The interactions between M +5 cations (Nb +5, Ta +5, or P +5) and anhydrous haplogranite melts

Abstract

The interactions of high-field-strength cations with haplogranite melts of peraluminous to peralkaline compositions were examined by determining the saturation of rutile as functions of the concentrations of the oxides of the pentavalent cations Nb, Ta, and P. The phase equilibria combined with existing spectroscopic data show that the competition of the high-field-strength cations for the alkali aluminosilicate components varies systematically from peraluminous to peralkaline melts. The data also show that the melt species of Nb +5, Ta +5, and P +5 are very similar. In peraluminous melts, these pentavalent cations have a strong affinity for Al and form very stable M +5AIO 4 complexes; in peralkaline melts, the governing species are characterized by M +5OK bond types, whereas both M +5OAI and M +5OK species exist in the transition between peraluminous to peralkaline melts. The solubility of rutile is significantly affected by the hierarchy of interactions that exist among M +5 cations, Ti +4, and the aluminosilicate network. The composition of rutile in peraluminous melts is a solid solution along the join Ti 2O 4-M +5AIO 4 where M +5 is Ta and/or Nb. Nb +5 and Ta +5 partition strongly into rutile with Al +3 playing the role of a charge balancing cation, Nb +5 + Al +3 = 2Ti +4. Rutile in peralkaline melts contains significantly smaller concentrations of the M +5AlO 4 component, reflecting the much lower activity of M +5AlO 4 species. P +5 does not enter into solid solution in rutile because the ionic radius of P +5 is much smaller than Ti. Nb/Ta in peraluminous pegmatites also enter into solid solution in rutile by a coupled mechanism except that Fe +3 assumes the role of Al +3. The main geochemical conclusions are that Nb, Ta, and P have similar speciations in peraluminous and peralkaline melts. It follows that fractionation of these cations, particularly the perturbation of the Nb/Ta ratios in silicic magmas, can be brought about only by interactions with a crystalline phase. The Nb/Ta ratio can be substantially changed only through the crystallization of Ti-rich crystalline phases and/or Nb/Ta oxides such as columbite.