Evolution of the lithospheric mantle in NE Angolaimplications for diamond exploration

  1. Castillo Oliver, Montgarri
Dirigida por:
  1. Salvador Galí Medina Director/a
  2. Joan Carles Melgarejo Draper Director/a
  3. Suzanne Y. O'Reilly Codirector/a
  4. William L. Griffin Codirector/a

Universidad de defensa: Universitat de Barcelona

Fecha de defensa: 30 de septiembre de 2016

Tribunal:
  1. Joan Martí Molist Presidente/a
  2. José Ignacio Gil Ibarguchi Secretario/a
  3. Alan Lynton Jaques Vocal

Tipo: Tesis

Teseo: 459514 DIALNET

Resumen

Angola is currently the 8th largest diamond producer in the world, but for historical reasons the subcontinental lithospheric mantle (SCLM) in the Congo-Kasai craton is still poorly studied. However, an understanding of the structure, composition and evolution of this SCLM, as well as the properties of the kimberlitic magmas themselves, is essential for the future of diamond exploration. A detailed textural and compositional study of groundmass perovskite (CaTiO3) in 6 kimberlitic intrusions in the Lunda Norte province (NE Angola) revealed a complex, multi-stage process that involved mingling of compositionally different melts. U-Pb dating of perovskite yielded Lower Cretaceous ages (116-133 Ma), which suggests that kimberlite magmatism resulted from reactivation of deep-seated translithospheric faults (>300km) during the break-up of Gondwana. A layered structure of this SCLM was inferred from the characterisation of mantle xenoliths and xenocrysts sampled by the Cat115 and Catoca kimberlites. The layering is defined by two distinct garnet-peridotite suites, separated by a zone rich in low-Mg eclogite at ~150km depth. Two origins have been proposed to explain the current structure of the SCLM (stacking subduction vs plume upwelling), although more data are required to decide which is the most likely explanation. Complex metasomatic processes have modified the composition of an originally depleted SCLM. In the model proposed in this thesis, the shallow SCLM was firstly affected by silicate melt metasomatism, which resulted in the introduction of garnet (± clinopyroxene). Subsequently, carbonatite-silicate melts refertilised the deep SCLM, introducing both clinopyroxene and garnet. Percolative fractional crystallisation of these melts would explain the late formation -or cryptic metasomatism- of clinopyroxene with a carbonatitic signature, as well as the crystallisation of phlogopite, in the shallower layer of the SCLM, shortly prior to kimberlite eruption. Finally, interaction with the kimberlitic magma led to the crystallisation of a second generation of phlogopite and fine- grained pyroxene along fractures and grain boundaries. The results presented in this thesis highlight the complexity of the processes that took place in the SCLM in NE Angola and how these may have modified the composition of the diamond indicator minerals (ilmenite, clinopyroxene and garnet) sampled by the kimberlites. This is especially relevant in the case of ilmenite, since interaction with a kimberlitic magma may lead to misinterpretation of the diamond potential of a prospective kimberlite if textures and trace-element composition are disregarded.