Natural Rhyolite Research Articles

Hydrothermal synthesis of zeolite-a from natural rhyolite for effective removal of Cr(VI) from wastewater

Hydrothermal synthesis of zeolite-a from natural rhyolite for effective removal of Cr(VI) from wastewater

Multiple Episodes of Rock-Melt Reaction at the Slab-Mantle Interface: Formation of High Silica Primary Magmas in Intermediate to Hot Subduction Zones

AbstractSiliceous slab-derived partial melts infiltrate the sub-arc mantle and cause rock-melt reactions, which govern the formation of diverse primary arc magmas and lithological heterogeneities. The effect of bulk water content, composition of reactants, and nature of melt infiltration (porous versus channelized) on the rock-melt reactions at sub-arc conditions have been investigated by previous studies. However, the effect of multiple episodes of rock-melt reactions in such scenarios has not been investigated before. Here, we explore mantle wedge modifications through serial additions of hydrous-silicic slab partial melts and whether such a process may ultimately explain the origin of high-Mg# andesites found in arcs worldwide. A series of piston-cylinder experiments simulate a serial addition of silicic slab melts in up to three stages (I through III) at 3 GPa and 800–1050°C, using rock-melt proportions of 75–25 and 50–50. A synthetic KLB-1 and a natural rhyolite (JR-1) represented the mantle and the slab components, respectively. Right from the first rock-melt interaction, the peridotite mantle transforms into olivine-free mica-rich pyroxenites ± amphibole ± quartz/coesite in equilibrium with rhyolitic-hydrous melts (72–80 wt% SiO2 and 40–90 Mg#). The formation of olivine-free pyroxenite seems to be controlled by complex functions of T, P, rock-melt ratio, wedge composition, and silica activity of the slab-melt. Remarkably, the pyroxenites approach a melt-buffered state with progressive stages of rock-melt reactions, where those rhyolitic melts inherit and preserve the major (alkalis, Fe, Mg, Ca) and trace element slab-signature. Our results demonstrate that lithological heterogeneities such as pyroxenites formed as products of rock-melt reactions in the sub-arc mantle may function as melt ‘enablers,’ implying that they may act as pathways that enable the infiltrating melt to retain their slab signature without undergoing modification. Moreover, the density contrast between the products of rock-melt reaction (melts and residues) and the average mantle wedge (~150 to 400 kg/m3) may help forming instabilities and diapiric rise of the slab components into the mantle wedge. However, the fate of the primitive slab-melts seems to be associated with the length of the pathway of mantle interaction which explains the evident wide magma spectrum as well as their degree of slab garnet-signature dilution. This work and the existence of high-Mg# Mexican-trondhjemites indicates that almost pristine slab-melts can make their way up to crustal levels and contribute to the arc magma diversity.

Rhyolite as a Naturally Sustainable Thermoluminescence Material for Dose Assessment Applications

Thermoluminescence characteristics of natural rhyolite have been studied. Dose response at a wide dose range of 0.5–2000 Gy has been determined. Minimum detectable dose and thermal fading rate are evaluated. Glow curve deconvolution is conducted after determining the best read-out conditions. The repeated initial rise (RIR) method is used to detect the overlapping peaks, and a glow curve deconvolution procedure is used to extract the thermoluminescence parameters of rhyolite. According to the findings, rhyolite glow curves show five interfering peaks corresponding to five electron trap levels at 142, 176, 221, 298, and 355 °C, respectively, at a heating rate of 3 °C/s. The obtained kinetic order for the deconvoluted peaks showed mixed-order kinetic. The reported results might be useful to introduce rhyolite as a natural sustainable material for radiation dosimetry applications.

The roles of microlites and phenocrysts during degassing of silicic magma

Silicic magmas span a wide range of eruptive styles between explosive and effusive, and transitions between these styles are commonplace. Yet the triggers of switches in eruptive style remain poorly understood. Eruptions are mostly driven by degassing of magmatic water and their eruption style - effusive or explosive - is likely governed by the efficiency of outgassing as well as magma ascent rate. Microlites and phenocrysts are often purported to promote heterogeneous bubble nucleation and outgassing, both key variables in the degassing dynamics that become crucial in controlling the eruptive style. Here, in order to shed light on the role of nature, size and abundance of crystals on degassing of silicic magma, we experimentally investigate (heating-induced) vesiculation in a multiphase (microlite- and phenocryst-bearing), low-water content, and bubble-free, natural rhyolite. The experiments were conducted at magmatic temperature (∼900-1100°C) and atmospheric pressure in an optical dilatometer. Our results indicate that microlites exert a large influence on bubble nucleation while the effect of phenocrysts is subordinate. Amongst the microlite phases, bubbles nucleate more easily on Fe-Ti oxides than other mineral phases. Bubble coalescence and connectivity are, in contrast, enhanced by the phenocrysts, more than microlites, in low-crystallinity magma. Comparing the bubble textures of the post-experimental samples with those produced in a phenocryst-free and microlite-poor silicic magma, we observe that the phenocryst- and microlite-bearing magma exhibits significantly more bubble coalescence and connectivity, as well as higher bubble number densities. These findings help to constrain the roles that pre- and syn-eruptive crystalline phases may play in degassing processes during ascent of silicic magma.

Experimental demonstration of continuous bubble nucleation in rhyolite

Bubble nucleation is the first step in magma degassing. The kinetics of bubble nucleation during eruptive magma ascent determines the rate of magma degassing, which in turn, has a significant control on the explosivity of eruptions. Here, we examine if bubble nucleation in rhyolite can be continuous over the inferred timescale of magma travel time from reservoir to the surface in explosive silicic eruptions. We performed homogeneous bubble nucleation experiments using cylinders cored from natural rhyolite. Samples were hydrated at an initial pressure and temperature, then decompressed to a final pressure, and held at that pressure for various amounts of time before being quenched. Within a given suite samples had the same temperature as well as initial and final pressures, but were held for different annealing times. The resulting bubble number density of samples within each suite increases by up to 2 orders of magnitude as annealing time increases, indicating that nucleation continued during annealing. We estimate the nucleation rate and duration from measured bubble number densities and show that nucleation in the majority of samples was still ongoing at the time of quenching. Our findings confirm theoretical predictions that bubble nucleation can be a continuous process during eruptive magma ascent in explosive eruptions.

Confirmation of slow Ti diffusion in quartz by diffusion couple experiments and evidence from natural samples

Titanium diffusion profiles in natural quartz crystals have become an increasingly popular tool to reconstruct the time scales of various magmatic, metamorphic, and hydrothermal processes. However, the original calibration of Ti diffusion rates in quartz has recently been challenged, and diffusivities were found to be about three orders of magnitude lower. We performed annealing experiments on crystal-crystal diffusion couples consisting of Ti-free synthetic quartz seeds over which Ti-rich quartz (100–3000 μg/g Ti) was grown hydrothermally. The annealing experiments were performed at 1000–1600 °C and 0.1 MPa to 2.0 GPa, and they lasted for 3–84 days. The resulting diffusion profiles were mapped by cathodoluminescence (CL), transmission electron microscope–energy-dispersive X-ray spectroscopy (TEM-EDXS), and, for the first time, by helium ion microscope–secondary ion mass spectrometry (HIM-SIMS). Obtained diffusion coefficients range from values similar to the lower range in previous research to values up to two orders of magnitude lower. In addition, inversely zoned quartz and sanidine phenocrysts in a natural rhyolite were studied. Comparison of the diffusion profiles suggests that at ~735 °C, the Ti diffusivity in quartz is ~1.5 and 3.0 orders of magnitude lower than that of Ba and Sr, respectively, in sanidine. The combined evidence confirms that Ti diffusion in quartz is very slow, potentially even slower than proposed earlier. Consequently, previous time scales derived from Ti diffusion profiles in quartz are likely orders of magnitude too short, and further experiments are necessary to fully clarify the issue.

Crystallization and partial melting of rhyolite and felsite rocks at Krafla volcano: A comparative approach based on mineral and glass chemistry of natural and experimental products

Rhyolite and felsite cuttings were collected at Krafla volcano during the perforation of the Iceland Deep Drilling Project Well 1 (IDDP-1). The perforation was stopped at a depth of 2100 m due to intersection with a rhyolite magma that intruded the felsite host rock. Rhyolite cuttings are vitrophiric (glass ~95%, RHL) and exhibit a mineral assemblage made of plagioclase + augite + pigeonite + titanomagnetite. Felsite cuttings display evidences of partial melting, responding to variable degrees of quartz + plagioclase + alkali feldspar + augite + titanomagnetite dissolution. The interstitial glass analyzed close to (i.e., FLS1) and far from (i.e., FLS2) the reaction surface of pyroxene from felsite cuttings shows continuous changes between the two end-members. FLS1 is compositionally similar to RHL, showing Na2O + K2O + REE depletions, counterbalanced by MgO + CaO enrichments. Conversely, FLS2 exhibits opposite chemical features. REE-exchange thermobarometric calculations reveal that plagioclase and augite cores from rhyolite and felsite formed under identical conditions, along a thermal path of 940–960 °C. However, in terms of major and trace element concentrations, plagioclase and augite crystal cores are not in equilibrium with the rhyolite magma, suggesting the incorporation of these minerals directly from the host felsite. To better understand the petrogenetic relationship between rhyolite and felsite, two sets of crystallization and partial melting experiments have been carried out at P = 150 MPa and T = 700–950 °C. Rhyolite crystallization experiments (RCE) reproduce the two-pyroxene assemblage of IDDP-1 rhyolite cuttings only at T ≤ 800 °C, when the crystal content (≥19%) is higher than that observed in the natural rhyolite (~5%). Under such conditions, the RCE glass is much more differentiated (i.e., marked CaO depletion and Eu anomaly) than RHL. On the other hand, felsite partial melting (FPM) experiments show interstitial glass with a bimodal composition (i.e., FPM1 and FPM2) comparable to FLS1 (≈RHL) and to FLS2, only at T = 950 °C. This effect has been quantified by fractional crystallization and batch melting modeling, denoting that FLS1 (≈RHL) and FLS2 reflect high (≥70%) and low (≤8%) degrees of felsite partial melting, respectively. In contrast, modeling RHL by crystal fractionation requires the removal of an amount (~22%) of solid material that is inconsistent with the low crystal content of the natural IDDP-1 rhyolite. It is therefore concluded that natural rhyolite and felsite represent, respectively, the near-liquidus and sub-solidus states of a virtually identical silicic magma, either feeding aphyric to subaphiric rhyolitic eruptions, or solidifying at depth as phaneritic quartzofeldspathic rocks. Felsite lenses from the Krafla substrate may explore variable degrees of remelting and remobilization processes. The intrusion into felsite of a fresh silicic magma from depth may lead to low degrees of partial melting, whereas the persistent heat release from intense basaltic intrusive events at Krafla may be the source of high degrees of felsite partial melting and consequent rejuvenation of the previously solidified silicic magma.

Vanadium magnetite–melt oxybarometry of natural, silicic magmas: a comparison of various oxybarometers and thermometers

To test a recently developed oxybarometer for silicic magmas based on partitioning of vanadium between magnetite and silicate melt, a comprehensive oxybarometry and thermometry study on 22 natural rhyolites to dacites was conducted. Investigated samples were either vitrophyres or holocrystalline rocks in which part of the mineral and melt assemblage was preserved only as inclusions within phenocrysts. Utilized methods include vanadium magnetite–melt oxybarometry, Fe–Ti oxide thermometry and -oxybarometry, zircon saturation thermometry, and two-feldspar thermometry, with all analyses conducted by laser-ablation ICP–MS. Based on the number of analyses, the reproducibility of the results and the certainty of contemporaneity of the analyzed minerals and silicate melts the samples were grouped into three classes of reliability. In the most reliable (n = 5) and medium reliable (n = 10) samples, all fO2 values determined via vanadium magnetite–melt oxybarometry agree within 0.5 log units with the fO2 values determined via Fe–Ti oxide oxybarometry, except for two samples of the medium reliable group. In the least reliable samples (n = 7), most of which show evidence for magma mixing, calculated fO2 values agree within 0.75 log units. Comparison of three different thermometers reveals that temperatures obtained via zircon saturation thermometry agree within the limits of uncertainty with those obtained via two-feldspar thermometry in most cases, whereas temperatures obtained via Fe–Ti oxide thermometry commonly deviate by ≥50 °C due to large uncertainties associated with the Fe–Ti oxide model at T-fO2 conditions typical of most silicic magmas. Another outcome of this study is that magma mixing is a common but easily overlooked phenomenon in silicic volcanic rocks, which means that great care has to be taken in the application and interpretation of thermometers and oxybarometers.

Magma mixing induced by particle settling.

A time series of experiments at high temperature have been performed to investigate the influence of particle settling on magma mixing. A natural rhyolite glass was held above a natural basalt glass in a platinum crucible. After melting of the glasses at superliquidus temperatures, a platinum sphere was placed on the upper surface of the rhyolitic melt and sank into the experimental column (rhyolitic melt above basaltic melt). Upon falling through the rhyolitic–basaltic melt interface, the Pt sphere entrained a filament of rhyolitic melt in its further fall. The quenched products of the experiments were imaged using X-ray microCT methods. The images of our time series of experiments document the formation of a rhyolite filament as it is entrained into the underlying basalt by the falling platinum sphere. When the Pt particle reached the bottom of the crucible, the entrained rhyolitic filament started to ascend buoyantly up to the initial rhyolitic–basaltic interface. This generated a significant thickness increase of a comingled “melange” layer at the interface due to “liquid rope coiling” and piling up of the filament. As a consequence, the basalt/rhyolite interface was greatly enlarged and diffusive hybridisation greatly accelerated. Further, bubbles, originating at the interface, are observed to have risen into the overlying rhyolite dragging basalt filaments with them. Upon crossing the basalt/rhyolite interface, the bubbles have non-spherical shapes as they adapt to the differing surface tensions of basaltic and rhyolitic melts. Major element profiles, measured across the rhyolite filaments, exhibit asymmetrical shapes from the rhyolite into the basalt. Na and Ti reveal uphill diffusion from the rhyolite towards the interface in the filament cross sections. These results reveal the potential qualitative complexity of the mingling process between rhyolitic and basaltic magmas in the presence of sinking crystals. They imply that crystal-rich magma mingling may be expected to be accelerated with respect to crystal-poor systems. We urge the further fluid dynamic analysis of these phenomena, obtainable for the first time using detailed tomographic imaging.Electronic supplementary materialThe online version of this article (doi:10.1007/s00410-016-1305-1) contains supplementary material, which is available to authorized users.

An updated calibration of the plagioclase-liquid hygrometer-thermometer applicable to basalts through rhyolites

An updated and expanded data set that consists of 214 plagioclase-liquid equilibrium pairs from 40 experimental studies in the literature is used to recalibrate the thermodynamic model for the plagioclase-liquid hygrometer of Lange et al. (2009); the updated model is applicable to metaluminous and alkaline magmas. The model is based on the crystal-liquid exchange reaction between the anorthite (CaAl 2 Si 2 O 8 ) and albite (NaAlSi 3 O 8 ) components, and all available volumetric and calorimetric data for the pure end-member components are used in the revised model. The activities of the crystalline plagioclase components are taken from Holland and Powell (1992). Of the 214 experiments, 107 are hydrous and 107 are anhydrous. Four criteria were applied for inclusion of experiments in the final data set: (1) crystallinities 2 O fluid saturated; (3) compositional totals (including H 2 O component) of 97–101% for hydrous quenched glasses and 98.5–101 for anhydrous quenched glasses; and (4) melt viscosities ≤5.2 log 10 Pa·s. The final data set spans a wide range in liquid composition (45–80 wt% SiO 2 ; 1–10 wt% Na 2 O+K 2 O), plagioclase composition (An 17–95 ), temperature (750–1244 °C), pressure (0–350 MPa), and H 2 O content (0–8.3 wt%). The water solubility model of Zhang et al. (2007) was applied to all hydrous experiments. The standard error estimate on the hygrometer model is 0.35 wt% H 2 O, and all liquid compositions are fitted equally well. Application of the model as a thermometer recovers temperatures to within ±12°, on average. Tests of the hygrometer on anhydrous piston-cylinder experiments in the literature, not included in the regression, show that the model is accurate at all pressures where plagioclase is stable. Applications of the hygrometer are made to natural rhyolites (Bishop Tuff, Katmai, and TobaTuff) with reported H 2 O analyses in quartz-hosted melt inclusions from the literature; the results show agreement. Applications of the hygrometer/thermometer are additionally made to natural rhyolites from Iceland and Glass Mountain, California. The updated model can be downloaded either as a program in Excel format or as a MatLab script from the Data Repository.

Determining the Provenance of Obsidian in Southern Patagonia Using Optical Properties

This paper aims to evaluate the optical properties of natural rhyolite obsidians that might be useful in the identification of different sources of archaeological samples. Even though colour, degree of alteration, crystaloclast composition and texture are important properties by which sources can be identified, all of these features do not always lead to an unambiguous distinction. In this regard, the refractive index (ND) becomes a helpful tool because it is not only sensitive to the chemical composition of natural glasses but also to their different thermal histories. Obsidians from six localities of Patagonia were analysed. Their refractive index measurements (ND) only range between 1.47 and 1.49, but the ND values of some of these sources are sufficiently constrained to allow discrimination between them.

Compositional and kinetic controls on liquid immiscibility in ferrobasalt–rhyolite volcanic and plutonic series

We present major element compositions of basalts and their differentiation products for some major tholeiitic series. The dry, low-pressure liquid lines of descent are shown to approach or intersect the experimentally-defined compositional space of silicate liquid immiscibility. Ferrobasalt–rhyolite unmixing along tholeiitic trends in both volcanic and plutonic environments is supported by worldwide occurrence of immiscible globules in the mesostasis of erupted basalts, unmixed melt inclusions in cumulus phases of major layered intrusions such as Skaergaard and Sept Iles, and oxide-rich ferrogabbros closely associated with plagiogranites in the lower oceanic crust. Liquid immiscibility is promoted by low-pressure, anhydrous fractional crystallization that drives the low Al2O3, high FeO liquids into the two-liquid field. Kinetic controls can be important in the development of two-liquid separation. The undercooling that occurs at the slow cooling rates of plutonic environments promotes early development of liquid immiscibility at higher temperature. In contrast rapid cooling in erupted lavas leads to large undercoolings and liquid immiscibility develops at significantly lower temperatures. Unmixing leads to the development of a compositional gap characterized by the absence of intermediate compositions, a feature of many tholeiitic provinces. The compositions of experimental unmixed silica-rich melts coincide with those of natural rhyolites and plagiogranites with high FeOtot and low Al2O3, suggesting the potential role of large-scale separation of immiscible Si-rich liquid in the petrogenesis of late-stage residual melts. No trace of the paired ferrobasaltic melt is found in volcanic environments because of its uneruptable characteristics. Instead, Fe–Ti±P-rich gabbros are the cumulate products of immiscible Fe-rich melts in plutonic settings. The immiscibility process may be difficult to identify because both melts crystallize the same phases with the same compositions. The two liquids might form incompletely segregated emulsions so that both liquids continue to exchange as they crystallize and remain in equilibrium. Even if segregated, both melts evolve on the binodal surface and exsolve continuously with decreasing temperature. The two liquids do not differentiate independently and keep crystallizing the same phases with differentiation. Further evolution by fractional crystallization potentially drives the bulk liquid out of the two-liquid field so that very late-stage liquids could evolve into the single melt phase stability field.

Magnesium isotope fractionation in silicate melts by chemical and thermal diffusion

Two types of laboratory experiments were used to quantify magnesium isotopic fractionations associated with chemical and thermal (Soret) diffusion in silicate liquids. Chemical diffusion couples juxtaposing a molten natural basalt (SUNY MORB) and a molten natural rhyolite (Lake County Obsidian) were run in a piston cylinder apparatus and used to determine the isotopic fractionation of magnesium as it diffused from molten basalt to molten rhyolite. The thermal diffusion experiments were also run in a piston cylinder apparatus but with a sample made entirely of molten SUNY MORB displaced from the hotspot of the assembly furnace so that the sample would have a temperature difference of about 100–200 °C from one end to the other. The chemical diffusion experiments showed fractionations of 26Mg/ 24Mg by as much as 7‰, which resulted in an estimate for the mass dependence of the self-diffusion coefficients of the magnesium isotopes corresponding to D 26 Mg / D 24 Mg = ( 24 / 26 ) β with β = 0.05. The thermal diffusion experiments showed that a temperature difference of about 100 °C resulted in the MgO, CaO, and FeO components of the basalt becoming slightly enriched by about 1 wt% in the colder end while SiO 2 was enriched by several wt% in the hotter end. The temperature gradient also fractionated the magnesium isotopes. A temperature difference of about 150 °C produced an 8‰ enrichment of 26Mg/ 24Mg at the colder end relative to the hotter end. The magnesium isotopic fractionation as a function of temperature in molten basalt corresponds to 3.6 × 10 −2‰/°C/amu.

Secondary ion mass spectrometry for the determination of δ37Cl : Part II. Intercalibration of SIMS and IRMS for aluminosilicate glasses

Secondary ion mass spectrometry (SIMS) offers unparalleled capability for the in situ determination of chlorine stable isotope ratios ( δ 37Cl) in small objects, such as igneous melt inclusions. However, SIMS analyses must be carefully calibrated for instrumental mass fractionation (IMF) of 37Cl/ 35Cl, which is highly dependent on the major-element composition of the sample. For the purpose of IMF calibration, eight synthetic aluminosilicate glasses, ranging from mafic to felsic composition, were fused from natural and synthetic rock powders with an internally heated pressure vessel (IHPV). This collection of standard materials was supplemented with one natural rhyolite glass. We used alkali fusion, followed by gaseous dual inlet isotope ratio mass spectrometry (AF-IRMS) as the reference technique for δ 37Cl determination of these glasses. Major-element and volatile contents of the glasses were measured by electron microprobe (EMPA) and infrared spectroscopy (FTIR). The absolute difference in IMF between basalt and rhyolite composition glasses can exceed 10‰. However, a strong linear correlation ( R 2=0.99) exists between IMF and expressions of the form (aSiO 2−bAl 2O 3−cCaO+dFeO), where oxides, based on EMPA analysis, are expressed in weight percent. Thus, calibrations based on a small set of glass standards allow SIMS determinations of δ 37Cl over a large range of silicate glass compositions, with an overall reproducibility of single spots better than ±1.0‰ (2 σ). Moreover, δ 37Cl data can be obtained even if the sample does not exactly match the composition of an individual reference glass.

Isotope fractionation by chemical diffusion between molten basalt and rhyolite

Experimental diffusion couples were used to study chemical diffusion between molten rhyolite and basalt with special emphasis on the associated fractionation of calcium and lithium isotopes. Diffusion couples were made by juxtaposing firmly packed powders of a natural basalt (SUNY MORB) and a natural rhyolite (Lake County Obsidian) and then annealing them in a piston cylinder apparatus for times ranging from 0.1 to 15.7 h, temperatures of 1350–1450°C, and pressures of 1.2–1.3 GPa. Profiles of the major elements and many trace elements were measured on the recovered quenched glasses. The diffusivities of all elements except lithium were found to be remarkably similar, while the diffusivity of lithium was two to three orders of magnitude larger than that of any of the other elements measured. Chemical diffusion of calcium from molten basalt into rhyolite was driven by a concentration ratio of ∼18 and produced a fractionation of 44Ca from 40Ca of about 6 ‰. Because of the relatively low concentration of lithium in the natural starting materials a small amount of spodumene (LiAlSi 2O 6) was added to the basalt in order to increase the concentration difference between basalt and rhyolite, which was expected to increase the magnitude of diffusive isotopic fractionation of lithium. The concentration ratio between Li-doped basalt and natural rhyolite was ∼15 and the resulting diffusion of lithium into the rhyolite fractionated 7Li from 6Li by about 40‰. We anticipate that several other major rock-forming elements such as magnesium, iron and potassium will also exhibit similarly larger isotopic fractionation whenever they diffuse between natural melts with sufficiently large differences in the abundance of these elements.

An experimental study of bromine behaviour in water-saturated silicic melts

To assess the effect of the melt composition on bromine concentrations in magmas, we have investigated bromide solubility for water-saturated, iron-free silicic melts with variable Na+K/Al and Si/Al molar ratios (albite, haplogranite, rhyolite, and pantellerite). The experiments were performed in rapid quench cold-seal autoclaves over a range of pressure (1, 1.5, and 2 kbar) and temperature (900, 1000, and 1080 °C) with run durations from 5 to 7 days. A series of natural volcanic glasses and melt inclusions hosted in magmatic minerals were analysed together with the synthetic glasses by PIXE (proton-induced X-ray emission). The Br concentrations range from 5360 to 7850 ppm for albite, from 2800 to 3900 ppm for haplogranite, from 4300 to 5900 ppm for rhyolite, and from 9745 to 11,250 ppm for pantellerite. Br concentrations are negatively correlated with pressure in H 2O-saturated silicic melts and vary with (Na+K)/Al molar ratio with a minimum value at the ratio close to unity. Br behaves similarly to chlorine for all of these melt compositions. The bromide solubility is similar in albitic and rhyolitic melts, which implies that D f/m is nearly the same for both compositions and is applicable for natural rhyolites as suggested in our previous study (Bureau et al., 2000). This means that the volcanic Br contribution to the atmosphere may be significant. In natural obsidian samples and MI hosted in quartz, olivine, and leucite, the Br concentration varies from < 3 to 28 ppm, with the highest concentrations in pantelleritic melts. We attribute the low Br concentrations of natural melts to a low initial abundance of this halogen in the Earth mantle. However, because Br behaves as an incompatible element before water exsolution, our results imply that magmas could contain much more dissolved Br before eruption and water degassing than the few ppm usually measured in volcanic rocks. Br behaviour during magma crystallisation is controlled by its partitioning into the H 2O-rich fluid phase when this occurs. In addition, its potential high solubility in silicate melts makes it a very sensitive chemical tracer of magma contamination by seawater and Br-rich material. This infers that the investigation of Br behaviour in subduction-zone samples may help for a better understanding of volatiles cycling between the Earth reservoirs.

Shear viscosity of rhyolite-vapor emulsions at magmatic temperatures by concentric cylinder rheometry

The viscosity of natural rhyolitic melt from Lipari, Aeolian Islands and melt-bubble emulsions (30–50 vol% porosity) generated from Lipari rhyolite have been measured in a concentric cylinder rheometer at temperatures and shear rates in the range 925–1150°C and 10 −3–10 −1.2 s −1, respectively, in order to better understand the dependence of emulsion shear viscosity on temperature and shear rate in natural systems. Bubble-free melt exhibits Newtonian–Arrhenian behavior in the temperature range 950–1150°C with an activation energy of 395±30 kJ/mol; the shear viscosity is given by log η m=−8.320+20624/ T. Suspensions were prepared from natural rhyolite glass to which small amounts of Na 2SO 4 were added as a ‘foaming agent’. Reasonably homogeneous magmatic mixtures with an approximate log-normal distribution of bubbles were generated by this technique. Suspension viscosity varied from 10 6.1 to 10 8.37 Pa s and systematically correlates with temperature and porosity in the shear stress range (10 4.26–10 5.46 Pa) of the experiments. The viscosity of melt-bubble emulsions is described in terms of the relative viscosity, η r= η e/ η m where η e is the emulsion viscosity and η m is the viscosity of melt of the same composition and temperature. The dependence of relative viscosity on porosity for magmatic emulsions depends on the magnitude of the capillary number Ca≡ G/( σr b −1 η m −1), the ratio of viscous forces acting to deform bubbles to interfacial forces resisting bubble deformation. For inviscid bubbles in magmatic flows three regimes may be identified. For Ca<0.1, bubbles are nearly spherical and relative viscosity is an increasing function of porosity. For dilute systems, η r=1+ φ given by the classical result of Taylor [Proc. R. Soc. London A 138 (1932) 41–48]. For Ca in the range 0.1<Ca<10, emulsions behave as power law fluids and the relative viscosity depends on shear rate (or Ca) as well as porosity. At high Ca (Ca>10) an asymptotic regime is reached in which relative viscosity decreases with increasing porosity and is independent of Ca. Our experiments were carried out for 30<Ca<925 in order to quantify the maximal effect of bubbles in reducing the viscosity of magmatic emulsions relative to single-phase melt at identical conditions of shear rate and temperature. The viscosity of a 50 vol% emulsion is a factor of five smaller than that of melt alone. Rheometric measurements obtained in this study are useful in constraining models of magma transport and volcanic eruption mechanics relevant to transport of volatile-saturated magma in the crust and upper mantle.

The solubility of H 2O and CO 2 in rhyolitic melts in equilibrium with a mixed CO 2–H 2O fluid phase

H 2O and CO 2 solubilities in a natural rhyolite melt (Erevan Dry Fountain (EDF), Armenia) in equilibrium with H 2O–CO 2 fluids were determined at 200 and 500 MPa and at 800°C and 1100°C. The composition of the fluid phase after experiment was determined by gravimetry, except for extreme CO 2-rich fluids for which mass balance was used. Water and CO 2 contents of the glasses were measured using IR spectroscopy. At 200 MPa, the water solubility changes from a square root dependence on mole fraction of H 2O in the fluid phase ( X H 2O f) at low X H 2O f to a linear dependence above X H 2O f=0.25. Up to about 5 wt.% dissolved water in the melt (corresponding to X H 2O f≈0.5), a similar trend is observed at 500 MPa. At higher X H 2O f, however, the dependence of water solubility on X H 2O f is more pronounced than at 200 MPa. A negative temperature dependence of water solubility is observed in the whole experimental range at 200 MPa (e.g. the water solubility at X H 2O f=1 decreases from 5.97 to 5.58 wt.% when temperature rises from 800°C to 1100°C). In contrast, at 500 MPa the temperature dependence of water solubility changes from positive at high X H 2O f (e.g., increase from 9.84 to 11.04 wt.% for a temperature increase from 800 to 1100°C) to negative at low X H 2O f. An empirical model to predict water solubility in rhyolitic melts in the P– T range 75–500 MPa and 800–1100°C was derived from our new solubility data and data from Blank et al. (1993). The model reproduces the data at 75 and 200 MPa within ±2.5% relative and the data at 500 MPa within ±5% relative. The CO 2 solubility shows a non-linear dependence on X H 2O f with deviation from linearity, which increases with pressure. The maximum CO 2 solubilities (equilibrium with pure CO 2) predicted from the data trends are 0.11±0.01 and 0.28±0.03 wt.% at 1100°C and 200 Mpa and 500 MPa, respectively. The temperature dependence of CO 2 solubility is found to be almost negligible at 200 MPa and slightly positive at 500 MPa. Our solubility data at 200 MPa are reproduced slightly better by the model of Papale (1999) than by that of Holloway and Blank (1994).

Water speciation in rhyolitic melt determined by in-situ infrared spectroscopy

Near-IR spectra obtained from hydrous silicate melts, close to natural rhyolite, were acquired at both high temperature using a heating stage and simultaneously at high pressure and temperature using a hydrothermal diamond cell. The temperature dependence of the extinction coefficients for the peaks due to OH and H 2 O is negligible. In both sets of experiments, the speciation reaction H 2 O + O = 2OH is shifted to the right with increasing temperature above the glass transition but changes below this are negligible, within experimental uncertainty. For a sample containing 3.93 wt% total water, the temperature dependence of the speciation equilibrium can be described by two equations with temperature in K: ln k = -36.74/T - 4.02 for the glass phase (giving enthalpy and entropy of reaction values of ΔH = 0.31 kJ/mol and ΔS = -33.45 J/mol-K), and ln k = -3821.83/T + 1.61 for the melt phase (where ΔH = 31.77 kJ/mol and ΔS = 13.41 J/mol-K) respectively). The glass transition temperature of 670 K is defined by the intersection of these curves. Our values are in good agreement with previously published glass transition temperatures for similar compositions. Similar values for the enthalpy and entropy of reaction were obtained from all the other experiments including those at pressures up to 4.5, 5, and 10 kbars, suggesting that the speciation depends negligibly on pressure for this pressure range.

The speciation of dissolved water in rhyolitic melt

Concentrations of water molecules and hydroxyl groups have been measured in rhyolitic glasses with 0.5% to 5.0% H 2O t using infrared spectroscopy at room temperature. The glasses were cooled at ∼10 2C/s after having been held at 400 to 600C, for sufficient time for the equilibrium distribution of species to have been reached. The speciation of water in samples with greater than 2.5 wt.% total dissolved water and quenched rapidly from temperatures ≥600C were shown to reequilibrate during quench. However, samples with less than 2.0% water and quenched rapidly from ≤600C, and those with less than 5.5% water and quenched rapidly from ≤500C, did not undergo changes on quench and record the equilibrium species concentrations of the experimental run conditions. Knowledge of the equilibrium speciation of water in samples at lower temperatures can be used to predict the species concentrations at magmatic temperatures and allow us to explore the effect of melt structure on the physical properties of natural hydrous magmas. Ideal mixing models can be used as a rough approximation for modeling the solution of water in rhyolitic melts with less than 2.5 wt.% total water: ln[(X OH melt ) 2/(X H 2 O m melt X O melt )]= ln K=1.89±0.05−(3120±40)/T, where X i melt is the mole fraction of species i on a single oxygen basis, H 2O m = water molecules, O = anhydrous oxygens, and T is temperature in Kelvin. This fit provides a standard state enthalpy and entropy of Δ H° = 25.9 ± 0.4 kJ/mol and Δ S° = 15.7 ± 0.4 J/mol · K for the mixing of water molecules in rhyolitic melt. At high water contents, either a modification to the infrared calibration or more complex models (such as a regular solution model) are required to fit the data. Our measurements differ with recent studies using in situ measurement techniques that show lesser concentrations of molecular species at magmatic temperatures, and we address concerns associated with the in situ method. Our study on quenched glasses can be applied to natural rhyolites; using measured species concentrations, the “apparent” equilibration temperature can be calculated to within 12°C (2σ uncertainty) which can be used to determine the cooling rate of a naturally quenched rhyolitic glass.