Abstract:
The unusual elemental enrichment typical of many carbonatites and their stable and radiogenic isotope signatures---which are unlike those of sedimentary limestones---forced researchers to abandon limestone melting theories in the early 1960s and to support mantle-related models of carbonatite genesis. However, the fluid compositions [CO2/(H2O + CO2) = 0.05] required to melt limestone near its eutectic in the CaO-MgO-CO2-H2O system (600 675 °C) are virtually identical to those found in infiltrative magmatic-hydrothermal, skarn-forming systems; therefore, carbonates within such systems would melt via volatile fluxing. Skarn-related decarbonation reactions produce the CO2 required to form the carbonic acid (H2CO3) in the infiltrative H2O-rich fluid essential to carbonate melting. In addition to H2O, other fluxes (HF, HCl, H3PO4) and related salts derived from fluid-phase saturation of silicate intrusions could further depress the carbonate-melting eutectic temperatures, as well as enhance mass transfer of mineralizing elements into a forming skarn system and any low-viscosity carbonate melts produced within the skarn. The isotopic signatures of the resultant carbonate melts should reflect the elemental mass transfer of constituents from the intrusion, as well as Rayleigh decarbonation and elemental mixing processes typical of contact-metasomatic (pneumatolytic) processes. Many intrusions exsolving volatiles into limestone during final stages of solidification should produce some carbonate melt. Only carbonatites with enrichments in F, P, Sr, Nb, U, Th, and rare-earth-elements, have been considered intrusive melts, whereas the solidified products of other melts may have been erroneously considered hydrothermal veins.
