ASSESSING SILICATE-LIQUID-IMMISCIBILITY IN INTERMEDIATE TO FELSIC SYSTEMS: EXAMPLE FROM THE RAFTSUND INTRUSION, VESTERÅLEN-LOFOTEN, NORTHERN NORWAY

Silicate-liquid immiscibility has gained acceptance in mafic tholeiitic systems in the last decade; however, the significance of this process in intermediate and felsic magmas remains unresolved. Here, we present new data suggesting that immiscibility could be more common than previously thought, especially in intermediate, alkalic ferroan systems. The Raftsund intrusion is a 1800 Ma monzonitic to granitic, alkalic ferroan batholith that forms part of the Lofoten-Vesterålen anorthosite-mangerite-charnockite-granite (AMCG) suite. The study focuses on the main unit of the intrusion, varying from augite-fayalite monzonite to pigeonite-augite syenite. The syenite part contains scattered Fe–Ti–P-rich rocks from cm-scale up to 50 x 200 m pods. The contact between the Fe-rich rocks and the host syenite is sharp or gradational. Fe–Ti–P-rich rocks are composed of euhedral apatite, augite, Fe-rich olivine, titanomagnetite, ilmenite and partly resorbed ternary feldspars locally surrounded by plagioclase films. Whole-rock chemistry indicates that the Fe–Ti–P-rich rocks are enriched in Mg, Ca and transition metals Sc, Zn and Mn and REE compared to the host syenite. Augite from the Fe–Ti–P-rich rocks is enriched in Ti and Sc, suggesting that the whole-rock geochemical anomalies are not only related to the accumulation of mafic minerals but that the minerals grew from a melt enriched in these elements. Furthermore, both augite and titanomagnetite in the Fe–Ti–P-rich rocks are enriched in Al, indicating that feldspar was not stable. This interpretation is consistent with the presence of partly resorbed ternary feldspar. Thin plagioclase films, which show significant Fe enrichment (0.1–0.3 wt% FeO) compared to the resorbed ternary feldspar (<0.12 wt% FeO) they are in contact with, crystallized from an Fe-rich melt. Mineral chemistry is not consistent with simple accumulation of mafic minerals but instead points to the presence of an Fe-rich melt enriched in many trace elements, consistent with liquid immiscibility. This study shows that liquid immiscibility is a viable process in magmatic systems of intermediate composition and that it can be tested using a combination of whole-rock and mineral chemistry and microtextural analyses.