Journal of Petrology Advance Access originally published online on April 17, 2007
Journal of Petrology 2007 48(6):1219-1241; doi:10.1093/petrology/egm015
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Intermediate Alkali–Alumino-silicate Aqueous Solutions Released by Deeply Subducted Continental Crust: Fluid Evolution in UHP OH-rich Topaz–Kyanite Quartzites from Donghai (Sulu, China)
1Dipartimento Di Scienze Della Terra, Università Degli Studi Di Siena, Via Laterina 8, I-53100 Siena, Italy
2Dipartimento Di Scienze Mineralogiche E Petrologiche, UniversitÀ degli studi di torino, Via Valperga Caluso 35, I-10125 Torino, Italy
3CNR–IGG, Istituto Di Geoscienze E Georisorse, Via G. Moruzzi 1, I-56124 Pisa, Italy
RECEIVED APRIL 7, 2006; ACCEPTED MARCH 8, 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONREGIONAL GEOLOGY AND SAMPLE...ANALYTICAL METHODSPETROGRAPHY AND MINERAL...FLUID INCLUSIONSOXYGEN AND HYDROGEN ISOTOPESDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Minerals, fluid inclusions and stable isotopes have been studied in ultrahigh-pressure (UHP) OH-rich topaz–kyanite quartzites from Hushan (west of Dongai), in southern Sulu (China). The quartzites underwent a metamorphic evolution characterized by a peak stage (3·5 GPa and 730–820°C) with the anhydrous assemblage coesite + kyanite I, followed by an early near-isothermal decompression stage (2·9 GPa and 705–780°C) with growth of kyanite II, muscovite, and OH-rich topaz, and by decompression-cooling stages, represented by paragonite (1·9 GPa and 700–780°C) and pyrophyllite (0·3 GPa and 400°C) on kyanite (I and II) and OH-rich topaz, respectively. These rocks may exhibit unusually low
18O and D values acquired before undergoing UHP metamorphism. Five distinct fluid generations are recognized. Type I: concentrated peak solutions rich in Si, Al, and alkalis, present within multiphase inclusions in kyanite I. Type II: CaCl2-rich brines present during the growth of early retrograde OH-rich topaz. Type III, IV, and V: late aqueous fluids of variable salinity, and rare CO2 present during amphibolite- and late greenschist-facies conditions. A number of conclusions may be drawn from these relationships that have an effect on fluid evolution in deeply subducted continental rocks. (1) At a pressure of about 3·5 GPa alkali–alumino-silicate aqueous solutions, with compositions intermediate between H2O fluid and melt (H2O > 25 and 
50 wt %) evolved from quartzites, probably generated by dehydration reactions. (2) During early decompression stages, at the transition from UHP to high-pressure (2·9 GPa) conditions, brines of external origin with higher water contents (82 wt % H2O) initiated the growth of OH-rich topaz and muscovite. (3) The subsequent decompression, at P <2 GPa, was defined by a limited circulation of NaCl aqueous fluids, and CO2 infiltration. Overall, fluid inclusions and stable isotopes highlight a metamorphic fluid–rock interaction characterized by internally derived intermediate aqueous solutions at UHP, followed by infiltration of Cl-rich brines with higher water activities. KEY WORDS: ultrahigh-pressure metamorphism; OH-rich topaz; fluid inclusions; stable isotopes; supercritical liquids
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGY AND SAMPLE...ANALYTICAL METHODSPETROGRAPHY AND MINERAL...FLUID INCLUSIONSOXYGEN AND HYDROGEN ISOTOPESDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Numerous discoveries of coesite and diamond in regional ultrahigh-pressure (UHP) rocks have demonstrated that crustal material can be subducted to mantle depths (Chopin, 1984
; Smith, 1984; Sobolev & Shatsky, 1990), and provided new understanding of subduction and continental collision processes. The presence of aqueous fluids (± CO2, N2, etc.) at high-pressure (HP) and UHP conditions was recognized as the driving mechanism for metamorphic reactions, and ultimately for melting in crustal lithologies subducted to mantle depth (e.g. Poli & Schmidt, 2002). Our knowledge of fluid chemical properties is poor, yet is critical for understanding the potential concentration and transport of elements in the mantle wedge during subduction. One of the open questions concerns the nature and the amount of the chemical species in solution: moderate- to low-salinity aqueous fluids vs concentrated solutions, or hydrous melts.
One approach to obtaining information on the nature of fluids evolving from deep-subducted rocks is provided by fluid inclusion analysis combined with stable isotope geochemistry of natural UHP rocks, although the study of inclusions in these rocks is not an easy task. Peak metamorphic conditions are largely outside the isochore fields even for the densest fluids, and the exhumation P–T paths of the host rocks strongly favor decrepitation of the early trapped fluid inclusions (Touret, 2001). Despite these potential problems, a number of documented examples show that fluid inclusions can be preserved, providing valuable information on the composition of UHP fluids, and, to some extent, on their evolution (for reviews, see Scambelluri & Philippot, 2001; Touret & Frezzotti, 2003; Ferrando et al., 2005a).
High-salinity aqueous fluid inclusions are often observed in UHP minerals. For example, in the UHP rocks from the Alps, fluid inclusions are characterized by high amounts of NaCl and MgCl2, and subordinate concentrations of CaCl2 and KCl (up to 50 wt % NaCl equiv.; Philippot & Selverstone, 1991; Selverstone et al., 1992; Philippot et al., 1995; Scambelluri et al., 2001). To explain the NaCl-dominated nature of such HP solutions, Scambelluri et al. (1997) advocated recycled sea-water, Cl and alkalis, whereas Philippot et al. (1998) suggested that Cl-rich inclusions are derived from hydrothermal alteration of the oceanic lithosphere. More recently, Sharp & Barnes (2004) presented a model for the generation of brines, via breakdown of subducted serpentinites, forming mobile high-salinity aqueous plumes at mantle depths.
In the Dabie-Shan and Sulu UHP continental metamorphic rocks, fluids preserved within inclusions are also aqueous and salt-rich, but generally CaCl2-dominated, and not NaCl-rich as would be expected if their ultimate origin was from past sea-water (Xiao et al., 2000, 2001; Fu et al., 2001, 2002, 2003; Zhang et al., 2005b). Xiao et al. (2000) and Fu et al. (2001, 2003) described Ca-rich brines (±N2), which may have originated during prograde and peak metamorphism. Zhang et al. (2005b) reported a spatial and temporal reconstruction of fluid composition within a vertical sequence of UHP rocks of different composition, and recognized primary CaCl2–NaCl-rich brines as peak fluids in both eclogite and quartzite lithologies. As stable isotope data showed that brines are internally derived, Fu et al. (2003) proposed that they represent significant amounts of meteoric water brought to mantle depths through continental collision. Fluids reveal UHP metamorphism with limited fluid mobility during subduction, peak metamorphism, and exhumation.
Whereas previous studies indicated a substantial enrichment of chlorides in aqueous solutions at HP and UHP conditions, the presence in many UHP rocks of multiphase (or polyphase) solid inclusions, containing considerable amounts of hydrous alumino-silicate phases (e.g. amphiboles and micas), indicates transport of Si and Al (Ferrando et al., 2005a, and references therein). These inclusions have been interpreted either as remnants of former UHP hydrous silicate melts, or as intermediate aqueous solutions (for a review, see Hermann et al., 2006). As experimentally shown, the solubility of silica in aqueous fluids increases significantly at increasing pressure, as a result of complexing as polymers (Manning, 2004). Polymerization of silica enhances the transport of Ca, Na, K, and of those metal components (e.g. Al and Ti) that generally have low solubility in aqueous fluids (Antignano & Manning, 2005; Tropper & Manning, 2005). At extreme pressures, the amount of silica (±alumina) may reach high concentrations and give rise to intermediate silicate aqueous solutions. As pressure increases, the fluid–melt miscibility gap narrows, until the apex of the miscibility gap intersects the endpoint of the water-saturated solidus. Above this pressure, a single supercritical ‘liquid’ [in the sense of Kessel et al. (2005)] is present, with an intermediate composition between a hydrous silicate melt and a concentrated aqueous solution (e.g. 70–30 wt % H2O), whose chemical and physical properties vary with temperature (Niggli, 1920; Stalder et al., 2000).
The UHP rocks occurring in the Dabie–Sulu terranes in eastern China are of great interest to understanding the nature of fluids evolved from deeply subducted rocks. Here, continental lithologies have been carried down to a depth of at least 120–150 km and converted to eclogite-facies rocks, but have retained a number of original features, such as fluid inclusions in peak metamorphic minerals. These fluid inclusions represent samples of pristine fluids formed at UHP conditions, with a pre-metamorphic oxygen isotope signature that is heterogeneous at the outcrop scale and locally extremely 18O-depleted (as low as 18O –10
), caused by interaction with meteoric water prior to subduction (Yui et al., 1995; Zheng et al., 1996; Rumble & Yui, 1998; Rumble et al., 2002).
The present study reports petrographic, fluid inclusion and stable isotope data for OH-rich topaz-bearing kyanite quartzites from the Donghai area, southern Sulu terrane, and shows that internally derived aqueous fluids, with high concentrations of Si, Al, S, Ca, K and Na, but devoid of Cl, evolved at UHP conditions from continental rocks. During the early stages of decompression, the fluid composition changed into Ca-rich brines, probably of external origin. The present data allow us to model the chemical behavior of major elements in aqueous solutions generated at UHP conditions, and the overall fluid–rock interaction during the subsequent decompression stages, facilitating the reconstruction of the quartzite exhumation history.
|
REGIONAL GEOLOGY AND SAMPLE LOCATION |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGY AND SAMPLE...ANALYTICAL METHODSPETROGRAPHY AND MINERAL...FLUID INCLUSIONSOXYGEN AND HYDROGEN ISOTOPESDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The Sulu terrane (Fig. 1) is the eastern end of the Qinling–Dabie orogen, which formed by the Triassic subduction and collision of the Yangtze craton beneath the Sino-Korean craton. Sulu is separated from Dabie Shan to the west by the Cretaceous left-lateral Tan-Lu fault (TLF) with a horizontal offset of about 500 km (Fig. 1). The Sulu terrane is bounded by the Yantai–Qingdao–Wulian fault (YQWF) to the NW and the Jianshan–Xiangshui fault (JXF) to the SE. It consists of a UHP, HP and migmatite metamorphic basement (Zhang et al., 1995
), intruded by post-orogenic Cretaceous granitic plutons. Mesozoic and more recent sediments constitute the youngest cover rocks (Wallis et al., 1999). In the Sino-Korean craton, syn- and post-orogenic lamproites and lamprophyres are reported (Lu et al., 1995; Guo et al., 2004).
|
In the southern Sulu terrane, near Donghai (Fig. 1), the UHP unit consists of amphibolite-facies orthogneisses, ultramafic rocks, eclogites, amphibolites retrograded from eclogites, and paragneisses. Generally, eclogites occur in orthogneisses as layers and/or boudins wrapped by the main foliation of the host rocks. The eclogites are locally interlayered with paragneisses and quartzites (Zhang, R. Y., et al., 1995
; Zhang, Z., et al., 2000). UHP eclogites from the Donghai area experienced a complex metamorphic history characterized by prograde, UHP peak, and retrograde metamorphism. The recent discovery of coesite in zircon from gneiss from Donghai and Taohang indicates that the country rocks also underwent regional UHP metamorphism (Ye et al., 2000; Liu et al., 2002). Prograde minerals (e.g. amphibole, paragonite, chlorite, staurolite, quartz, and phengite) may be preserved within peak garnet and pyroxene in eclogite (Wallis et al., 1999, and references therein) and within the cores of zircons in paragneiss (Liu et al., 2002).
Peak metamorphic conditions for eclogite are estimated at T = 700–890°C and P > 2·8 GPa (Zhang et al., 1995), and, more recently, at T = 600–700°C and P = 3·0–3·5 GPa (Mattinson et al., 2004), at T = 740–830°C and P = 3·0–3·9 GPa (Zhang et al., 2005a), and at T = 790–890°C and P = 3·5–4·0 GPa (Ferrando et al., 2005b). Garnet and omphacite, occurring together with coesite as inclusions within zircon in paragneiss, also indicate peak conditions at T = 815–850°C and P > 2·8 GPa (Liu et al., 2001). The early stages of the retrograde P–T path are characterized by decompression with moderate cooling (from UHP/HP eclogite- to amphibolite-facies conditions), whereas late exhumation stages (from amphibolite- to greenschist-facies conditions) are dominated by cooling and moderate decompression (Zhang et al., 1995; Ferrando et al., 2005b).
In the Sulu eclogites, metamorphic Sm/Nd ages from 210 to 240 Ma have been obtained (Li et al., 1993; Zhai et al., 2000), which are consistent with U–Pb dating of zircons at 217 Ma (Ames et al., 1996), and 228 Ma (Yang et al., 2003). The protolith age of an eclogite from the Donghai area is constrained at 762 ± 28 Ma by whole-grain analysis of zircon (Ames et al., 1996). Near Qinglongshan (Dongai area; Fig. 1), the rocks are characterized by oxygen and hydrogen isotope signatures acquired prior to subduction and preserved during subsequent prograde, UHP, and retrograde metamorphism. Extremely low 18O and D values suggest that the protolith of these rocks interacted with fluids in a hydrothermal meteoric water system during a period of cold climate (Yui et al., 1995; Zheng et al., 1996, 1998; Rumble & Yui, 1998). Combined zircon U–Pb geochronology and oxygen isotope analysis of metagranites from this area indicates that the hydrothermal fluid–rock interaction took place in connection with late Proterozoic intrusions, possibly during the Sturtian glacial episode (Rumble et al., 2002; Zheng et al., 2003).
In the Donghai area, the quartz-rich rocks include a garnet–quartz–jadeite lithology, jadeite quartzite, garnet–quartz–phengite schist, phengite–quartz schist, and kyanite quartzite (Zhang, R. Y., et al., 1995; Zhang, Z., et al., 2000). Among the kyanite quartzites, two types were identified by Zhang et al. (1996): Type 1: coesite pseudomorph-bearing quartzites interlayered with eclogites, and consisting of quartz, epidote, kyanite, and minor omphacite, garnet, phengite, and rutile; Type 2: kyanite quartzites, consisting of quartz, kyanite (including rutile and polycrystalline quartz aggregates after coesite), and accessory pyrite. In the latter, OH-rich topaz has been reported at Fushan, west of Dongai (Zhang et al., 2002). Five specimens of Type 2 kyanite quartzite from Hushan (west of Donghai; Fig. 1), which in all probability corresponds to the locality named Fushan by Zhang et al. (2002), were sampled from trenches 3 m deep to the SW of a ridge consisting of orthogneiss.
|
ANALYTICAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGY AND SAMPLE...ANALYTICAL METHODSPETROGRAPHY AND MINERAL...FLUID INCLUSIONSOXYGEN AND HYDROGEN ISOTOPESDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The chemical compositions of minerals were determined using both a JEOL-JXA 8600 electron microprobe, using wavelength dispersive spectrometry (WDS), at the CNR–IGG in Firenze, and a Cambridge Instruments Stereoscan 360 scanning electron microscope equipped with an energy-dispersive spectrometry (EDS) system at the Dipartimento di Scienze Mineralogiche e Petrologiche in Torino. WDS analyses were performed using 15 kV accelerating voltage, 20 nA beam current, 100 s counting time for each point analysis, and a beam diameter of 5 µm. Standards were: albite (Si, Na), ilmenite (Ti, Fe), plagioclase (Al), chromite (Cr), rhodonite (Mn), olivine (Mg), diopside (Ca), sanidine (K), fluorite (F), tugtupite (Cl). Operating conditions for EDS analyses were 15 kV accelerating voltage, 1·35 nA beam current, and 50 s counting time. Natural minerals and pure oxides were used as standards. The chemical composition of OH-rich topaz was calculated as described by Alberico et al. (2003
). Structural formulae of minerals were processed using the program by Ulmer (1986). Mineral abbreviations are after Kretz (1983). Microthermometry of fluid inclusions was performed on doubly polished, 100–150 µm thick, sections using a Linkam THMSG600 heating–freezing stage coupled with a microscope equipped with 40x or 100x objectives, at the Universities of Siena and Torino.
The stages were calibrated by a set of synthetic fluid inclusions with an estimated accuracy of about ±0·1°C at the triple point of CO2 (–56·6°C), at the triple point of H2O (0·015°C), and at the critical temperature of H2O (374°C). Freezing temperature (Tf), eutectic temperature (Te), final melting temperature (Tm), and homogenization temperature (Th) were measured during heating–freezing cycles. Heating rates were 0·1°C/min approaching Tm, and 0·5°C/min approaching Th. Microthermometric data to obtain fluid inclusion compositions, densities, and isochores were processed with the software packages FLUID 1 (Bakker, 2003).
Laser Raman analyses were made with a Labram microspectrometer (Horiba, Jobin Yvon Ltd) at the University of Siena. The excitation source was a polarized Ar+-ion laser operating at 514·5 nm wavelength, and 200–550 mW incident power. The laser spot size was focused to 1–2 µm with a 100x objective. Accumulation times were in the range 20–90 s. The 1332 cm–1 diamond band was used for the daily calibration.
Oxygen isotope compositions of mineral separates were measured at the CNR–IGG of Pisa, using the laser fluorination technique of Sharp (1990). Mineral separates were obtained from crushed, sieved (0·3 and 0·5 mm fraction), and ultrasonically cleaned samples. Pure mineral fractions (>99 vol. %) of quartz, kyanite, topaz and rutile were obtained by means of standard heavy liquid techniques, followed by hand-picking under a binocular microscope. Analyses on about 1–1·5 mg of material were duplicated and averaged. A 15 W Merchantek CO2 laser was used for heating the sample in a F2 atmosphere. The 18O values were measured on a Finnigan MAT Delta Plus mass spectrometer. The analytical precision was monitored by using the laboratory standards QMS and Lausanne1 (18O values were 14·05 and 18·1, respectively), and was always better than 0·12 (1
). During this study the average composition of NBS 28 at CNR–IGG was 9·54 ± 0·17 (n = 9). All isotope values are reported in the conventional 18O notation relative to SMOW. The D/H ratios of paragonite and topaz were also measured at the CNR–IGG, and duplicated at the University of Budapest for intercomparison by conventional vacuum fusion, following the method of Vennemann & O’Neil (1993). The D values were compared with those of an internal standard, calibrated relative to NBS-30 (D = –67).
|
PETROGRAPHY AND MINERAL CHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGY AND SAMPLE...ANALYTICAL METHODSPETROGRAPHY AND MINERAL...FLUID INCLUSIONSOXYGEN AND HYDROGEN ISOTOPESDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The studied quartzites are weakly foliated, medium- to coarse-grained, and consist of quartz (or coesite) (
70–80 vol. %), kyanite (10–20 vol. %), white mica (up to 5 vol. %) ±OH-rich topaz, ±pyrophyllite, and accessory rutile, pyrite, zircon, apatite, ±barite and diaspore (Fig. 2a and b). Two kyanite generations are present: an early coarse-grained kyanite (Ky I), which occurs as stubby crystals with no preferred orientation, and a later medium-grained prismatic kyanite (Ky II), whose preferred orientation defines a weak foliation (Fig. 2a). Kyanite I includes polycrystalline quartz (after coesite), rutile, zircon, and fluid inclusions, and rarely preserves a folded Si defined by alignment of very fine-grained quartz or zircon crystals (Fig. 2c). Kyanite II is locally deformed and has no inclusions of polycrystalline quartz, single quartz crystals, or fluids. In some samples, both kyanite generations show a weak bluish color, as a result of the presence of minor amounts of iron [Fetot 0·02 atoms per formula unit (a.p.f.u.); Table 1].
|
|
White mica consists of both muscovite and paragonite, in variable amounts. Muscovite occurs as medium- to fine-grained flakes typically grown on Ky I, and rarely on Ky II (Fig. 2b). Silicon ranges from 3·04 to 3·17 a.p.f.u., Na is up to 0·37 a.p.f.u., and Mg does not exceed 0·09 a.p.f.u. (Table 2). Muscovite shows high F contents for metamorphic rocks [up to 0·44 wt %: XOH = OH/(OH + F) = 0·95–0·96], but no Cl. Late paragonite is also present, and occurs as randomly oriented flakes forming more or less continuous coronae that mantle kyanite I and II, topaz, and locally muscovite (Fig. 2a, d and e). Minor paragonite is additionally found as fine-grained crystals within mineral fractures. The Na content ranges from 0·72 to 0·94 a.p.f.u., and the K content is <0·22 a.p.f.u.; Table 2). Fluorine ranges from 0·48 to 0·89 wt % (XOH = 0·91–0·95). In the most retrogressed samples, minor amounts of pyrophyllite containing F (
0·30 wt %; XOH 0·98; Table 2) developed at the expense of kyanite, locally overgrowing paragonite (Fig. 2d).
|
OH-rich topaz occurs sporadically in some horizons, where it may constitute up to 10 vol. %. The OH-rich topaz is clearly retrograde and developed at the expense of both Ky I and Ky II. In topaz, Ky I relics are often observed as corroded and oriented fine-grained crystals showing uniform extinction (Fig. 2a), whereas topaz completely replaces Ky II. Although a few OH-rich topaz grains appear to be deformed, they always show homogeneous extinction (Fig. 2f), indicating pseudomorphous replacement after former deformed Ky II, oriented along the main foliation. OH-rich topaz includes rutile and zircon, but no quartz. Rare diaspore, formed after relic kyanite, is locally present. Topaz is rich in fluid inclusions, which may be locally so abundant as to make it ‘dusty’ under the optical microscope. The significant amounts of OH-bonds in the topaz structure are revealed by Raman spectroscopic analyses with an intense vibration at 3637 cm–1 (not shown). WDS and X-ray diffraction analyses indicate a mean OH content with XOH = OH/(OH + F) = 0·28 (Table 1; Alberico et al., 2003
) without any evident zoning from the core to the rim. Such XOH value is in the range for topaz crystals from Fushan (XOH from 0·28 to 0·39) reported by Zhang et al. (2002). |
FLUID INCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGY AND SAMPLE...ANALYTICAL METHODSPETROGRAPHY AND MINERAL...FLUID INCLUSIONSOXYGEN AND HYDROGEN ISOTOPESDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Petrography, microthermometry, and Raman microspectroscopy
Aqueous fluid inclusions in quartzite are generally solute-rich with different solutes. Carbonic fluids are rare. Inclusion distribution is limited and localized within single minerals: primary inclusions are observed both in peak Ky I, and in OH-rich topaz, whereas late intra- and inter-granular inclusion trails occur in matrix quartz. Kyanite II and quartz after coesite do not contain fluid inclusions. Five fluid inclusion populations are discussed in the following sections, according to their relative timing of trapping. A selection of representative analytical results is reported in Table 3.
|
Primary alumino-silicate-rich multiphase solid inclusions in kyanite I (Type I)
Within porphyroclastic Ky I, multiphase solid inclusions are distributed randomly, or along short planes of original healed microfractures (Fig. 3a). These inclusions typically contain a variety of mineral phases, one or more deformed bubbles, but no liquid phase (Fig. 3b). Such inclusions, which are found also in peak phases of UHP eclogite from the same locality, have been the subject of a previous detailed petrographic study, which aimed at reconstructing their original chemical composition (Ferrando et al., 2005
a). Here, we summarize the major characteristics. Paragonite, muscovite, and anhydrite constitute the dominant mineral assemblage within the inclusions, with rather constant volume proportions, indicating that these are daughter minerals (Fig. 3b). A second sulfate phase containing Al and variable amounts of K and Na is often observed, which has been identified by Raman spectroscopy as an hydroxyl-bearing K–Na-sulfate [possibly alunite-type: KAl3(SO4)2(OH)6; Ferrando et al., 2005a]. Chlorides (e.g. halite) are absent. Additional phases may include corundum and diaspore, calcite, chlorite, barite, pyrite, and apatite, which are considered incidentally trapped phases [see Ferrando et al. (2005a) for a discussion of secondary processes]. A semi-quantitative estimate of volumetric proportions indicates that Type I inclusions contain approximately 20% paragonite, 20% ‘alunite-type’ sulfate, 20% muscovite, 15% anhydrite, and 20–30% of ‘fluid’. The remaining 5 vol. % consists of one or more incidentally trapped phases.
|
Primary Ca-rich brine inclusions in OH-rich topaz (Type II)
In OH-rich topaz, fluid inclusions are extremely abundant in the central portions of the crystals, whereas they are systematically absent from the rims (Fig. 3c; primary fluid inclusions, Roedder, 1984
). They are generally two- or three-phase aqueo-carbonic (LCO2–LH2O, and VCO2–LCO2–LH2O) and show rounded or negative crystal shapes (5–50 µm in size). Only some fluid inclusions contain daughter crystals, defining a four-phase assemblage (VCO2–LCO2–LH2O–S). The most common daughter phase is halite, followed by gypsum and anhydrite (Fig. 3d and e), both identified by Raman microspectroscopy (Fig. 4). Within each inclusion, CO2 varies from 5 to 50% of the total volume. There is an observed correlation between size of the inclusions, CO2, and daughter mineral content: larger inclusions generally contain daughter minerals and a larger volume per cent of low-density CO2 (i.e. two-phase, L + V) than smaller ones (Fig. 3d and e).
|
The aqueous part of Type II fluid inclusions freezes (Tf) between –70 and –58°C (Table 3). The first melting is observed between –54 and –42°C, close to the eutectic of the NaCl–CaCl2–H2O system (Fig. 5a). This result implies that divalent ions, notably Ca2+, may be present in solution. Hydrohalite melting (Tm Hhl) occurs in the range of –35·1 to –25·3°C (mean at –32·3°C), and the Tm ice varies from –19·0 to –7·1°C (mean at –13·0°C). The carbonic part of Type II fluid inclusions melts instantaneously at temperatures (Tm CO2) very close to the pure CO2 triple point (–56·6°C). The Th CO2 are reported in Fig. 5b and scatter between –7·7 and 31·0°C. Commonly CO2 homogenizes to the liquid (L + V
L), but both homogenization into the vapor (L + V V) and critical homogenization (L + V C) have also been measured. The total homogenization (LCO2 + Lbrine homogeneous fluid), and the dissolution of daughter phases could not be measured, as inclusions decrepitate during heating.
|
Secondary fluid inclusions in matrix quartz: Na-rich brines (Type III), CO2 (Type IV) and low-salinity H2O (Type V) inclusions
In matrix quartz, three distinct populations of fluid inclusions coexist. These inclusions have different compositions and represent distinct fluid trapping events.
Type III aqueous fluid inclusions are limited to short intragranular trails extending from kyanite and OH-rich topaz (Fig. 3f). They are biphase (L + V) aqueous inclusions with a degree of filling [df = L/(L + V)] equal to 0·90–0·95. Inclusions (3–15 µm in size) commonly show evidence for partial decrepitation. Figure 6a shows the range of freezing and melting temperatures for Type III fluid inclusions. Tf is between –38 and –35°C, Te H2O varies from –35 to –28°C. These temperatures point to the presence of Mg2+ and/or Fe2+ ions, which have eutectics in the range of –38 to –35°C. Tm Hhl is recorded between –24·8 and –19·0°C (mean at –23·3°C), and Tm ice varies between –7·6 and –4·8°C (mean at –7·4°C). The Th LH2O varies from 101·5 to 259·5°C and makes a symmetric histogram with a peak at around 190°C (Fig. 6b).
|
Rare Type IV CO2 fluid inclusions are present as intragranular trails within matrix quartz (Fig. 3g). These are high-density single-phase CO2 (L) inclusions (1–10 µm in size) with polygonal contours, which suggest re-equilibration after trapping in the quartz. Evidence for partial decrepitation is common. Timing relationships between Type IV CO2 inclusions and Type III aqueous inclusions are based on the observation of a few trails of Type IV inclusions cutting Type III inclusions. Initial melting of CO2 is recorded at –56·8 to –56·9°C (Table 3), and final melting occurs at –56·7°C. Raman analyses show traces of N2 (< 0·1 mol%). The Th LCO2 values, between –4·4 and 9·5°C (Fig. 6e), are strongly asymmetric, with most measurements around 7· 4°C.
Type V liquid-rich biphase (L + V) aqueous inclusions (1–5 µm) postdate all other inclusion types and are identified as intergranular trails, locally deformed, within matrix quartz (Fig. 3h). Inclusions freeze (Tf) between –36 and –32°C, whereas Te are between –23 and –21°C, close to the eutectic temperature of the NaCl–H2O system. Tm ice varies from –6·5 to –4·7°C (mean at –5·9°C), corresponding to 7·5–9·9 (mean = 9·1) NaCl wt % (Fig. 6c). Th LH2O ranges from 102·3 to 222·3°C, and makes an asymmetric histogram (Fig. 6d).
Fluid composition and density calculation
In Type I primary inclusions, paragonite, muscovite, anhydrite and ‘alunite’ represent the daughter minerals precipitated from the originally trapped fluid, as indicated by their ubiquitous presence and constant volumetric ratios. The constant relative volume proportions of these phases within inclusions show that the fluid was homogeneous at the time of trapping. The average bulk composition of the inclusions is: 30 wt % Al2O3, 24 wt % SiO2, 11 wt % SO3, 9 wt % CaO, 5 wt % K2O, 3 wt % Na2O, with minimum amounts of TiO2, Fe2O3, FeO, MgO, BaO, P2O5, and (CO3)2– (Ferrando et al., 2005a). The minimum calculated water content is 20–25 wt %, based on the volume of empty cavities in inclusions, which is, however, certainly underestimated, and consequently only a proxy for the original water contents. During decompression at high pressures, selective removal of water from the fluid inclusions is a very common process, which may occur by fluid-inclusion decrepitation, leakage through dislocations, H2 diffusion, and chemical reactions with host minerals at low temperature (Touret & Frezzotti, 2003). Presence of diaspore within the inclusions, however, indicates that liquid water was present at HP, and reacted with corundum to produce diaspore, during the exhumation of the host rocks.
Although the obtained compositions can only approximate those of the originally trapped fluids, they provide important constraints on the chemical nature of the fluids evolved during peak metamorphic conditions: (1) fluids are aqueous and contain high solute contents; (2) the dominant solutes are Si, Al, S, Ca, K and Na, whereas Mg and Fe are virtually absent; (3) Cl-ligands are negligible.
The calculated composition for preserved Type II inclusions is XH2O = 0·87, XCO2 = 0·04, Xsalt = 0·9; the aqueous part of the fluid is Ca-rich brine containing 14–15 CaCl2 and 2·5–3·5 NaCl in wt %. The presence of anhydrite and gypsum as daughter phases in some inclusions is consistent with a Ca-dominated solution, and further indicates that (SO4)2– and (HSO4)– represent important anions. A few of the smallest inclusions have an extremely high density (up to 1·16 g/cm3), indicating that the original fluid inclusion densities may be locally preserved in topaz, despite a remarkable pressure differential during decompression. Re-equilibrated Type II inclusions have a considerably lower density (0·76 g/cm3) and H2O/CO2 ratios (XH2O = 0·86, XCO2 = 0·07, XNaCl = 0·07), which suggest that stretching at high temperatures and selective water loss (Bakker & Jansen, 1990) might have been the main re-equilibration process for primary inclusions in OH-rich topaz during decompression.
Type III aqueous inclusions in matrix quartz are NaCl-rich fluids, containing one or more divalent cations. As inclusions freeze at temperatures above the eutectic of the NaCl–CaCl2–H2O system, we conclude that CaCl2 is not significant in the solution. In terms of the NaCl–MgCl2–H2O system, taken as representative for solutions with a eutectic in the observed temperature range, average salinity is 4·5 NaCl wt %, and 6·5 in MgCl2 equiv. wt % (Bakker, 2003). Extreme caution, however, should be taken when considering the nature of the divalent cations present in Type III inclusions, as Mg2+ should be present only at very low concentrations, much smaller than those of Fe2+ and Mn2+ in similar rock-equilibrated fluids. The fluid density has been calculated using the minimum Th LH2O, yielding values of 1· 02 g/cm3.
Type IV fluid inclusions contain CO2 with traces of N2 with a maximum density of 0·93 g/cm3. Such a value represents the minimum possible density of trapped fluids. Indeed, the widespread re-equilibration features observed in inclusions strongly suggest that all fluid densities might have been reset to lower values.
Type V aqueous fluid inclusions represent the latest trapped fluids in the studied rocks. They are NaCl solutions with a salinity between 7·5 and 10 NaCl wt %, and a density of 1 g/cm3.
|
OXYGEN AND HYDROGEN ISOTOPES |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGY AND SAMPLE...ANALYTICAL METHODSPETROGRAPHY AND MINERAL...FLUID INCLUSIONSOXYGEN AND HYDROGEN ISOTOPESDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The low
18O values of the minerals separated from the Sulu OH-rich topaz–kyanite quartzite (–0·3 < 18OQtz < 0·2; –2·1 < 18OKy < –1·3; 18OOH-Toz = –1·8 and –1·7; 18ORt = –6·4) are similar to those reported by Rumble & Yui (1998), and significantly lower than values expected for unaltered metapelitic rocks having undergone high-grade metamorphism. The complete dataset is reported in Table 4. The D values of OH-rich topaz and muscovite are –70 ± 12 and –115 ± 15, respectively, which are also similar to the D values reported for hydrous silicates from Dabie Shan (Zheng et al., 1998, 1999; Xiao et al., 2002). The 18OToz and DToz values are the first reported for this HP phase, and the large scatter in the D values probably results from the partial destabilization of the selected minerals (i.e. Fig. 2d: typically paragonite or pyrophyllite-mantled muscovite; Fig. 2e: paragonite flakes around OH-rich topaz). The isotope data for the minerals indicate that the UHP rocks from the Dabie–Sulu orogen are characterized by anomalously low oxygen and hydrogen isotope ratios and that the protoliths underwent meteoric–hydrothermal alteration before UHP metamorphism occurred (e.g. Zheng et al., 2003).
|
The 18O/16O ratios decrease as follows: quartz > kyanite
OH-rich topaz > rutile, as expected for equilibrium fractionation (Zheng, 1993a, 1993b), and suggest that the rocks experienced either (1) negligible open-system retrograde metamorphism, or (2) infiltration of a low 18O fluid, which did not alter significantly the peak metamorphic O-isotope composition.
Because of the refractory nature of the Al2SiO5 polymorphs to diffusional exchange upon cooling, even at high P–T metamorphic conditions (Fortier & Giletti, 1989; Young, 1993), oxygen isotope geothermometry provides accurate temperature estimates in alumino-silicate-bearing rocks that cooled under anhydrous conditions (Ghent & Valley, 1998; Moecher & Sharp, 1999; Vannay et al., 1999; Putlitz et al., 2002). Using the equation 1000 ln 
= a x 106/T2 (where is the fractionation factor related to the temperature = (1000 + 18OphaseA)/(1000 + 18OphaseB), O-isotope based temperatures were estimated for the following mineral pairs: Qtz–Ky (a(Qtz–Ky) = 2·25 ± 0·2; Sharp, 1995), Qtz–Toz (a(Qtz–Toz) = 2·25 ± 0·2; Zheng, 1993a, 1993b), and Qtz–Rt (a(Qtz–Rt) = 5·02; Matthews, 1994). At peak metamorphic conditions (stage A) the stable SiO2 polymorph was coesite; this was transformed into quartz during the stage B transition. The calculated temperature may be slightly lower than the original one because of pressure effects, which have been estimated to be of the order of 40°C at 4 GPa (Sharp et al., 1992).
Oxygen isotope thermometry was combined with the conventional ‘cation-based’ thermometric estimates for other rocks in the same area, to constrain the metamorphic conditions of the Sulu quartzites and to reconstruct the metamorphic evolution of this terrane (see following paragraph).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGY AND SAMPLE...ANALYTICAL METHODSPETROGRAPHY AND MINERAL...FLUID INCLUSIONSOXYGEN AND HYDROGEN ISOTOPESDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Metamorphic evolution
Although the studied quartzites show a relatively simple mineralogy, the scarce evidence of retrogression allows us to reconstruct the polyphase metamorphic evolution, summarized in Fig. 7. A P–T path characterized by four metamorphic stages—from the metamorphic peak to the late stages of exhumation—is shown in Fig. 8, combining microstructural relationships, mineralogical compatibilities, fluid inclusion isochores, stable isotope geothermometry, and theoretical phase relationships. The petrogenetic grid is based on selected phase relationships in the Na2O–K2O–Al2O3–SiO2–H2O–CO2 (NKASCH) system with excess SiO2, calculated with the thermodynamic approach of Connolly (1990
), using a modified ‘internally consistent’ thermodynamic database of Holland & Powell (1998). All equilibrium curves are calculated for aH2O = 1, except for those involving paragonite, and OH-rich topaz. The curve for the reaction Pg = Ky + Jd + fluid at aH2O = 0·75 is extrapolated from the experiments of Tropper & Manning (2004). The isopleth for the reaction Al-silicate + fluid = Toz is calculated considering a topaz with XOH = 0·30 (i.e. close to XOH = 0·28, obtained from WDS and X-ray diffraction analyses of topaz), and a H2O–NaCl–CO2 fluid phase with XH2O = 0·90 (XH2O = 0·87 in Type II inclusions). Thermodynamic models for low-pressure reactions involving topaz, with variable XOH between 0·05 and 0·40, were developed by Barton (1982), whereas those for high-pressure reactions (Wunder et al., 1993, 1999), are restricted to topaz with XOH = 1. To obtain the isopleth for topaz with XOH = 0·30, the thermodynamic database of Holland & Powell (1998) has been modified by the insertion of thermodynamic data for a topaz with XOH = 0·30 following the model proposed by Barton (1982) for the hydroxyl-topaz in solid solution with fluor-topaz, i.e. aToz = (XOH)2. The dry melting curve for Ms, with excess SiO2, to give Ky + liquid (L) is from Huang & Wyllie (1973), whereas the equilibrium curve involving diamond is from Bundy (1980). The calculated isochores for Type II, III, IV and V fluids, and the geothermometry by oxygen isotopes are also shown (Fig. 8).
|
The bulk chemical composition of the quartzite suggests a sedimentary protolith, such as a clay-rich sandstone (see also Zhang et al., 2000
, 2005a). Rare quartz or very fine-grained zircon crystals within Ky I preserve a folded pre-peak foliation, yet their occurrence is not sufficient to estimate the prograde P–T conditions.
The peak metamorphic stage (A in Fig. 8) is represented by the anhydrous assemblage Ky I + coesite + rutile. Temperatures calculated from the 18O values of quartz and kyanite (
18OQtz–Ky) in Toz-free samples are 730–820°C (Table 4). At these temperatures, the occurrence of coesite and the absence of diamond constrain pressures conditions between 2·9 and 3·5 GPa. These pressure conditions are in agreement with those obtained from nearby eclogites (3·5–4·0 GPa, Ferrando et al. 2005b). In contrast, O-isotope thermometry on Ms- and Toz-bearing samples RPC 546 and RPC 547 yields temperatures of 980 ± 30°C and 950 ± 100°C, respectively, which are not geologically plausible for the Donghai area. These T estimates probably result from O-isotope disequilibrium between quartz and Kyanite I (±Kyanite II).
|
The first decompression stage (stage B in Figs 7 and 8) is characterized by an earlier crystallization of Ky II (event B1) and a later growth of hydrous phases (event B2). OH-rich topaz possibly formed by the reaction Ky + fluid = Toz, similar to that invoked for the prograde Ky–Toz breakdown. For this reason, the transition from anhydrous (event B1) to hydrous (event B2) minerals during stage B is more probably related to an increase in the water activity rather than to a pronounced variation in the P–T conditions. Probably, the water entered the quartzite at the coesite–quartz transition, promoted by the volume change that accompanies this reaction.
The 18O value of kyanite, measured in Toz-bearing samples, possibly results from the combination of 18OKyI and 18OKyII, or more probably (mainly coarse-grained minerals were hand-picked), from the 18O value of Kyanite I acquired during hydration at the early decompression stage. It is worth noting that oxygen isotope geothermometry based on the Qtz–Toz pair constrains temperature conditions for metamorphic stage B at 705–780°C (Table 4), matching the temperature estimates based on the Qtz–Ky pair in Toz-free rocks. The fluid isochore, obtained from the extremely high-density Type II brines (1·16 g/cm3), indicates a pressure of 2·7 GPa, and locates stage B at the coesite–quartz transition (Fig. 8). The P–T estimates indicate a near-isothermal decompression, as inferred from the observed metamorphic parageneses.
Calculated P–T conditions for topaz growth are close to those proposed by Zhang et al. (2002) for prograde-peak OH-rich topaz from Fushan (700°C and 2·8 GPa). However, in our samples textural evidence clearly indicates that topaz is retrograde and replaces UHP peak Ky I, excluding the growth of topaz during stage A. Since the stability of the F–OH topaz solid solution is defined by P, T, OH content, and aH2O in the fluid phase, the absence of OH-rich topaz at the metamorphic peak may be explained by a variable water activity during the metamorphic evolution. As the equilibrium curves for the reaction Al-silicate + fluid = Toz, in the presence of a fluid phase with aH2O < 1, are not known, we have calculated a petrogenetic grid (Fig. 9), which shows some of the variations of the stability field of OH-topaz (XOH = 0·30) for different water activities in the fluid, using the thermodynamic approach described above. At stage B (705–780°C and 2·9 GPa), the water activity in Type II brines has been calculated at about 0·75 (Aranovich & Newton, 1999) and is consistent with the stability of topaz. Based on Fig. 9, during stage A at UHP conditions (730–820°C and 2·9–3·5 GPa), fluid water activity was low, and 0·4, to maintain the anhydrous assemblage Ky + Coe. At higher water contents, OH-rich topaz would have been the stable phase, instead of kyanite.
|
Stage C (Fig. 8) is characterized by growth of paragonite after both kyanite and OH-rich topaz, and by recrystallization of matrix quartz. In particular, both the destabilization of the OH-rich topaz and the crystallization of paragonite constrain the temperatures to values between 700 and 780°C [curves Toz (XOH = 0·30) = Al-silicate + fluid, and Pg = Al-silicate + Ab + fluid, in Fig. 8]. A minimum pressure of 1·6 GPa is obtained from Type III fluid isochores, and 1·9 GPa is calculated from the curve of destabilization of kyanite (Fig. 8; Ky + Jd + fluid = Pg at aH2O = 0·75; Tropper & Manning, 2004
). The last stage of retrogression (stage D in Fig. 8) is characterized by the growth of pyrophyllite and diaspore after kyanite. The reactions Ky + fluid = Prl and Ky + fluid = Dsp indicate retrograde temperatures lower than 410°C and 370°C, respectively. Pressure conditions of about 0·3 GPa, consistent with greenschist-facies conditions, are obtained by the fluid isochore calculated for Type V fluid inclusions.
The reconstructed P–T path shows that the quartzites are characterized by a UHP peak, strong near-isothermal decompression from UHP to HP conditions, and almost isobaric cooling from amphibolite- to greenschist-facies conditions (Fig. 8), similar to P–T paths for eclogite reported in previous studies of the Donghai area (e.g. Zhang et al., 1995; Hirajima & Nakamura, 2003; Ferrando et al., 2005b; Zhang et al., 2005a). Finally, the reconstructed P–T metamorphic evolution demonstrates that combined stable isotope geothermometry and fluid inclusion isochores may be a viable alternative to cation geothermobarometry in UHP–HP rocks characterized by unfavorable mineral assemblages.
Nature of fluid phases evolved at UHP and HP, and fluid–rock interaction
Fluids at UHP conditions: alkali-alumino-silicate aqueous solutions, intermediate between an aqueous fluid and a hydrous melt
The presence of primary Type I multiphase solid inclusions in peak Ky I indicates that the dominant peak mineral association of kyanite and coesite at 3·5 GPa and 730–820°C was formed in the presenc of a hydrous fluid phase in the quartzites. Peak aqueous fluids contain high amounts of Al, Si, S, Ca, Na, and K (Fig. 10). Conversely, ligands such as Cl seem to be irrelevant, and the rarity of carbonates indicates a very subordinate role for CO2 as well. The calculated range of Type I fluid water content, although speculative, is >25 and 48 wt %: the minimum value has been derived from Type I inclusions, and the maximum value has been calculated based on the stability field of OH-topaz derived from phase equilibria (see Figs 9 and 10).
|
The first remarkable feature of deep Type I fluids is represented by the dominant alkali–alumino-silicate (-sulfate) character of the solutes. This composition is different from the Cl-rich brines reported at peak conditions in other HP and UHP rocks from Sulu and other localities (Philippot & Selverstone, 1991
; Selverstone et al., 1992; Svensen et al., 1999; Fu et al., 2003; Zhang et al., 2005b, 2006; Xiao et al., 2006). In addition, Type I fluids are highly concentrated, resulting in an unusual composition, which is intermediate between a fluid and a melt. However, it is highly improbable that these fluids represent remnants of a water-rich silicate melt trapped as inclusions in peak kyanite. Based on previous experimental studies in metapelites, muscovite (e.g. phengite) is the major hydrous mineral potentially involved in melting at P > 3 GPa. In phengite-dominated melting, first melts are potassic high-silica rhyolites (Schmidt et al., 2004, and references therein). Type I fluids do not have, or even approximate, a rhyolitic composition, as they are too poor in SiO2, and too rich in Al2O3, CaO, and SO3. Furthermore, calculated water contents of Type I fluids are too high for a silicate melt, as the maximum water dissolved in a melt at the considered pressures cannot exceed 25–30 wt % (Kennedy et al., 1962; Boettcher & Wyllie, 1969).
None of the obtained compositions could result from selective major oxide enrichments during retrograde inclusion evolution (see Frezzotti, 2001). There is, in fact, no evidence for chemical interaction with the host kyanite (Ferrando et al., 2005a), and even if crystallization of the host kyanite occurred on the inclusion's wall, it would not modify the Al2O3/SiO2 ratio of the melt.
We propose that the Type I UHP fluids represent intermediate alkali–alumino-silicate aqueous solutions (H2O 50 wt %) at high pressures. A similar intermediate composition might indicate that fluids have evolved at pressures above the critical curve of the silicate–H2O system; that is, their composition reflects supercritical hydrous silicate liquids (Stalder et al., 2000; Kessel et al., 2005). The P–T conditions of the second critical end-point on the melting curve strongly depend on the nature of the silicate–H2O systems, ranging from 1 to >12 GPa, from silica-rich to mafic and ultramafic compositions (Kennedy et al., 1962; Anderson & Burnham, 1965; Eggler & Rosenhauer, 1978; Ryabchikov, 1993; Mysen, 1998; Bureau & Keppler, 1999; Stalder et al., 2000, 2001). In the simple NASH system, Bureau & Keppler (1999) have demonstrated that critical conditions might be attained at pressure as low as 1· 6 GPa, and experimental results in natural granitic rocks indicate pressures of 2·0–2· 6 GPa (Schreyer, 1999; Kawamoto, 2004). In general, the increase in Mg, Fe, K, and Ca contents greatly raises the pressure of the second critical point, whereas addition of volatile elements, such as B and F, has the opposite effect (Thomas et al., 2000; Sowerby & Keppler, 2002).
It is not known where the intersection between the critical curve and the H2O-saturated solidus lies, or even the existence of a system relevant to this bulk composition (note, however, that an end point need not exist for the fluid or liquid to be supercritical).
The existence of an intermediate fluid (H2O 50 wt %) enriched in Al, Si, S, Ca, K, and Na, leads us to propose that at 3–3·5 GPa and 780°C rocks in the (NK)ASH system may generate fluids of intermediate composition, which are at supercritical conditions.
Fluid alkali–alumino-silicate components mirror the bulk composition of the host rock, and suggest that Type I intermediate solutions may have been formed by progressive dehydration of muscovite ± paragonite during prograde and UHP peak metamorphism. To be really effective this process must have acted on a small volume of fluid locally, trending towards increasing solute content and a heterogeneous distribution of fluids. As Type I intermediate solutions are internally derived, we may speculate that they were ultimately (indirectly, via hydro-silicate breakdown) derived from the meteoric–hydrothermal event. This observation has been recently inferred also on the basis of a Fourier transform infrared investigation in nominally anhydrous minerals from Dabie-Shan eclogites (Xia et al., 2005).
Fluids at the transition from UHP to HP conditions: CaCl2 brines
The earliest isothermal decompression stages, at the transition from UHP to HP conditions (2·9 GPa and 705–780°C), led to a progressive growth of hydrated phases (such as muscovite and OH-rich topaz), and of minor barite and pyrite. Type II fluid inclusions indicate that OH-rich topaz grew from a Ca-rich Cl and (SO4)2– brine, where Si and Al are virtually absent (Fig. 10). Muscovite does not contain fluid inclusions, but it probably grew in the presence of the same fluid. The brines include traces of CO2 and should also have contained F, accommodated in both topaz and muscovite. Total fluid salinity is 18–20 in NaCl equiv. wt % as indicated by the Type II inclusions, which have preserved the original densities.
Assuming that silica, alumina and K2O were consumed during the growth of both muscovite and topaz, then the nature of most of the solutes present in Type II fluids is close to that of Type I intermediate solutions. Two major differences, however, are observed (Fig. 10): the presence of high amounts of Cl (12 wt %), and a sharp increase in the water content (82 wt % H2O). An evolution in a closed fluid system could not account for the observed variations in fluid chemistry. If, in fact, lowering of solubility as a result of decompression led to selective precipitation of a large part of the solutes from Type I fluids, initiating the growth of muscovite and OH-rich topaz, it seems unlikely that this represents the only process responsible for the observed increase of water activity. Most importantly, as much as 12 wt % Cl is present in Type II brines, which cannot be derived from Type I fluids because these do not contain significant chlorine.
In Sulu UHP rocks, similar HP CaCl2-rich brines are commonly observed in both low and high 18O rocks (Fu et al., 2003; Xiao et al., 2006). These brines are, therefore, unrelated to the O-isotope composition of UHP rocks, as expected for remnants of internally derived UHP and/or HP fluids. In the studied rocks, brines are present within topaz-bearing quartzite, whereas they are absent in topaz-free domains. As mentioned above, the paragenetic minerals from the topaz-bearing rocks show low 18O values similar to those of the topaz-free samples; however, quartz and Ky I from these domains are not in O-isotope equilibrium, as defined by unrealistic thermometric estimates, whereas O-isotope equilibrium is preserved in topaz-free rocks (Table 4). This observation clearly indicates that topaz-bearing domains were infiltrated by low-18O fluids, which are similar in O-isotope composition to the low-18O quartzites, although slightly higher. These necessarily external (Cl-rich) fluids infiltrated the quartzites upon topaz crystallization, thus after the UHP metamorphic peak.
If the new 18Oky results from external fluid infiltration, it may be qualitatively used to infer the water/rock (W/R) ratios in the topaz-bearing domains upon fluid infiltration according to the mass-balance equation
|
|
is the 18Oky of topaz-bearing domains, 
is the 18Oky of topaz-free domains, 18OiW is the O-isotope composition of the infiltrating fluid and ky–H2O is the isotopic fractionation between kyanite and water, that is –0·90 at T = 700°C (Zheng, 1993a). Because the ky–H2O is negative, the 18OiW needs to be >0. For a ky–H2O value between zero and +2 the resulting W/R ratios range from 1·8 to 0·3. These W/R ratios should be considered minimum values, because the water is assumed to have reacted completely with the rock.
The W/R values do not change significantly if an open-system infiltration model is assumed (= ln[W/Rclosed + 1]). It could be argued that the zero-dimensional model, which describes the simple mass-balance calculation used to estimate the W/R ratio, may not be adequate for a very low-porosity rock in which very low amounts of fluids entered during exhumation from UHP to HP conditions. What should be pointed out, however, is that at least discrete amounts of CaCl2-rich brines with higher 18O values entered the quartzite rock system during the early stages of decompression, indicating that the retrograde fluid–rock interaction locally occurred in a open system, probably generated at the coesite–quartz inversion. Brines probably originated from the mafic and ultramafic eclogitic rocks adjacent to the quartzites, where they have been recognized as dominant UHP and HP fluids (Xiao et al., 2000, 2001; Franz et al., 2001; Fu et al., 2001, 2003; Zhang et al., 2006).
Fluids at HP conditions: NaCl brines ± CO2
The final stages of the exhumation history preserve at least three distinct events of limited fluid flow within the quartzites. Fluid compositions were constrained by Na-rich brines with higher water activities (Type III), followed by minor pure CO2 fluids (Type IV), and by late low-salinity brines (Type V) (Fig. 10). The Na-rich brines (Type III) were present during the events that resulted in the growth of paragonite and the recrystallization of quartz at pressures below 2 GPa: their Na-rich nature may account for the onset of paragonite crystallization after both kyanite and topaz. Type III fluids are distinctly different from Type II brines (Fig. 10), mostly because of the absence of Ca and lower salinity. Whether this reflects post-entrapment diffusion of elements between fluid inclusions and host mineral, or initial geochemical differences is not clear. Despite the lack of petrographic evidence for the formation of calcite and/or gypsum by retrograde reactions in the rock, a decrease in Ca solubility is still a plausible mechanism to explain the disappearance of Ca from the fluid at lower pressure conditions.
CO2 fluids (Type IV) are rare and do not show any microstructural indication for heterogeneous trapping in inclusions along with Type III brines, excluding the possibility that CO2 might represent an immiscible fluid phase coeval with Type III brines. The distribution of CO2 inclusions suggests that their formation occurred at a later stage (Fig. 10). CO2 fluids are rather common late-stage fluids in UHP rocks from Dabie–Sulu. The general interpretation is that CO2 is external in origin and possibly locally derived from granulites (Touret, 2001; Touret & Frezzotti, 2003).
The last fluid event at greenschist-facies conditions is characterized by the presence of rare aqueous fluids, which were locally trapped as Type V inclusions (Fig. 10). These fluids, very common in the latest stages of retrogression of the studied metamorphic rocks, promote the growth of late hydrous minerals, such as pyrophyllite and diaspore, in the most retrogressed samples.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGY AND SAMPLE...ANALYTICAL METHODSPETROGRAPHY AND MINERAL...FLUID INCLUSIONSOXYGEN AND HYDROGEN ISOTOPESDISCUSSIONCONCLUSIONSREFERENCES |
|---|
OH-rich topaz–kyanite quartzites from the Sulu UHP metamorphic terrane are characterized by an anhydrous peak mineral assemblage that formed at 730–820°C and 2·9–3·5 GPa in the presence of intermediate fluids, containing significant amounts of Si, Al, SO4, Ca, K, and Na in solution, which probably represent supercritical liquids. They are probably derived from dehydration reactions, occurring during prograde and peak metamorphism, at UHP conditions. Peak alkali–alumino-silicate solutions are oxidized (sulfur present as sulfate) and do not contain significant CO2. Oxidized conditions and absence of CO2 might explain the extreme scarcity of micro-diamond in this area.
During early decompression, at the transition from UHP to HP conditions (2·9 GPa and 705–780°C), fluids are characterized by a high CaCl2 content, and by a sharp increase in the water activity (from 0·4 to 0·75). At these conditions, muscovite and OH-rich topaz crystallized at the expense of kyanite. Increase of water activity and addition of Cl– suggest that brines cannot have been derived from intermediate UHP solutions by decompression. Oxygen isotope data indicate discrete influx from neighboring lithologies, at the coesite–quartz transition.
Later stages at amphibolite- and greenschist-facies conditions are defined by limited circulation of Na-rich aqueous fluids. CO2 does not seem to be an important fluid phase during the overall metamorphic evolution of these crustal rocks.
We, therefore, propose a new model for fluid–rock interaction in the deeply subducted continental lithologies of the Sulu terrane, in which intermediate alumino-silicate solutions with low water activity represent internally derived fluids at UHP conditions, and early retrograde brines correspond to an external input of Cl-rich aqueous fluids. Such a model may explain the anhydrous nature of the peak mineral assemblage, the growth of hydrated phases during early decompression, and the 18O values of rocks in this area. This study provides evidence that subducted continental lithologies may release aqueous fluids, which are rich in alkalis, Al and Si, and Cl-poor, with intermediate compositions. An important consequence is that fluids with major differences in solutes and water content may be released during UHP evolution of continental metamorphic assemblages.
|
ACKNOWLEDGEMENTS |
|---|
Most of this research originates from the Ph.D. thesis of S.F. We acknowledge Shutong Xu, Y. Liu, W. Wu and F. Rolfo for assistance during field work. We are grateful to D. Castelli for discussion, and help in calculating the petrogenetic grid. The paper greatly benefited from reviews by C. Manning, Z. D. Sharp and an anonymous reviewer, and from editorial assistance by R. Gieré. Raman analytical facilities were provided by the Italian organization for research in Antarctica (PNRA). F. Olmi made available microprobe facilities in Firenze. This work was funded by national (MURST, CNR), and local (‘Finanziamento Convenzioni Interuniversitarie’, ‘Fondi Scambi Culturali’—University of Torino, and ‘Programmi Ricerca di Ateneo 2004’—University of Siena) grants.
*Corresponding author. Telephone: (+39)0577 233929. Fax: (+39)0577 233938. E-mail: frezzottiml{at}unisi.it
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGY AND SAMPLE...ANALYTICAL METHODSPETROGRAPHY AND MINERAL...FLUID INCLUSIONSOXYGEN AND HYDROGEN ISOTOPESDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Alberico A, Ferrando S, Ivaldi G, Ferraris G. X-ray single-crystal structure refinement of an OH-rich topaz from Sulu UHP terrane (Eastern China)—structural foundation of the correlation between cell parameters and fluorine content. European Journal of Mineralogy (2003) 15:875–881.
Ames L, Zhou G, Xiong B. Geochronology and isotopic character of ultrahigh-pressure metamorphism with implications for collision of the Sino-Korean and Yangtze cratons, central China. Tectonics (1996) 15:472–489.[CrossRef][Web of Science]
Anderson GM, Burnham CW. The solubility of quartz in supercritical water. American Journal of Science (1965) 263:494–511.[Abstract]
Antignano G, Manning CE. Rutile solubility in H2O–NaAlSi3O8 fluids at high T and P: implications for HFSE mobility in subduction zones. American Geophysical Union, Fall Meet. Suppl. Abstract (2005) 86:V31C–0620.
Aranovich LY, Newton RC. Experimental determination of CO2–H2O activity–composition relations at 600–1000°C and 6–14 kbar by reversed decarbonation and dehydration reactions. American Mineralogist (1999) 84:1319–1332.[Abstract]
Bakker RJ. Package FLUIDS 1. Computer programs for analysis of fluid inclusion data and for modelling bulk fluid properties. Chemical Geology (2003) 194:3–23.[CrossRef][Web of Science]
Bakker RJ, Jansen JBH. Preferential water leakage from fluid inclusions by means of mobile dislocations. Nature (1990) 345:58–60.[CrossRef]
Barton MD. The thermodynamic properties of topaz solid solutions and some petrologic applications. American Mineralogist (1982) 67:956–974.[Abstract]
Boettcher AL, Wyllie PJ. The system CaO–SiO2–CO2–H2O. III. Second critical end-point on the melting curve. Geochimica et Cosmochimica Acta (1969) 33:611–632.[CrossRef][Web of Science]
Bundy FP. The P, T phase and reaction diagram for elemental carbon. Journal of Geophysical Research (1980) 85:6930–6936.
Bureau H, Keppler H. Complete miscibility between silicate melts and hydrous fluids in the upper mantle: experimental evidence and geochemical implications. Earth and Planetary Science Letters (1999) 165:187–196.[CrossRef][Web of Science]
Chopin C. Coesite and pure pyrope in high-grade blueschists of the western Alps: a first record and some consequences. Contributions to Mineralogy and Petrology (1984) 86:107–118.[CrossRef][Web of Science]
Connolly JAD. Multivariable phase diagrams: an algorithm based on generalized thermodynamics. American Journal of Science (1990) 290:666–718.
Eggler DH, Rosenhauer M. CO2 in silicate melts: II. Solubilities of CO2 and H2O in CaMgSi2O6 (diopside) liquids and vapors at pressures to 40 kbar. American Journal of Science (1978) 278:64–94.
Ferrando S, Frezzotti ML, Dallai L, Compagnoni R. Multiphase solid inclusions in ultrahigh pressure metamorphic rocks from Su-Lu belt (China): evidence for critical silicate-rich aqueous fluids released during continental subduction. Chemical Geology (2005a) 223:68–81.[CrossRef][Web of Science]
Ferrando S, Frezzotti ML, Dallai L, Compagnoni R. Fluid–rock interaction in UHP Phe–Ky–Ep eclogite from the Su-Lu orogen (Eastern China). International Geology Review (2005b) 47:750–774.[Web of Science]
Fortier SM, Giletti BJ. An empirical model for predicting diffusion coefficients in silicate minerals. Science (1989) 245:1481–1484.
Franz L, Romer RL, Klemd R, Schmid R, Oberhänsli R, Wagner T, Shuwen D. Eclogite-facies quartz veins within metabasites of the Dabie Shan (eastern China): pressure–temperature–deformation path, composition of the fluid phase and fluid flow during exhumation of high-pressure rocks. Contributions to Mineralogy and Petrology (2001) 141:322–346.[Web of Science]
Frezzotti ML. Silicate-melt inclusions study in magmatic rocks: applications to petrology. Lithos (2001) 55:273–299.[CrossRef][Web of Science]
Fu B, Touret JLR, Zheng Y-F. Fluid inclusions in coesite-bearing eclogites and jadeite quartzite at Shuanghe, Dabie Shan (China). Journal of Metamorphic Geology (2001) 19:529–545.[Web of Science]
Fu B, Zheng Y-F, Touret JLR. Petrological, isotopic and fluid inclusion studies of eclogites from Sujiahe, NW Dabie Shan (China). Chemical Geology (2002) 187:107–128.[CrossRef][Web of Science]
Fu B, Touret JLR, Zheng Y-F. Remnants of pre-metamorphic fluid and oxygen-isotopic signatures in eclogites and garnet clinopyroxenite from the Dabie–Sulu terranes, eastern China. Journal of Metamorphic Geology (2003) 21:561–578.[CrossRef][Web of Science]
Ghent ED, Valley JW. Oxygen isotope study of quartz–Al2SiO5 pairs from the Mica Creek area, British Columbia: implications for the recovery of peak metamorphic temperatures. Journal of Metamorphic Geology (1998) 16:223–230.[CrossRef][Web of Science]
Guo F, Fan W, Wang Y, Zhang M. Origin of early Cretaceous calc-alkaline lamprophyres from the Sulu orogen in eastern China: implications for enrichment processes beneath continental collision belt. Lithos (2004) 78:291–305.[CrossRef][Web of Science]
Hermann J, Spandler C, Hack A, Korsakov AV. Aqueous fluids and hydrous melts in high-pressure and ultra-high pressure rocks: implications for element transfer in subduction zones. Lithos (2006) 92:399–417.[CrossRef][Web of Science]
Hirajima T, Nakamura D. The Dabie Shan–Sulu orogen. In: Ultrahigh-pressure Metamorphism: European Mineralogical Union Notes in Mineralogy, 5—Carswell DA, Compagnoni R, eds. (2003) Budapest: Eötvös University Press. 105–144.
Holland TJB, Powell R. An internally consistent thermodynamic dataset for phases of petrological interest. Journal of Metamorphic Geology (1998) 16:309–343.[CrossRef][Web of Science]
Huang WL, Wyllie PJ. Muscovite dehydration and melting in deep crust and subducted ocean sediments. Earth and Planetary Science Letters (1973) 18:133–136.[CrossRef][Web of Science]
Kawamoto T. Critical phenomena between magmas and aqueous fluid (in Japanese with English abstract). Review in High Pressure Science and Technology (2004) 14:235–241.
Kennedy GC, Wasserburg GJ, Heard HC, Newton RC. The upper three-phase region in the system SiO2–H2O. American Journal of Science (1962) 260:501–521.
Kessel R, Schmidt MW, Ulmer P, Pettke T. The trace element signature of subduction zone fluids, melts and supercritical liquids at 120–180 km depth. Nature (2005) 437:724–727.[CrossRef][Medline]
Kretz R. Symbols for rock-forming minerals. American Mineralogist (1983) 68:277–279.[Abstract]
Li S, Xiao Y, Liou D, Chen Y, Ge N, Zhang Z, Sun S, Cong B, Zhang R, Hart SR, Wang S. Collision of the North China and Yangtze blocks and formation of coesite-bearing eclogites: timing and processes. Chemical Geology (1993) 109:89–111.[CrossRef][Web of Science]
Liu F, Xu Z, Katayama I, Yang J, Maruyama S, Liou JG. Mineral inclusions in zircons of para- and ortho-gneiss from pre-pilot drillhole CCSD-PP1, Chinese Continental Scientific Drilling Project. Lithos (2001) 59:199–215.[CrossRef][Web of Science]
Liu F, Xu Z, Liou JG, Katayama I, Masago H, Maruyama S, Yang J. Ultrahigh-pressure mineral inclusions in zircons from gneissic core samples of the Chinese Continental Scientific Drilling Site in eastern China. European Journal of Mineralogy (2002) 14:499–512.
Lu F, Zhao L, Deng J, Zheng J. Discussion on the ages of kimberlitic magma activity in the North China platform. Acta Petrologica Sinica (1995) 11:365–374.
Manning CE. The chemistry of subduction-zone fluids. Earth and Planetary Science Letters (2004) 223:1–16.[CrossRef][Web of Science]
Matthews A. Oxygen-isotope geothermometers for metamorphic rocks. Journal of Metamorphic Geology (1994) 12:211–219.[CrossRef][Web of Science]
Mattinson CG, Zhang RY, Tsujimori T, Liou JG. Epidote-rich talc–kyanite–phengite eclogites, Sulu terrane, eastern China: P–T–fO2 estimates and the significance of the epidote–talc assemblage in eclogite. American Mineralogist (2004) 89:1772–1783.
Moecher DP, Sharp ZD. Comparison of conventional and garnet–aluminosilicate–quartz O isotope thermometry: insights for mineral equilibration in metamorphic rocks. American Mineralogist (1999) 84:1287–1303.[Abstract]
Mysen BO. Interaction between aqueous fluid and silicate melt in the pressure and temperature regime of the Earth's crust and upper mantle. Neues Jahrbuch für Mineralogie, Abhandlungen (1998) 172:227–244.[Web of Science]
Niggli P. Die leichtflüchtigen Bestandteile im Magma (1920) Leipzig: B. G. Teubner.
Philippot P, Selverstone J. Trace-element-rich brines in eclogitic veins: implications for fluid composition and transport during subduction. Contributions to Mineralogy and Petrology (1991) 106:417–430.[CrossRef][Web of Science]
Philippot P, Chevallier P, Chopin C, Dubessy J. Fluid composition and evolution in coesite-bearing rocks (Dora–Maira massif, Western Alps): implications for element recycling during subduction. Contributions to Mineralogy and Petrology (1995) 121:29–44.[CrossRef][Web of Science]
Philippot P, Agrinier P, Scambelluri M. Chlorine cycling in the subducted oceanic lithosphere. Earth and Planetary Science Letters (1998) 161:33–44.[CrossRef][Web of Science]
Poli S, Schmidt MW. Petrology of subducted slabs. Annual Review of Earth and Planetary Sciences (2002) 30:207–335.[CrossRef][Web of Science]
Putlitz B, Valley JW, Matthews A, Katzir Y. Oxygen isotope thermometry of quartz–Al2SiO5 veins in high-grade metamorphic rocks on Naxos island (Greece). Contributions to Mineralogy and Petrology (2002) 143:350–359.[Web of Science]
Roedder E. Fluid Inclusions. Mineralogical Society of America, Reviews in Mineralogy (1984) 12:644.
Rumble D, Yui T-F. The Qinglongshan oxygen and hydrogen isotope anomaly near Donghai in Jiangsu Province, China. Geochimica et Cosmochimica Acta (1998) 62:3307–3321.[CrossRef][Web of Science]
Rumble D, Giorgis D, Ireland T, Zhang Z, Xu H, Yui TF, Yang J, Xu Z, Liou JG. Low 18O zircons, U–Pb dating, and the age of the Qinglongshan oxygen and hydrogen isotope anomaly near Donghai in Jiangsu Province, China. Geochimica et Cosmochimica Acta (2002) 66:2299–2306.[CrossRef][Web of Science]
Ryabchikov ID. Fluid transport of ore metals in ultramafic mantle rocks. In: Proceedings of the Eight Quadrennial IAGOD Symposium (1993) Stuttgart: Schweizerbart. 425–433.
Scambelluri M, Philippot P. Deep fluids in subduction zones. Lithos (2001) 55:213–227.[CrossRef][Web of Science]
Scambelluri M, Piccardo GB, Philippot P, Robbiano A, Negretti L. High salinity fluid inclusions formed from recycled seawater in deeply subducted alpine serpentinites. Earth and Planetary Science Letters (1997) 148:485–500.[CrossRef][Web of Science]
Scambelluri M, Bottazzi P, Trommsdorff V, Vannucci R, Hermann J, Gomez-Pugnaire MT, Lòpez-Sànchez Vizcaìno V. Incompatible element-rich fluids released by antigorite breakdown in deeply subducted mantle. Earth and Planetary Science Letters (2001) 192:457–470.[CrossRef][Web of Science]
Schmidt MW, Vilzeuf D, Auzanneau E. Melting and dissolution of subducting crust at high pressures: the key role of white mica. Earth and Planetary Science Letters (2004) 228:65–84.[CrossRef][Web of Science]
Schreyer W. Experimental aspects of UHP metamorphism: granitic systems. International Geology Review (1999) 41:701–710.[Web of Science]
Selverstone J, Franz G, Thomas S, Getty S. Fluid variability in 2 GPa eclogites as indicator of fluid behavior during subduction. Contributions to Mineralogy and Petrology (1992) 112:341–357.[CrossRef][Web of Science]
Sharp ZD. A laser-based microanalytical method for the in situ determination of oxygen isotope ratios of silicates and oxides. Geochimica et Cosmochimica Acta (1990) 54:1353–1357.[CrossRef][Web of Science]
Sharp ZD. Oxygen isotope geochemistry of the Al2SiO5 polymorphs. American Journal of Science (1995) 295:1058–1076.
Sharp ZD, Barnes JD. Water soluble chlorides in massive seafloor serpentinites: a source of chloride in subduction zones. Earth and Planetary Science Letters (2004) 226:243–254.[CrossRef][Web of Science]
Sharp ZD, Essene EJ, Smyth JR. Ultra-high temperatures from oxygen isotope thermometry of a coesite–sanidine grospydite. Contributions to Mineralogy and Petrology (1992) 112:358–370.[CrossRef][Web of Science]
Smith DC. Coesite in clinopyroxene in the Caledonides and its implications for geodynamics. Nature (1984) 310:641–644.[CrossRef]
Sobolev NV, Shatsky VS. Diamond inclusions in garnets from metamorphic rocks: a new environment for diamond formation. Nature (1990) 343:742–746.[CrossRef]
Sowerby JR, Keppler H. The effect of fluorine, boron and excess sodium on the critical curve in the albite–H2O system. Contributions to Mineralogy and Petrology (2002) 143:32–37.[Web of Science]
Stalder R, Ulmer P, Thompson AB, Gunther D. Experimental approach to constrain second critical end point in fluid/silicate systems: near-solidus fluids and melts in the system albite–H2O. American Mineralogist (2000) 85:68–77.
Stalder R, Ulmer P, Thompson AB, Günther D. High pressure fluids in the system MgO–SiO2–H2O at upper mantle conditions. Contributions to Mineralogy and Petrology (2001) 140:607–618.[Web of Science]
Svensen H, Jamtveit B, Yardley B, Engvik AK, Austrheim H, Broman C. Lead and bromine enrichment in eclogite-facies fluids: extreme fractionation during lower-crustal hydration. Geology (1999) 27:467–470.
Thomas R, Webster JD, Heinrich W. Melt inclusions in pegmatite quartz: complete miscibility between silicate melts and hydrous fluids at low pressure. Contributions to Mineralogy and Petrology (2000) 139:394–401.[CrossRef][Web of Science]
Touret JLR. Fluids in metamorphic rocks. Lithos (2001) 55:1–27.[CrossRef][Web of Science]
Touret JLR, Frezzotti ML. Fluid inclusions in high pressure and ultrahigh pressure metamorphic rocks. In: Ultrahigh-pressure Metamorphism. European Mineralogical Union Notes in Mineralogy, 5—Carswell DA, Compagnoni R, eds. (2003) Budapest: Eötvös University Press. 467–487.
Tropper P, Manning CE. Paragonite stability at 700°C in the presence of H2O–NaCl fluids: constraints on H2O activity and implications for high pressure metamorphism. Contributions to Mineralogy and Petrology (2004) 147:740–749.[Web of Science]
Tropper P, Manning CE. The solubility of kyanite in H2O–SiO2 fluids at 800°C and 1 GPa: implications for Al–Si complexing in subduction-zone fluids. EOS Transactions, American Geophysical Union (2005) 86(52). Fall Meeting Supplement, Abstract V31C-0621.
Ulmer P. NORM-Program for cation and oxygen mineral norms (1986) Zürich: Computer Library, Institute für Mineralogie und Petrographie, ETH-Zentrum.
Vannay JC, Sharp ZD, Grasemann B. Himalayan inverted metamorphism constrained by oxygen isotope thermometry. Contributions to Mineralogy and Petrology (1999) 137:90–101.[CrossRef][Web of Science]
Vennemann TW, O’Neil JR. A simple and inexpensive method of hydrogen isotope and water analysis of minerals and rocks based on zinc reagent. Chemical Geology (1993) 103:227–234.[CrossRef][Web of Science]
Wallis S, Enami M, Banno S. The Sulu UHP Terrane—a review of the petrology and structural geology. International Geology Review (1999) 41:906–920.[Web of Science]
Wunder B, Rubie DC, Ross CR, Medenbach O, Seifert F, Schreyer W. Synthesis, stability, and properties of Al2SiO4(OH)2: a fully hydrated analogue of topaz. American Mineralogist (1993) 78:285–297.[Abstract]
Wunder B, Andrut M, Wirth R. High-pressure synthesis and properties of OH-rich topaz. European Journal of Mineralogy (1999) 11:803–813.
Xia Q-K, Sheng Y-M, Yang X-Z, Yu H-M. Heterogeneity of water in garnets from UHP eclogites, eastern Dabieshan, China. Chemical Geology (2005) 224:237–246.[CrossRef][Web of Science]
Xiao Y, Hoefs J, van den Kerkhof AM, Fiebig J, Zheng Y. Fluid history of UHP metamorphism in the Dabie Shan, China: a fluid inclusion and oxygen isotope study on the coesite-bearing eclogite from Bixiling. Contributions to Mineralogy and Petrology (2000) 139:1–16.[CrossRef][Web of Science]
Xiao Y, Hoefs J, van den Kerkhof AM. The Raobazhai complex in North Dabie Shan, China: eclogite- to granulite-facies metamorphism of subducted oceanic crust. Journal of Metamorphic Geology (2001) 19:3–19.[CrossRef][Web of Science]
Xiao YL, Hoefs J, van den Kerkhof AM, Simon K, Fiebig J, Zheng Y. Fluid evolution during HP and UHP metamorphism in Dabie Shan, China: constraints from mineral chemistry, fluid inclusions and stable isotopes. Journal of Petrology (2002) 43:1505–1527.
Xiao YL, Zhang Z, Hoefs J, van den Kerkhof AM. Ultrahigh-pressure metamorphic rocks from the Chinese Continental Scientific Drilling Project: II oxygen isotope and fluid inclusion distributions through vertical sections. Contributions to Mineralogy and Petrology (2006) 152:443–458.[CrossRef][Web of Science]
Yang JS, Wooden JL, Wu CL, Liu FL, Xu ZQ, Shi RD, Katayama I, Liou JG, Maruyama S. SHRIMP U–Pb dating of coesite-bearing zircon from the ultrahigh-pressure metamorphic rocks, Sulu terrane, east China. Journal of Metamorphic Geology (2003) 21:551–560.[CrossRef][Web of Science]
Ye K, Yao Y, Katayama I, Cong B, Wang Q, Maruyama S. Large areal extent of ultrahigh-pressure metamorphism in the Sulu ultrahigh-pressure terrane of East China: new implications from coesite and omphacite inclusions in zircon of granitic gneiss. Lithos (2000) 52:157–164.[CrossRef][Web of Science]
Young ED. On the 18O/16O record of reaction progress in open and closed metamorphic systems. Earth and Planetary Science Letters (1993) 117:147–167.[CrossRef][Web of Science]
Yui T-F, Rumble D, Lo C-H. Unusually low 18O ultra-high-pressure metamorphic rocks from the Sulu Terrain, eastern China. Geochimica et Cosmochimica Acta (1995) 59:2859–2864.[CrossRef][Web of Science]
Zhai M, Cong J, Guo J, Liu Y, Wang Q. Sm–Nd geochronology and petrography of garnet pyroxene granulites in the northern Sulu region of China and their geotectonic implication. Lithos (2000) 52:23–33.[CrossRef][Web of Science]
Zhang RY, Hirajima T, Banno S, Cong B, Liou JG. Petrology of ultrahigh-pressure rocks from the southern Su-Lu region, eastern China. Journal of Metamorphic Geology (1995) 13:659–675.[Web of Science]
Zhang RY, Liou JG, Ye K. Petrography of UHPM rocks and their country rock gneisses. In: Ultrahigh-pressure Metamorphic Rocks in the Dabieshan–Sulu Region, China—Cong B, ed. (1996) Beijing: Science Press. 49–68.
Zhang RY, Liou JG, Shu JF. Hydroxyl-rich topaz in high-pressure and ultrahigh-pressure kyanite quartzites, with retrograde woodhouseite, from the Sulu terrane, eastern China. American Mineralogist (2002) 87:445–453.
Zhang Z, Xu Z, Xu H. Petrology of ultrahigh-pressure eclogites from the ZK703 drillhole in the Donghai, eastern China. Lithos (2000) 52:35–50.[CrossRef][Web of Science]
Zhang Z, Xiao Y, Liu F, Liou JG, Hoefs J. Petrogenesis of UHP metamorphic rocks from Quinglongshan, southern Sulu, east–central China. Lithos (2005a) 81:189–207.[CrossRef][Web of Science]
Zhang Z, Kun S, Xiao Y, van den Kerkhof AM, Hoefs J. Fluid composition and evolution attending UHP metamorphism: study of fluid inclusions from drill cores, southern Sulu belt, eastern China. International Geological Review (2005b) 47:297–309.[CrossRef]
Zhang Z, Shen K, Xiao Y, Hoefs J, Liou JG. Mineral and fluid inclusions in zircon of UHP metamorphic rocks from the CCSD-main drill hole: a record of metamorphism and fluid activity. Lithos (2006) 92:378–398.[CrossRef][Web of Science]
Zheng YF. Calculation of oxygen isotope fractionation in anhydrous silicate minerals. Geochimica et Cosmochimica Acta (1993a) 57:1079–1091.[CrossRef][Web of Science]
Zheng YF. Calculation of oxygen isotope fractionation in hydroxyl-bearing silicates. Earth and Planetary Science Letters (1993b) 120:247–263.[CrossRef][Web of Science]
Zheng YF, Fu B, Gong B, Li S. Extreme 18O depletion in eclogite from the Su-terrane in East China. European Journal of Mineralogy (1996) 8:317–323.
Zheng YF, Fu B, Li Y, Xiao Y, Li S. Oxygen and hydrogen isotope geochemistry of ultrahigh-pressure eclogites from the Dabie Mountains and the Sulu terrane. Earth and Planetary Science Letters (1998) 155:113–129.[CrossRef][Web of Science]
Zheng YF, Fu B, Xiao Y, Li Y, Gong B. Hydrogen and oxygen isotope evidence for fluid–rock interactions in the stage of pre- and post-UHP metamorphism in the Dabie Mountains. Lithos (1999) 46:677–693.[CrossRef][Web of Science]
Zheng YF, Fu B, Gong B, Li S. Stable isotope geochemistry of ultrahigh pressure metamorphic rocks from the Dabie–Sulu orogen in China: implications for geodynamics and fluid regime. Earth-Science Reviews (2003) 1276:1–57.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  Y.-F. Zheng Fluid regime in continental subduction zones: petrological insights from ultrahigh-pressure metamorphic rocks Journal of the Geological Society, July 1, 2009; 166(4): 763 - 782. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-B. Wu, S. Gao, H.-F. Zhang, S.-H. Yang, X.-C. Liu, W.-F. Jiao, Y.-S. Liu, H.-L. Yuan, H.-J. Gong, and M.-C. He U-Pb age, trace-element, and Hf-isotope compositions of zircon in a quartz vein from eclogite in the western Dabie Mountains: Constraints on fluid flow during early exhumation of ultrahigh-pressure rocks American Mineralogist, February 1, 2009; 94(2-3): 303 - 312. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||