Vasyukova and Williams-Jones reported in 2014 that pegmatites and adjacent subsolvus granites (two alkali feldspars) of the Mid-Proterozoic Strange Lake pluton (Que´bec Labrador, Canada) host potentially economic concentrations of high field strength elements (HFSE), including the rare earth elements (REE), zirconium and niobium. Previous studies have proposed that these concentrations were the result of a combination of extreme fractionation of a peralkaline granitic magma and later hydrothermal remobilization. However, the recent discovery of melt inclusions which, after heating to 900 or 950 C, quench to immiscible fluoride and silicate glasses, suggests
an additional mechanism of HFSE concentration.
Crystallized melt inclusions are common in quartz phenocrysts of the early hypersolvus and transsolvus granite. After heating, four types of inclusions were discerned. Type 1 inclusions quench to a single silicate glass containing a high concentration of Zr, Nb and Ti. Quenching of Type 2 inclusions produces a Zr, Nb, Ti-rich silicate glass containing a globule of calcium-rich fluoride glass with a high concentration of the REE. The third inclusion type is similar to Type 2, except that the calcium-rich fluoride glass contains a globule of REE-fluoride glass (up to 50 wt.% total REE). Type 4 inclusions contain calcium-fluoride glass with multiple silicate globules.
The authors proposed that during or soon after emplacement, the early granitic magma exsolved a calcium fluoride melt into which rare earth elements (REE) partitioned preferentially. The conjugate silicate melt was consequently depleted in Ca and the REE, and enriched in Zr, Nb and Ti. Crystallization of the fluoride melt occurred late in the crystallization history of the silicate magma allowing it to accumulate in the volatile-rich residual magma that formed the pegmatites. This played a major role in the extreme enrichment of the pegmatites in Ca, F and REE. Crystallization of the pegmatites proceeded inwards from an outer zone in which feldspars, quartz, arfvedsonite and zirconosilicate minerals dominate to a core where the immiscible fluoride liquid collected and crystallized fluorite and REE minerals. Fluid exsolved from the silicate melt reacted with the products of both the silicate and fluoride melts, remobilizing most of the HFSE including the REE. This study reports a rare example of silicate fluoride melt immiscibility and the first in which such immiscibility played a role in concentrating the REE to potentially economic levels.
The authors have documented that a rare example of natural fluoride–silicate melt immiscibility and the first example in which exsolution of a fluoride melt has provided a new means of metal concentration. Analogous to the sulfide liquids that are responsible for the formation of magmatic Ni-Cu-PGE deposits, this liquid not only preferentially partitioned the REE from its conjugate silicate liquid but continued to scavenge REE from the residual silicate melts eventually concentrating the REE to potentially economic levels in the pegmatite cores.
Immiscibility of the type described by the authors will likely continue to be a rarely observed phenomenon. Further examples will most likely be found in lithium-rich systems, as the presence of lithium serves to reduce the concentration of fluorine required to produce an immiscible fluoride melt. In such systems, as at Strange Lake, the Ca-fluoride melt likely evolved by exsolving a REE-fluoride melt, potentially further enriching the REE, and the conjugate silicate melt will concentrate HFSE like Zr and Nb.