Studio di nanograniti nei granati di metapeliti anatettiche (kinzigiti della zona Ivrea-Verbano e Migmatiti Himalayane)

Melt inclusions have been since their discovery a powerful tool to gain useful data on petrological studies, both magmatic and metamorphic. The work of this thesis focuses on the investigation of metapelitic Kinzigites from the Ivrea-Verbano Zone (Italy) and migmatitic paragneisses from Kaligandaki Valley (Central Nepal) and on the characterization of melt inclusions found in peritectic garnet from these rocks. In both cases inclusions are very small (<15 m), polycrystalline aggregates of quartz + feldspar + micas (“nanogranites”); no finding of glassy inclusions was observed. Quantitative analysis of the polycrystalline inclusions was difficult to obtain due to analytical limits and sample preparation difficulty; in fact inclusions were often mechanically removed during polishing. The final aim of this study was the re-homogenization of the inclusions to obtain the bulk composition of the trapped melt: this was undertaken using both a high temperature and room pressure stage and a high pressure piston-cylinder. Usually anatectic melt composition is inferred from migmatitic leucosomes, that are affected by fractional crystallization and differentiation phenomena that can change the original composition, or assumed using experimental analogues that cannot totally reproduce natural processes; therefore the study of glassy or re-homogenized nanogranitic inclusions found in peritectic minerals can be the only way to obtain the bulk composition of the melt during anatexis. Heating stage experiments were not successful, since the inclusions showed evidences of decrepitation and oxidation (probably due to volatile loss) but did not melt at all. The inclusions were successfully re-melted at 900° C and 8 Kbar using a piston cylinder apparatus, proving they are nanogranites trapped within garnet during incongruent melting reactions. The melt obtained shows in both cases a granitic composition clearly affected by interactions with the host garnet (decrepitation, growth of peritectic orthopiroxene) due to over-heating, thus the data obtained cannot be used for quantitative interpretation. New experiments should be performed in the future at lower temperature and higher pressure to find the chemical composition of the melt and its trapping temperature.