GEMOC ARC National Key CentreTrevor H. Green, GEMOC Macquarie
Granitic magma series commonly show variation in K/Na ratio between different suites, possibly affecting the saturation level, at a given silica content, of accessory minerals such as apatite (phosphorus-saturation) or titanite or ilmenite (Ti-saturation). In turn, this will have important impact on trace element distribution in further fractionated products because of the ability of these minerals to take up rare earth elements (REE) (apatite, titanite) or high-field-strength elements (HFSE) (rutile, ilmenite). Direct determination of apatite and Ti-rich mineral solubility in silicic melts as a function of K content and pressure should allow assessment of the potential role of these accessory minerals in controlling trace element distribution in the crust, as governed by intra-crustal melting and granite evolution at different crustal depths.
Accordingly, the solubility of Ti and P-rich accessory minerals has been examined as a function of pressure and K/Na ratio in two series of highly evolved silicate systems. These systems correspond to (a) alkaline, varying from metaluminous to peralkaline with increasing K/Na ratio (b) strongly metaluminous (essentially trondhjemitic at the lowest K/Na ratio), and remaining metaluminous with increasing K/Na ratio (to 3).The experiments were conducted as far as possible at a fixed temperature of 1000°C, with water contents varying from 3-5% wt. at low pressure (0.5GPa), increasing through 5-10% wt. at 1-2.5 GPa to 10-15%wt. at 3.5 GPa. Pressure was extended outside the normal crustal range so that results would also apply to derivation of hydrous silicic melts from subducted oceanic crust.
For the alkaline composition series the Ti02 content of the melt at Ti-rich mineral saturation decreases with increasing pressure, and is little affected by K content until the melts are peralkaline, when Ti-saturation levels at fixed pressure increase as K content increases. The P205 content of the alkaline melts at apatite saturation increase with increasing pressure, but decrease with increasing K content (and peralkalinity). These contrasting results for P and Ti saturation levels point to contrasting behaviour of Ti and P in the structure of silicate melts. For the trondhjemite composition series the Ti02 content of the melt at Ti-rich mineral saturation also decreases with increasing pressure, but contrasts with the alkaline compositions, by showing decreased Ti-solubility with increasing K content, at fixed pressure. The P205 content of the trondhjemitic series melts at apatite saturation decrease with increasing K content. No consistent variation in apatite solubility with varying pressure was observed. Thus P and Ti do not show contrasting behaviour in the metaluminous trondhjemitic melts, differing from their behaviour in the alkaline melts, and reflecting different structural attributes in the different melt series. The results have application to A-type granite suites that are typically peralkaline, and to contrasting I-type metaluminous suites that frequently exhibit differing K/Na ratios from one suite to another.