rock assemblage displaying a prograde medium-grade, medium- to high-pressure metamorphic overprint (0·8–0·9 GPa, 600°C; Sbu, 2000). The host assemblage most probably represents a subduction–accretion mélange, in which the incorporation of HP blocks is also consistent with a syn-subduction exhumation scenario. Transient heating may be related to thermal perturbations in the subduction channel. This geodynamic feature was part of a Neoproterozoic–Early Palaeozoic subduction–collision scenario outlined in detail by Sbu & Massonne (2003) for other basement units of the South Carpathians.
In the case of the occurrence reported here, the appearance of Mg-rich chloritoid coexisting with kyanite and garnet occurred along the decompressional path during thorough textural resetting; thereby displaying an unusually pervasive retrograde overprint still under HP conditions. The P–T array followed by the rock between the growth of the Mg-rich chloritoid-bearing assemblage and the formation of a lower pressure, higher temperature assemblage indicated by garnet outer zone and matrix minerals marks a significant change in the thermal gradient; this can be related to the closing stages of subduction and involvement of rocks from a subduction-channel environment in syn-collisional stacking of the rock units along a continental margin affected by continent–continent collision.
|
ACKNOWLEDGEMENTS |
|---|
This contribution is a revised version of part of the first author's PhD thesis, and resulted during an Alexander von Humboldt partnership between the institutions of the authors. Electron microprobe work at Universität Stuttgart was financially supported by Deutsche Forschungsgemeinschaft (Ma1160/13). Anita Czambor kindly supported whole-rock XRF (major element) and ICP-MS (trace element) analyses. Reviews by Mark Caddick, Thomas Fockenberg and an anonymous reviewer improved the manuscript considerably. The editor, Reto Gieré, is acknowledged for suggestions concerning the presentation of the electron microprobe analytical data. Nigel Cook kindly improved the quality of the English presentation.
*Corresponding author. Telephone: +40-21-3181328. Fax: +40-21-3181326. E-mail: elinegu{at}yahoo.com
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDBULK-ROCK COMPOSITIONAL DATAMINERAL ASSEMBLAGES AND TEXTURAL...MINERAL COMPOSITIONS AND ZONINGP-T CONDITIONSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
Abraham K., Schreyer W. A talc–phengite assemblage in piemontite schist from Brezovica, Serbia, Yugoslavia. Journal of Petrology (1976) 17:421–439.
Berman R. G. Internally-consistent thermodynamic data for minerals in the system Na2O–K2O–CaO–MgO–FeO–Fe2O3–Al2O3–SiO2–TiO2–H2O–CO2. Journal of Petrology (1988) 29:445–522.
Berman R. G. Mixing properties of Ca-Mg-Fe-Mn garnets. American Mineralogist (1990) 75:328–344.[Abstract]
Bosse V., Ballevre M., Vidal O. Ductile thrusting recorded by the garnet isograd from blueschist-facies metapelites of the Ile de Groix, Armorican Massif, France. Journal of Petrology (2002) 43:485–510.
Brandelik A., Massonne H.-J. PTGIBBS—an ExcelTM Visual Basic program for computing and visualizing thermodynamic functions and equilibria of rock-forming minerals. Computers and Geosciences (2004) 30:909–923.[CrossRef]
Caddick MJ, Thompson AB. Quantifying the tectono-metamorphic evolution of pelitic rocks from a wide range of tectonic settings: mineral compositions in equilibrium. Contributions to Mineralogy and Petrology (2008) doi:10.1007/s00410-008-0280-6.
Chopin C. Talc–phengite: a widespread assemblage in high-grade pelitic blueschists of the Western Alps. Journal of Petrology (1981) 22:628–650.
Chopin C. Coesite and pure pyrope in high-grade pelitic blueschists of the Western Alps: a first record and some consequences. Contributions to Mineralogy and Petrology (1984) 86:107–118.[CrossRef][Web of Science]
Chopin C., Schreyer W. Magnesiocarpholite and magnesiochloritoid: two index minerals of pelitic blueschists and their preliminary phase relations in the model system MgO–Al2O3–SiO2–H2O. American Journal of Science (1983) 283-A:72–96.
Coggon R., Holland T. J. B. Mixing properties of phengitic micas and revised garnet–phengite thermobarometers. Journal of Metamorphic Geology (2002) 20:683–696.[CrossRef][Web of Science]
Connolly J. A. D. Multivariable phase diagrams: an algorithm based on generalized thermodynamics. American Journal of Science (1990) 290:666–718.
Costin G., Luffi P. Cr-rich minerals in eclogites from Râul Târgului (Iezer Mts. South Carpathians). Anuarul Institutului Geologic al României (2001) 72:11–12.
El-Shazly A. K., Liou J. G. Glaucophane chloritoid-bearing assemblages from NE Oman: petrologic significance and a petrogenetic grid for high P metapelites. Contributions to Mineralogy and Petrology (1991) 107:180–201.[CrossRef][Web of Science]
Fockenberg T. New experimental results up to 100 kbar in the system MgO–Al2O3–SiO2–H2O (MASH): Preliminary stability fields of chlorite, chloritoid, staurolite, MgMgAl-pumpellyite, and pyrope. Bochumer Geologische und Geotechnische Arbeiten (1995) 44:39–44.
Fockenberg T. An experimental investigation on the P–T stability of Mg-staurolite in the system MgO–Al2O3–SiO2–H2O. Contributions to Mineralogy and Petrology (1998) 130:187–198.[CrossRef][Web of Science]
Fockenberg T., Schreyer W. Yoderite, a mineral with essential ferric iron: its lack of occurrence in the system MgO–Al2O3–SiO2–H2O. American Mineralogist (1991) 76:1052–1060.[Abstract]
Gabriele P., Ballèvre M., Jaillard E., Hernandez J. Garnet–chloritoid–kyanite metapelites from Raspas Complex (SW Ecuador): a key eclogite-facies assemblage. European Journal of Mineralogy (2003) 15:977–989.
Ghent E. D., Stout M. Z., Black P. M., Brothers R. N. Chloritoid-bearing rocks associated with blueschists and eclogites, northern New Caledonia. Journal of Metamorphic Geology (1987) 5:239–254.[Web of Science]
Gherasi N., Dimitrescu R., Kasper U., Vulpescu G. Contributions au problème des éclogites. Les éclogites des Monts Ezer et Leaota (Carpathes méridionales, Roumanie). Tschermak’. s Mineralogische und Petrographische Mitteilungen (1971) 15:151–158.[CrossRef]
Gheuca I., Dinic I. Litostratigrafia 
i tectonica cristalinului Leaotei între Albeti–valea Ghimbav–valea Bdeanca (Iezer–Leaota). Dri de Seam ale edinelor (1986) 70–71/5:87–95.
Gheuca, I. & Dinic I. The metamorphic basement of the Getic Nappe in the eastern margin of the South Carpathians (Leaota and Iezer Mountains). Anuarul Institutului Geologic al României (1996) 69:15. Supplement 5: IGR 90 Symposium, Excursion Guide C3.
Gil Ibarguchi J. I. G., Mendia M., Girardeau J. Mg- and Cr-rich staurolite and Cr-rich kyanite in high-pressure ultrabasic rocks (Cabo Ortegal, northwestern Spain). American Mineralogist (1991) 76:501–511.[Abstract]
Godard G., Mabit J.-L. Peraluminous sapphirine formed during retrogression of a kyanite-bearing eclogite from Pays de Léon, Armorican Massif, France. Lithos (1998) 43:15–29.[CrossRef][Web of Science]
Godard G., Smith D. Preiswerkite and Na–(Mg,Fe)-margarite in eclogites. Contributions to Mineralogy and Petrology (1999) 136:20–32.[CrossRef][Web of Science]
Hawthorne F., Ungaretti L., Oberti R., Caucia F., Callegari A. The crystal chemistry of staurolite. I. Crystal structure and site populations. Canadian Mineralogist (1993) 31:551–582.[Web of Science]
Holdaway M. J., Dutrow B. L., Shore P. A model for the crystal chemistry of staurolite. American Mineralogist (1986) 71:1142–1159.[Abstract]
Holland T. J. B. High water activities in the generation of high-pressure kyanite eclogites of the Tauern Window, Austria. Journal of Geology (1979) 87:1–27.[Web of Science]
Holland T. J. B., Powell R. An internally consistent thermodynamic data set for phases of petrological interest. Journal of Metamorphic Geology (1998) 16:309–343.[CrossRef][Web of Science]
Koch-Müller M., Wirth R. An experimental study of the effect of iron on magnesiochloritoid–talc–clinochlore–kyanite stability. Contributions to Mineralogy and Petrology (2001) 141:546–559.[Web of Science]
Konopásek J. Eclogitic micaschists in the central part of the Kru
né hory Mountains (Bohemian Massif). European Journal of Mineralogy (2001) 13:87–100.
Kretz R. Symbols for rock-forming minerals. American Mineralogist (1983) 68:277–279.[Abstract]
Liou J. G., Zhang R. Y., Ernst W. G., Rumble D. III, Maruyama S. High-pressure minerals from deeply subducted metamorphic rocks. In: Hemley, R. J. (ed.). Ultrahigh-pressure Mineralogy: Physics and Chemistry of the Earth Deep Interior. Mineralogical Society of America, Reviews in Mineralogy (1998) 37:33–96.
Massonne H-J. Thermochemical determination of water activities relevant to eclogitic rocks. In:. Water–Rock Interaction WRI-7—Kharaka YK, Maest A, eds. (1992) 1523–1526. Proceedings of the 7th International Symposium, Park City, Utah, USA. Rotterdam: Balkema.
Massonne H-J. Experimental and petrogenetic study of UHPM. In:. In: Ultrahigh Pressure Metamorphism—Coleman RG, Wang X, eds. (1995) Cambridge: Cambridge University Press. 33–95.
Massonne H.-J., Czambor A. Geochemical signatures of Variscan eclogites from the Saxonian Erzgebirge, central Europe. Chemie der Erde (2007) 67:69–83.[CrossRef][Web of Science]
Massonne H.-J., Schreyer W. Stability field of the high-pressure assemblage talc + phengite and two new phengite barometers. European Journal of Mineralogy (1989) 1:391–410.
Massonne H.-J., Szpurka Z. Thermodynamic properties of white micas on the basis of high-pressure experiments in the systems K2O–MgO–Al2O3–SiO2–H2O and K2O–FeO–Al2O3–SiO2–H2O. Lithos (1997) 41:229–250.[CrossRef][Web of Science]
Massonne H.-J., Willner A. P., Gerya T. Densities of metapelitic rocks at high to ultrahigh pressure conditions: what are the geodynamic consequences? Earth and Planetary Science Letters (2007) 256:12–27.[CrossRef][Web of Science]
Mruniu M., Johan V., Iancu V., Ledru P., Cocherie A. Kyanite-bearing eclogite from the Leaota Mountains (South Carpathians, Romania): mineral associations and metamorphic evolution. Earth and Planetary Sciences (1997) 325:831–838.
Medaris G. Jr, Ducea M., Ghent E., Iancu V. Conditions and timing of high-pressure Variscan metamorphism in the South Carpathians, Romania. Lithos (2003) 70:141–161.[CrossRef][Web of Science]
Negulescu E. Semnificaia mineralelor i a asociaiilor minerale în determinarea istoriei metamorfice a cristalinului masivului Leaota. In: PhD thesis (2006) University of Bucharest. 461.
Negulescu E., Sbu G. The Bughea Level (Leaota Massif, Romania): a subduction mélange proven by LILE-enrichment in eclogites and their anisofacial matrix. Analele Universitii Bucureti (1999) XLVIII:55–58.
Pognante U., Compagnoni R., Gosso G. Micromesostructural relationships in the continental eclogitic rocks of the Sesia–Lanzo zone (Italian Western Alps): a record of subduction cycle. Rendiconti della Società Italiana di Mineralogia e Petrologia (1980) 36:169–186.
Sbu G. A possible UHP-eclogite in the Leaota Mts. (South Carpathians) and its history from high-pressure melting to retrograde inclusion in a subduction melange. Lithos (2000) 52:253–276.[CrossRef][Web of Science]
Sbu G., Massonne H.-J. Relationships among eclogite bodies and host rocks in the Lotru Metamorphic Suite (South Carpathians, Romania): Petrological evidence for multistage tectonic emplacement of eclogites in a medium pressure terrain. International Geology Review (2003) 45:225–262.[Web of Science]
Sbu G., Alberico A., Negulescu E. Peraluminous sapphirine in retrogressed kyanite-bearing eclogites from the South Carpathians: Status and implications. International Geology Review (2002) 44:859–876.[CrossRef][Web of Science]
Schreyer W. Subduction of continental crust to mantle depths: petrological evidence. Episodes (1988a) 11:97–104.[Web of Science]
Schreyer W. Experimental studies on metamorphism of crustal rocks under mantle pressures. Mineralogical Magazine (1988b) 52:1–26.[CrossRef][Web of Science]
Schreyer W., Baller T. Talc–muscovite: synthesis of a new high-pressure phyllosilicate assemblage. Neues Jahrbuch für Mineralogie, Monatshefte (1977) 1977:421–425.
Smith DC. A review of the peculiar mineralogy of the ‘Norwegian coesite-eclogite province’, with crystal-chemical, petrological, geochemical and geodynamical notes and an extensive bibliography. In:. In: Eclogites and Eclogite-facies Rocks—Smith DC, ed. (1988) Amsterdam: Elsevier. 1–206.
Stöckhert B., Massonne H.-J., Nowlan E. U. Low differential stress during high-pressure metamorphism: The microstructural record of a metapelite from the Eclogite Zone, Tauern Window, Eastern Alps. Lithos (1997) 41:103–118.[CrossRef][Web of Science]
Sun S.-S., McDonough W. F. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In:. Magmatism in the Ocean Basins. Geological Society, London, Special Publications—Saunders A. D, Norry MJ, eds. (1989) 42:313–345.
Taylor, S. R. & McLennan SM. The Continental Crust: its Composition and Evolution (1985) 312 pp: Oxford: Blackwell.
Tropper P., Essene E. J. Thermobarometry in eclogites with multiple stages of mineral growth: an example from the Sesia–Lanzo Zone (Western Alps, Italy). Schweizerische Mineralogische und Petrographische Mitteilungen (2002) 82:487–514.
Tropper P., Essene E. J., Sharp Z. D., Hunziker J. C. Application of K-feldspar–jadeite–quartz barometry to eclogite facies metagranites and metapelites in the Sesia Lanzo Zone (Western Alps, Italy). Journal of Metamorphic Geology (1999) 17:195–209.[CrossRef][Web of Science]
Vidal O., Parra T., Trotet F. A thermodynamic model for Fe–Mg aluminous chlorite using data from phase equilibrium experiments and natural pelitic assemblages in the 100° to 600°C, 1 to 25 kb range. American Journal of Science (2001) 301:557–592.
Wei C., Powell R. Calculated phase relations in high-pressure metapelites in the system NKFMASH (Na2O–K2O–FeO–MgO–Al2O3–SiO2–H2O). Journal of Petrology (2004) 45:183–202.
Wood B. J., Walther J. V. Fluid flow during metamorphism and its implications for fluid–rock ratios. In:. Fluid–Rock Interactions during Metamorphism. Advances in Physical Geochemistry—Walther JV, Wood BJ, eds. (1986) 5:89–108.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||