Phase relations in peralkaline Cl- and F-rich phonolitic melts
Abstract
High concentrations of Cl and F are known to affect phase relations in magmatic rocks. In this study, phase equilibrium experiments were conducted using a peralkaline iron-rich phonolitic starting composition with variable Cl and F content. The starting material represents the composition of a dyke rock, which may resemble a potential parental melt composition of the peralkaline Ilímaussaq plutonic complex/South Greenland (Marks & Markl 2003). Recent phase equilibrium experiments reproduced the phenocryst mineral assemblage of the dyke at 850 - 800 °C at low fO2 and nominally dry conditions (Giehl et al. 2013). The influence of P (100 - 150 MPa), H2O content (nominally dry - H2O saturated) and fO2 (ΔFMQ -4 to +1) was investigated from 1000 to 650 °C in an internally heated argon pressure vessel and in hydrothermal rapid-quench cold seal pressure vessels. To cover the large T interval of crystallization we applied a three-step fractional crystallization strategy. Stable mineral phases are titano magnetite, fayalitic olivine, clinopyroxene (Cpx), alkali feldspar, nepheline and aenigmatite (× native iron) coexisting with residual melt (Fig. 1). High Cl and F concentrations additionally stabilize fluorite (CaF2), hiortdahlite ((Ca,Na)6(Zr,Ti)2Si4O14F4), sodalite (Na8Al6Si6O24Cl2) and eudialyte (Na4Ca2(Fe,Mn)ZrSi8O22(Cl,OH)2). Our experiments show that Na/(Na+Ca) in Cpx is mainly controlled by the Na/(Na+Ca) melt ratio. In the Ilímaussaq rocks, magmatic Cpx shows a gap between Ca-rich and Na-rich compositions. This gap can be explained by increasing F content in the residual melt: Ca-rich Cpx is destabilized at the expense of Ca- and F-rich hiortdahlite and fluorite. At lower T, Na-rich Cpx is restabilized due to the increasing Na/(Na+Ca) melt ratio as the stability of Na-rich Cpx is unaffected by high F melt concentrations. Partitioning of Mn between clinopyroxene and eudialyte and between clinopyroxene and aenigmatite can be fitted as a linear functions of T from 800 to 650 °C. These ratios may have the potential for geothermometers, rarely available for peralkaline phase assemblages: log Kd (Mn, Eud/Cpx) = 0.376 * T - 3.858 (n = 7, R2 = 0.94) log Kd (Mn, Ae/Cpx) = 0.292 * T - 2.715 (n = 6, R2 = 0.87) where T = 10000/T (K) and n is the number of experiments used for the fit. References: Giehl C, Marks M, Nowak M (2013) Contrib Mineral Petr 165: 283-304 Marks M, Markl G (2003) Mineral Mag 67: 893-919 Fig. 1: Experimental products at 100 MPa, 700 °C, hydrated starting glass (backscattered electron image): clinopyroxene (Cpx), aenigmatite (Ae), eudialyte (Eud), alkali feldspar (Afs) and residual glass (Gl).
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.V51C2677G
- Keywords:
-
- 3630 MINERALOGY AND PETROLOGY Experimental mineralogy and petrology;
- 3640 MINERALOGY AND PETROLOGY Igneous petrology;
- 3612 MINERALOGY AND PETROLOGY Reactions and phase equilibria;
- 3651 MINERALOGY AND PETROLOGY Thermobarometry