Stability of tourmaline in pei aluminous granite systems: the boron cycle from anatexis to hydrothermal aureoles
European Journal of Mineralogy Volume 11 Number 2 (1999), p. 253 - 262
published: Apr 19, 1999
manuscript accepted: Nov 30, 1998
manuscript received: Mar 16, 1998
ArtNo. ESP147051102003, Price: 29.00 €
Abstract Peraluminous granites and pegmatites derived from the anatexis of metasedimentary rocks commonly possess elevated B contents, which are manifested as disseminated accessory tourmaline within the igneous bodies or as locally abundant tourmaline or other borates/borosilicates at their margins, in miarolitic cavities, within internal breccias, or in the host wallrocks. Because of the incompatibility of B in all but a few phases, the history of B accumulation and depletion in granites and pegmatites is difficult to establish from whole-rock chemistry. A new experimentally based model provides understanding of how tourmaline (principally schorl/dravite-olenite-foitite solid solutions) will behave during anatexis of metapelites, through the fractionation of peraluminous S-type melt, to the magmatic-hydrothermal transition in evolved melts. High B contents of melts at saturation in tourmaline (together with other AFM phases) imply that tourmaline should be completely consumed with > - 5 % melting of metapelites. Such melts are initially saturated in biotite, cordierite, etc., and the eventual saturation in tourmaline entails the back-reaction of earlier AFM phases with the B-enriched melt. The B content of melts will not normally remain buffered because of the limited quantity of AFM silicates to form tourmaline. Consequently, B increases to the end stages of melt fractionation. If Li content increases sufficiently, then elbaite may form. Whether late-stage elbaite crystallizes or not, most magmas do not conserve B. Much of the boron originally present in magma is lost to wallrocks, wherein B-precipitating reactions (of tourmaline, axinite, danburite, etc.) proceed.