Metapelitic Host Rocks Research Articles

Applicability of Na/K geothermometer to the metapelitic non-volcanic geothermal fields in the Taiwan orogenic belt

Taiwan is situated at one of the world's most active orogenic belts with arc-continent collision. This event resulted in the burial and uplift of mudstones along the continental margin, transforming them into metapelitic mountains. These mountains intercept rainfall and serve as the driving force for deep water circulation, enabling meteoric fluids to carry residual heat to the surface, thus giving rise to hot springs. The Taiwan orogenic belt is home to over 100 hot springs, and some of the hot spring reservoir temperatures such as Chingshui, Tuchang-Jentse, Lushan, Chihpon, and Chinlun exceeds 160 °C. Geothermal development in Taiwan has been carried out in such hot spring areas without any volcanic activities.In order to assess reservoir temperatures and confirm the applicability of the Na/K geothermometer to such high temperature hot spring system, the concentrations of Na+, K+, and SiO2 were analyzed from geothermal wells. Considering the mineral composition of metapelitic host rock, a new Na/K geothermometer is proposed based on equilibrium between albite and illite:T>180∘C,T(∘C)=2474log(NaK)+3.73−273.15This formula is only applicable in a reducing fluid and does not involve the formation of clay minerals except illite. It also requires several years to reach equilibrium. When using this formula to estimate reservoir temperatures, careful consideration of various conditions is necessary. It is advisable to incorporate both quartz geothermometer and down-hole temperature measurements for a more accurate assessment. Combining the Na/K geothermometer with the quartz geothermometer, along with other geochemical data such as Cl−concentrations and δD and δ18O of fluid, helps to understand whether hydrothermal fluids have undergone evaporation, dilution, and/or mixing processes.

Industrial garnet as an unconventional heavy rare earth element resource: Preliminary insights from a literature survey of worldwide garnet trace element concentrations

Current global supply of the critically important Heavy Rare Earth Elements (HREE) may be insufficient to meet future demand unless new sources are identified. This study utilizes trace element data in the literature to evaluate the potential of garnet, a common industrial mineral, to be further developed as a HREE-enriched mineral resource. A compilation of garnet REE data (n > 1,900 individual garnet analyses) demonstrates that garnet in metapelitic host rocks can have higher concentrations of the HREE dysprosium (Dy) through lutetium (Lu) in comparison to many of the existing and prospective sources of the Rare Earth Elements (REE). Furthermore, consideration of a smaller dataset containing garnet REE in addition to uranium and thorium concentrations indicates that many of the garnets with elevated HREE content also have uranium and thorium concentrations below those of the prospective and existing sources. As a case study, data from an actively mined garnet placer in northern Idaho (Emerald Creek), indicates that its proven garnet reserve contains ∼ 12 Mt of Lu alone despite its relatively small size. In total, the data indicate that experiments focused on the economics of extracting REE from garnet are warranted.

High-Density Upper Amphibolite/Granulite Facies Fluid Inclusions in Magmatic Garnet from the Koralpe Mountains (Eastern Alps, Austria)

Fluid and solid inclusions in magmatic garnet from Permian pegmatites of the Koralpe Mountains were investigated. On the basis of MnO/(MnO + FeO) ratios, different degrees of melt fractionation during garnet growth were linked with fluid inclusion densities and chemistries. It is shown that garnet indicating low-melt fractionation trends contained primary CO2 ± N2-rich fluid inclusions of the highest densities, up to 1.15 g/cm3, compared to garnet samples of increased fractionation trends comprising CO2-N2-rich fluid inclusions with lower densities up to 0.85 g/cm3. This fluid composition is interpreted as a part of an unmixed CO2 ± N2-H2O-rich fluid that was present during garnet crystallization. Variabilities in the nitrogen composition up to 40.83 mol% resulted from different degrees of partial melting of mica and plagioclase from the metapelitic host rock. Densities, fluid chemistries, and mineral chemical data enabled a continuous upward trend for garnet crystallization during anatexis from lower (ca. 25 km) up to middle crustal levels (12–15 km). Resulting amphibolite/granulite facies conditions of 7.6 kbar/700 °C for garnet crystallization in spodumene-free pegmatites were significantly higher than previously suggested for pegmatite formation in the Koralpe Mountains.

A secondary precious and base metal mineralization in chromitites linked to the development of a Paleozoic accretionary complex in Central Chile

Platinum-group element (PGE) and gold inclusions are usually present in peridotites and chromitite deposits associated with ophiolites. Here, we present the first detailed study of the mineralogy of precious metals in ultramafic rocks hosted in the Paleozoic Coastal Accretionary Complex of Central Chile. In these ultramafic rocks the mineralization of precious metals is associated with small meter-size pods and veins of massive chromitite hosted in serpentinite-filled shear zones. Crystallographic orientation maps of single chromite grains, obtained using the Electron-Backscattered Secondary Diffraction technique, allow us to identify two types of chromite in the precious-metal bearing chromitites: (1) Type A chromite, characterized by an average misorientation per grain of ≤2° and chemically homogeneous cores surrounded by a porous rim with abundant inclusions of chlorite, and (2) Type B chromite, which exhibits higher degrees of misorientation (2–8°) and porosity, and abundant silicate inclusions, but a relatively homogeneous chemical composition. In situ analyses using EMPA and LA-ICP-MS for major, minor and trace elements indicate that composition of the magmatic chromite is only preserved in the cores of Type A chromite grains. Core to rim chemical trends in these Type A chromites are characterized by a progressive increase of the Cr# with a decrease of the Mg#, loss of Al and addition of Fe2+ in the porous rim. The observed changes in the microstructure and chemistry of chromite are associated with the infiltration of external fluids through shear zones filled with antigorite (±talc) developed in partly serpentinized peridotites (i.e., olivine–lizardite dunites). Thermodynamic calculations using the phase equilibria relations in the system Cr2O3–MgO–FeO–Al2O3–SiO2–H2O (CrMFASH) indicate that Fe2+-rich porous chromite+chlorite replaced the original assemblage chromite+olivine in the chromitite while prograde antigorite was formed. According to our results this transformation occurred at ~510–560°C when external fluids penetrated the ultramafic/chromitite bodies through shear zones. These temperatures are slightly higher than estimated for the metamorphic peak in the host metapelitic rocks (i.e., ~420°C at 9.3kbar), suggesting that a hotter ultramafic body was captured by the metasediments of the accretionary prism during their exhumation through subduction channel. Chlorite geothermometry yielded a wide range of lower temperature from 430 to 188°C, for chlorite present in the porous chromite rims. These results are in agreement with the retrograde overprint under greenchist-facies metamorphism conditions recorded by metapelitic host rocks and minor volcanogenic massive sulphide deposits in the area (300–400°C, ~3–4kbar). We suggest that although initially decoupled, the chromitite-bearing ultramafic rocks and their metasedimentary host undergone a common metamorphic PT pathway of exhumation during the formation and evolution of the subduction-related accretionary complex.The chromitites contain appreciable amounts of the platinum-group elements (up to 347ppb total) and gold (up to 24ppb), present as inclusions of platinum-group minerals (PGM) and alloys as well as native gold. The PGM identified include native osmium, laurite (RuS2), irarsite (IrAsS), osarsite (OsAsS), omeiite (OsAs2), Pt–Fe alloy (possibly isoferroplatinum) and a suite of inadequately identified phases such as PtSb (possibly stumpflite), PdHg (possibly potarite), RhS, Ir–Ni and Ir–Ni–Ru compounds. Only a few grains of osmium and laurite were identified in unaltered cores of chromite and therefore considered as magmatic in origin formed during the high-T event of chomite crystallisation in the upper mantle. The other PGM were located in the porous chromite associated with chlorite or base-metal minerals (BMM) that often fill the pores of this altered chromite or are intergrowth with antigorite in the host serpentinized ultramafic rock. The assemblage of BMM identified in the studied rocks include sulphides [millerite (NiS), polydymite (Ni3S4), violarite (FeNi2S4), galena (PbS), sphalerite (ZnS), chalcocite (CuS)], arsenides [(orcelite (Ni5−xAs2) and maucherite (Ni11As8)], the sulpharsenide gersdorfitte (NiAsS), and native bismuth. The irregular shape of several PGM grains observed in porous chromite suggest disequilibrium, whereas others exhibit perfectly developed crystal faces with the associated secondary silicate or base-metal mineral suggesting neoformation of PGMs in situ from metamorphic fluids. We suggest that the origin of these PGM inclusions is magmatic, but some grains were reworked in situ when metalloid (i.e., As, Sb, Pb, Zn and Hg)-rich fluids released from metasediments penetrated the ultramafic rocks through active shear zones, once the ultramafic bodies became tectonically mixed with the host metasedimentary host rocks. During this event, gold sourced from the (meta)sediments was also precipitated within chromitites and serpentinites.

Tourmaline occurrences within the Penamacor-Monsanto granitic pluton and host-rocks (Central Portugal): genetic implications of crystal-chemical and isotopic features

Tourmalinization associated with peraluminous granitic intrusions in metapelitic host-rocks has been widely recorded in the Iberian Peninsula, given the importance of tourmaline as a tracer of granite magma evolution and potential indicator of Sn-W mineralizations. In the Penamacor-Monsanto granite pluton (Central Eastern Portugal, Central Iberian Zone), tourmaline occurs: (1) as accessory phase in two-mica granitic rocks, muscovite-granites and aplites, (2) in quartz (±mica)-tourmaline rocks (tourmalinites) in several exocontact locations, and (3) as a rare detrital phase in contact zone hornfels and metapelitic host-rocks. Electron microprobe and stable isotope (δ18O, δD, δ11B) data provide clear distinctions between tourmaline populations from these different settings: (a) schorl–oxyschorl tourmalines from granitic rocks have variable foititic component (X□ = 17–57 %) and Mg/(Mg + Fe) ratios (0.19–0.50 in two-mica granitic rocks, and 0.05–0.19 in the more differentiated muscovite-granite and aplites); granitic tourmalines have constant δ18O values (12.1 ± 0.1 ‰), with wider-ranging δD (−78.2 ± 4.7 ‰) and δ11B (−10.7 to −9.0 ‰) values; (b) vein/breccia oxyschorl [Mg/(Mg + Fe) = 0.31–0.44] results from late, B- and Fe-enriched magma-derived fluids and is characterized by δ18O = 12.4 ‰, δD = −29.5 ‰, and δ11B = −9.3 ‰, while replacement tourmalines have more dravitic compositions [Mg/(Mg + Fe) = 0.26–0.64], close to that of detrital tourmaline in the surrounding metapelitic rocks, and yield relatively constant δ18O values (13.1–13.3 ‰), though wider-ranging δD (−58.5 to −36.5 ‰) and δ11B (−10.2 to −8.8 ‰) values; and (c) detrital tourmaline in contact rocks and regional host metasediments is mainly dravite [Mg/(Mg + Fe) = 0.35–0.78] and oxydravite [Mg/(Mg + Fe) = 0.51–0.58], respectively. Boron contents of the granitic rocks are low (<650 ppm) compared to the minimum B contents normally required for tourmaline saturation in granitic melts, implying loss of B and other volatiles to the surrounding host-rocks during the late-magmatic stages. This process was responsible for tourmalinization at the exocontact of the Penamacor-Monsanto pluton, either as direct tourmaline precipitation in cavities and fractures crossing the pluton margin (vein/breccia tourmalinites), or as replacement of mafic minerals (chlorite or biotite) in the host-rocks (replacement tourmalinites) along the exocontact of the granite. Thermometry based on 18O equilibrium fractionation between tourmaline and fluid indicates that a late, B-enriched magmatic aqueous fluid (av. δ18O ~12.1 ‰, at ~600 °C) precipitated the vein/breccia tourmaline (δ18O ~12.4 ‰) at ~500–550 °C, and later interacted with the cooler surrounding host-rocks to produce tourmaline at lower temperatures (400–450 °C), and an average δ18O ~13.2 ‰, closer to the values for the host-rock. Although B-metasomatism associated with some granitic plutons in the Iberian Peninsula seems to be relatively confined in space, extending integrated studies such as this to a larger number of granitic plutons may afford us a better understanding of Variscan magmatism and related mineralizations.

Magnetic evidence of a crustal fault affecting a linear laccolith: The Guadiana Fault and the Monchique Alkaline Complex (SW Iberian Peninsula)

Magnetic anomalies can help reveal the structure of the upper crust in regions with intermediate or basic igneous rocks, and their continuity is essential to determine the position of crustal faults. The southwestern Iberian Peninsula constitutes the foreland of the Betic Cordillera and is characterized by an elongated E-W dipole extending 200km toward its external zones. The anomaly is related to the outcropping Monchique Alkaline Complex, characterized by rocks of moderate magnetic susceptibility (0.029 SI) intruding into the metapelitic host rock of the South Portuguese Zone. Analysis of aeromagnetic and field magnetic anomalies serves to constrain the geometry of this laccolith. Toward the east, the magnetic dipole has a 60km long N-S sharp boundary that coincides with the southern part of the Guadiana River. Field magnetic and gravity anomalies confirm the presence of this structure. It is produced by a sharp step in the elongated anomalous body, with an E downthrown block, interpreted as the offset produced by a deep N-S crustal fault—the Guadiana Fault. Therefore, the Guadiana River has three long linear segments near its mouth, locally coinciding with a N-S trending joint set, that support the presence of this major fault. To date, no evidence of this tectonic discontinuity, coinciding with the Spanish-Portuguese border, has been reported. Magnetic research is essential for understanding the structure of wide regions intruded by intermediate and/or basic igneous rocks.

Geochemistry and Geochronology of Peraluminous High-K Granitic Leucosomes of Yaound&amp;#233; Series (Cameroon): Evidence for a Unique Pan-African Magmatism and Melting Event in North Equatorial Fold Belt

Geochemical and geochronological studies have been carry out on the leucosomes of Yaounde series with the aims to identify the magma sources and to indicate their production periods and emplacement within the formations of the Pan-African North-Equatorial Fold Belt (PANEFB) in Cameroon. The Yaounde series belongs to the Southern domain of the PANEFB and it is composed of migmatites in which two types of granitic leucosomes (in situ leucosomes and injected leucosomes) have been distinguished. These rocks display characteristic of calc-alkaline (in situ leucosomes) and high-K calc-alkaline to shoshonitic series (injected leucosomes). All the rocks are peraluminous with in situ leucosomes conform to S-type and injected leucosomes conform to I- and S-type granitoids. Major and trace elements composition reveal that in situ leucosomes derived from the partial melting of the host metapelite whereas injected leucosomes derived from the melting of metagreywacke. These sources are similar to those of granitoids from central and northern domains of the PANEFB. Th-U-Pb dating by electron microprobe (EMP) and LA-ICP-MS U-Pb dating on zircon have been used to constraints the melting event and emplacement of leucosomes in Yaounde series. Th-U-Pb monazite dating, undertaken in two samples of leucosomes, gives two groups of monazite ages. The older group gives an age of 658 Ma whereas the age of younger group is 592 Ma. U-Pb dating of zircons from the leucosomes reveals a Pan-African age ranging from 626 to 654 Ma whereas zircons from metapelitic host rock reveal the overprinting of an early Pan-African event 911 - 1127 Ma on Palaeoproterozoic (2127 Ma) inheritance. These data clearly indicate that the host rocks of leucosomes of Yaounde series have been firstly metamorphosed during Tonien-Stenien period (911 - 1127 Ma) and reveal the existence of extended unique melting event (592 and 658 Ma) in the Yaounde series which is contemporaneous with the magmatism responsible for the emplacement of granitoids in the other domains of the PANEFB.

Assessing Bulk Assimilation in Cordierite-bearing Granitoids from the Central System Batholith, Spain; Experimental, Geochemical and Geochronological Constraints

This paper describes in detail an example of a calc-alkaline batholith, located in the Gredos massif (Spanish Central system batholith, Iberian Variscan massif), intruded into pelitic metasediments at shallow crustal levels. The igneous rocks of the study area are divided into three main groups according to their petrographic and geochemical (major, trace and rare earth element) characteristics: calc-alkalic (Qtz diorites and tonalites), transitional (Bt granodiorites), and alkali-calcic series (typically composed of Crd monzogranites). A paradoxical feature of the Crd monzogranites is the presence of euhedral Crd crystals in strong disequilibrium within a non-anatectic monzogranite. The field relations and geochemical data strongly suggest a process of bulk assimilation of the host metasediments by the intruding Bt granodiorites to generate the Crd monzogranites. Geochemical trends defined by the Crd monzogranites clearly depart from cotectic-like evolutions of liquid composition, and point to the composition of the metapelitic host-rocks. Laboratory experiments have been performed to check this hypothesis, and the results show that cordierite forms in a local domain where the reactants of the peritectic reaction Bt + Qtz + Pl + Als → Crd + Kfs + Melt are located. The process includes the isolation of metapelitic xenoliths in the granodiorite magma (a feature observed in the field), partial melting of these xenoliths with generation of cordierite as a peritectic phase in local equilibrium with the melt inside the entrapped xenolith, and the survival of these crystals because the residual melt of the granodiorite magma converges compositionally toward the peraluminous leucogranite generated in the relic xenolith. In turn, the granodiorite gains K released from the xenoliths, inducing the crystallization of K-feldspar and shifting to a Crd monzogranitic composition. Inherited zircons are present in the Crd monzogranites, similar to those found in the host metasediments; these are virtually absent in the intrusive Bt granodiorites. Mass-balance calculations and estimates based on the amount of Crd present have been applied to quantify the extent of assimilation in the petrogenesis of the granitoids; the volume of assimilated material can be high, depending on the proximity to the contacts with the xenoliths and the amount of pelitic material in the metasediments. © The Author 2010. Published by Oxford University Press. All rights reserved.

Li-bearing, disordered Mg-rich tourmaline from a pegmatite-marble contact in the Austroalpine basement units (Styria, Austria)

Pale-blue to pale-green tourmalines from the contact zone of Permian pegmatites to mica schists and marbles from different localities of the Austroalpine basement units (Rappold Complex) in Styria, Austria, are characterized. All these Mg-rich tourmalines have small but significant Li contents, up to 0.29 wt% Li2O, and can be characterized as dravite, with FeO contents of ~ 0.9–2.7 wt%. Their chemical composition varies from X (Na0.67Ca0.19 K0.02☐0.12) Y (Mg1.26Al0.97Fe2+ 0.36Li0.19Ti4+ 0.06Zn0.01☐0.15) Z (Al5.31 Mg0.69) (BO3)3 Si6O18 V (OH)3 W [F0.66(OH)0.34], with a = 15.9220(3), c = 7.1732(2) A to X (Na0.67Ca0.24 K0.02☐0.07) Y (Mg1.83Al0.88Fe2+ 0.20Li0.08Zn0.01Ti4+ 0.01☐0.09) Z (Al5.25 Mg0.75) (BO3)3 Si6O18 V (OH)3 W [F0.87(OH)0.13], with a = 15.9354(4), c = 7.1934(4) A, and they show a significant Al-Mg disorder between the Y and the Z sites (R1 = 0.013–0.015). There is a positive correlation between the Ca content and distance for all investigated tourmalines (r ≈ 1.00), which may reflect short-range order configurations including Ca and Fe2+, Mg, and Li. The tourmalines have XMg (XMg = Mg/Mg + Fetotal) values in the range 0.84–0.95. The REE patterns show more or less pronounced negative Eu and positive Yb anomalies. In comparison to tourmalines from highly-evolved pegmatites, the tourmaline samples from the border zone of the pegmatites of the Rappold Complex contain relatively low amounts of total REE (~8–36 ppm) and Th (0.1–1.8 ppm) and have low LaN/YbN ratios. There is a positive correlation (r ≈ 0.91) between MgO of the tourmalines and the MgO contents of the surrounding mica schists. We conclude that the pegmatites formed by anatectic melting of mica schists and paragneisses in Permian time. The tourmalines crystallized from the pegmatitic melt, influenced by the metacarbonate and metapelitic host rocks.

A zircon study from the Rhodope metamorphic complex, N-Greece: Time record of a multistage evolution

Zircons from an eclogite and a diamond-bearing metapelite near the Kimi village (north-eastern Rhodope Metamorphic Complex, Greece) have been investigated by Micro Raman Spectroscopy, SEM, SHRIMP and LA-ICPMS to define their inclusion mineralogy, ages and trace element contents. In addition, the host rocks metamorphic evolution was reconstructed and linked to the zircon growth domains. The eclogite contains relicts of a high pressure stage (ca. 700 °C and > 17.5 kbar) characterised by matrix omphacite with Jd 40–35. This assemblage was overprinted by a lower pressure, higher temperature metamorphic event (ca. 820 °C and 15.5–17.5 kbar), as indicated by the presence of clinopyroxene (Jd 35–20) and plagioclase. Biotite and pargasitic amphibole represent a later stage, probably related to an influx of fluids. Zircons separated from the eclogite contain magmatic relicts indicating Permian crystallization of a quartz-bearing gabbroic protolith. Inclusions diagnostic of the high temperature, post-eclogitic overprint are found in metamorphic zircon domain Z2 which ages spread over a long period (160 – 95 Ma). Based on zircon textures, zoning and chemistry, we suggest that the high-temperature peak occurred at or before ca. 160 Ma and the zircons were disturbed by a later event possibly at around 115 Ma. Small metamorphic zircon overgrowths with a different composition yield an age of 79 ± 3 Ma, which is related to a distinct amphibolite-facies metamorphic event. The metapelitic host rock consists of a mesosome with garnet, mica and kyanite, and a quartz- and plagioclase-bearing leucosome, which formed at granulite-facies conditions. Based on previously reported micro-diamond inclusions in garnet, the mesosome is assumed to have experienced UHP conditions. Nevertheless, (U)HP mineral inclusions were not found in the zircons separated from the diamond-bearing metapelite. Inclusions of melt, kyanite and high-Ti biotite in a first metamorphic zircon domain suggest that zircon formation occurred during pervasive granulite-facies metamorphism. An age of 171 ± 1 Ma measured on this zircon domain constrains the high-temperature metamorphic event. A second, inclusion-free metamorphic domain yielded an age of 160 ± 1 Ma that is related to decompression and melt crystallization. The similar age data obtained from the samples indicate that both rock types recorded a high-T metamorphic overprint at granulite-facies conditions at ca. 170 – 160 Ma. This age implies that any high pressure or even ultra-high pressure metamorphism in the Kimi Complex occurred before that time. Our findings define new constraints for the geodynamic evolution for the Alpine orogenic cycle within the northernmost Greek part of the Rhodope Metamorphic Complex. It is proposed that the rocks of the Kimi Complex belong to a suture zone squeezed between two continental blocks and result from a Paleo-ocean basin, which should be located further north of the Jurassic Vardar Ocean.

Formation of corundum in metapelites around ultramafic bodies. An example from the Saualpe region, Eastern Alps

This paper describes corundum formation in a metasomatic reaction zonation around an ultramafic body within a metapelitic sequence. The investigated body is about 100 m in diameter and is located in the Saualpe of the Austroalpine nappe complex in the Eastern Alps. The body is surrounded by a 10 m wide reaction zone (here called zone 3) containing the paragenesis garnet–staurolite–biotite–margarite–chlorite–corundum. Beyond a further metasomatic transition zone (here called zone 2), there are undisturbed metapelitic host rocks (zone 1) that have the metamorphic peak paragenesis garnet–biotite–plagioclase–staurolite–muscovite–quartz. It is shown that reaction zonation formed around 7.2 kbar and 615 °C during regional metamorphism, just above the serpentine breakdown reactions in the system MgO–SiO2–H2O. Detailed analysis of the whole rock compositions shows that the reaction zonation formed by infiltration metasomatism that caused significant mass loss in the two alteration zones. These zones are particularly depleted in SiO2, Na2O and possibly K2O. An XNa2O–XSiO2 thermodynamic pseudosection is presented that includes the parageneses of both the unaltered metapelitic host rock and the corundum-bearing parageneses. This suggests that the metasomatic process can be explained by the transfer of SiO2 and Na2O alone. We interpret that the process is driven by water liberated from the previously serpentinised ultramafic body during prograde dehydration during regional Eo-Alpine metamorphism. This fluid flowed outwards from the ultramafic body depleting the surrounding pelites in silica and causing margarite formation from plagioclase and muscovite. This interpretation of the driving mechanism is consistent with our knowledge of the low water activities of the Saualpe during the Eo-Alpine orogenic cycle.

Assimilation and crystal accumulation in a mid-crustal magma chamber: the Sausfjellet pluton, north-central Norway

The Sausfjellet pluton is made up of two intrusive units emplaced into high-grade metamorphic rocks of the Helgeland Nappe Complex of the Uppermost Allochthon in the Norwegian Caledonides. The eastern part of the pluton intruded marble and less voluminous calc-silicate and pelitic rocks. The western half is hosted predominantly by semi-pelitic migmatite with intercalated marble. Remelting of the migmatite during pluton emplacement occurred in a thermal aureole as much as 1000 m wide. The early gabbroic unit forms the southeastern part of the body; it consists of hornblende-bearing to hornblende-rich gabbro and diorite which is thought to have crystallized from an H 2O-rich andesitic parental magma. The younger dioritic unit underlies the central and western parts of the pluton, as well as a zone as much as 200 m wide that separates the rest of the pluton from its host rocks (herein the “annular zone”). The interior or central zone of the dioritic unit is pyroxene diorite that is locally interlayered with anorthosite. The western and annular zones are, by comparison, mineralogically heterogeneous. They range from diorite to quartz monzonite and from biotite-bearing two- and three-pyroxene assemblages to biotite–hornblende assemblages. Neither rock type nor mafic assemblage is correlated with position in the pluton or proximity to a contact. Stoped blocks of a distinctive coarse-grained diorite, as well as pyroxene-rich calc-silicates, are present in the gabbroic unit and the central zone of the dioritic unit. The few stoped blocks observed in the western zone of the dioritic unit are predominantly quartz-rich gneiss. Chemical variation in the central zone of the dioritic unit is interpreted to result from accumulation of pyroxenes+plagioclase from an H 2O-poor andesitic parent. These rocks have approximately constant δ 18O of +6.6±0.2‰ and lack evidence of in situ assimilation. Heterogeneities in the western and annular zones of the dioritic unit are reflected in variable, anomalously enriched incompatible element contents and in δ 18O, which ranges from +6.7‰ to +8.6‰. Petrologic models indicate that the magma parental to the central zone could also be parental to the western and annular zones. If so, evolution of the western and annular zone magma was by crystal accumulation and assimilation of metapelitic host rocks. As much as 20% of the mass of the western and annular zones can be ascribed to assimilated material, which apparently entered the magma by stoping. Therefore, the asymmetrical zoning of the pluton is due to differences in host rock compositions and the relative ability of the magma to assimilate its host rocks.

Boron metasomatism and behaviour of rare earth elements during formation of tourmaline rocks in the eastern Arunta Inlier, central Australia

Tourmaline rocks of previously unclear genesis and spatially associated with W- (Cu)-bearing calc-silicate rocks occur in Palaeoproterozoic supracrustal and felsic intrusive rocks in the Bonya Hills in the eastern Arunta Inlier, central Australia. Tourmalinisation of metapelitic host rocks postdates the peak of regional low-pressure metamorphism (M1/D1, ~500 °C, ~0.2 GPa), and occurred synkinematically between the two main deformation events D1 and D2, coeval with emplacement of Late Strangways (~1.73 Ga) tourmaline-bearing leucogranites and pegmatites. Tourmaline is classified as schorl to dravite in tourmaline–quartz rocks and surrounding tourmaline-rich alteration zones, and as Fe-rich schorl to foitite in the leucogranites. Boron metasomatism resulted in systematic depletion of K, Li, Rb, Cs, Mn and enrichment of B, and in some samples of Na and Ca, in the tourmaline rocks compared to unaltered metasedimentary host rocks. Whole-rock REE concentrations and patterns of unaltered schist, tourmalinised schist and tourmaline–quartz veins—the latter were the zones of influx of the boron-rich hydrothermal fluid—are comparable to those of post-Archaean shales. Thus, the whole-rock REE patterns of these rocks are mostly controlled by the metapelitic precursor. In contrast, REE concentrations of leucogranitic rocks are low (≤10 times chondritic), and their flat REE patterns with pronounced negative Eu anomalies are typical for fractionated granitic melts coexisting with a fluid phase. REE patterns for tourmalines separated from metapelite-hosted tourmaline–quartz veins and tourmaline-bearing granites are very different from one another but each tourmaline pattern mirrors the REE distribution of its immediate host rock. Tourmalines occurring in tourmaline–quartz veins within tourmalinised metasediments have LREE-enriched (LaN/YbN=6.3–55), shale-like patterns with higher ΣREE (54–108 ppm). In contrast, those formed in evolved leucogranites exhibit flat REE patterns (LaN/YbN=1.0–5.6) with pronounced negative Eu anomalies and are lower in ΣREE (5.6–30 ppm). We therefore conclude that REE concentrations and patterns of tourmaline from the different tourmaline rocks studied are controlled by the host rock and not by the hydrothermal fluid causing boron metasomatism. From the similarity of the REE pattern of separated tourmaline with the host rock, we further conclude that incorporation of REEs in tourmaline is not intrinsically controlled (i.e. by crystal chemical factors). Tourmaline does not preferentially fractionate specific REEs or groups of REEs during crystallisation from evolved boron- and fluid-rich granitic melts or during alteration of clastic metasediments by boron-rich magmatic-hydrothermal fluids.