Ti-in-quartz thermobarometry and TiO2 solubility in rhyolitic melts: New experiments and parametrization

https://doi.org/10.1016/j.epsl.2020.116213Get rights and content

Highlights

  • New Ti-in-quartz thermobarometry has been experimentally calibrated.

  • Simultaneously, a new model of TiO2 solubility in rhyolitic melt has been calibrated.

  • Calibration conditions were relevant for high-silica magmas stored at crustal depths.

  • Our new Ti-in-quartz thermobarometry has been successfully applied to natural silicic eruptions.

Abstract

The Ti-in-quartz thermobarometer has a wide potential for constraining crystallization pressure and temperature of quartz in natural geological systems. However, there is a long-lasting debate on the applicability of two models that were proposed previously, based on the equilibration of quartz with Ti-bearing aqueous fluids. In this study, the Ti-in-quartz thermobarometer was calibrated based on partitioning data of Ti between quartz and aluminosilicate melt in the pressure and temperature range of 0.5−4 kbar and 700−900 °C, which are conditions relevant for high-silica magmas stored at crustal depths.

For seventeen experiments, in which both quartz, rutile and high-silica glass are present as experimental products (i.e., activity of TiO2 in silicate melt equals to unity), the Ti concentrations in quartz can be modeled with the following equation:logCTiQtz=5.32261948.4/T981.4P0.2/T, in which CTiQtz is the Ti concentration (ppm) in quartz, T is temperature in kelvin and P is pressure in kbar. Based on the data from this study and a previous work of Hayden and Watson (2007), we modeled the dependence of rutile (TiO2) solubility in silicic melt on temperature, pressure and melt composition, which can be expressed aslog(STiliq)=6.51893006.5/T461.0P0.2/T+0.1155FM, in which STiliq is Ti solubility (ppm) at rutile saturation and FM is a parameter accounting for melt compositional effect, computed asFM=(Na+K+2Ca+2Mg+2Fe)/(SiAl), in which the chemical symbols denote molar fractions of each cation.

Combining the two models presented above as well as some additional experimental data at activity of TiO2 <1, and assuming an ideal behavior for the activity of TiO2, the following Ti-in-quartz thermobarometer is proposed:log(CTiQtz/CTiliq)=1.1963+(1058.1520.4P0.2)/T0.1155FM, in which CTiliq is Ti concentration (ppm) in melt. Assuming an uncertainty of input temperature of ±25 °C, the corresponding pressure can be determined within ±0.2 kbar. However, the Ti concentrations in quartz and glass need to be determined with a high precision. Typical values of the ratio CTiQtz/CTiliq in natural systems vary in the range from ∼0.09 to ∼0.13, corresponding to a change of pressure from ∼5 to ∼1 kbar assuming a temperature of ∼800 °C.

The model above was applied to natural datasets obtained for several silicic eruptions (i.e. Oruanui Rhyolite, Early Bishop Tuff, Toba Tuff, Upper Bandelier Tuff). The analyses of quartz and glass inclusions in quartz indicate that the pre-eruptive magma storage pressures are mainly in the range 2–4 kbar. These pressures are consistent to or slightly higher than the maximum value estimated previously from the analysis of H2O-CO2 in glass inclusions, indicating a possible post-entrapment loss of hydrogen from melt inclusions and that gas saturation provides a minimum estimation of pressure.

Introduction

Pre-eruptive magma storage conditions are key information for understanding the mechanism of volcanic eruptions and evolution of the underlying magma plumbing systems (e.g., Bachmann and Huber, 2016; Putirka, 2017). Developments in petrological methods, particularly geothermobarometers based on mineral and melt compositions, have been widely employed to provide constraints on intensive variables (e.g., pressure, temperature and volatile activities) prevailing in magma reservoirs of volcanic systems (e.g., Andersen et al., 1993; Blundy and Cashman, 2008). Compared to basaltic and andesitic systems, much less thermobarometers are available for silicic systems, because suitable minerals (e.g., olivine, pyroxene and amphibole) that have been regularly involved in calibrating thermobarometers are lacking. In addition, due to potential loss of H from melt inclusion after entrapment (e.g., Severs et al., 2007; Tollan et al., 2019), gas saturation pressure estimated primarily based on measured H2O concentration in rhyolitic melt inclusions, in which CO2 concentrations are usually extremely low (e.g. Liu et al., 2006), may significantly underestimate the depth of magma pre-eruption storage if raw data are used. Alternatively, Myers et al. (2016, 2018, 2019) restored H2O contents in quartz-hosted melt inclusions based on an assumed linear correlation between H2O and incompatible elements, which would result in much appropriate estimation of gas saturation pressure.

As one of the major phases in silicic magmas, quartz accommodates various trace elements (such as Ti, Al, Fe, K; Götze, 2009), and several experimental studies showed that the concentration of Ti in quartz is pressure- and temperature-dependent (Wark and Watson, 2006; Thomas et al., 2010; Huang and Audétat, 2012; Thomas et al., 2015). Combined with the geospeedmeter based on diffusion of Ti in quartz (Cherniak et al., 2007), concentrations and distribution patterns of Ti in quartz can potentially offer spectacular information about magma storage conditions and timescales relevant to magmatic events (e.g., Wark et al., 2007; Gualda et al., 2018; Shamloo and Till, 2019).

Currently, there are two available independent calibrations of Ti-in-quartz thermobarometer, i.e. one from Thomas et al. (2010) and another one from Huang and Audétat (2012). However, they have large and irreconcilable discrepancies between each other, resulting in widespread confusion about the application of Ti-in-quartz thermobarometer for geological issues (e.g., Wilson et al., 2012). Both calibrations are based on experiments in which Ti-bearing quartz was crystallized from H2O-dominated fluids (i.e., hydrothermal quartz). Thomas et al. (2015) discussed critically the experimental design of Huang and Audétat (2012) and suggested that their approach may inevitably result in disequilibrium between newly formed quartz and fluid in terms of TiO2 activity. Other the other hand, the calibration of Thomas et al. (2010) was performed at pressures of 5–20 kbar, which does not cover the main range of storage pressures of silicic magmas in the continental crust that is usually below 5 kbar (e.g., Liu et al., 2006; Wilke et al., 2017). Considering that it is difficult to reconcile both datasets published so far, we decided to test another experimental approach, consisting in the crystallization of quartz from a silicate melt (rather than from a fluid). The experiments were designed to obtain quartz and co-existing Ti-bearing melt, as well as rutile in some cases. The experimental data are used to propose an improved Ti-in-quartz thermobarometer for rhyolitic magmas at crustal depths.

Section snippets

Starting materials

For investing the solubility of Ti in igneous quartz, we performed crystallization experiments using four synthesized high-silica glasses containing different TiO2 contents. Except for TiO2, these starting glasses have similar relative mass proportions for other present cations (Si, Al, Na and K). The normative Qz-Ab-Or (SiO2-NaAlSi3O8-KAlSi3O8) proportion is Qz70-Ab20-Or10, and the glasses are peraluminous with an aluminum saturation index [calculated as molar ratio Al2O3/(Na2O+K2O)] of

Experimental products

Experimental conditions and products using the two low-Ti starting glasses (S3 and S4) are listed in Table 1). Representative back-scattered electron (BSE) images are shown in Fig. 1. All experimental samples contained glass, quartz and an aluminosilicate phase. The ratio of Al/Si in the aluminosilicate phase clearly shows that the chemical formulae of the phase is Al2SiO5 and that should be sillimanite considering the P-T range of the experiments. For experimental runs using starting glass S3

Attainment of equilibrium

The calibration of a Ti-in-quartz thermobarometer requires that experimental data reflect equilibrium distribution of Ti between quartz and silicate melt. Different independent observations indicate that conditions close to equilibrium were reached.

In our experiments the Ti content of the melt is significantly higher than that of the quartz and possible problems leading to disequilibrium distribution of Ti may occur if this element cannot diffuse fast enough away from the interface quartz-melt.

Conclusions

Experiments designed to attain near-equilibrium conditions between felsic melt, quartz and rutile were used to re-calibrate the Ti-in-quartz thermobarometer at rutile-saturated conditions and the TiO2 solubility in melt, which are functions of both pressure and temperature. The combination of these two models allowed us to propose a new Ti-in-quartz thermobarometer without an independent estimation of aTiO2liqRt provided that Ti concentrations in quartz and in a coexisting melt are available.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We thank Colin Wilson and Jörg Hermann for their thoughtful reviews that have substantially improved this paper. We also thank Heather Handley for her editorial handling. This work was supported by German Research Foundation (DFG) project HO1337/40 in the frame of ICDP program.

References (54)

  • S. Klemme et al.

    Experimental constraints on major and trace element partitioning during partial melting of eclogite

    Geochim. Cosmochim. Acta

    (2002)
  • S. Klemme et al.

    Partitioning of trace elements between rutile and silicate melts: implications for subduction zones

    Geochim. Cosmochim. Acta

    (2005)
  • K. Kularatne et al.

    Rutile solubility in hydrous rhyolite melts at 750–900 °C and 2 kbar, with application to titanium-in-quartz (TitaniQ) thermobarometry

    Geochim. Cosmochim. Acta

    (2014)
  • Q. Liu et al.

    Acoustic velocity measurements on Na2O–TiO2–SiO2 liquids: Evidence for a highly compressible TiO2 component related to five-coordinated Ti

    Geochim. Cosmochim. Acta

    (2007)
  • M.L. Myers et al.

    Inferring magma ascent timescales and reconstructing conduit processes in explosive rhyolitic eruptions using diffusive losses of hydrogen from melt inclusions

    J. Volcanol. Geotherm. Res.

    (2019)
  • M.L. Myers et al.

    Prolonged ascent and episodic venting of discrete magma batches at the onset of the Huckleberry Ridge supereruption, Yellowstone

    Earth Planet. Sci. Lett.

    (2016)
  • F.J. Ryerson et al.

    Rutile saturation in magmas: implications for TiNbTa depletion in island-arc basalts

    Earth Planet. Sci. Lett.

    (1987)
  • M.J. Severs et al.

    Experimental determination of H2O loss from melt inclusions during laboratory heating: evidence from Raman spectroscopy

    Chem. Geol.

    (2007)
  • X.L. Xiong et al.

    Rutile stability and rutile/melt HFSE partitioning during partial melting of hydrous basalt: implications for TTG genesis

    Chem. Geol.

    (2005)
  • X. Xiong et al.

    Partitioning of Nb and Ta between rutile and felsic melt and the fractionation of Nb/Ta during partial melting of hydrous metabasalt

    Geochim. Cosmochim. Acta

    (2011)
  • R.J. Angel et al.

    A simple and generalised P–T–V EoS for continuous phase transitions, implemented in EosFit and applied to quartz

    Contrib. Mineral. Petrol.

    (2017)
  • A. Audétat et al.

    Characterisation of a natural quartz crystal as a reference material for microanalytical determination of Ti, Al, Li, Fe, Mn, Ga and Ge

    Geostand. Geoanal. Res.

    (2014)
  • O. Bachmann et al.

    Silicic magma reservoirs in the Earth's crust

    Am. Mineral.

    (2016)
  • J. Berndt et al.

    A combined rapid-quench and H2-membrane setup for internally heated pressure vessels: description and application for water solubility in basaltic melts

    Am. Mineral.

    (2002)
  • J. Blundy et al.

    Petrologic reconstruction of magmatic system variables and processes

    Rev. Mineral. Geochem.

    (2008)
  • W. Dennen et al.

    Aluminum in quartz as a geothermometer

    Contrib. Mineral. Petrol.

    (1970)
  • J. Götze

    Chemistry, textures and physical properties of quartz - geological interpretation and technical application

    Mineral. Mag.

    (2009)
  • Cited by (32)

    View all citing articles on Scopus
    1

    Present address: Department of Geology, Northwest University, 710069 Xi'an, China.

    View full text