Hydrous Mantle Melting
Water plays a key role in mantle melting processes, particularly in subduction zone environments. There is, however, very little experimental data on the influence of water on mantle melting, because of the technical difficulties associated with water-bearing silicate melts at high temperature: diffusion of water through capsules, diffusion of cations and melt modification during quench, and loss of iron to sample containers. Over the past years, I have developed experimental techniques for the extraction of hydrous silicate melts during mantle melting experiments, and used them to determine the composition of hydrous mantle melts. I am also looking in detail at the influence of water on melt fraction and melting processes.
Melt extraction in hydrous experiments
Due to the increase in diffusion rates caused by dissolved H2O, hydrous silicate melts in contact with crystals cannot be preserved during quench and are modified by the quench-related growth of the crystals. The idea behind the extraction techniques is to segregate the melt from the crystals during the experiment, while maintaining bulk equilibrium. If melts are sufficiently segregated, they will not be modified during the quench process. I have been developping two new techniques to segregate melts during hydrous experiments. The first technique involves melting of a thin gold foil placed in the middle of the experiment. When the foil melts, it creates a space where the silicate melt can be collected (Fig. 1). The second technique builds on existing work, where extraction is provided by capillary forces in crimped noble metal capsules (Fig. 2).
Fig. 1. Melt extraction using a molten gold foil Fig. 2. Melt extraction using a crimped capsule
Mantle melting processes
The most common model of mantle melting on Earth is adiabatic decompression melting, where an ascending diapir of dry mantle melts over a large range of pressure. Adiabatic decompression melting is the dominant melting process at mid-ocean ridges and ocean islands, and is probably a major melting process in subduction zones and continental settings too. However, since water strongly partitions into the liquid phase during melting, the solid residue of hydrous melting is highly refractory. Adiabatic decompression melting thus becomes difficult when water is present, and other melting processes can develop in the mantle:
- in the flux melting (or water-saturated melting) model, fluids generated by dehydration of the subducting slab move into the inverted thermal gradient of a subduction zone. These fluids strongly depress the mantle solidus, and water-rich melts are generated at temperatures significantly lower than typical basalts. High-Mg andesites are typical examples of magmas generated by flux-melting of the mantle.
- in the dehydration melting (or water-undersaturated) model, water is present in the mantle as hydrous minerals. Melt is generated when the mantle is heated to temperatures above the stability limit of the hydrous phases. A typical example are the "bajaites" of Baja California, which tap a previously metasomatized mantle domain.
- in the flux melting (or water-saturated melting) model, fluids generated by dehydration of the subducting slab move into the inverted thermal gradient of a subduction zone. These fluids strongly depress the mantle solidus, and water-rich melts are generated at temperatures significantly lower than typical basalts. High-Mg andesites are typical examples of magmas generated by flux-melting of the mantle.
- in the dehydration melting (or water-undersaturated) model, water is present in the mantle as hydrous minerals. Melt is generated when the mantle is heated to temperatures above the stability limit of the hydrous phases. A typical example are the "bajaites" of Baja California, which tap a previously metasomatized mantle domain.
Fig. 3. Principles of flux melting (adapted from the work of Tim Grove at MIT)
Influence of water on melt fraction
Experiments at variable temperature and water content can be used to construct models of the influence of water on melt fractions. However, due to experimental uncertainties, extrapolations of the experimental data have limited accuracy. In order to get accurate models of the influence of water on mantle melting, I have constructed models based on the relation between melt fraction and temperature in dry peridotites, combined with the liquidus depression caused by water on silicate melts (Medard and Grove 2008). Preliminary models (Fig. 4) reproduce the experimental data well, and can be extrapolated to lower water contents (< 0.1 wt%) that are difficult to reach experimentally, but are the most relevant for mantle melting outside of subduction zones.
Fig. 4. Variation of melt fraction versus temperature as a function of water content in the source. Comparison of experimental data (Hirose and Kushiro 1995, Sorbadère et al. 2003, and unpublished data) with a theoretical model based on liquidus depression.
Relevant publications
Baasner A, Médard E, Laporte D, Hoffer G (2016) Partial melting of garnet lherzolite with water and carbon dioxide at 3 GPa using a new melt extraction technique: implications for intraplate magmatism. Contribution to Mineralogy and Petrology 171: 45, doi: 10.1007/s00410-016-1233-0.
Condamine P, Médard E (2014) Experimental melting of phlogopite-bearing mantle at 1.0 GPa: implications for potassic magmatism. Earth and Planetary Science Letters 397: 80-92, doi: 10.1016/j.epsl.2014.04.027.
Condamine P, Médard E, Devidal JL (2016) Experimental melting of phlogopite-peridotite in the garnet stability field. Contribution to Mineralogy and Petrology 171: 95, doi: 10.1007/s00410-016-1306-0.
Grove TL, Chatterjee N, Parman SW, Médard E (2006) The influence of H2O on mantle wedge melting. Earth and Planetary Science Letters 249: 74-89, doi: 10.1016/j.epsl.2006.06.043.
Grove TL, Till CB, Lev E, Chatterjee N, Médard E (2009) Kinematic variables and water transport control the formation and location of arc volcanoes. Nature 459: 694-697, doi: 10.1038/nature08044.
Médard E, Grove TL (2008) The effect of H2O on the olivine liquidus of basaltic melts: experiments and thermodynamic models. Contribution to Mineralogy and Petrology 155: 417-432, doi: 10.1007/s00410-007-0250-4.
Médard E, Schmidt MW, Schiano P, Ottolini L (2006) Partial melting of amphibole-bearing wehrlites: an experimental study on the origin of ultracalcic nepheline-normative melts. Journal of Petrology 47: 481-504, doi: 10.1093/petrology/egi083.
Sorbadère F, Médard E, Laporte D, Schiano P (2013) Experimental melting of hydrous peridotite-pyroxenite mixed sources: constraints on the genesis of silica-undersaturated magmas beneath volcanic arcs. Earth and Planetary Science Letters 384: 42-56, doi: 10.1016/j.epsl.2013.09.026.
Condamine P, Médard E (2014) Experimental melting of phlogopite-bearing mantle at 1.0 GPa: implications for potassic magmatism. Earth and Planetary Science Letters 397: 80-92, doi: 10.1016/j.epsl.2014.04.027.
Condamine P, Médard E, Devidal JL (2016) Experimental melting of phlogopite-peridotite in the garnet stability field. Contribution to Mineralogy and Petrology 171: 95, doi: 10.1007/s00410-016-1306-0.
Grove TL, Chatterjee N, Parman SW, Médard E (2006) The influence of H2O on mantle wedge melting. Earth and Planetary Science Letters 249: 74-89, doi: 10.1016/j.epsl.2006.06.043.
Grove TL, Till CB, Lev E, Chatterjee N, Médard E (2009) Kinematic variables and water transport control the formation and location of arc volcanoes. Nature 459: 694-697, doi: 10.1038/nature08044.
Médard E, Grove TL (2008) The effect of H2O on the olivine liquidus of basaltic melts: experiments and thermodynamic models. Contribution to Mineralogy and Petrology 155: 417-432, doi: 10.1007/s00410-007-0250-4.
Médard E, Schmidt MW, Schiano P, Ottolini L (2006) Partial melting of amphibole-bearing wehrlites: an experimental study on the origin of ultracalcic nepheline-normative melts. Journal of Petrology 47: 481-504, doi: 10.1093/petrology/egi083.
Sorbadère F, Médard E, Laporte D, Schiano P (2013) Experimental melting of hydrous peridotite-pyroxenite mixed sources: constraints on the genesis of silica-undersaturated magmas beneath volcanic arcs. Earth and Planetary Science Letters 384: 42-56, doi: 10.1016/j.epsl.2013.09.026.
Proudly powered by Weebly