A set of thermodynamic models is presented that, for the first time, allows partial melting equilibria to be calculated for metabasic rocks. The models consist of new activity–composition relations combined with end‐member thermodynamic properties from the Holland & Powell dataset, version 6. They allow for forward modelling in the system Na2O–CaO–K2O–FeO–MgO–Al2O3–SiO2–H2O–TiO2–Fe2O3. In particular, new activity–composition relations are presented for silicate melt of broadly trondhjemitic–tonalitic composition, and for augitic clinopyroxene with Si–Al mixing on the tetrahedral sites, while existing activity–composition relations for hornblende are extended to include K2O and TiO2. Calibration of the activity–composition relations was carried out with the aim of reproducing major experimental phase‐in/phase‐out boundaries that define the amphibolite–granulite transition, across a range of bulk compositions, at ≤13 kbar.

No abstract

Rocks of basic and intermediate bulk composition occur in orogenic terranes from all geological time periods and are thought to represent significant petrological components of the middle and lower continental crust. However, the former lack of appropriate thermodynamic models for silicate melt, amphibole and clinopyroxene that can be applied to such lithologies at high temperature has inhibited effective phase equilibrium modelling of their petrological evolution during amphibolite‐ and granulite facies metamorphism. In this work, we present phase diagrams calculated in the Na2O–CaO–K2O–FeO–MgO–Al2O3–SiO2–H2O–TiO2–O2 (NCKFMASHTO) compositional system for a range of natural basic and intermediate bulk compositions for conditions of 2–12 kbar and 600–1050 ∘C using newly parameterized activity–composition relationships detailed in a companion paper by Green et al. in this issue. Particular attention is given to mid‐ocean ridge basalt (MORB) and diorite protolith bulk compositions. Calculated subsolidus mineral assemblages in all basic and intermediate rock types are modally dominated by hornblende and plagioclase, with variable proportions of epidote, clinopyroxene, garnet, biotite, muscovite, quartz, titanite or ilmenite present at different pressures. The H2O‐saturated (wet) solidus has a negative P−T slope and occurs between ∼620–690 ∘C at mid‐ to lower‐crustal pressures of 5–10 kbar. The lowest‐T melts generated close to the wet solidus are calculated to have granitic major‐element oxide compositions. Melting at higher temperature is attributed primarily to multivariate hydrate‐breakdown reactions involving biotite and/or hornblende. Partial melt compositions calculated at 800–1050 ∘C for MORB show good correlation with analysed compositions of experimental glasses produced via hydrate‐breakdown melting of natural and synthetic basic protoliths, with Niggli norms indicating that they would crystallize to trondhjemite or tonalite. Diorite is shown to be significantly more fertile than MORB and is calculated to produce high‐T melts (>800 ∘C) of granodioritic composition. Subsolidus and suprasolidus mineral assemblages show no significant variation between different members of the basalt family, although the P−T conditions at which orthopyroxene stabilizes, thus defining the prograde amphibolite–granulite transition, is strongly dependent on bulk‐rock oxidation state and water content. The petrological effects of open‐ and closed‐system processes on the mineral assemblages produced during prograde metamorphism and preserved during retrograde metamorphism are also examined via a case‐study analysis of a natural Archean amphibolite from the Lewisian Complex, northwest Scotland.

No abstract

We report the results of experiments on two natural marine sediments with different carbonate contents (calcareous clay: CO 2 ¼ 6Á1 wt %; marl: CO 2 ¼ 16Á2 wt %) at subduction-zone conditions (3 GPa, 750-1200 C). Water (7-15 wt %) was added to the starting materials to simulate the effects of external water addition from within the subducting slab. The onset of melting is at 760 C in water-rich experiments; melt becomes abundant by 800 C. In contrast, the onset of melting in published, water-poor experiments occurs at variable temperatures with the production of significant melt fractions being restricted to more than 900 C (phengite-out). The different solidus temperatures (T solidus ) can be ascribed to variable fluid X H2O [H 2 O/(CO 2 þ H 2 O)], which, in turn, depends on bulk K 2 O, H 2 O and CO 2 . Partial melts in equilibrium with residual garnet, carbonate, quartz/coesite, epidote, rutile, kyanite, phengite, and clinopyroxene are granitic in composition, with substantial dissolved volatiles. Supersolidus runs always contain both silicate melt and solute-rich fluid, indicating that experimental conditions lie below the second critical endpoint in the granite-H 2 O-CO 2 system. Carbonatite melt coexists with silicate melt and solute-rich fluid above 1100 C in the marl. The persistence of carbonate to high temperature, in equilibrium with CO 2 -rich hydrous melts, provides a mechanism to both supply CO 2 to arc magmas and recycle carbon into the deep Earth. The trace element compositions of the experimental glasses constrain the potential contribution of calcareous sediment to arc magmas. The presence of residual epidote and carbonate confers different trace element characteristics when compared with the trace element signal of Ca-poor marine sediments (e.g. pelagic clays). Notably, epidote retains Th and light rare earth elements, such that some melts derived from calcareous sediments have elevated Ba/Th and U/Th, and low La/Sm PUM , thereby resembling fluids conventionally ascribed to altered oceanic crust. Our results emphasize the importance of residual mineralogy, rather than source lithology, in controlling the trace element characteristics of slab-derived fluids.