Peralkaline Melts Research Articles

On the Controls of Oxygen Fugacity in the Generation and Crystallization of Peralkaline Melts

The formation of peralkaline melts is generally recognized to be related to enrichment of the mantle in sodium or potassium during a metasomatic event that predated magma generation. The wide range in mineral assemblages, mineral compositions, and fluid compositions in peralkaline melts are, to a large extent, a function of the sodium/potassium content of these mantle-metasomatizing fluids, which also governs redox conditions. We propose that sodic fluids will reduce the mantle assemblage by depleting ferric iron from garnet to form the aegirine component in pyroxene. In contrast, potassic fluids would oxidize the mantle assemblage by extracting Al2O3 from the garnet to make phlogopite. We suggest that during differentiation of peralkaline melts, once Fe^Ti oxides have been depleted from the assemblage, simple crystallization reactions of common solid phases such as aegirine or arfvedsonite control the oxygen fugacity by equilibria such as

Solubility of columbite, (Mn, Fe)(Nb, Ta)2O6, in granitoid and alkaline melts at 650–850°C and 30–400 MPa: An experimental investigation

This paper reports experimental data on columbite solubility in model water-saturated Li- and F-rich silicic melts with different contents of alumina and alkalis. It was found that the columbite solubility is strongly affected by melt composition and is maximal in peralkaline melt. The maximum contents of Ta and Nb in subaluminous and peraluminous melts at the contact with columbite are lower by at least an order of magnitude. The peralkaline melt is relatively enriched in Nb, and the peraluminous melt is enriched in Ta. The temperature dependence of solubility is positive but less pronounced than the effect of melt composition. It is most distinct in the subaluminous melts. The Nb/Ta ratio of melt usually decreases with decreasing temperature. The effect of pressure is relatively small. It was shown that columbite cannot crystallize on the liquidus of both peralkaline and peraluminous magmas. Perhaps, columbite crystallization from a melt is possible only at final near-solidus stages at the high degrees of crystallization of strongly evolved low-temperature melts.

Shear modulus, heat capacity, viscosity and structural relaxation time of Na2O–Al2O3–SiO2 and Na2O–Fe2O3–Al2O3–SiO2 melts

The configurational heat capacity, shear modulus and shear viscosity of a series of Na2O–Fe2O3–Al2O3–SiO2 melts have been determined as a function of composition. A change in composition dependence of each of the physical properties is observed as Na2O/(Na2O + Al2O3) is decreased, and the peralkaline melts become peraluminous and a new charge-balanced Al-structure appears in the melts. Of special interest are the frequency dependent (1 mHz–1 Hz) measurements of the shear modulus. These forced oscillation measurements determine the lifetimes of Si–O bonds and Na–O bonds in the melt. The lifetime of the Al–O bonds could not, however, be resolved from the mechanical spectrum. Therefore, it appears that the lifetime of Al–O bonds in these melts is similar to that of Si–O bonds with the Al–O relaxation peak being subsumed by the Si–O relaxation peak. The appearance of a new Al-structure in the peraluminous melts also cannot be resolved from the mechanical spectra, although a change in elastic shear modulus is determined as a function of composition. The structural shear-relaxation time of some of these melts is not that which is predicted by the Maxwell equation, but up to 1.5 orders of magnitude faster. Although the configurational heat capacity, density and shear modulus of the melts show a change in trend as a function of composition at the boundary between peralkaline and peraluminous, the deviation in relaxation time from the Maxwell equation occurs in the peralkaline regime. The measured relaxation times for both the very peralkaline melts and the peraluminous melts are identical with the calculated Maxwell relaxation time. As the Maxwell equation was created to describe the timescale of flow of a mono-structure material, a deviation from the prediction would indicate that the structure of the melt is too complex to be described by this simple flow equation. One possibility is that Al-rich channels form and then disappear with decreasing Si/Al, and that the flow is dominated by the lifetime of Si–O bonds in the Al-poor peralkaline melts, and by the lifetime of Al–O bonds in the relatively Si-poor peralkaline and peraluminous melts with a complex flow mechanism occurring in the mid-compositions. This anomalous deviation from the calculated relaxation time appears to be independent of the change in structure expected to occur at the peralkaline/peraluminous boundary due to the lack of charge-balancing cations for the Al-tetrahedra.

CRYSTALLIZATION OF THE KYMI TOPAZ GRANITE STOCK WITHIN THE WIBORG RAPAKIVI GRANITE BATHOLITH, FINLAND: EVIDENCE FROM MELT INCLUSIONS

The crystallization of the late-stage F-enriched Kymi topaz granite, in southeastern Finland, has been investigated by remelting and analyzing crystallized melt inclusions in quartz and topaz grains from porphyritic and equigranular topaz granites and marginal pegmatite. The conditions of rehomogenization were 300 MPa and 700°C for the granites, and 100 MPa and 700°C for pegmatite, as well as 1 atm and 900°C for all phases of the stock. After rehomogenization, quartz from the granites and pegmatite contains four types of melt inclusion: two coexisting types of melt inclusion varying in vapor content, extremely high-F melt inclusions and melt inclusions with very low F contents. The compositions of the melt inclusions show that the topaz granites and the stockscheider pegmatite of the Kymi stock crystallized from very F- and H2O-rich melts, and that Li and B were present in the granitic melt at high concentrations. They also confirm that the equigranular granite and pegmatite crystallized from a more evolved melt than the porphyritic granite. The melt inclusions in quartz and topaz grains from the granites and pegmatite contain indirect indications of melt separation and formation of a minor peralkaline melt fraction in addition to the prevailing peraluminous melt at the final stages of the crystallization. Combined petrological and field observations as well as melt-inclusion studies of the Kymi stock indicate that the zoned structure of the stock is a result of the migration of a highly evolved residual melt from the deeper parts of the magma chamber to the carapace of the chamber. This migration may be related to an opening of the contact between the surrounding rapakivi granite bedrock and residual crystal mush of the porphyritic granite, and to a migration of the residual melt into the carapace along contacts and fractures, or to a migration of highly evolved interstitial magma and fluid through the crystallizing porphyritic granite and along the walls of the magma chamber to the carapace.

Peralkaline silicic melts of island arcs, active continental margins, and intraplate continental settings: Evidence from the investigation of melt inclusions in minerals and quenched glasses of rocks

Based on the analysis of data on the composition of melt inclusions in minerals and quenched glasses of igneous rocks, we considered the problems of the formation of peralkaline silicic magmas (i.e., whose agpaitic index, the molar ratio AI = (Na2O + K2O)/Al2O3, is higher than one). The mean compositions of peralkaline silicic melts are reported for island arcs and active continental margins and compared with the compositions of melts from other settings, primarily, intraplate continental areas. Peralkaline silicic rocks are rather common in the latter. Such rocks are rare in island arcs and active continental margins, but agpaitic melts were observed in inclusions in phenocrysts of plagioclase, quartz, pyroxene, and other minerals. Plagioclase fractionation from an alkali-rich melt with AI < 1 is considered as a possible mechanism for the formation of peralkaline silicic melts (Bowen’s plagioclase effect). However, the analysis of available experimental data on plagioclase-melt equilibria showed that natural peralkaline melts are almost never in equilibrium with plagioclase. For the same reason, the melting of the majority of crustal rocks, which usually contain plagioclase, does not produce peralkaline melts. The existence of peralkaline silicic melt inclusions in plagioclase phenocrysts suggests that plagioclase can crystallize from peralkaline melts, and the plagioclase effect may play a certain role. Another mechanism for the formation of peralkaline silicic magmas is the melting of alkali-rich basic and intermediate rocks, including the spilitized varieties of subalkali basalts.

The role of fractional crystallization and late-stage peralkaline melt segregation in the mineralogical evolution of Cenozoic nephelinites/phonolites from Saghro (SE Morocco)

The Saghro Cenozoic lavas form a bimodal suite of nephelinites (with carbonatite xenoliths) and phonolites emplaced in the Anti-Atlas belt of Morocco. Despite the paucity of samples with intermediate composition between the two main types of lava (only one phonotephrite flow is reported in this area), whole-rock major element modelling shows that the two main lithologies can be linked by fractional crystallization. The most primitive modelled cumulates are calcite-bearing olivine clinopyroxenites, whereas the final stages of differentiation are characterized by the formation of nepheline-syenite cumulates. This evolution trend is classically observed in plutonic alkaline massifs associated with carbonatites. Late-stage evolution is responsible for the crystallization of hainite- and delhayelite-bearing microdomains, for the transformation of aegirine-augite into aegirine (or augite into aegirine-augite), and for the crystallization of lorenzenite and a eudialyte-group mineral as replacement products of titanite. These phases were probably formed, either by crystallization from late residual peralkaline melts, or by reaction of pre-existing minerals with such melt, or hydrothermal peralkaline fluid.

The extreme diversity of uranium deposits

Most available classifications of uranium deposits are based on the characteristics of the host rocks or on the morphology of the ore deposits. The aim of the present paper is to propose the basis for a genetic classification of these deposits. After a short introduction on the geochemical behavior of uranium in fluids and silicate melts and on the main uranium fractionation mechanisms operating in uranium-rich peraluminous, metaluminous, and peralkaline melts, the most recent metallogenic models of the main types of uranium deposits are shortly reviewed.

H2O diffusion in peralkaline to peraluminous rhyolitic melts

The diffusion of water in a peralkaline and a peraluminous rhyolitic melt was investigated at temperatures of 714–1,493 K and pressures of 100 and 500 MPa. At temperatures below 923 K dehydration experiments were performed on glasses containing about 2 wt% H2O t in cold seal pressure vessels. At high temperatures diffusion couples of water-poor (<0.5 wt% H2O t ) and water-rich (~2 wt% H2O t ) melts were run in an internally heated gas pressure vessel. Argon was the pressure medium in both cases. Concentration profiles of hydrous species (OH groups and H2O molecules) were measured along the diffusion direction using near-infrared (NIR) microspectroscopy. The bulk water diffusivity ( $$D_{{{\text{H}}_{2} {\text{O}}_{t} }}^{{}}$$ ) was derived from profiles of total water ( $$C_{{{\text{H}}_{2} {\text{O}}_{t} }}^{{}}$$ ) using a modified Boltzmann-Matano method as well as using fittings assuming a functional relationship between $$D_{{{\text{H}}_{2} {\text{O}}_{t} }}^{{}}$$ and $$C_{{{\text{H}}_{2} {\text{O}}_{t} }}^{{}}.$$ Both methods consistently indicate that $$D_{{{\text{H}}_{2} {\text{O}}_{t} }}^{{}}$$ is proportional to $$C_{{{\text{H}}_{2} {\text{O}}_{t} }}^{{}}$$ in this range of water contents for both bulk compositions, in agreement with previous work on metaluminous rhyolite. The water diffusivity in the peraluminous melts agrees very well with data for metaluminous rhyolites implying that an excess of Al2O3 with respect to alkalis does not affect water diffusion. On the other hand, water diffusion is faster by roughly a factor of two in the peralkaline melt compared to the metaluminous melt. The following expression for the water diffusivity in the peralkaline rhyolite as a function of temperature and pressure was obtained by least-squares fitting: $$ \log D_{{{\text{H}}_{2} {\text{O}}_{t} }}^{{1\,{\text{wt}}\% }} = ( - 7.09 \pm 0.15) - \frac{(4,\!788 \pm 166) + (0.56 \pm 0.21) \times P}{T} $$ where $$ D_{{{\text{H}}_{2} {\text{O}}_{t} }}^{{1\,{\text{wt}}\% }} $$ is the water diffusivity at 1 wt% H2O t in m2/s, T is the temperature in K and P is the pressure in MPa. The above equation reproduces the experimental data (14 runs in total) with a standard fit error of 0.15 log units. It can be employed to model degassing of peralkaline melts at water contents up to 2 wt%.

The miarolitic pegmatites from the Königshain: a contribution to understanding the genesis of pegmatites

In this paper, we show that the crystallization of miarolitic pegmatites at Konigshain started at about 700°C, in melts containing up to 30 mass% water. Such high water concentration at low pressures (1–3 kbar) is only possible if the melts are peralkaline. Such peralkaline melts are highly corrosive, and reacted with the wall rock—here the granite host—forming the graphic granite zone, in part via a magmatic–metasomatic reaction. With cooling, the water concentration in some melt fractions increased up to 50 mass% H2O. The melt-dominated system ends below 600°C and passes into a fluid-dominated system, the beginning of which is characterized by strong pressure fluctuations, caused by the change of OH and CO3 2− in the melt, to molecular water and CO2. We note two generations of smoky quartz, one crystallized above the β–α-transition of quartz (≈573°C), and one below, both of which contain melt inclusions. This indicates that some melt fraction remains during at least the higher-temperature portion of the growth of minerals into the miarolitic cavity, contradicting the view that minerals growing into a pegmatite chamber only do so from aqueous fluids. We show that the Konigshain miarolitic pegmatites are part of the broad spectrum of pegmatite types, and the processes active at Konigshain are representative of processes found in most granitic pegmatites, and are thus instructive in the understanding of pegmatite formation in general, and constraining the composition and characteristics of pegmatite-forming melts.

Experimental study of loparite formation in complex fluid-magmatic systems

Loparite is a mineral-concentrator of Nb, Ti, and REE. Its general formula is (Ce, Na, Ca) 2 (Ti, Nb) 2 O 6 [1]. Loparite is the major mineral of the REE‐Nb ore in the Lovozero peralkaline pluton in the Kola Peninsula. The ore mineralization is largely related to units of urtite, malignite, and less frequent lujavrite of the differentiated complex. The deposit has a long history of its geological investigation, including experimental works. The experiments on the loparite‐nepheline [2] and the lujavrite melt‐loparite [4] systems have established a wide field of loparite crystallization from the peralkaline melt. However, the studies were performed without the involvement of volatile components, although the fluid regime often is an important factor of magmatic evolution and the related ore formation [5‐7]. Only few experiments were carried out at T = 850 and 750 ° C and P = 1 kbar of water pressure [4]. Our experimental investigations were conducted with the aim of approaching natural conditions of the evolution of magmatism and ore formation closely related to fluid impact on magmatic systems. Therefore, crystallization of loparite was studied in complex silicate‐salt systems [8, 9] that contain sodium carbonate, chloride, fluoride, or sulfate. The runs were performed on a high-pressure apparatus with external heating and cold seal ( T = 400 and 450 ° C, P = 1 kbar, during 30 days; T = 840 ° C, P = 1.5 kbar, during 5‐7 days) and in high gas pressure vessel ( T = 1200 ° C, P = 2 kbar, during 6 h or 1 day) with subsequent quenching. The charge consisted of an eutectic mixture of albite and nepheline gels with the addition of oxides of Nb, La, Ce, and Y; carbonates of Ca and Sr; and excess of titanium oxide, which was commonly placed on the ampule bottom. The respective salt or a combination of salts was also added. The runs were performed in the

Solubility of tin in (Cl, F)-bearing aqueous fluids at 700 °C, 140 MPa: A LA-ICP-MS study on synthetic fluid inclusions

Synthetic fluid inclusions in quartz were grown from cassiterite-saturated fluids in cold-seal pressure vessels at 700 ° C / 140 MPa / f O 2 ∼ NNO and subsequently analyzed by laser ablation-ICP-MS. Most inclusions were synthesized using a new technique that allows entrapment of fluid that had no immediate contact to the capsule walls, such that potential disequilibrium effects due to alloying could be avoided. Measured Sn solubilities increase with increasing ligand concentration in the fluid, ranging from 100 to 800 ppm in NaCl-bearing fluids (5–35 wt% NaCl), from 70 to 2000 ppm in HF-bearing fluids (0.5–3.2 m HF), and from 0.8 to 11 wt% in HCl-bearing fluids (0.5–4.4 m HCl). Two runs performed with the in-situ cracking method after 1 week of pre-equilibration demonstrate that the speed of hydrogen diffusion through the capsule wall relative to that of fluid inclusion formation is a critical factor in f O 2 -dependent solubility studies. Graphical evaluation of the solubility data suggests that Sn may have been dissolved as Sn(OH)Cl in the NaCl-bearing fluids, as Sn(OH)Cl and SnCl 2 in the HCl-bearing fluids, and as SnF 2 in the HF-bearing fluids. Experiments with NaF-bearing fluids produced an additional melt phase with an approximate composition of 53 wt% SiO 2, 25 wt% H 2O, 14 wt% NaF and 8 wt% SnO, which caused the composition of the coexisting fluid to be buffered at 0.5 wt% NaF and 150 ppm Sn. Fluorine-rich, peralkaline melts may therefore serve as important transport media for Sn in the final crystallization stages of tin granites. Based on the available cassiterite-solubility data in fluids and melts, D Sn fluid/melt in natural granite systems is estimated to be in the order of 0.1–4 (depending on their aluminosity), suggesting that Sn is not easily mobilized by magmatic-hydrothermal fluids. This interpretation is in accordance with the high degrees of Sn-enrichment commonly observed in highly fractionated melt inclusions. D Sn fluid/melt is primarily controlled by the HCl concentration in the fluid, which in turn is a function of the aluminum saturation index of the magma. Compared to HCl, the effect of fluorine on D Sn fluid/melt is subordinate.

Fluorapatite-monazite relationships in granulite-facies metapelites, Schwarzwald, southwest Germany

Abstract Fluorapatite grains with monazite inclusions and/or rim grains are described in two of four samples from a set of granulite-facies metapelites collected from the Variscan Schwarzwald, southern Germany. Fluorapatite in all four samples appears to have experienced some dissolution in the partial granitic melt formed during granulite-facies metamorphism. Monazite inclusions and rim grains are highly deficient in Th and are presumed to have formed from fluorapatite in association with partial melting during granulite-facies metamorphism. Monazite inclusions range from very small (&lt;1 μm) and very numerous to small (1–2 μm), sometimes elongated, and less numerous; both types are evenly distributed throughout the fluorapatite grain interior. Monazite rim grains tend to be 1–10 μm. The formation of monazite inclusions is proposed to be due to dissolution-reprecipitation of the fluorapatite by the aqueous fluids inherent in the granitic melt. We propose that an increase in inclusion size coupled with a decrease in inclusion number is due to Ostwald ripening (interfacial energy reduction), which is greatly facilitated by the presence of an interconnected, fluid-filled porosity in the metasomatized fluorapatite. We further propose that monazite rim grains formed principally during partial dissolution of the fluorapatite in the granitic melt and to a lesser extent by partial dissolutionreprecipitation of the fluorapatite grain rim area allowing for the partial removal of (Y+REE). We conclude that fluorapatite, with monazite inclusions and rim grains, experienced partial dissolution in a H2O-rich peraluminous granitic melt compared to fluorapatite with monazite rim grains and no inclusions which reacted with a similar, relatively less H2O-rich melt. In contrast, monazite-free fluorapatite experienced partial dissolution in a comparatively H2O-poor, subaluminous, possibly peralkaline melt.

The combined effects of chlorine and fluorine on the viscosity of aluminosilicate melts

The effect of fluorine and fluorine + chlorine on melt viscosities in the system Na 2O–Fe 2O 3–Al 2O 3–SiO 2 has been investigated. Shear viscosities of melts ranging in composition from peraluminous [(Na 2O + FeO) < (Al 2O 3 + Fe 2O 3)] to peralkaline [(Na 2O + FeO) > (Al 2O 3 + Fe 2O 3)] were determined over a temperature range 560–890 °C at room pressure in a nitrogen atmosphere. Viscosities were determined using the micropenetration technique in the range of 10 8.8 to 10 12.0 Pa s. The compositions are based on addition of FeF 3 and FeCl 3 to aluminosilicate melts with a fixed amount of SiO 2 (67 mol%). Although there was a significant loss of F and Cl during glass syntheses, none occurred during the viscometry experiments. The presence of fluorine causes a decrease in the viscosity of all melts investigated. This is in agreement with the structural model that two fluorines replace one oxygen; resulting in a depolymerisation of the melt and thus a decrease in viscosity. The presence of both chlorine and fluorine results in a slight increase in the viscosity of peraluminous melts and a decrease in viscosity of peralkaline melts. The variation in viscosity produced by the addition of both fluorine and chlorine is the opposite to that observed in the same composition melts, with the addition of chlorine alone (Zimova M. and Webb S.L. (2006) The effect of chlorine on the viscosity of Na 2O–Fe 2O 3–Al 2O 3–SiO 2 melts. Am. Mineral. 91, 344–352). This suggests that the structural interaction of chlorine and fluorine is not linear and the rheology of magmas containing both volatiles is more complex than previously assumed.

The transition from peraluminous to peralkaline granitic melts: Evidence from melt inclusions and accessory minerals

Fractional crystallization of peraluminous F- and H 2O-rich granite magmas progressively enriches the remaining melt with volatiles. We show that, at saturation, the melt may separate into two immiscible conjugate melt fractions, one of the fractions shows increasing peraluminosity and the other increasing peralkalinity. These melt fractions also fractionate the incompatible elements to significantly different degrees. Coexisting melt fractions have differing chemical and physical properties and, due to their high density and viscosity contrasts, they will tend to separate readily from each other. Once separated, each melt fraction evolves independently in response to changing T/P/X conditions and further immiscibility events may occur, each generating its own conjugate pair of melt fractions. The strongly peralkaline melt fractions in particular are very reactive and commonly react until equilibrium is attained. Consequently, the peralkaline melt fraction is commonly preserved only in the isolated melt and mineral inclusions. We demonstrate that the differences between melt fractions that can be seen most clearly in differing melt inclusion compositions are also visible in the composition of the resulting ore-forming and accessory minerals, and are visible on scales from a few micrometers to hundreds of meters.

The behavior of trace elements during the chemical evolution of the H2O-, B-, and F-rich granite–pegmatite–hydrothermal system at Ehrenfriedersdorf, Germany: a SXRF study of melt and fluid inclusions

Detailed melt and fluid inclusion studies in quartz hosts from the Variscan Ehrenfriedersdorf complex revealed that ongoing fractional crystallization of the highly evolved H2O-, B-, and F-rich granite magma produced a pegmatite melt, which started to separate into two immiscible phases at about 720°C, 100 MPa. With cooling and further chemical evolution, the immiscibilty field expanded. Two conjugate melts, a peraluminous one and a peralkaline one, coexisted down to temperatures of about 490°C. Additionally, high-salinity brine exsolved throughout the pegmatitic stage, along with low-density vapor. Towards lower temperatures, a hydrothermal system gradually developed. Boiling processes occurred between 450 and 400°C, increasing the salinities of hydrothermal fluids at this stage. Below, the late hydrothermal stage is dominated by low-salinity fluids. Using a combination of synchrotron radiation-induced X-ray fluorescence analysis and Raman spectroscopy, the concentration of trace elements (Mn, Fe, Zn, As, Sb, Rb, Cs, Sr, Zr, Nb, Ta, Ag, Sn, Ta, W, rare earth elements (REE), and Cu) was determined in 52 melt and 8 fluid inclusions that are representative of distinct stages from 720°C down to 380°C. Homogenization temperatures and water contents of both melt and fluid inclusions are used to estimate trapping temperatures, thus revealing the evolutionary stage during the process. Trace elements are partitioned in different proportions between the two pegmatite melts, high-salinity brines and exsolving vapors. Concentrations are strongly shifted by co ncomitant crystallization and precipitation of ore-forming minerals. For example, pegmatite melts at the initial stage (700°C) have about 1,600 ppm of Sn. Concentrations in both melts decrease towards lower temperatures due to the crystallization of cassiterite between 650 and 550°C. Tin is preferentially fractionated into the peralkaline melt by a factor of 2–3. While the last pegmatite melts are low in Sn (64 ppm at 500°C), early hydrothermal fluids become again enriched with about 800 ppm of Sn at the boiling stage. A sudden drop in late hydrothermal fluids (23 ppm of Sn at 370°C) results from precipitation of another cassiterite generation between 400 and 370°C. Zinc concentrations in peraluminous melts are low (some tens of parts per million) and are not correlated with temperature. In coexisting peralkaline melts and high-T brines, they are higher by a factor of 2–3. Zinc continuously increases in hydrothermal fluids (3,000 ppm at 400°C), where the precipitation of sphalerite starts. The main removal of Zn from the fluid system occurs at lower temperatures. Similarly, melt and fluid inclusion concentrations of many other trace elements directly reflect the crystallization and precipitation history of minerals at distinctive temperatures or temperature windows.

Redox state of iron in peralkaline rhyolitic glass/melt: X-ray absorption micro-spectroscopy experiments at high temperature

Assessing the ferric–ferrous ratio in magmas prior to eruption remains a challenging task. X-ray absorption near-edge structure (μXANES) spectra were collected at the iron K-edge in water-rich peralkaline silicic melt/glass inclusions trapped in quartz. These experiments were carried out between 800 and 20 °C. The chemical environment of iron was also determined in the naturally quenched samples (glass inclusions and matrix glass) and in the peralkaline rhyolitic reference glasses, with variable [Fe 3+ / ∑Fe] ratios. In the reference glasses, both the intensity of the pre-peaks (Fe 2+, Fe 3+) and site geometry of iron change as the oxidation state increases. Fourfold-coordinated Fe 3+ prevails in highly oxidised peralkaline silicic glasses, using alkalis for charge balance. The position of the pre-edge centroid of the 1s–3d transition correlates with the Fe 3+ / ΣFe ratios that allowed calibration of the redox state of iron of our natural samples. At high temperatures, Fe 2+ dominates in the pre-edge structure of melt inclusions. Upon cooling down to 20 °C, the intensity of the Fe 3+ peak increases, the centroid position of the pre-edge features shifts by nearly 0.5 eV and the main edge moves slightly towards higher energies. The slower the cooling rate, the higher the ferric iron contribution. Iterative μXANES experiments performed on the same samples show that the process is reversible. However, this apparent oxidation of iron upon cooling is an artefact of changes in Fe coordination. It implies that the [Fe 3+ / ΣFe] ratio of glassy samples, measured at 20 °C, may be overestimated by a factor > 1.7, and that this ratio cannot be reliably retrieved by probing naturally cooled glass inclusions, and most silicate glasses. High temperature μXANES experiments led first to an assessment of the ferric–ferrous ratio in the water-rich peralkaline melt in pre-eruptive magmatic conditions and second to the determination of the corresponding oxygen fugacity at 740 °C.

Experimental and Thermodynamic Constraints on the Sulphur Yield of Peralkaline and Metaluminous Silicic Flood Eruptions

Many basaltic flood provinces are characterized by the existence of voluminous amounts of silicic magmas, yet the role of the silicic component in sulphur emissions associated with trap activity remains poorly known. We have performed experiments and theoretical calculations to address this issue. The melt sulphur content and fluid/melt partitioning at saturation with either sulphide or sulphate or both have been experimentally determined in three peralkaline rhyolites, which are a major component of some flood provinces. Experiments were performed at 150MPa, 800–900 � C, fO2 in the range NNO – 2 to NNO þ 3 and under water-rich conditions. The sulphur content is strongly dependent on the peralkalinity of the melt, in addition to fO2, and reaches 1000ppm at NNO þ 1 in the most strongly peralkaline composition at 800 � C. At all values of fO2, peralkaline melts can carry 5–20 times more sulphur than their metaluminous equivalents. Mildly peralkaline compositions show little variation in fluid/melt sulphur partitioning with changing fO2 (DS � 270). In the most peralkaline melt, DS rises sharply at fO2 > NNO þ 1 to values of >500. The partition coefficient increases steadily for Sbulk between 1 and 6wt % but remains about constant for Sbulk between 0� 5 and 1wt %. At bulk sulphur contents lower than 4wt %, a temperature increase from 800 to 900 � C decreases DS by � 10%. These results, along with (1) thermodynamic calculations on the behaviour of sulphur during the crystallization of basalt and partial melting of the crust and (2) recent experimental constraints on sulphur solubility in metaluminous rhyolites, show that basalt fractionation can produce rhyolitic magmas having much more sulphur than rhyolites derived from crustal anatexis. In particular, hot and dry metaluminous silicic magmas produced by melting of dehydrated lower crust are virtually devoid of sulphur. In contrast, peralkaline rhyolites formed by crystal fractionation of alkali basalt can concentrate up to 90% of the original sulphur content of the parental magmas, especially when the basalt is CO2-rich. On this basis, we estimate the amounts of sulphur potentially released to the atmosphere by the silicic component of flood eruptive sequences. The peralkaline Ethiopian and Deccan rhyolites could have produced � 10 17 and � 10 18 g of S, respectively, which are comparable amounts to published estimates for the basaltic activity of each province. In contrast, despite similar erupted volumes, the metaluminous Parana´–Etendeka silicic eruptives could have injected only 4� 6 · 10 15 g of S in the atmosphere. Peralkaline flood sequences may thus have greater environmental effects than those of metaluminous affinity, in agreement with evidence available from mass extinctions and oceanic anoxic events.

Solution mechanisms of H 2O in depolymerized peralkaline melts

Solubility mechanisms of water in depolymerized silicate melts quenched from high temperature (1000°–1300°C) at high pressure (0.8–2.0 GPa) have been examined in peralkaline melts in the system Na 2O-SiO 2-H 2O with Raman and NMR spectroscopy. The Na/Si ratio of the melts ranged from 0.25 to 1. Water contents were varied from ∼3 mol% and ∼40 mol% (based on O = 1). Solution of water results in melt depolymerization where the rate of depolymerization with water content, ∂(NBO/Si)/∂X H2O, decreases with increasing total water content. At low water contents, the influence of H 2O on the melt structure resembles that of adding alkali oxide. In water-rich melts, alkali oxides are more efficient melt depolymerizers than water. In highly polymerized melts, Si-OH bonds are formed by water reacting with bridging oxygen in Q 4-species to form Q 3 and Q 2 species. In less polymerized melts, Si-OH bonds are formed when bridging oxygen in Q 3-species react with water to form Q 2-species. In addition, the presence of Na-OH complexes is inferred. Their importance appears to increase with Na/Si. This apparent increase in importance of Na-OH complexes with increasing Na/Si (which causes increasing degree of depolymerization of the anhydrous silicate melt) suggests that water is a less efficient depolymerizer of silicate melts, the more depolymerized the melt. This conclusion is consistent with recently published 1H and 29Si MAS NMR and 1H- 29Si cross polarization NMR data.

Fluorite solubility in hydrous haplogranitic melts at 100 MPa

Fluorite is the most common fluoride mineral in magmatic silicic systems and its crystallization can moderate or buffer fluorine concentrations in these settings. We have experimentally determined fluorite solubility and speciation mechanisms in haplogranitic melts at 800–950 °C, 100 MPa and aqueous-fluid saturation. The starting haplogranite compositions: peraluminous (alumina saturation index, ASI = 1.2), subaluminous (ASI = 1.0) and peralkaline (ASI = 0.8) were variably doped with CaO or F 2O −1 in the form of stoichiometric mineral or glass mixtures. The solubility of fluorite along the fluorite–hydrous haplogranite binaries is low: 1.054 ± 0.085 wt.% CaF 2 (peralkaline), 0.822 ± 0.076 wt.% (subaluminous) and 1.92 ± 0.15 wt.% (peraluminous) at 800 °C, 100 MPa and 10 wt.% H 2O, and exhibits a minimum at ASI ∼ 1. Fluorite saturation isotherms are strongly hyperbolic in the CaO–F 2O −1 space, suggesting that fluorite saturation is controlled by the activity product of CaO and F 2O −1, i.e., these components are partially decoupled in the melt structure. The form of fluorite liquidus isotherms implies distinct roles of fluorite crystallization: in Ca-dominant systems, fluorite crystallization is controlled by the fluorine concentration in the melt only and remains nearly independent of calcium contents; in F-rich systems, the crystallization of fluorite is determined by CaO contents and it does not buffer fluorine concentration in the melt. The apparent equilibrium constant, K, for the equilibrium CaO + cF 2O −1 = CaF 2 (+ associates) is log K= − (2.449 ± 0.085)·Al 2O 3 exc + (4.902 ± 0.066); the reaction-stoichiometry parameter varies as follows: c= − (0.92 ± 0.11)·Al 2O 3 exc + (1.042 ± 0.084) at 800 °C, 100 MPa and fluid saturation where Al 2O 3 exc are molar percent alumina in excess over alkali oxides. The reaction stoichiometry, c, changes at subaluminous composition: in peralkaline melts, competition of other network modifiers for excess fluorine anions leads to the preferential alkali–F short-range order, whereas in peraluminous compositions, excess alumina associates with calcium cations to form calcioaluminate tetrahedra. The temperature dependence of fluorite solubility is described by the binary symmetric Margules parameter, W = 36.0 ± 1.4 kJ (peralkaline), 39.7 ± 0.5 kJ (subaluminous) and 32.8 ± 0.7 kJ (peraluminous). The strong positive deviations from ideal mixing imply the occurrence of CaF 2–granite liquid–liquid immiscibility at temperatures above 1258 °C, which is consistent with previous experimental data. These experimental results suggest very low solubilities of fluorite in Ca-rich melts, consistent with the lack of fluorine enrichment in peralkaline rhyolites and calc-alkaline batholiths. On the other hand, high CaO concentrations necessary to crystallize fluorite in F-rich peraluminous melts are not observed in nature and thus magmatic crystallization of fluorite in topaz-bearing silicic suites is suppressed. A procedure for calculating fluorite solubility and the liquidus isotherms for a whole-rock composition and temperature of interest is provided.

The effect of water on accessory phase solubility in subaluminous and peralkaline granitic melts

The solubilities of columbite, tantalite, wolframite, rutile, zircon and hafnon were determined as a function of the water contents in peralkaline and subaluminous granite melts. All experiments were conducted at 1035 °C and 2 kbar and the water contents of the melts ranged from nominally dry to approximately 6 wt.% H 2O. Accessory phase solubilities are not affected by the water content of the peralkaline melt. By contrast, solubilities are affected by the water content of the subaluminous melt, where the solubilities of all the accessory phases examined increase with the water content of the melt, up to ∼2 wt.% H 2O. At higher water contents, solubilities are nearly constant. It can be concluded that water is not an important control of accessory phase solubility, although the water content will affect diffusivities of components in the melt, thus whether or not accessory phases will be present as restite material. The solubility behaviour in the subaluminous and peralkaline melts supports previous spectroscopic studies, which have observed differences in the coordination of high field strength elements in dry vs. wet subaluminous granitic glasses, but not for peralkaline granitic glasses. Lastly, the fact that wolframite solubility increases with increasing water content in the subaluminous melt suggests that tungsten dissolved as a hexavalent species.