Rare earth element distributions among minerals in a granodiorite and their petrogenetic implications

A study of the distribution of lanthanide rare earths in a granodiorite from the eastern Peninsular Ranges batholith, southern California, reveals that a large fraction of the REE in this rock resides in the accessory phases sphene and allanite. The minerals plagioclase, alkali feldspar, biotite, epidote and apatite each contribute approximately 1% or less of each REE to the whole rock, with the exception of Eu for which plagioclase contributes 7%. Sphene and allanite together contain 80% to 95% of each REE. Each of these phases is zoned in REE concentration with substantial decreases from core to margin. Textural observations argue for relatively early saturation and precipitation of sphene and allanite in the magma. REE zoning trends in sphene and allanite, and unexpectedly low REE concentrations in largely later crystallizing minerals such as feldspar, indicate that the precipitation of sphene and allanite significantly reduced REE concentrations in residual melts. These results illustrate the potential that sphene and allanite have for controlling the behavior of REE in granitic magmas.Available information collectively suggest that the sampled granodiorite existed as a complete melt, that the REE contained in the assemblage of phases were derived by direct crystallization from the melt, and that the melt behaved essentially as a closed system once crystallization of the phases now present began. Close correspondences between the major and trace element chemistries of the granodiorite and phenocryst-poor lavas from similar tectonic settings support these conclusions. The REE pattern of the granodiorite melt appears to have originated at depth and is characteristic of its source regions and derivation mechanism. The high liquidus temperature of a granodiorite melt (~ 1000°C) indicates the importance of mantle-derived components within the sources of batholithic magmas in the Peninsular Ranges.     

Rare earth element geochemistry of the Scourian complex N.W. Scotland — Evidence for the granite-granulite link

Medium-to high-pressure granulite facies complexes represent samples of lower crustal material and are, therefore, important in the study of crustal processes. New rare earth element data for the Scourian granulite facies terrain of the Precambrian Lewisian complex of N.W. Scotland indicate that:

1. Overall, the Scourian complex has a light rare earth enriched pattern with a small but distinct positive Eu anomaly;
2. While some rare earth element trends for the complex as a whole (e.g., Σ REE vs SiO₂) are similar to those observed in upper crustal cogenetic sequences, others (e.g., Eu/Eu* vs SiO₂) are reversed;
3. Compared to average upper crust, the Scourian complex is depleted in REE (except Eu) by a factor of 2 to 3.

These new data, along with previously reported major and trace element data, isotopic abundances, and trace element modelling support the hypothesis that the Scourian terrain is the residuum left after genesis and removal of granitic melts.     

Metamorphism of the Scourian Complex, NW Scotland

The rocks of the Scourian Complex have been intensively studied, but there is still no consensus as to the conditions of the granulite-facies metamorphism preserved in these rocks. Recent estimates of these conditions fall into two groups, one at 820-920°C and ca. 11 kbar and the second at ca. 1000°C and >12 kbar. Investigation of a variety of rocks shows that the recorded conditions vary with grain-size, with higher-grade conditions recorded by the cores of coarser ( ca. 10 mm) crystals, and lower-grade conditions recorded by the rims of coarser grains and by finer grains. This observation suggests that re-equilibration during recovery of these rocks to the surface has been important which may account for the discrepancy in estimated P-T conditions. Revised estimates of the equilibration conditions of the Scourian Complex of T > 1000°C and P > 8.5 kbar are presented. The conditions suggested for the peak of metamorphism mean that the role of anatexis in the genesis of these rocks must be considered and the nature of the fluid phase thoroughly investigated.     

Rare earth element mobility during prograde granulite facies metamorphism: significance of fluorine

 Mafic gneisses occur as lenses or thin layers in spatial association with tonalitic leucosomes in a granulite zone of the Quetico subprovince of the Superior Province, Ontario, Canada, and exhibit concentric zoning with a biotite-rich margin, orthopyroxene-rich outer zone, clinopyroxene-rich central zone, and, occasionally, patches of relict amphibolites within the clinopyroxene-rich zone. The granulites (biotite-, orthopyroxene- and clinopyroxene-rich zones) in the mafic gneisses are characterized by significant amounts of rare earth element (REE)-bearing fluorapatite (1–10 vol.%) and other REE-rich minerals (allanite, monazite and zircon). Fluorapatite shows an increase in modal abundance from the biotite- and orthopyroxene-rich zones to the clinopyroxene-rich zone, but is rare in the relict amphibolites. Textural evidence and element partitioning indicate that fluorapatite (and other REE-rich minerals) was part of the peak metamorphic assemblages. Whole-rock geochemical analyses confirm that the granulites in the mafic gneisses contain anomalously high contents of REE and high field strength elements (HFSE), whereas the relict amphibolites are geochemically typical of tholeiitic basalts. Mass-balance calculations reveal that REE and HFSE were introduced into the mafic gneisses during the prograde granulite facies metamorphism, pointing to REE mobility under granulite facies metamorphic conditions. The presence of high F contents in the REE-rich minerals and their associated minerals (e.g. biotite and hornblende) suggests that REE and HFSE may have been transported as fluoride complexes during the granulite facies metamorphism. This conclusion is supported by previously published results of hydrothermal experiments on the partitioning of REE between fluorapatite and F-rich fluids at 700°C and 2 kbar. Received: 2 May 1995 / Accepted: 28 September 1995     

The trace element geochemistry of clinopyroxenes from pyroxenites in the Lewisian of NW Scotland: insights into light rare earth element mobility during granulite facies metamorphism

Clinopyroxenes from layered pyroxenites and from pyroxenite pods in felsic gneisses of the Lewisian granulite complex, NW Scotland, have distinctive chemistries suggestive of different origins. Clinopyroxenes in the layered pyroxenites crystallised from mafic melts in a magma chamber located in the middle to shallow crust, whereas clinopyroxenes in pods in the felsic gneisses crystallised from the tonalitic protolith to the felsic gneisses. In detail clinopyroxenes in the layered pyroxenites are variably enriched in the light REE. Inversion modelling shows that this is not a primary feature inherited from their parent magmas. Rather selective light rare earth element enrichment took place through reaction with a felsic melt generated by the localised partial melting of the hornblende pyroxenites during granulite facies metamorphism. Published isotopic evidence suggests that the light REE mobilisation took place at ca 2.7 Ga, about 200 Ma after the time of crust formation. This observation provides an explanation for the scattered pattern of whole-rock isochron ages from the Lewsian granulites.     

Interpreting granulite facies events through rare earth element partitioning arrays

The use of rare earth element (REE) partition coefficients is an increasingly common tool in metamorphic studies, linking the growth or modification of accessory mineral geochronometers to the bulk silicate mineral assemblage. The most commonly used mineral pair for the study of high‐grade metamorphic rocks is zircon and garnet. The link from U–Pb ages provided by zircon to the P–T information recorded by garnet can be interpreted in relation to experimental data. The simplistic approach of taking the average REE abundances for zircon and garnet and comparing them directly to experimentally derived partition coefficients is imperfect, in that it cannot represent the complexity of a natural rock system. This study describes a method that uses all the zircon analyses from a sample, and compares them to different garnet compositions in the same rock. Using the most important REE values, it is possible to define zircon–garnet equilibrium using an array rather than an average. The array plot describes partitioning between zircon and garnet using D_Yb and D_Yb/D_Gd as the defining features of the relationship. This approach provides far more sensitivity to mineral reactions and diffusional processes, enabling a more detailed interpretation of metamorphic history of the sample.     

Geochemistry and origin of granitic rocks,Scourian Complex,NW Scotland

Concordant granite sheets from the granulite facies Scourian Complex, N.W. Scotland exhibit the following features:

1. a common planar fabric with their host pyroxene granulites;
2. the presence of an exsolved ternary feldspar phase;
3. a low-pressure, water-saturated minimum composition;
4. K/Rb ratios (450–1,350) distinctly higher than most upper crustal granites but similar to the surrounding granulites;
5. low absolute concentrations of the rare earth elements (REEs), light REE enrichment, and large positive Eu anomalies.

It is proposed that the granite sheets have originated by anatexis of gneisses undergoing granulite facies metamorphism — gneisses that were already essentially dry and depleted in incompatible elements. Their unusual trace element chemistry may be explained by either disequilibrium melting and/or sub-solidus reequilibration of the granite sheets with the surrounding gneisses. Isotopic and trace element data suggest that cross-cutting, potash-rich pegmatites represent reworking of the granite sheets during a later amphibolitization.     

Large-ion lithophile element characteristics of an amphibolite facies to granulite facies transition at Gruinard Bay, North-west Scotland

Abstract Lewisian grey gneisses from Gruinard Bay, North-west Scotland retain mineralogical and geochemical evidence for Scourian horn-blende-granulite facies metamorphism, and they may be used to assess current models of elemental depletion at granulite grade. Their 'immobile'major and trace element geochemistry is indistinguishable from that of Lewisian amphibolite and pyroxene-granulite facies counterparts. The K, Rb, Th and U contents of the Gruinard Bay gneisses are depleted relative to amphibolite facies gneisses, but generally the abundances of these elements are above those of comparable pyroxene granulites. U and Th have reached an advanced stage of depletion, but allanite appears to be crucial in maintaining significantly higher U and Th abundances at Gruinard Bay than in pyroxene granulites. K and Rb loss is less extreme, and depends on the stability of the rock-forming minerals: K-feldspar; biotite; and, amphibole. Early removal of K and Rb has resulted in a small rise in K/Rb, but further preferential Rb loss would have been required to generate the characteristically high K/Rb ratios of Lewisian pyroxene granulites.
The residence of U and Th in the accessory minerals of granulite facies gneisses, which are often correlated with the residua of intracrustal partial melting, renders unlikely their extreme incompatibility required by such models. Even if such phases are ignored, high mineral-melt partition coefficients for silicic melts argue against partial fusion as an efficient depletion mechanism. On the other hand, the advanced stage of U and Th depletion reached in Gruinard Bay gneisses, which were still partly hydrous, severely restricts the role played by CO₂-dominated fluids and a hydrous medium is preferred.     

Rare earth element partition between sphene,apatite and other coexisting minerals of the Kangerdlugssuaq intrusion,E. Greenland

Cumulus apatite, sphene, feldspar, amphibole and biotite from the pulaskite of the Kangerdlugssuaq alkaline intrusion have been analysed for rare earth elements (REE) by instrumental neutron activation analysis. The apatite is particularly rich in REE, contains 3.6% Ce and shows a steep, light REE-enriched, chondrite-normalised pattern. The other minerals have light REE enrichment but with sphene showing a peak at Ce on a chondrite-normalised plot. REE partition coefficient values show that the light REE are preferentially accommodated by apatite relative to sphene. The differences in these coefficients result from differences in the co-ordination of the REE in the two minerals.     

Significance of rare earth distributions in coexisting minerals of peralkaline undersaturated rocks

Rare earth element (REE) abundances in eudialytes and co-existing feldspars and mafic minerals from six alkaline intrusions ranging from mildly to strongly peralkaline indicate large scale complexing of the REE and yttrium. The eudialyte-bearing stages of these intrusions are characterized by a low Ce/Y ratio. This enrichment in the heavy REE is not due to the high REE content of the mineral eudialyte but appears to be related to the importance, duration and type of REE and other element complexing during differentiation. Feldspar and mafic mineral REE abundances reflect this low Ce/Y ratio and are often altered by late stage magmatic, metasomatic or hydrothermal fluids enriched in the heaviest REE and yttrium.     

Rare earth element distributions in a proto-stratiform ultramafic intrusion

Rare earth element (REE) abundances are reported for ten whole rock and eight mineral samples from the Preacher Creek ultramafic intrusion of southeastern Wyoming. Chondrite-normalized distribution patterns for the whole rocks exhibit a broad maximum between Sm and Gd and reflect the REE pattern of clinopyroxene, the major REE-bearing phase. Alteration of the primary mineral assemblages to actinolite and chlorite, which is generally minor, does not appear to have significantly affected the REE distributions. Absolute abundances of the REE in the rocks and constituent minerals increase as a function of differentiation, and relative abundances suggest an accompanying light REE enrichment. Trapped-liquid phases, which may be relatively enriched in REE, possibly account for some or all of the observed REE trends. The REE data, interpreted in terms of crystal-melt fractionation, suggest derivation of the intrusion by crystallization from a gabbroic magma having a REE distribution pattern similar to the parent magma of the Skaergaard stratiform complex. The results of this study are in accord with and complement a previous proposal that the Preacher Creek body formed in a manner analogous to major stratiform intrusions.     

The geochemistry of Precambrian granulite facies rocks from the Lewisian complex of Tiree,Inner Hebrides,Scotland

The metamorphism and geochemistry of the major components of a small area of granulite facies rock are described and discussed, and a chemical model for the evolution of anomalous trace element distributions in such materials is suggested. The local complex was subjected to medium to high pressure granulite facies metamorphism between 2,900 and 2,600 m.y. All the analysed granulite facies rocks from Tiree; acid to intermediate gneisses, basic metamorphic rocks, and granitic rocks, have anomalous chemistries, being depleted in K, Rb, Nb, Y and Th, and have high K/Rb, Ba/Rb and Ca/Y ratios, and very low K/Ba and Rb/Sr ratios relative to normal portions of the upper continental crust. The gneisses seem to have been enriched in Ba and Sr.The chemical features of the rocks are considered to reflect their stable mineral assemblages in the granulite facies, and to be representative of deep-level crustal materials. The geochemical peculiarities of the complex may have been largely controlled by an upward intergranular diffusion, or “degassing” caused by high-grade metamorphism. It is suggested that such diffusion may have been active at the crust/upper mantle interface, some diffused material of mantle origin accounting for certain chemical oddities typical of Lewisian and some other Precambrian granulite facies rocks.     

Rare earth element partitioning between minerals from anhydrous spinel peridotite xenoliths

REE abundances in minerals from spinel peridotite xenoliths from West Germany, the south-western U.S. and Mongolia decrease in the order clinopyroxene > orthopyroxene > olivine > spinel. While clinopyroxenes are similar in absolute chondrite-normalized concentrations to those known from other studies, orthopyroxenes and olivines are significantly lower in LREE although comparable in HREE. Spinels are much lower in all REE than any previously reported values and are completely negligible for the REE budget of peridotites.Partition coefficients for most orthopyroxene/clinopyroxene pairs increase systematically from La to Lu. Olivine/clinopyroxene and spinel/clinopyroxene partition coefficients increase from the intermediate rare earth elements to Lu and normally are higher for La compared to Sm.The application of Nagasawa's (1966) elastic lattice model suggests that all heavy but only minor amounts of the light REE substitute into structural positions of orthopyroxene and olivine.Significant differences between orthopyroxene/clinopyroxene partition coefficients for various xenoliths may be assigned to dependences upon equilibration temperature and bulk chemistry.Apart from grain surface contaminations, fluid inclusions which are practically always present in mantle minerals, can highly concentrate light rare earth elements and thus may be responsible for unexpectedly high concentrations of incompatible elements frequently reported for mantle olivines or orthopyroxenes.     

Rare earth element geochemistry of the Fen alkaline complex,Norway

Determination of rare earth element (REE) abundances in rocks of the Fen complex has shown that within rocks of the first magmatic series REE abundances increase in the order urtiteFen magmas are discussed and it is considered that parental magmas had relatively high La/Yb ratios (40–60). Utilizing petrological evidence from other alkaline complexes coupled with experimental studies it is considered that the parental magma was a carbonated nephelinite produced by limited (<10%) partial melting of the mantle. All the Fen rocks are placed in a petrogentic scheme in which a carbonated nephelinite magma undergoes liquid immiscibility, differentiation and volatile transport.     

Rare earth element distributions in some Precambrian rocks and their phyllosilicates,Numedal, Norway

Mass spectrometric analyses for rare earth elements (REE) have been carried out on some Precambrian mica schists, gneisses and granites from the Precambrian Numedal area, Norway and on their phyllosilicates. The rocks, which are metamorphosed in the upper greenschist to amphibolite facies, were originally partly sedimentary, partly magmatic.The total REE contents for rocks varies from 145 to 761 ppm. An average of 16 phyllosilicate samples gave 417 ppm REE (max. of 1809 ppm, min. of ca. 50 ppm). Coexisting light and dark phyllosilicates have similar abundances of REE. For the micas of high REE content most of the REE was extractable by rinsing with EDTA. The data thus support the possibility of an extensive adsorption of REE ions on micaceous minerals. The REE distribution patterns do not provide a clear distinction between the sedimentary and magmatic origin for the rocks examined.     

Archaean granulite facies metamorphism of the Lewisian of Tiree, Inner Hebrides, north-west Scotland

The Lewisian of Tiree, north-west Scotland, underwent granulite facies metamorphism prior to 2.4 Ga. The temperatures and pressures estimated from garnet–clinopyroxene, garnet–orthopyroxene, hornblende–plagioclase and garnet–biotite geothermometers and clinopyroxene–plagioclase–garnet–quartz and orthopyroxene–plagioclase–garnet–quartz geobarometers are 810 ± 50° C and 10.5 ± 1.5 kbar. The imprecision of pressure estimates stems largely from uncertainties in garnet activity models. Calculations of blocking temperatures for Fe–Mg interdiffusion in clinopyroxene and garnet suggest that these temperatures and pressures represent only slightly reset peak-metamorphic conditions.
Down-temperature re-equilibration resulted in chemical zoning over the outer 50–100 μm of the mafic minerals. P–T paths calculated from this mineralogical zoning suggest nearly isobaric cooling. However, the growth of late sillimanite in metapelites requires that the retrograde P–T path had a significant decompression component, suggesting that the mineralogical zonation does not define the retrograde P–T path. The discrepancy between the P–T path calculated from mineralogical zonation and that implied by mineral reactions probably results from the net-transfer geobarometry reactions closing at higher temperatures than the exchange geothermometers.
The Tiree rocks have a similar history to the mainland Scourian complex. Granulite facies metamorphism accompanied by partial melting occurred prior to the intrusion of the Scourie dykes at c. 2.4 Ga, and the rocks underwent retrogression both prior to and after dyke emplacement. However, peak metamorphic temperatures and pressures on Tiree were lower than those recorded in the Scourian complex, and the Tiree rocks may have been at a different crustal level at that time.     

Rare earth element distribution in some hydrothermal minerals: evidence for crystallographic control

Rare earth element (REE) abundances were measured by neutron activation analysis in anhydrite (CaSO₄), barite (BaSO₄), siderite (FeCO₃) and galena (PbS). A simple crystal-chemical model qualitatively describes the relative affinities for REE substitution in anhydrite, barite, and siderite. When normalized to ‘crustal’ abundances (as an approximation to the hydrothermal fluid REE pattern), log REE abundance is a surprisingly linear function of (ionic radius of major cation—ionic radius of REE)² for the three hydrothermal minerals, individually and collectively. An important exception, however, is Eu, which is anomalously enriched in barite and depleted in siderite relative to REE of neighboring atomic number and trivalent ionic radius. In principle, REE analyses of suitable pairs of co-existing hydrothermal minerals, combined with appropriate experimental data, could yield both the REE content and the temperature of the parental hydrothermal fluid.The REE have only very weak chalcophilic tendencies, and this is reflected by the very low abundances in galena—La, 0.6 ppb; Sm, 0.06 ppb; the remainder are below detection limits.     

Pumpellyites and coexisting minerals in different low-grade metamorphic facies of Liguria, Italy

Pumpellyite from four-phase assemblages (pumpellyite + epidote + prehnite + chlorite; pumpellyite + epidote + actinolite + chlorite; pumpellyite + epidote + Na-amphibole + chlorite, together with common excess phases), considered to be low variance in a CaO-(MgO + FeO)-Al₂O₃-Fe₂O₃ (+Na₂O + SiO₂+ H₂O) system, have been examined in areas which underwent metamorphism in the prehnite-pumpellyite, pumpellyite-actinolite and low-temperature blueschist facies respectively. The analysed mineral assemblages are compared for nearly constant (basaltic) chemical composition at varying metamorphic grade and for varying chemical composition (basic, intermediate, acidic) at constant metamorphic conditions (low-temperature blueschist facies). In the studied mineral assemblages, coexisting phases approached near chemical equilibrium. At constant (basaltic) bulk rock composition the MgO content of pumpellyite increases, and the X_Fe3+ of both pumpellyite and epidote decreases with increasing metamorphic grade, the Fe³⁺ being preferentially concentrated in epidote. Both pumpellyite and epidote compositions vary with the bulk rock composition at isofacial conditions; pumpellyite becomes progressively enriched in Fe and depleted in Mg from basic to intermediate and acidic bulk rock compositions. The compositional comparison of pumpellyites from high-variance (1–3 phases) assemblages in various bulk rock compositions (basic, intermediate, acidic rocks, greywackes, gabbros) shows that the compositional fields of both pumpellyite and epidote are wide and variable, broadly overlapping the compositional effects observed at varying metamorphic grade in low-variance assemblages. The intrinsic stability of both Fe- and Al-rich pumpellyites extends across the complete range of the considered metamorphic conditions. Element partitioning between coexisting phases is the main control on the mineral composition at different P-T conditions.     

Rare earth element geochemistry of Thetford Mines ophiolite complex,Northern Appalachians,Canada

The Thetford Mines complex is a complete ophiolite which is part of an ultramafic-mafic belt within Québec Appalachians. These allochtonous bodies were emplaced during the Early Ordovician. The Thetford Mines complex comprises a lower unit of metamorphic harzburgite (in which tabular, dyke-like, dunitic bodies occur) overlain successively by ultramafic cumulates, mafic cumulates, ophitic gabbros, diabase sills and dykes, and basaltic volcanic rocks. Field evidence, petrography and chemical data indicate that the tabular dunitic bodies formed when fractures in the metamorphic harzburgite (which constituted the floor of the magma chamber) filled with early cumulates (i.e., olivine±chromite). Representative rocks from all units were analyzed for major and rare earth elements (REE). Metamorphic harzburgite samples from Thetford Mines complex have U-shaped chondrite-normalized REE patterns. Pyroxenites and wehrlites of the cumulate sequence are all strongly light-REE depleted and have heavy REE ranging from 0.4 to 1.5 times chondrite. REE data from ultramafic and volcanic rocks of Thetford Mines complex and geochemical modelling indicate that the metamorphic harzburgite has the chemical characteristics of depleted upper mantle residues with U-shaped patterns, and that the ultramafic cumulates crystallized from magmas having different La/Yb ratios.