Deep and persistent melt layer in the Archaean mantle

Transcription

1 SUPPLEMENTARY INFORMATION Articles In the format provided by the authors and unedited. Deep and persistent melt layer in the Archaean mantle Denis Andrault 1 *, Giacomo Pesce 1,4, Geeth Manthilake 1, Julien Monteux 1, Nathalie Bolfan-Casanova 1, Julien Chantel 1, Davide Novella 1, Nicolas Guignot 2, Andrew King 2, Jean-Paul Itié 2 and Louis Hennet 3 1 Université Clermont Auvergne, CNRS, IRD, OPGC, Laboratorie Magmas et Volcans, Clermont-Ferrand, France. 2 Synchrotron SOLEIL, Gif-sur-Yvette, France. 3 Conditions Extrêmes et Matériaux: Haute Température et Irradiation, CNRS, Orléans, France. Present address: 4 School of Geosciences, The University of Edinburgh, Edinburgh, UK. * denis.andrault@uca.fr Nature Geoscience Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

2 Supplementary Information Supplementary Figures Fig. S1: Old and new solidus compared with the current adiabatic profile of the mantle Fig. S2: Water and CO2 contents extracted from Fourier transform infrared spectroscopy Fig. S3: (a) High pressure assembly used for electrical conductivity measurements; (b) Thermal gradient simulations for the 18/11 mm assembly Fig. S4: Typical impedance profiles recorded for the chondritic sample Fig. S5: (a) Arrhenius plot showing different heating and cooling cycles (b) Comparison between our EC measurements and previous results Fig. S6: (a) Close-up of the 6-8 DIA set-up used for in-situ X-ray diffraction measurements; (b) Schematic drawing of the 7/3 assembly used for X-ray diffraction experiments; (c) Typical X-ray radiography of an assembly after compression to ~15 GPa Fig. S7: Example of diffraction patterns and associated mineral contents Fig. S8: Fast variation with time of diffraction peaks intensity Fig. S9: Degree of partial melting retrieved from experimental works. Supplementary Tables Table S1. Composition of our chondritic-type starting material Table S2. Summary of experimental runs performed in this work Table S3: Simon&Glatzel parameters refined for the solidus and liquidus

3 Fig. S1: The new solidus profile (green) is compared with the present-day adiabatic profile (red), which room pressure potential mantle temperature (Tp) could range between 1600 K and 1750K 4. In the shallow mantle, the temperature profile can be typically oceanic-type (blue) or cratonic-type (mauve), depending on the type of geological settings.

4 Fig. S2: Fourier transform infrared (FTIR) spectra of the glass starting material (in red) and of a sample recovered after melting at 5 GPa (black line). The triplet between 3000 and 2800 cm -1 is characteristic of the epoxy used to embed the sample for the chemical analysis. Spectra display overtones of silicate bands at cm -1. We can assess absence of CO 2 in our samples, as spectra show no evidence of CO 3 2- bands at cm -1.

5 Fig. S3a: High pressure assembly used for electrical conductivity measurements. Fig. S3b: Thermal gradient simulations for the 18/11 mm assembly and for thermocouple temperatures of 1473 (left), 1873 (centre) and 2273 K (right). The sample size is 1.5 mm (only ¼ of the sample is drawn here). The maximum difference between the thermocouple reading and the sample temperature is ~80 K, a value almost independent of the sample temperature.

6 Fig. S4: Typical impedance profiles recorded for the chondritic sample at increasing temperatures. (a) After compression, electrical conduction in our samples occurs through the grains interior, which can be modelled as a simple R/CPE circuit. (b) Upon melting, electrical conduction also occurs through the interconnected melt channels located at grain boundaries. It yields the very low resistances observed at high temperatures. This behaviour can be modelled by a parallel circuit between the solid and molten sample fractions.

7 Fig. S5a: Arrhenius plot showing the different cycles of heating and cooling, before the final heating cycle #4, which induces the sample melting. Fig. S5b: Comparison between our values of electrical conductivity (all measured at 5 GPa) and previous measurements performed on major mantle minerals 1-3 or various types of melts 8,9. Opx and Cpx stand for orthopyroxene and clinopyroxene, respectively.

8 Fig. S6a: Close-up of the 6-8 DIA set-up used for in situ X-ray diffraction measurements at the SOLEIL synchrotron. The finger orientation mimics the X-ray path. The red wire is the thermocouple. Fig. S6b: Schematic drawing of the 7/3 assembly used for X-ray diffraction experiments, in lateral (left) and axial (right) views. The axis of the tubular Re furnace is parallel to the X-ray beam. Fig. S6c: Typical X-ray radiography of an assembly after compression to ~15 GPa. The octahedral orientation is the front view (Fig. S6b). The white circle at the centre of the image is the sample located inside the tubular Re-furnace (darkest ring). Further away laterally from the Re-furnace, the two wires of the W-Re thermocouple are clearly visible.

9 Fig. S7: Example of fitted diffraction patterns. Experimental measurements (black crosses) are fitted using a Rietveld-type multiphase model. The diffraction peaks intensities are fixed to the values expected from the atomic topology in each mineral. Only cell parameters and phase contents are adjusted. The theoretical model (Red line) is the sum of the contributions of each phases (yellow, blue and green). Above the crystallization temperature, we could detect the mixture of olivine (or its high pressure polymorphs), garnet, OPx and/or CPx. Due to the limited sample size, diffraction peaks of the MgO capsule material are sometimes visible. Higher pressures favour the dissolution of OPx and CPx in the majoritic garnet. At the maximum pressure of ~28 GPa performed in this work, we observed a mixture of garnet plus ferropericlase, as previously reported for comparable pressure and sample composition 6.

10 Fig. S8: Fast variation with time of the intensity of different dhkl diffraction peaks (e.g. (222), (-131) and (-112) for olivine), recorded at a constant experimental temperature. Along the time axis, each point corresponds to a spectrum with an exposure time of 10 sec. The intensity expected for each dhkl peak from the atomic topology (normalized to the maximum peak intensity for each phase) is also reported (e.g. 74% for the (222) diffraction peak of olivine).

11 Fig. S9: Degree of partial melting as a function of temperature and at different pressure ranges from 3 to 23 GPa. Our determination of the solidus temperature (purple dots at F~0) is plotted together with previous works on mantle partial melting (blue 5, orange 10 and grey 6 ) and liquidus (brown 12 and blue 5 dots at F=1). The CMAS data points appear systematically ~200K more refractory than our composition, possibly due to lack of Fe, Na, Ti, Cr.

12 Table S1. Composition of our chondritic-type starting material.

13 Table S2. Summary of experimental runs performed in this work. X-ray diffraction Exp. # P(GPa) Tobs ( C) T & grad T H2O T Solidus(K) T cryst(k) Phases MA2_ Gt, Oli MA2_ Gt, Oli MA2_ Gt, Oli MA2_ Gt, Oli MA2_ Gt, Wad MA3_ Gt, Oli MA3_ Gt, Oli MA4_ Gt, Oli MA4_ Gt, Wad MA4_ Gt, MgO MA4_ Gt, Oli MA5_ Gt, Oli MA5_ Bg, Gt, MgO Electrical Conductivity LMV * LMV * LMV * LMV * LMV * LMV * P(GPa) is the experimental pressure at the onset of melting. Tobs is the raw experimental temperature when the melting criterion is detected. T grad is the temperature correction for presence of thermal gradients in the cell assembly (Fig. S3b). T H2O is the melting temperature correction for presence of 90 ppm water in the sample (Fig. S2). T Solidus is the corrected solidus temperature. T cryst is the temperature when occurs the crystallisation of the glassy starting material. Phases is the list of minerals present in the sample, as refined from X-ray diffraction patterns (Fig. S7). Oli, Gt, Wad and Bg stand for olivine, garnet, wadsleyite and brigmanite, respectively. * During conductivity experiments, onset of sample crystallization was observed below 1273 K. Then, the temperature was keep constant between 1373 and 1573 K for ~3h.

14 Table S3: Simon&Glatzel parameters refined for the solidus and the liquidus 5,7 in the ranges of pressure corresponding to the upper mantle (UM), and also to the lower (LM) mantle, based on a previous work 11. SG-param. UMSolidus UMLiquidus LMSolidus LMLiquidus T * a 0.821(80) c 6.85(25) *Fixed 13

15 References cited: 1 Dai, L. D. & Karato, S. I. Electrical conductivity of orthopyroxene: Implications for the water content of the asthenosphere. Proceedings of the Japan Academy Series B-Physical and Biological Sciences 85, , (2009). 2 Yang, X. Orientation-related electrical conductivity of hydrous olivine, clinopyroxene and plagioclase and implications for the structure of the lower continental crust and uppermost mantle. Earth Planet. Sci. Lett , , (2012). 3 Wang, D. J., Mookherjee, M., Xu, Y. S. & Karato, S. The effect of water on the electrical conductivity of olivine. Nature 443, , (2006). 4 Dalton, C. A., Langmuir, C. H. & Gale, A. Geophysical and Geochemical Evidence for Deep Temperature Variations Beneath Mid-Ocean Ridges. Science 344, 80-83, (2014). 5 Iwamori, H., McKenzie, D. & Takahashi, E. Melt generation by isentropic mantle upwelling. Earth Planet. Sci. Lett. 134, , (1995). 6 Litasov, K. & Ohtani, E. Phase relations and melt compositions in CMAS pyrolite H2O system up to 25 GPa. Phys. Earth Planet. Inter. 134, , (2002). 7 Zhang, J. & Herzberg, C. Melting experiments on anhydrous peridotite KLB-1 from 5.0 to 22.5 GPa. J. Geophys. Res. 99, , (1994). 8 Ni, H., Keppler, H. & Behrens, H. Electrical conductivity of hydrous basaltic melts: implications for partial melting in the upper mantle. Contributions to Mineralogy and Petrology 162, , (2011). 9 Sifre, D. et al. Electrical conductivity during incipient melting in the oceanic low-velocity zone. Nature 509, 81-85, (2014). 10 Walter, M. J. Melting of garnet peridotite and the origin of komatiite and depleted lithosphere. Journal of Petrology 39, 29-60, (1998). 11 Andrault, D. et al. Melting curve of the deep mantle applied to properties of early magma ocean and actual core-mantle boundary. Earth Planet. Sci. Lett. 304, , (2011). 12 Herzberg, C. & Zhang, J. Melting experiments on anhydrous peridotite KLB-1: compositions of magmas in the upper mantle and transition zone. J. Geophys. Res. 101, , (1996). 13 Hirschmann, M. M. Mantle solidus: Experimental constraints and the effects of peridotite composition. Geochem., Geophys., Geosyst. 1, (2000).