Journal of Petrology Advance Access published online on February 3, 2009
Journal of Petrology, doi:10.1093/petrology/egn077
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mineral-scale Trace Element and U–Th–Pb Age Constraints on Metamorphism and Melting during the Petermann Orogeny (Central Australia)
1Research School of Earth Sciences, The Australian National University, Canberra, Act 0200, Australia
2Department of Applied Geology, Curtin University of Technology, Perth 6845, WA, Australia
3Department of Geology, Geography and Environmental Studies, Stellenbosch University, Private Bag X1, Stellenbosch, Matieland 7602, South Africa
Received November 22, 2007; Revised typescript accepted December 15, 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL SETTINGSAMPLE DESCRIPTIONANALYTICAL PROCEDURESRESULTSDISCUSSIONCONCLUSIONSAPPENDIX A: AVERAGE MAJOR...APPENDIX B: AVERAGE MAJOR...REFERENCES |
|---|
High-pressure amphibolite-facies migmatitic orthogneisses from the Cockburn Shear Zone (CSZ), northern Musgrave Block in central Australia, were formed during the 580–520 Ma intraplate Petermann Orogeny. The shear-zone hosted orthogneisses are of an intermediate bulk composition that promoted the growth of rare earth element (REE)-bearing major phases (garne and hornblende), as well as numerous accessory phases (zircon, titanite, apatite, epidote and allanite), all of which are potential U–Th–Pb geochronometers and are involved in the distribution of REEs. We have integrated petrology and detailed in situ trace element analysis of major and accessory phases in samples collected outside and inside the CSZ to establish the relative timing of metamorphic mineral growth. This paper presents one of the first applications of newly developed in situ dating protocols on metamorphic allanite. Sensitive high-resolution ion microprobe geochronology on metamorphic zircon and allanite indicate that metamorphism and partial melting occurred between 559 ± 6 and 551 ± 6 Ma. Peak temperatures of 720–750°C, determined from rutile included in garnet, necessitate the presence of fluids to flux partial melting in the CSZ quartzofeldspathic rocks. Metamorphic zircon formed during cooling in the presence of melt near the granitic wet solidus at T
700°C. In contrast, allanite formed at different stages of the CSZ P–T path: (1) as a prograde sub-solidus phase (T < 650°C) formed in the presence of fluids, and (2) as melt-precipitated Th- and REE-rich overgrowths on pre-existing allanite. The ages of the two growth episodes are not isotopically resolvable by allanite dating. Trace element compositions indicate that in both melted and unmelted rocks, garnet and hornblende growth was primarily controlled by prograde sub-solidus hydration reactions that consumed feldspar below the metamorphic peak. REE compositions of the metamorphic zircon and allanite overgrowths that formed in the presence of melt also suggest disequilibrium with garnet. Thus, the major period of garnet and hornblende growth was not coeval with partial melting. KEY WORDS: allanite; ion microprobe dating; zircon; sub-solidus mineral growth; disequilibrium
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGSAMPLE DESCRIPTIONANALYTICAL PROCEDURESRESULTSDISCUSSIONCONCLUSIONSAPPENDIX A: AVERAGE MAJOR...APPENDIX B: AVERAGE MAJOR...REFERENCES |
|---|
The integration of the rapidly developing field of in situ trace element geochemistry with well-established U–Th–Pb dating techniques has proved to be a powerful tool for understanding complex metamorphic pressure–temperature (P–T)–time paths (e.g. Rubatto, 2002
; Hermann & Rubatto, 2003; Kelly & Harley, 2005; Rubatto et al., 2006; Buick et al., 2007). The relative resistance of trace elements, such as the rare earth elements (REE), to diffusive re-equilibration compared with (divalent) major elements in minerals (e.g. Pyle & Spear, 1999; Otamendi et al., 2002; Van Orman et al., 2002) makes them an effective recorder of metamorphic processes, including potential mineral thermometers (Pyle & Spear, 1999, 2000; Degeling, 2003; Zack et al., 2004; Watson & Harrison, 2005). The trace element composition of different mineral phases is a function of P, T and bulk composition, but it is also sensitive to the competitive co-crystallization of trace element-rich minerals (e.g. Pyle & Spear, 2000). The latter can be used to infer apparent disequilibrium or equilibrium of mineral phases within a metamorphic assemblage. For example, light REE (LREE)-depleted patterns imply simultaneous growth of a mineral of interest with a phase that strongly fractionates the LREE (e.g. Sorensen & Grossman, 1989). Similarly, heavy REE (HREE) depletion is generally consistent with the presence of garnet (e.g. Sorensen & Grossman, 1989; Rubatto, 2002; Hoskin & Schaltegger, 2003).
Characteristic trace element compositions of minerals from amphibolite- and granulite-facies terranes have been used to constrain the processes of their formation. Few of these studies, however, have examined the trace element distribution between multiple REE-rich major and accessory phases during amphibolite-grade metamorphism at sub-solidus conditions (e.g. Sorensen & Grossman, 1989; Mulrooney & Rivers, 2005). The use of trace elements to determine the process of formation is particularly relevant for accessory minerals that can be dated via the U–Th–Pb system. Such accessory minerals are major hosts of trace elements (e.g. Hermann, 2002) and in metamorphic rocks often the correct interpretation of their U–Th–Pb age depends on how well their formation can be related to P–T conditions. In this paper we contribute to the expanding field of mineral trace element geochemistry and geochronology with a study of high-pressure (HP) migmatitic orthogneisses of the Cockburn Shear Zone (CSZ) within the Mann Terrane, central Australia. Understanding the growth histories of metamorphic accessory phases is facilitated when samples of the same bulk composition with different extents of metamorphic overprint can be compared (e.g. Sorensen, 1991; Rubatto et al., 2001; Storkey et al., 2005; Buick et al., 2007; Clarke et al., 2007). The investigated rocks are particularly suited for such a study because they include a suite of granodioritic orthogneisses that are cut by shear zones, in which the same bulk compositions have undergone partial melting at 
700°C (Scrimgeour & Close, 1999).
Dating amphibolite-grade rocks is commonly attempted using zircon and monazite. However, in relatively low-Al, high-Ca bulk compositions (for example, metamorphosed calc-alkaline granites, tonalites and granodiorites, such as those investigated here), monazite is largely absent and metamorphic zircon typically forms at higher temperatures (> 700°C, e.g. Rubatto et al., 2001, 2006). Instead, accessory (epidote)–allanite is stable, and has great potential both to date high-grade events (Oberli et al., 2004) and to act as a tracer of mineral-scale processes. As a chemically complex mineral, allanite should ideally be dated by spot analysis rather than bulk dilution methods, as this may average any chemical or isotopic zoning. Gregory et al. (2007) developed laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) and sensitive high-resolution ion microprobe (SHRIMP) micro-analytical techniques for the Th–Pb dating of igneous allanite. Here we examine the potential of dating metamorphic allanite using a SHRIMP and LA-ICP-MS approach and the sensitivity of the allanite REE composition to metamorphic processes. U–Pb dating of zircon provided an independent age constraint for comparison with the allanite isotopic system. These U–Th–Pb ages are then linked to stages of mineral growth through mineral-scale trace element geochemistry.
|
GEOLOGICAL SETTING |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGSAMPLE DESCRIPTIONANALYTICAL PROCEDURESRESULTSDISCUSSIONCONCLUSIONSAPPENDIX A: AVERAGE MAJOR...APPENDIX B: AVERAGE MAJOR...REFERENCES |
|---|
The largest HP metamorphic terrane documented in Australia is exposed in the Mann Terrane of central Australia. The Mann Terrane represents part of the basement of the Musgrave Block, located south of the Amadeus Basin, and is delineated by two crustal-scale, south-dipping structures, the Woodroffe Thrust and the Mann Fault (Fig. 1a). Both structures were formed during the Neoproterozoic to early Palaeozoic intracratonic Petermann Orogeny, which resulted in the reworking of anhydrous Mesoproterozoic granulite-facies orthogneisses and igneous rocks of the Musgrave Block, and the northward exhumation of deep crustal mylonites along the Woodroffe Thrust (Camacho & Fanning, 1995
; Camacho et al., 1997; Scrimgeour & Close, 1999). North of the Mann Fault, basement outcrop is dominated by the 1190–1120 Ma Umutju Granite Suite (Fig. 1b; Camacho & Fanning, 1995), which was intruded by 1080–1050 Ma and 800 Ma mafic dykes (e.g. Zhao et al., 1994; Sun et al., 1996). During the Petermann Orogeny, basement rocks were metamorphosed to transitional garnet-granulite- to eclogite-facies conditions at 11–13 kbar and 700–750°C (Scrimgeour & Close, 1999). Metamorphic grade varies across the orogen with the highest P (and T) attained in the south, immediately north of the Mann Fault (Fig. 1b). Metamorphic grade decreases upsection, with P–T conditions of 6–7 kbar and 600–650°C recorded in shear zones crosscutting the Pottoyu and Mantarurr Granite Suites north of the Woodroffe Thrust (Fig. 1b; Scrimgeour & Close, 1999).
Despite the widespread presence of generally anhydrous and clinopyroxene-bearing granite in the Mann Terrane (Scrimgeour et al., 1999), melting occurred within discrete shear zones active during the Petermann Orogeny, such as the Cockburn Shear Zone (CSZ). These zones, however, represent < 1% of exposed outcrop (Scrimgeour et al., 1999). The CSZ is a shallow, WSW-dipping ductile thrust located immediately north of the Mann Fault (Fig. 1b), which crosscuts a porphyritic K-feldspar- and clinopyroxene-bearing granodiorite. P–T estimates taken from recrystallized mafic dykes immediately south of the CSZ record near eclogite-facies conditions of 13 kbar and 750°C (Fig. 1b; Srimgeour & Close, 1999).
|
Localized fabric inside the CSZ truncates the regional proto-mylonitic Petermann-age fabric outside the CSZ. Deformed porphyritic granodiorite outside the CSZ (referred to here as orthogneiss) does not show evidence of melting and contains fine-grained garnet and hornblende. The orthogneiss grades relatively sharply into the shear zone-hosted partial melt zone (Fig. 2a), where rocks of the same protolith contain decimetre-scale alkali feldspar + quartz + plagioclase leucosomes (referred to here as migmatitic orthogneiss). Leucosomes preserve varying degrees of syn- to post-metamorphic strain, from strongly mylonitized and concordant leucosomes (Fig. 2a and b) to relatively undeformed solidified melt pods (Fig. 2c). Migmatitic orthogneiss of the CSZ contain significantly coarser (centimetre-sized) garnet and hornblende grains (Fig. 2c), as well as abundant accessory minerals compared with the orthogneiss. Mafic dykes in orthogneiss were partially recrystallized during the Petermann Orogeny, whereas dykes located within the CSZ were transformed into amphibolite boudins, which preserve solidified melt pods in strain shadows and stringers of melt entrained from the surrounding migmatitic orthogneiss (Fig. 2b). Scrimgeour & Close (1999
) proposed a mechanism of focused fluid-fluxed melting within the shear zones to explain the mineralogical differences between rocks located inside and outside the CSZ.
|
The timing of high-grade metamorphism associated with the Petermann Orogeny is poorly constrained (
560 and 520 Ma; Maboko et al., 1992; Camacho et al., 1997). This is the result of an absence of extensive high-T accessory mineral U–Th–Pb geochronology partly because of a lack of new zircon formation in rocks metamorphosed during the Petermann Orogeny, poor outcrop exposure, and restricted land access (Scrimgeour et al., 1999). Available mineral Sm–Nd and Ar–Ar ages for basement rocks in the eastern and western Musgrave Block indicate cooling of these rocks below 550°C by c. 550–530 Ma (Maboko et al., 1992; Camacho et al., 1997). There remains, however, a lack of geochronological constraints on the high-T history of the Petermann Orogeny in the central to northern Musgrave Block. |
SAMPLE DESCRIPTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGSAMPLE DESCRIPTIONANALYTICAL PROCEDURESRESULTSDISCUSSIONCONCLUSIONSAPPENDIX A: AVERAGE MAJOR...APPENDIX B: AVERAGE MAJOR...REFERENCES |
|---|
Four representative samples taken from outside and inside the CSZ were investigated to compare the extent of metamorphic overprinting between the melted and unmelted rock counterparts. These are: samples Pe1, 6 and 11 (25°58·35'S, 129°25·85'E) and Pe13 (25°57·98'S, 129°25·94'E). Mineral assemblage evolution determined from chemical information and textural observations of Pe1, Pe11 and Pe13 is summarized in Fig. 3.
|
Pe1
Sample Pe1 was collected from a location 20 m outside the CSZ and is representative of the unmelted and largely anhydrous, although strongly deformed, orthogneiss. This sample was investigated to characterize trace element distribution during prograde to peak metamorphism in unmelted equivalents of the CSZ migmatitic orthogneisses. It contains variably recrystallized igneous quartz + K-feldspar + plagioclase + clinopyroxene + ilmenite + apatite (e.g. relict igneous clinopyroxene; fig. 4c of Scrimgeour & Close, 1999
). No REE-rich magmatic accessory phase (e.g. monazite or allanite) was detected, and on a regional scale, monazite has not been reported in Mann Terrane granites (Scrimgeour et al., 1999). The relict primary assemblage is overprinted by a fine-grained (< 300 µm) secondary metamorphic assemblage of garnet + hornblende + biotite + titanite + epidote/allanite (Fig. 4a and b). The two feldspars are present as millimetre-size relict grains (K-feldspar is cloudy; Fig. 4c), and as small, recrystallized grains. Pe1 exhibits a domainal distribution of metamorphic minerals (Fig. 4c). Hornblende forms coronal replacement textures around igneous clinopyroxene and is also found near garnet and plagioclase. Subhedral garnet forms as aggregates of small euhedral grains surrounding either hornblende or ilmenite, and adjacent to feldspar (Fig. 4b and c), and contains inclusions of ilmenite. Hornblende and garnet do not form mutual inclusion relationships. Titanite grains form as clusters around relict ilmenite (Fig. 4b), or as lone grains near garnet. Ilmenite contains magnetite exsolution features. Small epidote–allanite grains (< 25 µm) are included in, or adjacent to, garnet, plagioclase, titanite and apatite. Apatite and zircon are present, although it is difficult to distinguish whether they are magmatic or metamorphic phases, based on petrography alone.
|
Pe6
Pe6 is an amphibolite boudin (Fig. 2b), a representative mafic dyke that has been reworked within the CSZ. This rutile-bearing sample was analysed to provide constraints on peak metamorphic T. It contains leucosomes of plagioclase and quartz, within a matrix of hornblende + plagioclase + garnet + clinopyroxene + rutile + titanite + zircon + traces of quartz (Fig. 4d and e). Rutile grains are 50 µm to 1 mm in diameter, inclusion-free and are present both in the matrix and commonly as inclusions in millimetre- to centimetre-sized garnet, where they are rimmed by titanite.
Pe11 and Pe13
Sample Pe13 is a deformed leucosome containing garnet and hornblende from inside the CSZ. Sample Pe11 contains both leucosome and mesosome; however, the following description and compositional analyses are taken only from the leucosome segment. Unlike Pe1, the migmatitic orthogneisses are generally devoid of clinopyroxene and do not preserve the earlier igneous assemblage, with the exception of relict K-feldspar and minor plagioclase (Fig. 4h). Both samples contain a relatively hydrous major mineral assemblage of quartz + K-feldspar + plagioclase + garnet + hornblende + biotite + epidote (Figs 3 and 4f–j); Pe11 additionally contains minor scapolite. Both feldspars are present as > 500 µm-sized porphyroclasts and as < 250 µm-sized recrystallized grains. Garnets are millimetre- to centimetre-sized and commonly poikiloblastic and anhedral. Garnet cores contain variably sized inclusions of quartz and accessory epidote, zircon, titanite and apatite. Garnet rims are typically inclusion-free; they locally contain larger (100 µm) grains of epidote, titanite and apatite, but lack the abundant small inclusions found in garnet cores (Fig. 4h). Two hornblende generations were distinguished by texture: millimetre- to centimetre-sized sub-euhedral hornblende (in Pe11 only), and a subsequent generation of foliated ‘matrix’ hornblende (Fig. 4f and i). Biotite is fine-grained, and both biotite and matrix hornblende appear to be (sub-solidus) retrograde phases and have a fabric preferred orientation.
Compared with Pe1, the migmatitic orthogneisses contain a notable amount of accessory mineral growth, including < 500 µm-sized, inclusion-free titanite, epidote–allanite, zircon and apatite. Allanite and titanite grains are subhedral to euhedral, and are commonly associated with matrix hornblende and biotite within the fabric, or alongside garnet (Fig. 4f and i). REE-rich allanite is a matrix phase, whereas epidote is a common inclusion in garnet (Fig. 4g and j). Compositional zoning of allanite is observed in thin section; grains show pale yellow cores and yellow–brown rims (Fig. 4f).
|
ANALYTICAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGSAMPLE DESCRIPTIONANALYTICAL PROCEDURESRESULTSDISCUSSIONCONCLUSIONSAPPENDIX A: AVERAGE MAJOR...APPENDIX B: AVERAGE MAJOR...REFERENCES |
|---|
Imaging and electron probe microprobe analysis (EPMA)
Compositional zoning was identified by high-contrast backscatter electron (BSE) imaging using a Cambridge S360 scanning electron microscope (SEM) at the ANU Electron Microscopy Unit (EMU;
2 nA, 15 kV and 15 mm working distance). Wavelength-dispersive analysis (WDS) of major elements was carried out on a Cameca SX100 electron microprobe at the Research School of Earth Sciences, Australian National University (RSES, ANU) and on a Cameca SX50 at the University of Melbourne. Analytical conditions were 15 kV and 20–25 nA, with the beam defocused to 5 µm to avoid specimen damage. An analytical method specific to allanite (REE) analysis was used at RSES (sample Pe13; Gregory et al., 2007). In-house silicate mineral and allanite standards were analysed to monitor internal consistency. Fe3+ contents were recalculated based on charge balance assuming 8 cations and 12·5 oxygens for epidote–allanite. Zircon mounts were imaged using a cathodoluminescence (CL) detector at the ANU EMU on a HITACHI S2250-N SEM with operating conditions of 15 kV, 60 µm and 20 mm working distance.
Bulk-rock analysis
Procedures followed are those described by Buick et al. (2006). Samples Pe1 and Pe11 were ground in a tungsten carbide mill to a grain size of 25 µm, and fused into (La2O3-doped) lithium borate glass discs (0·84 g of sample to 4·5 g of flux) for major-element determinations, and REE-free lithium borate glass discs (Sigma® 12:22 X-ray flux; sample:flux = 1:2) for trace-element determinations. Major elements were analysed by X-ray spectrometry (XRF) on a Siemens SR303AS system with Rh end-window X-ray tube at La Trobe University (Melbourne, Australia) and trace elements were analysed using the LA-ICP-MS facility at RSES, ANU, using a spot size of 142 µm and SiO2 as the internal standard. The resulting data are an average of 4–5 ablation spots; 1 standard deviation from the average is < 5% relative for most elements.
Mineral LA-ICP-MS trace element analysis
Polished thin sections used for EPMA were used to determine mineral trace element contents. Analyses were performed using a pulsed ArF Excimer laser system (193 nm wavelength) coupled to a quadrupole ICP-MS system (Agilent 7500S) at RSES, ANU. The instrumental setup generally followed that described by Eggins et al. (1998). Depending on the target mineral, the laser was focused to produce an ablation pit ranging in diameter from 24 to 84 µm, with 100 mJ energy at a repetition rate of 5 Hz. Data acquisition for each element during a single analysis included a total of 70–80 mass spectrometer sweeps, comprising a gas background of 20–25 sweeps. During the time-resolved analysis, contamination or alteration was detected by monitoring several elements and only the relevant part of the signal was integrated. NIST612 (Pearce et al., 1997) was used as the external standard, except for Zr in rutile when NIST610 was used. Internal standards were major elements (SiO2, CaO, ZrO2, TiO2) measured by EPMA or determined stoichiometrically. Reproducibility of results for BCR glasses using ANU analytical protocols is generally 2–5% 1
for multiple analyses. REE plots are normalized to the values of McDonough & Sun (1995).
Zircon and allanite ion microprobe dating
CL and BSE images served as a basis for selection of zircon and allanite grains for isotopic analysis, respectively. U–Th–Pb analyses were conducted on a SHRIMP (SHRIMP II and RG) at RSES. Epoxy-mounted grains of allanite and zircon from Pe13 were analysed during different analytical sessions under similar operating conditions of a 3–4 nA, 10 kV primary O2– beam focused to a 20 µm diameter spot. Instrumental conditions and data acquisition and treatment were generally as described by Compston et al. (1992) and Williams (1998), with isotope data collected from sets of six scans through the masses. The measured 206Pb/238U ratio was corrected using the ANU reference zircon FC1 (1099 Ma). Because of 204Pb overcounts, Th-deficient zircon analyses were corrected for common Pb based on measured 208Pb/206Pb calculated from Th/U, assuming equilibrium between Th and U systems. This was satisfied by all analyses by plotting ThO/UO vs 208Pb/206Pb. Because of their low content in radiogenic Pb, zircon rims had high proportions of common Pb (2–8% 206Pbc) relative to the 204Pb-free standard. Therefore a model common Pb composition (Stacey & Kramers, 1975) was assumed for a population age of 550 Ma. A detailed description for allanite Th–Pb analysis has been given by Gregory et al. (2007). The measured 208Pb/232Th ratio was corrected using the allanite standard CAP (276 Ma; Barth et al., 1994). All analyses were corrected for common Pb based on measured 207Pb/206Pb and 208Pb/206Pb assuming a model Pb composition at 550 Ma (Stacey & Kramers, 1975). Equilibrium between Th and U systems was satisfied by plotting ThO/UO vs 208Pb/206Pb. Th–Pb isochrons regressed from uncorrected data gave initial 208Pb/206Pb intercept values of 2·14 ± 0·24 for cores and 2·12 ± 0·12 for rims, which are within error of the model 208Pb/206Pb composition. EPMA chemical compositions were acquired for each domain analysed by SHRIMP. Age calculation was carried out using RSES internal software for allanite and using the software Isoplot/Ex (Ludwig, 2000) for zircon and allanite.
Allanite laser ablation ICP-MS dating
The analytical procedure for U–Th–Pb analyses by LA-ICP-MS was generally as described by Gregory et al. (2007). Each isotopic analysis on a 32 µm spot took 65 s in time-resolved (peak hopping) analysis mode, including 25 s background. Data were processed off-line using an in-house macro-based EXCEL reduction spreadsheet, allowing selective integration of ‘clean’ isotope signals. Inter-elemental fractionation of U–Th–Pb ratios was corrected for using an external AVC allanite standard (276 Ma, Barth et al., 1994) and calculating a matrix normalization factor F [F = (208Pb/232Thknown)/(208Pb/232Thmeasured)], from the average of replicate standard measurements (12), for each down-hole mass sweep, and applying this to each unknown for the same depth interval. Unknowns were referenced directly to NIST610 for 232Th/238U, Si, P, Ca and REE. The standard allanite data were vetted for outliers (those deviating by > 2% from the mean 208Pb/232Th). Isochron plots were constructed from uncorrected data using Isoplot/Ex software (Ludwig, 2000).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGSAMPLE DESCRIPTIONANALYTICAL PROCEDURESRESULTSDISCUSSIONCONCLUSIONSAPPENDIX A: AVERAGE MAJOR...APPENDIX B: AVERAGE MAJOR...REFERENCES |
|---|
Bulk-rock composition
Chondrite- normalized bulk-rock REE plots for Pe1 and Pe11 are shown in Fig. 5 and bulk-rock compositional data are given in Table 1. Both samples display identical bulk-rock REE patterns characterized by strong LREE enrichment (
200 x chondrite) with respect to the HREE, and show negligible Eu anomalies. The major element compositions are also similar outside and inside the shear zone. The similarity of bulk-rock REE contents outside and inside the shear zone indicates that there was a lack of (or very limited) metasomatism of rocks within the CSZ. This suggests that trace elements were redistributed within a closed system during prograde metamorphism and partial melting, and that the trace element budget was not likely to have been modified by an external source (e.g. fluids).
|
|
Mineral composition
Average major element mineral analyses are provided in Appendices A and B. An electron microprobe traverse of garnet in Pe11 is shown in Fig. 6. In general, major element zoning is limited. As the composition of minerals in Pe11 and Pe13 are very similar, we discuss them together.
|
Pe1
Magmatic relicts. Clinopyroxene shows only minor internal zoning from core to rim, with an XMg of 0·58–0·60 and 0·12–0·15 Al cations per formula unit (c.p.f.u.), respectively. Plagioclase is predominantly andesine (An36 at core to An33 at rim), with minor orthoclase. K-feldspar is orthoclase (Or88) in composition, with a minor decrease in Na and Ca from core to rim. Amphibole replacing clinopyroxene is pargasite hornblende according to the classification of Leake (1978), with 6·13–6·16 Si c.p.f.u. and an XMg of
0·50. Titanite overgrowths on ilmenite contain 0·10 Al c.p.f.u. Metamorphic minerals. Garnet aggregates adjacent to both hornblende and ilmenite are predominantly an almandine–grossular solid solution (Alm54·5Gro31·5Py12·5Sp0·02). Amphibole that occurs as rare grains in the feldspar-rich matrix is pargasite hornblende in composition (6·07–6·10 Si c.p.f.u.; Leake, 1978) and has a lower XMg of 0·45 compared with hornblende replacing clinopyroxene. Recrystallized plagioclase is oligoclase (An23–28) and less calcic than relict plagioclase. Recrystallized K-feldspar is orthoclase (Or76–90) in composition. Titanite in the feldspar-rich matrix has a variable Al content of 0·08–0·13 c.p.f.u.
Pe11 and Pe13
Garnet grains are almandine–grossular-dominant solid solutions (Alm53Gro32–36Py7–11Sp7–1). EPMA traverses of garnet reveal subtle zoning in the major elements from core to rim, characterized by a compositionally homogeneous core and a broad rim that shows a small but progressive change in composition towards the edge, with increasing per cent grossular and XMg and decreasing per cent spessartine (Fig. 6). Garnet rim major-element compositions overlap with those of the fine-grained garnet in Pe1. Amphibole is pargasitic hornblende (Leake, 1978). Both hornblende generations have XMg 0·37 and lack significant major element zoning. Plagioclase is oligoclase (An22), similar to recrystallized plagioclase in Pe1, with minor orthoclase. Titanite is unzoned in the major elements and no clear core to rim relationship is observed for Al content (0·07–0·09 Al c.p.f.u.). Epidote–allanite grains show major element zoning from core (REE-poor epidote) to rim (REE-rich allanite) consistent with the coupled substitution Ca2+ + Fe3+[Al3+] 
REE3+ + Fe2+[Mg, Mn2+]. Here we describe grains having REE + Th > 0·2 c.p.f.u. as allanite rather than REE-rich epidote to distinguish between garnet inclusions (epidote) and matrix grains (allanite) (see Gieré & Sorensen, 2004, for discussion). Epidote in garnet has Fe3+/Fetotal 1, a clinozoisite component of 2·3 Al c.p.f.u. and a REE + Th content < 0·15 c.p.f.u. Matrix allanite compositions vary in the range of Fe3+/Fetotal of 0·8–0·6 and REE + Th of 0·22–0·31 c.p.f.u. (or All22–30), from cores to rims, respectively. Strontium content decreases from epidote cores (2000 ppm) to allanite rims (600 ppm).
REE and trace element chemistry
Thin sections and epoxy mounts of allanite and zircon were used for LA-ICP-MS trace element analysis. Analyses of mineral domains are given in Table 2. Chondrite-normalized REE plots of minerals from sample Pe1 are shown in Fig. 7 and from Pe11 and 13 in Fig. 8.
|
|
|
Pe1 (orthogneiss)
Magmatic relicts. Clinopyroxene REE analyses were taken from a single relict grain and are characterized by a relatively enriched middle REE (MREE) content and moderate negative Eu anomalies (Eu/Eu*
0·6) (Fig. 7b). Plagioclase is noted for its strong enrichment in the LREE relative to HREE, up to 350 times chondrite, compared with K-feldspar (Fig. 7b). Both plagioclase and K-feldspar strongly fractionate Eu over all other phases. In addition, the feldspars are primary Sr repositories (plagioclase 599–850 ppm; K-feldspar 594–901 ppm). Rubidium occurs primarily in K-feldspar (344–393 ppm) and biotite (798–834 ppm), and Ba is mostly abundant in the feldspars (plagioclase 75–192 ppm; K-feldspar 6482–9054 ppm) and biotite (3307–3807 ppm; Table 2). Zircon shows a typical steep MREE–HREE pattern and strong positive Ce anomaly. Apatite and zircon REE patterns show negative Eu anomalies (Eu/Eu* = 0·46–0·61 and Eu/Eu* = 0·31–0·35, respectively; Fig. 7c). Titanite grains replacing ilmenite are LREE–MREE enriched compared with HREE and also have negative Eu anomalies (Fig. 7c). Similarly, hornblende overgrowths on clinopyroxene have clinopyroxene-like REE patterns (Fig. 7c), characterized by a negative Eu anomaly (Eu/Eu* 0·65). Metamorphic minerals. Fine-grained garnet is characterized by small negative, to significant positive, Eu anomalies (Eu/Eu* = 0·8–3·3, mostly > 0·90; Fig. 7a). Compared with the LREE, garnet displays relatively enriched but variable MREE–HREE contents. In contrast to hornblende near relict clinopyroxene, hornblende in leucocratic (K-feldspar + plagioclase + quartz) layers is characterized by a relative enrichment in MREE and has no negative Eu anomaly (Fig. 7a). Titanite in leucocratic layers is enriched in LREE–MREE relative to HREE (Fig. 7a). Unlike titanite on ilmenite, these titanite grains either have no Eu anomaly or show a small positive Eu anomaly. Similarly, some apatite grains also lack a negative Eu anomaly (Eu/Eu* = 1·0; Fig. 7a). BSE imaging revealed zoned epidote–allanite grains, which formed in association with the new metamorphic mineral growth. The grains were too small, however, to be analysed for trace elements by LA-ICP-MS.
Pe11 and Pe13 (migmatitic orthogneiss)
Garnet. Garnet from both samples is heterogeneous with respect to trace element content (Fig. 6) and displays a general decrease of Y and HREE from core to rim (Fig. 8b and f). In general, garnet also lacks an appreciable negative Eu anomaly (Eu/Eu* 1·3; Fig. 6).
Amphibole. Hornblende grains occur in two textural settings, as coarse (millimetre- to centimetre-sized) grains (Pe11), and as smaller (< 1 mm) grains, located in garnet strain shadows or aligned within the S2b fabric (Pe11 and Pe13). Both hornblende types show LREE-depleted REE patterns and lack a negative Eu anomaly (Fig. 8b and g). Hornblende in Pe11 has slightly lower HREE abundances, compared with matrix hornblende, although overall mineral-scale trace element zoning is variable between grains (e.g. Y–HREE, Ti, Zr).
Plagioclase and K-feldspar. Plagioclase and relict K-feldspar strongly fractionate Eu from the other REE and hence, have large positive Eu anomalies (Eu/Eu* > 15; Fig. 8d and h). Both feldspars are relatively enriched in the LREE, and the MREE–HREE are commonly below detection levels. Plagioclase grains in migmatitic orthogneiss are significantly REE-depleted compared with magmatic plagioclase in Pe1 (compare Figs 7b and 8h). This can be explained by differences in the relative abundance of REE-bearing accessory minerals (e.g. allanite) between sample Pe1 and samples Pe11 and Pe13. Feldspar competes with allanite and apatite for Sr (Table 2).
Scapolite. Scapolite was found in only one sample (Pe11) and contains inclusions of titanite. It has feldspar-like REE patterns, at higher concentrations, characterized by a decrease in relative normalized abundance from LREE to HREE (Fig. 8d), and a moderate positive Eu anomaly (Eu/Eu* > 2·0).
Epidote–allanite. BSE imaging of epidote–allanite grains (Fig. 9a) reveals LREE-controlled internal zoning that is correlated with major and minor elemental substitution. In the matrix, allanite grains (intermediate BSE intensity) are commonly overgrown or truncated by allanite rims (high relative BSE intensity) of higher REE and Th content, which appear to be discontinuous overgrowths (Fig. 9a). Epidote cores (low relative BSE intensity) are also observed. Epidote in garnet is relatively LREE-enriched compared with the MREE–HREE (although overall LREE contents are lower than allanite in the matrix) and has a positive Eu anomaly (Eu/Eu* 1·5) (Fig. 8b and e). Rims of intermediate epidote–allanite composition are locally observed on epidote grains in garnet (Fig. 8a). Allanite is LREE-enriched by up to two orders of magnitude, more than any other mineral in the migmatitic orthogneisses, and shows REE patterns of decreasing chondrite-normalized abundance with increasing atomic number. Like epidote, allanite lacks a negative Eu anomaly (Eu/Eu* 1·1). Allanite rims are less depleted in HREE with respect to the MREE (GdN/LuN < 150) than allanite cores (GdN/LuN 240). Of the four accessory phases, allanite incorporates the most radiogenic elements (Th > 600 ppm to > 900 ppm from core to rim). The textural context of epidote–allanite and garnet can provide information on the relative timing of mineral growth and therefore estimates for the P–T conditions at the time of allanite crystallization, as discussed below.
|
Titanite. Three titanite generations, identified from textural location and BSE imaging, are also observed: (1) titanite included in garnet; (2) cores in matrix titanite; (3) rims on matrix titanite. It is a major host of the MREE–HREE and shows considerable variation in these elements (Fig. 8c and g). Titanite grains in garnet are markedly depleted in HREE with respect to the LREE–MREE. Like allanite, titanite does not show a negative Eu anomaly (Eu/Eu*
1·3). Notably, LREE contents of titanite in these rocks are significantly lower than those of titanite in the orthogneiss (compare Fig. 7a, c and Fig. 8c). This is due to the direct influence of allanite as a strongly LREE-fractionating phase. Garnet and titanite show opposing compositional growth zoning in Y. Yttrium content increases from titanite in garnet (< 600 ppm) to rims on matrix titanite (> 1370 ppm). Titanite is the major repository for Nb and Ta (Table 2). It is insufficiently radiogenic for dating, being hampered by a greater initial Pb content than zircon (Table 2).
Zircon. CL images of zircon grains from Pe13 (Fig. 9b) reveal partially resorbed, oscillatory-zoned or weakly zoned cores, typical of an igneous origin, which are inclusion-rich. Zircon cores are commonly overgrown by rounded, unzoned, inclusion-free rims. Grains are commonly crosscut by fractures filled by new zircon (Fig. 9b). Zircon rims are chemically distinct from igneous cores in terms of their low trace element contents, including Y, REE, P, Nb, Th and U (Table 2). Zircon cores in the migmatitic orthogneiss have identical REE characteristics to zircon from the orthogneiss. Compared with magmatic cores, zircon rims have no negative Eu anomaly (Eu/Eu* 1·2), a reduced Ce anomaly, and lower HREE abundances. They are also depleted in U and show low Th/U (< 25 ppm and < 0·05, respectively).
Apatite. Matrix apatite has relatively LREE-depleted, MREE–HREE-enriched patterns (Fig. 8c and h). In contrast, apatite included in garnet is relatively HREE-depleted, suggesting equilibrium with its host phase. Like titanite, apatite in the migmatitic orthogneiss also contains lower absolute LREE contents than grains in the orthogneiss. Whereas some apatite grains display a small positive Eu anomaly (Eu/Eu* 1·1–1·3), others show a negative Eu anomaly (Eu/Eu* 0·4), depending on their textural position in the rock. Apatite included in garnet is richer in Sr, Ba and Pb than matrix apatite. Overall, apatite is poor in radiogenic elements (Th and U content < 1 ppm).
Accessory phase thermometry
Trace-element thermometry of accessory minerals was applied to the Pe13 migmatitic orthogneiss and Pe6 amphibolite boudin from the CSZ. Metamorphic rims of zircon grains analysed by SHRIMP contained 2–5 ppm Ti (Table 2). Using the Ti-in-zircon thermometer calibrated by Watson et al. (2006), temperatures of 664–707 ± 15°C were obtained for crystallization of metamorphic zircon rims, assuming Ti buffering by titanite and 
TiO2 of 0·7 (Lowery Claiborne et al., 2006). Amphibolite boudins (recrystallized former mafic dykes) provided the best estimate of peak metamorphic temperatures within the shear zone. Zirconium thermometry on rutile in garnet gave 720–747 ± 24°C, based on the calibration of Watson et al. (2006) for measured Zr concentrations of 722–955 ppm (Table 2). Quartz and zircon are present in the Pe6 assemblage (Fig. 4e) therefore the Zr in rutile was buffered. These constraints are consistent with previous T estimates for recrystallized mafic dykes of 700–750°C (Scrimgeour & Close, 1999).
Geochronology
U–Pb in zircon
It was not the purpose of this study to investigate zircon inheritance so only seven analyses of zircon cores were obtained from migmatitic orthogneiss Pe13. Four concordant analyses (>96%) gave ages from 1085 to 1108 Ma (Table 3). Unzoned, U-poor rims (< 27 ppm) were found on the terminations of nearly all zircon grains. Poor counting statistics for U and Pb isotopes resulted in large within-spot errors (Table 3). A 208Pb-corrected concordia age of 555 ± 7 Ma (probability of concordance = 0·9, MSWD of single 206Pb/238U age population = 1·1), was obtained from 14 of 19 analyses (Fig. 10a). Three of the remaining analyses were discarded because of poor precision (> 4% error, <1 ppm radiogenic 206Pb), and two analyses were rejected as statistical outliers (588 and 525 Ma).
|
|
Th–Pb in allanite
BSE images and REE chemistry indicate more than one allanite generation (Fig. 9a). High- to intermediate-intensity BSE zones (All20–30), which represent allanite rims and cores, respectively, were analysed by SHRIMP. The spots analysed contain between 1600 and 600 ppm ThO2 and have a Th/U of 9–15. SHRIMP Pb/Th analyses were pooled based on core–rim textural relationships observed in BSE imaging and trace element composition (Table 4). Two weighted mean 207Pb-corrected 208Pb/232Th ages of 551 ± 6 Ma (MSWD = 0·4, n = 12) and 559 ± 6 Ma (MSWD = 0·8, n = 13) were obtained for rims and cores, respectively (Fig. 10c and d). The allanite rims have a common Pb contribution of 40–55% of the total 208Pb content and the allanite cores have a 58–70% common Pb contribution to the total 208Pb content. Figure 10b is a Th–Pb isochron plot showing uncorrected SHRIMP analyses of allanite rims (white ellipses), which provide an independent constraint on the initial Pb composition. The age of the allanite rim and that of the cores were not statistically resolvable by SHRIMP dating because of the error introduced from the common Pb correction; however, the calculation of two ages is justified by the observed core–rim textural and compositional relationships.
|
Distinguishing between LA-ICP-MS analyses of allanite rims and cores proved to be more difficult. Given that the depth of drilling for the LA-ICP-MS method in this study is
25 µm, compared with 2 µm for SHRIMP analysis, it is possible that the laser drilling may have sampled more than one allanite domain. Additionally, despite the larger volume sampled, LA-ICP-MS analyses resulted in an inferior analytical precision compared with SHRIMP analyses. Pooled core and rim analyses fell, within error, along a 232Th–208Pb isochron whose slope gave an age of 560 ± 36 Ma (MSWD = 0·2, n = 26) and an initial 208Pb/206Pb intercept of 2·2 ± 0·1 (Table 5, Fig. 10b).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGSAMPLE DESCRIPTIONANALYTICAL PROCEDURESRESULTSDISCUSSIONCONCLUSIONSAPPENDIX A: AVERAGE MAJOR...APPENDIX B: AVERAGE MAJOR...REFERENCES |
|---|
Metamorphic evolution of the Cockburn Shear Zone
Magmatic stage
The magmatic minerals preserved in Pe1 are relicts of the anhydrous rock protolith (Walytjatjata Granodiorite). Relict igneous phases comprising clinopyroxene, apatite and zircon cores have moderate negative Eu anomalies. This is attributed to the strong Eu fractionation of plagioclase and K-feldspar (Fig. 7b and c). Notably, bulk-rock REE patterns of granodiorite outside and inside the shear zone have virtually no Eu anomaly. The titanite that replaces ilmenite has a negative Eu anomaly, and therefore we infer that it also coexisted with feldspar as a late-stage magmatic mineral. Similarly, hornblende replacing clinopyroxene has a negative Eu anomaly and is probably a late-stage magmatic phase. The almost identical REE patterns of hornblende and clinopyroxene support the textural observation that clinopyroxene was being replaced by hornblende. The similar REE concentrations of the two phases also imply that they are not in trace element equilibrium, based on predicted equilibrium partitioning values D
of 2–5 (Storkey et al., 2005; Buick et al., 2007) at similar P–T conditions to those in this study. Hornblende formation in a late magmatic stage may have resulted from fluid release during magma cooling and crystallization. The ages of the few zircon cores measured (1085–1108 Ma) indicate that the protolith of the orthogneisses crystallized in the Late Mesoproterozoic.
Sub-solidus metamorphic stage
Outside the shear zone, sub-solidus metamorphic reactions were not complete, as indicated by the preferential domainal development of new metamorphic minerals on relict igneous phases. Despite the presence of two feldspars, metamorphic minerals have small negative, negligible or even positive Eu anomalies (Fig. 7a). The corona textures in Pe1 indicate that chemical exchange pathways during metamorphism were short or limited, which suggests that the Eu signatures of the metamorphic minerals were inherited from the minerals they replaced. For example, metamorphic garnet, hornblende, titanite and apatite in leucocratic layers do not have negative Eu anomalies, and probably formed when Eu was liberated by the breakdown of feldspar. The transition from an anhydrous igneous assemblage to a hydrous metamorphic assemblage is an indication that fluid was added to the rock. In the presence of K-feldspar + quartz + plagioclase, fluid addition above the wet solidus would lead to melting. The absence of melt outside the CSZ, however, indicates that there was fluid addition at sub-solidus conditions.
Partial melting
Migmatitic orthogneisses in the CSZ contain abundant garnet. Garnet in felsic, metapelitic and metabasic migmatites is commonly interpreted to be a peritectic phase formed through incongruent melting reactions that consume other Fe–Mg-bearing minerals, such as biotite and hornblende (Vielzeuf & Schmidt, 2001; Patiño Douce, 2005). However, there is evidence that this is not the case for the rocks of the Mann Terrane, as explained below.
The approximate position of rocks from the CSZ in P–T space with respect to the fluid-present and fluid-absent solidus in quartzo-feldspathic rocks is shown in Fig. 11 (after Patiño Douce, 2005, fig. 11). Over a range of 5–12 kbar, wet melting for a two-feldspar- plus quartz-bearing rock occurs at 650 ± 25°C. In contrast, a temperature of at least 850°C is needed to produce garnet from dehydration melting of biotite and hornblende in a broadly granitic composition. Such temperatures are far in excess of estimates for the Petermann Orogeny, obtained from accessory phase thermometry (720–750°C; this study) and previous conventional thermobarometry (Camacho et al., 1997; Scrimgeour & Close, 1999). Both approaches to thermometry yielded comparable results for the central Musgrave Block (Fig. 11). Therefore, garnet in the CSZ migmatites must have formed by a process other than dehydration melting.
|
It has been documented extensively that garnet formed as a product of fluid-absent melting commonly has a strong negative Eu anomaly as a result of the co-production of K-feldspar during dehydration melting reactions (Bea et al., 1994
, 1997; Bea & Montero, 1999; Buick et al., 2006; Jung & Hellebrand, 2006; Rubatto et al., 2006). It is possible that biotite breakdown reactions during partial melting may occur under non-equilibrium conditions leading to the dissolution of feldspar (Barbey, 2007). The temperature required for such non-equilibrium melting would, however, still be too high compared with that indicated by our accessory phase thermometry. Therefore, on the basis of the low metamorphic temperature and lack of Eu anomaly in metamorphic garnet and hornblende from orthogneiss inside and adjacent to the CSZ (Fig. 8), we suggest that these minerals formed during sub-solidus reactions that involved feldspar and liberated Eu; for example, via a general sub-solidus hydration reaction K-feldspar + plagioclase + clinopyroxene + H2O 
garnet + biotite + epidote + quartz ± hornblende. Alternatively, garnet would also lack a significant negative Eu anomaly if it formed as a result of feldspar breakdown at or above the wet solidus. Therefore, although we can demonstrate that it did not form through fluid-absent melting reactions, the lack of a negative Eu anomaly alone is not sufficient to establish whether it grew under sub-solidus or near supra-solidus conditions. The relationships with accessory zircon place additional constraints on the relative timing of garnet growth.
Metamorphic zircon overgrowths were found only in the melted rocks of the CSZ. The identical bulk-rock composition inside and outside the shear zone suggests that the presence of fluid or melt was the driving force behind new zircon precipitation. This is in agreement with previous observations regarding the influence of melt on zircon behaviour (Rubatto et al., 2001, 2006). These studies document that in amphibolite- to granulite-facies metapelites, new zircon formed only at the onset of melting (
700°C) and, unlike monazite, occurred exclusively in the melt field. To help determine whether metamorphic zircon and garnet were in trace element equilibrium we calculated empirical REE zircon–garnet distribution coefficients (DREE) from migmatitic orthogneiss Pe13 (Fig. 12). The absolute values of distribution coefficients, and trends in their values as a function of atomic number, are in disagreement with inferred equilibrium distributions obtained empirically from HP (e.g. Hermann & Rubatto, 2003) or ultrahigh-temperature (UHT) metamorphic granulites (e.g. Hokada & Harley, 2004), or determined experimentally [800°C partitioning experiments of Rubatto & Hermann (2007)]. This suggests that zircon and garnet in the CSZ migmatitic orthogneiss are not in trace element equilibrium. Moreover, accepting that zircon is a melt product, the metamorphic zircon–garnet REE partitioning evidence supports the interpretation that within the CSZ garnet formed under sub-solidus conditions at T < 650°C.
|
It must be cautioned here that it is difficult to confidently identify an equilibrium assemblage for geothermobarometry of the CSZ migmatitic orthogneisses because the trace element partitioning clearly shows that (local) chemical equilibrium between coexisting minerals during prograde metamorphism cannot be assumed. This is highlighted by garnet and plagioclase, which are commonly used for garnet–hornblende–plagioclase–quartz geobarometry (Kohn & Spear, 1989
). These phases were in apparent disequilibrium for the entire rock history, as inferred from petrographic observations and trace element geochemistry. Plagioclase is present as a relict magmatic phase and a retrograde sub-solidus phase and therefore is not in equilibrium with any stage of garnet growth (Fig. 3).
|
Allanite formation in response to metamorphic P–T conditions
The value of trace elements for reconstructing complicated metamorphic histories is well demonstrated in high-grade terranes, where they are more sensitive monitors of geological processes than major element compositions based on the preservation of mineral-scale zoning (e.g. Hickmott & Shimuzu, 1990
; Lanzirotti, 1995; Otamendi et al., 2002; Van Orman et al., 2002). In such a way trace elements have provided a tool to help link U–Th–Pb ages of trace element-rich accessory minerals (principally zircon and monazite) to metamorphic P–T conditions (e.g. Rubatto, 2002; Hermann & Rubatto, 2003; Hokada & Harley, 2004; Kelly & Harley, 2005; Rubatto et al., 2006). In this section, the trace element compositions of dated allanite domains are used to relate these domains to major mineral phases in the rock.
Allanite preserves ‘reverse’ core to rim REE zoning (Fig. 9a), previously interpreted in the Catalina Schist to represent prograde growth zoning developed during sub-solidus metasomatic reactions (Sorenson, 1991). The occurrence of metamorphic allanite in both the melted and unmelted counterparts of the CSZ suggests that allanite growth initiated prior to the development of metamorphic zircon. In fact, metamorphic epidote appears as an early phase in the prograde metamorphic sequence as inclusions in garnet and as cores in allanite grains.
Trace element analyses of the two dated allanite domains identified using BSE images indicate compositional differences in Lu content, the size of the Eu anomaly, and initial Pb content (Fig. 13a and b). In igneous rocks, primary magmatic allanite typically contains 0–15% common 208Pb of the total 208Pb (Gregory et al., 2007; Gregory, 2008). In contrast, initial Pb concentrations in sub-solidus metamorphic (or hydrothermal) allanite are commonly substantially higher (above 60% common 208Pb of total 208Pb), particularly in allanite from (ultra)-HP metamorphic terranes (Davis et al., 1994; Catlos et al., 2000; Spandler et al., 2003; Frei et al., 2004; Romer & Xiao, 2005; Rubatto et al., 2008; Janots et al., 2009). Allanite crystallizing from a melt is involved in competitive partitioning of Pb with feldspar. Thus, melt-precipitated allanite commonly displays relatively low initial Pb contents. On the other hand, the high initial Pb contents in sub-solidus allanite can be explained in some cases by the breakdown of another sub-solidus phase, such as feldspar that releases Pb2+ (effective ionic radius of 1·19 Å; Shannon, 1967), which is subsequently incorporated into the Ca2+ (1·00 Å) A2-site of crystallizing allanite. Likewise, allanite strongly partitions Sr2+ (1·18 Å) in HP rocks where plagioclase is no longer stable (Sorensen, 1991; Nagasaki & Enami, 1998; Spandler et al., 2003; Frei et al., 2004; Rubatto et al., 2008). Because migmatization involves a combination of metamorphic and igneous processes we would expect the common 208Pb content of melt-precipitated allanite to reflect this (i.e. sub-solidus metamorphic > migmatitic > igneous).
The observed correlation between initial Pb content in allanite and the geological environment in which it forms may be relevant for understanding allanite growth history in the CSZ migmatitic orthogneiss. In this study, allanite cores show elevated initial Pb contents relative to rims and a small positive Eu anomaly (Fig. 13b). We interpret both features to reflect allanite core formation during prograde sub-solidus reactions that led to the breakdown of plagioclase and growth of garnet. Allanite formation may also be related to garnet with respect to Lu content. Allanite cores contain < 1 ppm Lu, whereas allanite rims are relatively enriched in the HREE (Fig. 13a). A comparison of the allanite domains indicates that the rim compositions may not have been in equilibrium with an HREE-rich phase (i.e. garnet). On this basis, we suggest that the allanite rims formed during incipient garnet breakdown related to melt crystallization, which would have involved the liberation of Lu. This hypothesis is further supported by the absence of allanite rim compositions in unmelted orthogneiss outside the shear zone.
Considerations on the dating of complex allanite
Performing geochronology on the CSZ migmatitic orthogneiss is dependent on zircon and allanite, both of which display multiple stages of formation. Therefore, to extract U–Th–Pb isotopic information that discriminates between distinct growth zones, a high-spatial resolution dating approach is necessary. In this case, analysis by thermal ionization mass spectrometry is inappropriate and would lead to a mixing of different populations, irrespective of the timescale over which the formation of the mineral zone occurred. Two micro-analytical techniques have been proposed for allanite dating; that is, SHRIMP and LA-ICP-MS (Gregory et al., 2007). The smaller sampling volume and higher sensitivity afforded by the ion microprobe made SHRIMP preferable for dating the CSZ allanite grains, which are strongly zoned and exhibit moderate to high amounts of common Pb.
Two episodes of allanite growth in the CSZ have been identified on the basis of textural and chemical evidence. This requires that the isotopic analyses be grouped as separate populations, from which ages can be calculated. In the specific case of the CSZ, the analytical uncertainty on the ages does not allow the age difference of the two episodes of allanite growth to be resolved. However, because the two allanite domains distinguished chemically can be related to different assemblages and their relative P–T stages, their unresolvable ages yield additional information (i.e. an estimate of the maximum duration of prograde to peak metamorphism).
The small volume sampled for isotopic analysis, combined with the low contents of radiogenic elements (particularly for zircon) has the inevitable repercussion of relatively large uncertainties on single U–Th–Pb data: < 6% and < 5% at the 2 level for zircon and allanite, respectively (Tables 3 and 4). Particularly for allanite, which is relatively non-radiogenic, these uncertainties do not allow for recognition of minor anomalies in the form of Pb loss or inheritance of exotic Pb (e.g. Romer, 2001; Romer & Siegesmund, 2003). A robust approach for correcting samples with significant amount of initial Pb (e.g. allanite, titanite, apatite) involves regressing uncorrected U–Pb data on a Tera–Wasserburg diagram (Rubatto & Hermann, 2001; Aleinikoff et al., 2002; Gregory et al., 2007). An alternative approach would be to determine the initial Pb composition from other minerals in the rock (e.g. leached K-feldspar; Romer, 2001). For the present study, however, this approach would be incorrect, as we have shown, using trace elements, that the predominant K-feldspar present in the rock is inherited from the 1 Ga igneous protolith. As allanite is a high Th/U mineral, we obtained an estimate of the initial Pb composition from a Th–Pb isochron regression of uncorrected data (Fig. 10b) and then ascertained that the Th–Pb and U–Pb systems had not been significantly disturbed. The uncertainty on single spots may limit the accuracy of this regression, and potential uncertainties in initial Pb and variable initial Pb through time could account for a few million years apparent age shift (Romer, 2001). However, this would have limited impact on the conclusions reached above on the timing of metamorphism and partial melting.
For the investigated samples, the general agreement of allanite and zircon ages provides evidence that allanite with a relatively high initial Pb content can be successfully dated by in situ methods. The suitability of metamorphic allanite for geochronology, however, is likely to remain sample dependent, based on the assessment of initial Pb content and composition.
Timing of metamorphism and partial melting
Determining the timing of shear zone development within a ductile regime (550–750°C) is a non-trivial task because of the difficulty in finding suitable geochronometers. The Ar–Ar method of dating remains one of the best ways to directly date deformation fabrics (e.g. Camacho & Fanning, 1995; Camacho et al., 1997). However, for HT shear zones the 40Ar/39Ar system typically records the timing of closure of micas or amphiboles to volume diffusion under relatively low temperature (< 550°C). In the relatively high-Ca, low-Al rocks of the CSZ, accessory allanite and zircon are the most important U–Th–Pb chronometers. Titanite and apatite are sufficiently low in radiogenic elements as to make them unsuitable for in situ dating. In particular, the Th-depleted compositions of titanite, apatite and metamorphic zircon are a sign that these phases formed in the presence of Th-bearing allanite.
New zircon growth associated with the Petermann Orogeny is rare and limited to discrete partial melt shear zones (Scrimgeour et al., 1999), which indicates that zircon formation within the CSZ was promoted by partial melting. We demonstrated above that zircon was not in REE equilibrium with sub-solidus garnet and, based on the Ti-in-zircon saturation thermometry, probably formed at T 700°C, implying that zircon crystallized during cooling, close to the wet solidus. We thus interpret the zircon rim U–Pb age of 555 ± 7 Ma to constrain the timing of partial melting and crystallization on cooling within the CSZ. This is in agreement with a metamorphic zircon SHRIMP U–Pb age of 561 ± 11 Ma obtained from a migmatite north of the CSZ (Scrimgeour et al., 1999). The metamorphic zircon age provides a useful constraint with which to examine allanite isotopic behaviour.
From petrography and trace element analysis we have established that, in contrast to zircon, two episodes of metamorphic allanite formation occurred along the CSZ P–T path: allanite core compositions are present in melted and unmelted counterparts of the CSZ and have high Gd/Lu suggestive of concomitant garnet growth. In comparison, allanite rim compositions are limited to migmatitic orthogneisses, are relatively HREE-enriched and show small Eu anomalies and initial 208Pb contents, attributed to the competitive formation of feldspar and allanite from a crystallizing melt. Petrography and mineral chemistry allow calculation of ages for each allanite domain, which bracket the zircon age. Allanite cores formed at 559 ± 6 Ma during sub-solidus hydration reactions occurring below the granitic wet solidus, and allanite rims formed at 551 ± 6 Ma during the initial stages of cooling and melt crystallization near or at the wet solidus (650°C). This implies that the prograde path from sub-solidus hydration to melt crystallization probably lasted of the order of 10 Myr.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGSAMPLE DESCRIPTIONANALYTICAL PROCEDURESRESULTSDISCUSSIONCONCLUSIONSAPPENDIX A: AVERAGE MAJOR...APPENDIX B: AVERAGE MAJOR...REFERENCES |
|---|
Amphibolite-grade metamorphism and melting in the CSZ occurred around
559 and 551 Ma as indicated by zircon and allanite U–Th–Pb dating. Notably, fluid-present melting (> 650°C) in the CSZ was not synchronous with the major episode of garnet and hornblende formation. Instead, garnet growth occurred from prograde sub-solidus hydration reactions at the expense of feldspar below the metamorphic peak. Peak temperature of 720–750°C from Zr-in-rutile thermometry supports the requirement of fluid-induced partial melting in the quartzo-feldspathic rocks.
Detailed trace element investigation has allowed the dated U–Th–Pb accessories to be related to garnet. Outside the CSZ in the orthogneiss, metamorphic zircon was absent and metamorphic allanite growth was limited. Inside the CSZ, metamorphic zircon crystallized from a cooling melt close to the wet solidus at T 700°C based on Ti-in-zircon thermometry. In comparison, metamorphic allanite formed over an extended P–T range: allanite cores formed on the prograde sub-solidus path in equilibrium with garnet rims and zircon and allanite overgrowths formed after garnet on initial melt crystallization. The two periods of allanite growth are not resolvable by the allanite dating as they occurred within a period of 10 Myr.
|
APPENDIX A: AVERAGE MAJOR ELEMENT COMPOSITION (WT %) OF MINERALS FROM Pe1, Pe11 AND Pe13 DETERMINED BY EPMA |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGSAMPLE DESCRIPTIONANALYTICAL PROCEDURESRESULTSDISCUSSIONCONCLUSIONSAPPENDIX A: AVERAGE MAJOR...APPENDIX B: AVERAGE MAJOR...REFERENCES |
|---|
|
|
APPENDIX B: AVERAGE MAJOR ELEMENT COMPOSITION (WT %) OF EPIDOTE–ALLANITE IN Pe13 DETERMINED BY EPMA |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGSAMPLE DESCRIPTIONANALYTICAL PROCEDURESRESULTSDISCUSSIONCONCLUSIONSAPPENDIX A: AVERAGE MAJOR...APPENDIX B: AVERAGE MAJOR...REFERENCES |
|---|
|
|
ACKNOWLEDGEMENTS |
|---|
C.J.G. thanks the EMU (ANU) for access to SEM facilities, and C. Allen and C. Magee for technical assistance. Dorothy Close, Christine Edgoose and Ian Scrimgeour of the Northern Territory Geological Survey are thanked for logistic support. The reviews of P. Barbey, F. Bea and three anonymous reviewers helped to improve the manuscript. The editorial handling of R. Gieré is acknowledged. Research was financially supported by the Research School of Earth Sciences and an Australian Research Council Grant DP0556700 (D.R. and J.H.). I.S.B. acknowledges an Australian Professorial Fellowship, Large Grant A39700909 and Discovery Grant DP0342473 from the Australian Research Council.
*Corresponding author. Present address: Department of Applied Geology, GPO Box U1987, Perth 6845, WA, Australia. Telephone: +61 (0)8 9266 7969. Fax: +61 (0)8 9266 3153. E-mail: C.Gregory{at}curtin.edu.au
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGSAMPLE DESCRIPTIONANALYTICAL PROCEDURESRESULTSDISCUSSIONCONCLUSIONSAPPENDIX A: AVERAGE MAJOR...APPENDIX B: AVERAGE MAJOR...REFERENCES |
|---|
Aleinikoff J. N., Wintsch R. P., Fanning C. M., Dorais M. J. U–Pb geochronology of zircon and polygenetic titanite from the Glastonbury Complex, Connecticut, USA: an integrated SEM, EMPA, TIMS, and SHRIMP study. Chemical Geology (2002) 188:125–147.[CrossRef][Web of Science]
Barbey P. Diffusion-controlled biotite breakdown reaction textures at the solid/liquid transition in the continental crust. Contributions to Mineralogy and Petrology (2007) 154:707–716.[CrossRef][Web of Science]
Barth S., Oberli F., Meier M. Th–Pb versus U–Pb isotope systematics in allanite from co-genetic rhyolite and granodiorite: implications for geochronology. Earth and Planetary Science Letters (1994) 124:149–159.[CrossRef][Web of Science]
Bea F., Montero P. Behaviour of accessory phases and redistribution of Zr, REE, Y, Th, and U during metamorphism and partial melting of metapelites in the lower crust: An example from the Kinzigite Formation of Ivrea–Verbano, NW Italy. Geochimica et Cosmochimica Acta (1999) 63:1133–1153.[CrossRef][Web of Science]
Bea F., Pereira M. D., Stroh A. Mineral/leucosome trace-element partitioning in a peraluminous migmatite (a laser ablation-ICP-MS study). Chemical Geology (1994) 117:291–312.[CrossRef][Web of Science]
Bea F., Montero P., Stroh A., Baasner J. Pressure dependence of REE distribution in metamorphic garnets from the Kinzigite formation (Ivrea–Verbano Zone, NW Italy). A LA-ICP-MS study. Geostandards Newsletters (1997) 21:253–270.[CrossRef]
Buick I. S., Hermann J., Williams I. S., Gibson R., Rubatto D. A SHRIMP U–Pb and LA-ICP-MS trace element study of the petrogenesis of garnet–cordierite–orthoamphibolite gneisses from the Central Zone of the Limpopo Belt, South Africa. Lithos (2006) 88:150–172.[CrossRef][Web of Science]
Buick I. S., Hermann J., Mass R., Gibson R. L. The timing of sub-solidus hydrothermal alteration in the Central Zone, Limpopo Belt (South Africa): Constraints from titanite U–Pb geochronology and REE partitioning. Lithos (2007) 98:97–117.[CrossRef][Web of Science]
Camacho A., Fanning C. M. Some isotopic constraints on the evolution of the granulite and upper amphibolite facies terrains in the eastern Musgrave Block, central Australia. Precambrian Research (1995) 71:155–181.[CrossRef][Web of Science]
Camacho A., Compston W., McCulloch M., McDougall I. Timing and exhumation of eclogite facies shear zones, Musgrave Block, central Australia. Journal of Metamorphic Geology (1997) 15:735–751.[CrossRef][Web of Science]
Catlos E. J., Sorensen S. S., Harrison T. M. Th–Pb ion-microprobe dating of allanite. American Mineralogist (2000) 85:633–648.
Clarke G. L., White R. W., Lui S., Fitzherbert J. A., Pearson N. J. Contrasting behaviour of rare earth element and major elements during partial melting in granulite facies migmatites, Wuluma Hills, Arunta Block, central Australia. Journal of Metamorphic Geology (2007) 25:1–18.[CrossRef][Web of Science]
Compston W., Williams I. S., Kirschvink J. L., Zha. Zircon U–Pb ages for the Early Cambrian time-scale. Journal of the Geological Society, London (1992) 149:171–184.
Davis D. W., Schandl E. S., Wasteneys H. A. U–Pb dating of minerals in alteration halos of Superior Province massive sulphide deposits: syngenesis versus metamorphism. Contributions to Mineralogy and Petrology (1994) 115:427–437.[CrossRef][Web of Science]
Degeling H. Zr Equilibria in Metamorphic Rocks (2003) Canberra: Australian National University. 231 pp.
Eggins S. M., Kinsley L. P. J., Shelley J. M. G. Deposition and element fractionation processes during atmospheric pressure laser sampling for analysis by ICP-MS. Applied Surface Science (1998) 127–129.
Frei D, Liebscher A, Franz G, Dulski P. Trace element geochemistry of epidote minerals. In:. Epidotes. Mineralogical Society of America, Reviews in Mineralogy and Geochemistry—Liebscher A, Franz G, eds. (2004) 56:553–605.
Gieré R., Sorensen S. S. Allanite and other REE-rich epidote-group minerals. In: Epidotes. Mineralogical Society of America, Reviews in Mineralogy and Geochemistry —Liebscher A., Franz G., eds. (2004) 56:431–493.
Gregory C. J., Rubatto D., Allen C. M., Williams I. S., Hermann J., Ireland T. R. Allanite microgeochronology: a SHRIMP and LA-ICP-MS study. Chemical Geology (2007) 245:162–182.[CrossRef][Web of Science]
Gregory C. J. Allanite chemistry and U-Th-Pb geochronology. PhD thesis, Research School of Earth Sciences. In: The Australian National University Canberra (2008) p. 271.
Hermann J. Allanite: thorium and light rare earth element carrier in subducted crust. Chemical Geology (2002) 192:289–306.[CrossRef][Web of Science]
Hermann J., Rubatto D. Relating zircon and monazite domains to garnet growth zones: age and duration of granulite facies metamorphism in the Val Malenco lower crust. Journal of Metamorphic Geology (2003) 21:833–852.[Web of Science]
Hickmott D. D., Shimuzu N. Trace element zoning in garnet from the Kwoiek Area, British Columbia: disequilibrium partitioning during garnet growth? Contributions to Mineralogy and Petrology (1990) 104:619–630.[CrossRef][Web of Science]
Hokada T., Harley S. L. Zircon growth in UHT leucosome: constraints from zircon–garnet rare earth element (REE) relations in Napier Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences (2004) 99:180–190.[CrossRef][Web of Science]
Hoskin P. W. O., Schaltegger U. The composition of zircon and igneous and metamorphic petrogenesis. In: Zircon. Mineralogical Society of America, Reviews in Mineralogy and Geochemistry —Hanchar J. M., Hoskin P. W. O., eds. (2003) 53:27–62.
Janots E, Engi M, Rubatto D, Berger A, Gregory C, Rahn M. Metamorphic rates in collisional orogeny from in situ allanite and monazite dating. Geology (2009) 37:11–14.
Jung S., Hellebrand E. Trace element fractionation during high-grade metamorphism and crustal melting—constraints from ion microprobe data of metapelitic, migmatitic and igneous garnets and implications for Sm–Nd garnet chronology. Lithos (2006) 87:193–213.[CrossRef][Web of Science]
Kelly N., Harley S. An integrated microtextural and chemical approach to zircon geochronology: refining the Archean history of the Napier Complex, east Antarctica. Contributions to Mineralogy and Petrology (2005) 149:57–84.[CrossRef][Web of Science]
Kohn M. J., Spear F. S. Empirical calibration of geobarometers for the assemblage garnet + hornblende + plagioclase + quartz. American Mineralogist (1989) 74:77–84.[Abstract]
Lanzirotti A. Yttrium zoning in metamorphic garnets. Geochimica et Cosmochimica Acta (1995) 59:4105–4110.[CrossRef][Web of Science]
Leake B. E. Nomenclature of amphiboles. American Mineralogist (1978) 63:1023–1052.[Abstract]
Lowery Claiborne L., Miller C. F., Walker B. A., Wooden J. L., Mazdab F. K., Bea F. Tracking magmatic processes through Zr/Hf ratios in rocks and Hf and Ti zoning in zircons: An example from the Spirit Moutain batholith, Nevada. Mineralogical Magazine (2006) 70:517–543.
Ludwig KR. Isoplot/Ex Version 2.4. A Geochronological Toolkit for Microsoft Excel. Berkeley Geochronological Centre Special Publications (2000) 1a.
Maboko M. A. H., McDougall I., Zeitler P. K., Williams I. S. Geochronological evidence for 530–550 Ma juxtaposition of two Proterozoic metamorphic terranes in the Musgrave Ranges, central Australia. Australian Journal of Earth Sciences (1992) 39:457–471.[Web of Science]
McDonough W. F., Sun S. S. The composition of the Earth. Chemical Geology (1995) 120:223–253.[CrossRef][Web of Science]
Mulrooney D., Rivers T. Redistribution of the rare-earth elements among coexisting minerals in metamafic rocks across the epidote-out isograd: an example from the St. Anthony Complex, Northern Newfoundland, Canada. Canadian Mineralogist (2005) 43:263–294.[CrossRef][Web of Science]
Nagasaki A., Enami M. Sr-bearing zoisite and epidote in ultra-high pressure (UHP) metamorphic rocks from the Su-Lu province, eastern China: An important Sr reservoir under UHP conditions. American Mineralogist (1998) 83:240–247.[Abstract]
Oberli F., Meier M., Berger A., Rosenberg C. L., Gieré R. U–Th–Pb and 230Th/238U disequilibrium isotope systematics: Precise accessory mineral chronology and melt evolution tracing in the Alpine Bergell intrustion. Geochimica et Cosmochimica Acta (2004) 68:2543–2560.[CrossRef][Web of Science]
Otamendi J. E., de la Rosa J. D., Patiño Douce A. E., Castro A. Rayleigh fractionation of heavy rare earths and yttrium during metamorphic garnet growth. Geology (2002) 30:159–162.
Patiño Douce A. E. Vapor-absent melting of a tonalite at 15–32 kbar. Journal of Petrology (2005) 46:275–290.
Pearce N. J. G., Perkins W. T., Westgate J. A., Gorton M. P., Jackson S. E., Neal C. R., Chenery S. P. A compilation of new and published major and trace element data for NIST SRM 610 and NIST SRM 612 glass reference materials. Geostandards Newsletter (1997) 21:115–144.[CrossRef][Web of Science]
Pyle J. M., Spear F. S. Yttrium zoning in garnet: coupling of major and accessory phases during metamorphic reactions. Geological Materials Research (1999) 1:1–49.[Medline]
Pyle J. M., Spear F. S. An empirical garnet (YAG)–xenotime thermometer. Contributions to Mineralogy and Petrology (2000) 138:51–58.[CrossRef][Web of Science]
Romer R. L. Lead incorporation during crystal growth and the misinterpretation of geochronological data from low-238U/204Pb metamorphic minerals. Terra Nova (2001) 13:485–500.
Romer R. L., Siegesmund S. Why allanite may swindle about its true age. Contributions to Mineralogy and Petrology (2003) 146:297–307.[CrossRef][Web of Science]
Romer R. L., Xiao Y. Initial Pb–Sr(–Nd) isotopic heterogeneity in a single allanite–epidote crystal: implications of reaction history for the dating of minerals with low parent-to-daughter ratios. Contributions to Mineralogy and Petrology (2005) 148:662–674.[CrossRef][Web of Science]
Rubatto D. Zircon trace element geochemistry: partitioning with garnet and the link between U–Pb ages and metamorphism. Chemical Geology (2002) 184:123–138.[CrossRef][Web of Science]
Rubatto D., Hermann J. Exhumation as fast as subduction? Geology (2001) 29:3–6.
Rubatto D., Hermann J. Experimental zircon/melt and zircon/garnet trace element partitioning and implications for the geochronology of crustal rocks. Chemical Geology (2007) 241:38–61.[CrossRef][Web of Science]
Rubatto D., Williams I. S., Buick I. S. Zircon and monazite response to prograde metamorphism in the Reynolds Range, central Australia. Contributions to Mineralogy and Petrology (2001) 140:458–468.[CrossRef][Web of Science]
Rubatto D., Hermann J., Buick I. S. Temperature and bulk compositional control on the growth of monazite and zircon during low-pressure anatexis (Mount Stafford, Central Australia). Journal of Petrology (2006) 47:1973–1996.
Rubatto D., Müntener O., Barnhoorn A., Gregory C. Dissolution–reprecipitation of zircon at low-temperature, high-pressure conditions (Lanzo Massif, Italy). American Mineralogist (2008) 93:1519–1529.
Scrimgeour I., Close D. Regional high-pressure metamorphism during intracratonic deformation: the Petermann Orogeny, central Australia. Journal of Metamorphic Petrology (1999) 17:557–572.[CrossRef]
Scrimgeour IR, Close DF, Edgoose CJ. Explanatory notes: Petermann Ranges SG 52–7, scale 1:250 000 (1999) Darwin: Northern Territory Geological Survey.
Shannon R. D. Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides. Acta Crystallographica (1967) A32:751–767.
Sorensen S. S. Petrogenetic significance of zoned allanite in garnet amphibolites from a paleo-subduction zone; Catalina Schist, Southern California. American Mineralogist (1991) 76:589–601.[Abstract]
Sorensen S. S., Grossman J. N. Enrichment of trace-elements in garnet amphibolites from a paleosubduction zone: Catalina Schist, southern California. Geochimica et Cosmochimica Acta (1989) 76:3155–3177.
Spandler C., Hermann J., Arculus R., Mavrogenes J. Redistribution of trace elements during prograde metamorphism from lawsonite blueschist to eclogite facies; implications for deep subduction-zone processes. Contributions to Mineralogy and Petrology (2003) 146:205–222.[CrossRef][Web of Science]
Stacey J. S., Kramers J. D. Approximation of terrestrial lead isotope evolution by a two-stage model. Earth and Planetary Science Letters (1975) 26:207–221.[CrossRef][Web of Science]
Storkey A., Hermann J., Hand M., Buick I. S. Using in situ trace element determination to monitor partial melt process in metabasites. Journal of Petrology (2005) 46:1283–1308.
Sun S.-S., Sheraton J. W., Glickson A. Y., Stewart A. J. A major magmatic event during 1050–1080 in central Australia and an emplacement age for the Giles Complex. AGSO Newsletter (1996) 24:13–15.
Van Orman J. A., Grove T. L., Shimuzu N., Layne G. D. Rare earth element diffusion in a natural pyrope single crystal at 2·8 GPa. Contributions to Mineralogy and Petrology (2002) 142:416–424.[Web of Science]
Vielzeuf D., Schmidt M. W. Melting reactions in hydrous systems revisited: applications to metapelites, metagreywackes and metabasalts. Contributions to Mineralogy and Petrology (2001) 141:251–267.[Web of Science]
Watson E. B., Harrison T. M. Zircon thermometer reveals minimum melting conditions on earliest Earth. Science (2005) 308:841–844.
Watson E. B., Wark D. A., Thomas J. B. Crystallisation thermometers for zircon and rutile. Contributions to Mineralogy and Petrology (2006) 151:413–433.[CrossRef][Web of Science]
Williams IS. U–Th–Pb geochronology by ion microprobe. In:. Applications of Microanalytical Techniques to Understanding Mineralizing Processes. Reviews in Economic Geology—McKibben MA, Shanks WC III, Ridley WI, eds. (1998) 7:1–35.
Zack T., Moraes R., Kronz A. Temperature dependence of Zr in rutile: empirical calibration of a rutile thermometer. Contributions to Mineralogy and Petrology (2004) 148:471–488.[Web of Science]
Zhao J.-X., McCulloch M. T., Korsch R. J. Characterisation of a plume related 800 Ma magmatic event and its implications for basin formation in central–southern Australia. Earth and Planetary Science Letters (1994) 121:349–367.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Acosta-Vigil, I. Buick, J. Hermann, B. Cesare, D. Rubatto, D. London, and G. B. Morgan VI Mechanisms of Crustal Anatexis: a Geochemical Study of Partially Melted Metapelitic Enclaves and Host Dacite, SE Spain J. Petrology, February 6, 2010; (2010) egp095v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||