Professor Peter J. Wyllie, Division of Geological and Planetary Sciences, Caltech
FROM DIVISION RESEARCH REPORT, 1992-94, with current references (April 1999).

The Origin of Archean Tonalites and Trondhjemites

The tonalites, trondhjemites and granites (TTGs) of the Archean grey gneisses are characterized by highly fractionated rare earth elements, and this appears to require that amphibole, or garnet or both were residual minerals during generation of the magmas. Recently, attention has been focussed on vapor-absent dehydration-melting, during which H₂O released by the breakdown of amphibolite dissolves directly into a H₂O-undersaturated silicate liquid, and it appears that this process can yield liquids corresponding to the Archean TTG rocks through a range of pressures and temperatures. Dr. M. B. Wolf has completed a detailed study at 10 kbar (40 km depth) of the effects of temperature, time and texture on the progressive melting of vapor-absent amphibolite. His results are consistent with the phase diagram for vapor-absent amphibolite shown in figure 1A. The new feature is the sharp backbend of the solidus near 9 kbar and 900 C, associated with the formation of garnet. This expands the field for liquid generation with garnet-amphibole residues to much lower temperatures and pressures than in the other recent experimental results. Figure 1B shows a generalized map of the residual mineral assemblages to be expected after a small fraction of melt has been removed during dehydration-melting of amphibolite.

Figure 1. (A). Selected phase boundaries for fully- hydrated, simple amphibolite. Vapor-absent melting occurs where hornblende begins to break down forming augite in the reaction interval [1], and forming garnet (and other minerals) in the reaction interval [3-5] (Wyllie and Wolf, 1993). (B) Residual minerals left behind after dehydration-melting of amphibolite and magma segregation. Abbreviations: Hb = hornblende, Pl = plagioclase, Cpx = clinopyroxene, Ga = garnet.

The forward approach of melting possible source rocks under known conditions is complemented by the inverse approach of crystallizing the magmas to determine the liquidus mineralogy as a function of pressure, temperature, H₂O content, and other variables. If the tonalites and trondhjemites are primary magmas from amphibolite, then for the pressure and temperature of their generation, the near-liquidus minerals should match the residual minerals in partially melted rocks. Experiments on the Nuk trondhjemite from Greenland by Drs. A. D. Johnston and S. R. van der Laan illustrate the approach. Figure 2 shows the liquidus surface in terms of pressure and temperature, with (1) a third dimension shown by the contours for H₂O content (figure 2A), and (2) the distribution of liquidus mineral fields and field boundaries (figure 2B). According to these results, the trondhjemite is near multiple-saturation with the assemblage [Cpx + Pl + Hb + Ga] at 13.5 kb, 900 C, with about 7.5% dissolved H₂O. The liquidus minerals match the residual minerals in area (3) in figure 1B.

Figure 2. The H₂0-undersaturated liquidus surface for the Nuk gneiss (trondhjemite). Abbreviations: see fig. 1, and Ep = epidote. (A) The liquidus surface mapped with approximate contours for H₂0 content. (B) Primary minerals on the liquidus surface. Compare the boundaries for hornblende, and for garnet, with the residual minerals in fig. 1 (van der Laan and Wyllie, 1992).
Experimental data are also available for the liquidus minerals on the H₂O-undersaturated liquidus surfaces of two tonalites with different SiO₂ contents. The liquidus field boundaries for amphibole and garnet occur in similar positions for all three magmas. One can read the pressures, temperatures and H₂O contents of magmas that would leave residual garnet, amphibole or both in the source rocks. Residual amphibole requires moderate temperatures but high H₂O; residual garnet requires depths greater than about 50 km, lower H₂O, and higher temperatures than for residual amphibole. There is a only a limited area where garnet and amphibole coexist on the liquidi. Further refinement of these phase boundaries will place tighter constraints on processes in terms of depth, temperature and H₂O contents, and help to define the tectonic environment in which the magmas were formed and emplaced.

Immiscibity between Silicate and Carbonate Melts

There is good petrological evidence that some alkaline igneous rocks and carbonatites are the crystalline products of conjugate immiscible liquids. Experimental studies have defined miscibility gaps between a variety of silicate and carbonate liquids through a range of pressures, but the conditions have not been defined to determine the controls on whether a CO₂-enriched silicate magma precipitates silicate minerals with evolution of CO₂, differentiates to a residual carbonate-rich melt, or yields a carbonate-rich melt as an immiscibile phase.

Woh-jer Lee has studied in this and related problems in several systems spanning the depth interval from lower lithosphere to shallow crust. He is locating the position of the phase boundary between liquidus fields for primary silicate minerals, and primary carbonates. The latter phase boundary is a critical factor in controlling liquid paths, and types of igneous rocks that can form. He has determined that the positions of this phase boundary, and of the two-liquid phase boundaries, shift significantly as a function of pressure, and of Na/(Ca+Mg+Fe) and Ca/(Mg+Fe). The results have applications to mantle metasomatism, the possible formation of carbonatic fluids above subduction zones, and the generation of carbonatites and associated ore deposits.

In another related investigation, Dr. B. McInnes is determining the compositions of liquids that can be generated during subduction from water-altered basalt, or from coexisting basalt and pelgic limestone. The liquids are enriched in alkalis and carbonate components, and if formed during subduction they should have significant geochemical consequences.

Figure 3. Scanning electron microscope image of the polished surface of an experimental run (from 1200C, 15 kbar) showing spherical silicate glass (with some silicate minerals) enclosed in carbonate liquid which has quenched to an intimate mixture of dendritic calcite in silicate glass (Woh-jer Lee, unpublished).

SELECTED RELEVANT PUBLICATIONS. (September 1997)

Archean Amphibolites, Tonalites and Trondhjemites

Wyllie, P.J. 1986. A petrologic viewpoint, p. 41-43 in deWit, M. J., and Ashwal, L. D. (eds), in "Workshop on evolution of greenstone belts". LPI Technical Report 86-10, Lunar and Planetary Institute, Houston.

Johnston, A. D., and Wyllie, P. J., 1988. Constraints on the origin of Archean trondhjemites based on phase relationships of Nuk Gneiss with H₂O at 15 kbar. Contr. Miner. Petrol., 100, 35-46.

Carroll, M. R., and Wyllie, P. J., 1990. The system tonalite-H₂O at 15 kbar and the genesis of calc-alkaline magmas. Amer. Miner., 75, 345-357.

Wolf, M. B., and Wyllie, P. J. 1991. Dehydration-melting of solid amphibolite at 10 kbar: textural development, liquid interconnectivity and applications to the segregation of magmas. Miner. and Petrol., 44, 151-179.

Van der Laan, S. R., and Wyllie, P. J., 1992. Constraints on archean trondhjemite genesis from hydrous crystallization experiments on Nuk gneiss at 10-17 kbar. Jour. Geol., 100, 57-68.

Wyllie, P. J., and Wolf, M. B., 1993. Amphibolite-dehydration melting: sorting out the solidus. p. 405-416 in: (eds), H.M.Pritchard, T. Alabaster, N.B.W.Harris, and C.R.Neary. Magmatic processes and plate tectonics. Geol. Soc. Spec. Paper No. 76, London.

Wolf, M. B., and Wyllie, P. J. 1994. Dehydration-melting of amphibolite at 10 kbar: effects of temperature and time. Contr. Miner. Petrol., 115, 369-383.

Wyllie, P. J., Wolf, M. B., and van der Laan, S. R., 1997. Conditions for formation of tonalites and trondhjemites.: magmatic sources and products. Chapter 3.3.1, p. 258-267, in Eds. de Wit, M. J., and Ashwahl, L. D. Tectonic evolution of greenstone belts. Oxford University Press.

Silicate-Carbonate Liquid Immiscibility

Baker, M. B., and Wyllie, P. J., 1990. Liquid immiscibility in a nephelinite-carbonate system at 25 kbar and implications for carbonatite origin. Nature, 346, 168-170.

Lee, W.-L, Wyllie, P.J., and Rossman, G.R. 1994. CO₂-rich glass, round calcite crystals, and no liquid immiscibility in the system CaO-SiO₂-CO₂ at 2.5 GPa. Amer. Miner., 79, 1135-1144.

Lee, W.-J., and Wyllie, P. J. 1994. Experimental data bearing on liquid immiscibility, crystal fractionation, and the origin of calciocarbonatites and natrocarbonatites. Intern. Geology Review, 36, 797-819.

Lee, Woh-jer, and Wyllie, P. J., 1996. Liquid immiscibility in the join NaAlSi₃O₈-CaCO₃ to 2.5 GPa and the origin of calciocarbonatite magmas. Jour. Petrology, 37, 1125-1152.

Lee, W.-J., and Wyllie, P. J., 1997. Liquid immiscibility between nephelinite and carbonatite from 2.5 to 1.0 GPa compared with mantle melt compositions. Contr. Miner. Petrol., 127, 1-16.

Lee, Woh-jer, and Wyllie, P. J., 1997. Liquid immiscibility in the join NaAlSiO₄-NaAlSi₃O₈-CaCO₃ at 1 GPa: implications for crustal carbonatites. Jour. Petrol., 38, 1113-1135.

Lee, Woh-jer, and Wyllie, P. J., 1998. Petrogenesis of carbonatite magmas from mantle to crust, constrained by the system CaO-(MgO+FeO*)-(Na₂O+K₂O)-(SiO₂+Al₂O₃+TiO₂)-CO₂. Jour. Petrol., 39, 495-577..

Wyllie, P. J., and Lee, Woh-Jer, 1998. Model system controls on conditions for formation of magnesiocarbonatite and calciocarbonatite magmas from the mantle. Jour. Petrol., 39, 1885-1893.

Lee, Woh-Jer and Wyllie, P. J., 1998. Processes of crustal carbonatite formation by liquid immiscibility and differentiation, elucidated by model systems. Jour. Petrol., 39, 2005-2013.

Wyllie, P. J., and Lee, Woh-Jer, 199x. Kimberlites, carbonatites, peridotites and silicate-carbonate liquid immiscibility explained in parts of the system CaO-(MgO+FeO*)-(Na₂O+K₂O)-SiO₂+Al₂O₃+TiO₂)-CO₂. Proceedings 7th Intern. Kimberlite Conference, April 1998, Cape Town. In press.

Lee Woh-jer, Fanelli, M. F., Cava, N., and Wyllie, P. J., 199x. Calciocarbonatite and magnesiocarbonatite rocks and magmas representeed in the system CaO-MgO-CO₂-H₂O at 0.2 GPa. Mineral. and Petrol. Submitted July 1998. In press.

Lee Woh-jer, Huang Wuu-Liang, and Wyllie, P. J., 199x. Carbonate-rich melts in the mantle modeled in the system CaO-MgO-SiO₂-CO₂ at 2.7 GPa. Contrib. Mineral. Petrol. Submitted April 1998. In revision..

Lee Woh-jer, and Wyllie, P. J., 199x. The system CaO-MgO-SiO₂-CO₂ at 1.0Gpa, metasomatic wehrlites, and calciocarbonatite magmas. Contrib. Mineral. Petrol. Submitted June 1998. In revision.

Wyllie, P. J., 199x. Two centuries on granite, basalt and marble/carbonatite after Hutton's theory and Hall's experiments. "Journal of China University of Geosciences" (title to be corrected). In press.

Back to Peter Wyllie .