A ROLE FOR LOW TEMPERATURE HYDROUS PERALKALINE MELT IN GRANITE FORMATION AND SILICIC IGNIMBRITE ERUPTION?

The relationship between silicic calderas and granitic plutons remains enigmatic. Recent work points to granitic plutons reflecting long slow incremental emplacement; in contrast, the highly aphyric nature of many ignimbrites argues for fast ascent from magma source followed by immediate eruption. Above all, differentiation trends for plutonic and volcanic rock suites are identical arguing for a common magma evolution process. Notably, seismic tomography has found no blobs of mostly melt in Earth’s upper crust, yet electrical conductivities in presumed “magma chambers” can be very large.

One recent finding that may bear on these issues is the behavior of magmatic water in a temperature gradient. Huang et al. (GCA 2009) showed experimentally that andesite with 4 wt. % H₂O placed in temperature gradient from 950-350°C evolved to a granite at the cold end of the gradient. Lundstrom (GCA 2009) incorporated this into a top down sill injection model for granitoid assembly called thermal migration zone refining (TMZR). Both these works inferred that a hydrous interstitial peralkaline melt existed at temperatures <600°C to serve as transport medium for component migration. New experiments examining melt-fluid-crystal equilibria at <600°C and 0.5-2 kbars have been performed in cold seal vessels. A peralkaline glass (high Na₂O, K₂O and SiO₂ but 1 wt. % Al₂O₃) was synthesized and then added to platinum capsules with water, pre-fractured quartz and various feldspar components. An experiment (at 0.5 kbar and 330°C) that used Or₅₀Ab₅₀ glass produced albite and Kspar crystals as products indicating a fluid coexisting with a granitic assemblage. Other qtz + feldspar + melt experiments at 1 kbar produced qualitative observations suggesting an immiscibility boundary between peralkaline melt and water occurs between 400 and 600°C. Analyses of water contents of melt inclusions trapped in quartz at 400°C by confocal laser Raman spectroscopy indicate >15 wt % H₂O dissolved in the peralkaline melt; analyses of melt in the 600°C charge are forthcoming.

If zoned silicic mushes form by the TMZR process, the curious property of retrograde immiscibility between peralkaline melts and water could lead to a potentially important triggering mechanism for erupting silicic mush, thus providing a link between granitoids and ignimbrites.