Basalt solidus

FIGURE 4.19 Pressure-temperature phase diagram showing the results of the thermal modeling experiment of Peacock et al. (1994) for the subduction of ocean crust at a rate of 1 cm/yr. The curves labeled 0 Ma to 50 Ma show the P-T path for subducted ocean floor 0-50 Ma old. Also shown are the wet and dry solidi for basalt melting and the hornblende-out curve. It can be seen that only the P-T trajectories of young ocean floor intersect the wet basalt solidus (after Rollinson, 2006). 

Pertermann M. and Hirschmann M. M. (2003) Partial melting experiments on a MORB-like pyroxenite between 2 and 3 GPa constraints on the presence of pyroxenite in basalt source regions from solidus location and melting rate. J. Geophys. Res. 10.1029/2000JB000118. 

The typical concept of low-density H2O-CO2 fluids (with small to moderate amounts of solute) in the sub-solidus and high-density silicate liquids (with typically 1-15 wt.% H2O dissolved) alDove the solidus does not apply to the subduction environment at pressures above 5-6 GPa. At higher pressures, a chemical continuum between fluids and melts exists (Eoettcher and Wyllie, 1969) and, depending on fluid-rock ratios, a continuous dissolution process leaches hydrophile species out of sediments, basalts, and serpentine. 

Nevertheless, this hypothesis has been unpopular since the 1980s, because geodynamic models, incorporating either constant mantle viscosity or a rigid upper plate of prescribed thickness, predicted that solidus temperatures could not be reached in basalt or sediment at the top of the subducting plate, except under unusual circumstances (see reviews in Kelemen et al., 2003a Peacock, 1996, 2003 Peacock et al., 1994). For example (Filer et al., 2000, p. 247), discounted their successful model involving H20-rich silicate melt because [the] successful melt-fluxed [model]. .. require[ s]... temperatures... that are... not obviously compatible with...thermal models... (Peacock, 1996). ... 

---