P-T phase relations of silicic, alkaline, aluminous liquids: new results and applications to mantle melting and metasomatism

David S. Draper*, Trevor H. Green

*Corresponding author for this work

    Research output: Contribution to journalArticlepeer-review

    53 Citations (Scopus)


    We report new experimental results obtained under nominally anhydrous conditions at 1.0-1.5 GPa on a synthetic melt whose composition is typical of extreme-composition xenolith glasses. These results demonstrate that part of this extreme compositional range is in equilibrium with a lherzolitic assemblage (olivine, orthopyroxene, and clinopyroxene on the liquidus), extending our earlier findings [D.S. Draper, T.H. Green P-T phase relations of silicic, alkaline, aluminous mantle-xenolith glasses under anhydrous and C-O-H fluid-saturated conditions, J. Petrol. 38 (1997) 1187-1224] showing saturation with harzburgite minerals (olivine and orthopyroxene on the liquidus). The new results strengthen the view that such liquids can readily coexist with upper mantle rocks. Our results also bear on the current debate regarding the nature of low-degree mantle melts between proponents of the diamond-aggregate technique [who argue for comparatively silica- and alkali-rich low-degree melts; e.g., M.B. Baker, M.M. Hirschmann, M.S. Ghiorso, E.M. Stolper, Compositions of near-solidus peridotite melts from experiments and thermodynamic calculations. Nature 375 (1995) 308-311; M.B. Baker, M.M. Hirschmann, L.E. Wasylenki, E.M. Stolper, M.S. Ghiorso, Quest for low-degree mantle melts, Nature 381 (1996) 286] and those favoring the sandwich technique [who question the value of the diamond-aggregate work and argue that near-solidus melts must be nepheline- and olivine-normative; T.J. Falloon, D.H. Green, H.St.C. O'Neill, C.G. Ballhaus, Quest for low-degree mantle melts. Nature 381 (1996) 285; T.J. Falloon, D.H. Green, H.St.C. O'Neill, W.O. Hibberson, Experimental tests of low degree peridotite partial melt compositions: implications for the nature of anhydrous near-solidus peridotite melts at 1 GPa, Earth Planet, Sci. Lett. 152 (1997) 149-162]. Our results support aspects of both views. The sandwich-technique view is supported, for example, because all our liquids coexisting with mantle minerals are nepheline- and olivine-normative: and our olivine-liquid Fe-Mg exchange K(D) values fall on a trend similar to that supported by those workers. The diamond-aggregate view is supported, for example, because we find equilibrium between highly silicic, alkaline liquids and mantle minerals, showing the effect of high alkali contents to allow high silica contents at silica activities buffered by magnesian olivine and orthopyroxene at low pressure [M.M. Hirschmann, M.B. Baker, E.M. Stolper. The effect of alkalis on the silica content of mantle-derived melts, Geochim. Cosmochim. Acta 62 (1998) 883-902]. Additionally, the melting trends put forward by the sandwich-technique workers include revised low-degree melt compositions, as reported by Hirschmann et al., and our compositions fall on extensions of these trends. These new analyses also yield an olivine-liquid K(D) that more closely follows the trend of K(D) vs. melt alkali contents. The views of both sides of this controversy appear to permit, under certain conditions, the existence of small amounts of melt in the upper mantle with compositions similar to the extreme-compositions xenolith glasses that are the focus of our work. On the basis of our new results, we conclude that extreme-composition xenolith glasses can act as agents of cryptic metasomatism in the upper mantle.

    Original languageEnglish
    Pages (from-to)255-268
    Number of pages14
    JournalEarth and Planetary Science Letters
    Issue number3
    Publication statusPublished - 15 Jul 1999


    • glasses
    • mantle
    • melting
    • metasomatism
    • P-T conditions
    • xenoliths


    Dive into the research topics of 'P-T phase relations of silicic, alkaline, aluminous liquids: new results and applications to mantle melting and metasomatism'. Together they form a unique fingerprint.

    Cite this