GSA Annual Meeting in Phoenix, Arizona, USA - 2019

Paper No. 136-11
Presentation Time: 4:25 PM

THREE-DIMENSIONAL NETWORK OF CHROMITE CRYSTALS IN LAYERED INTRUSIONS AS EVIDENCE FOR SELF-NUCLEATION AND IN SITU GROWTH IN BASALTIC MAGMA CHAMBERS


LATYPOV, Rais1, CHISTYAKOVA, Sofia1, RADERMACHER, Viktor2 and JAKATA, Kudakwashe2, (1)School of Geosciences, Wits University, Private Bag 3, Johannesburg, 2050, South Africa, (2)Evolutionary Studies Institute, Wits University, Private Bag 3, Johannesburg, 2050, South Africa

The massive chromitites of the Bushveld Complex in South Africa occur as monomineralic layers up to a few metres thick and hundreds of kilometres in length. We have undertaken a textural study of one of these layers (UG1 chromitite). It consists of up to 40-50 vol.% of small chromite grains (<1 mm in size) that are hosted by large oikocrysts of plagioclase (up to 5-10 cm in size). Traditionally, such chromitite layers are attributed to gravity-induced deposition of chromite crystals on the chamber floor with the subsequent growth of plagioclase oikocrysts from the interstitial liquid. Since chromite is almost twice as dense as a basaltic melt (4.8 g/cm3 versus 2.6 g/cm3), it is expected to settle to the floor of the chamber. However, chromite mainly appears in thin sections as individual grains separated from each other by large ‘empty’ spaces occupied by plagioclase oikocrysts. One challenge that arises from the gravity-settling model is its inability to explain why settling grains of chromite failed to reach the chamber floor despite being much denser than parental melt. To address this discrepancy, we have visualised multiple chromitite specimens using X-ray computed tomography and discovered that chromite grains do not actually occur as separate grains but rather form a continuous three-dimensional network of interconnected crystals. The formation of such delicate three-dimensional aggregates is difficult to reconcile with gravity-induced settling of chromite crystals and is best explained by their nucleation and growth directly at the magma-cumulate interface. This occurs because self-nucleation or heterogeneous nucleation on pre-existing crystals at the chamber floor is more favourable due to energy considerations compared to homogenous nucleation in the main magma body. Our study provides strong support for the concept that basaltic magma chambers predominantly develop by in situ crystallization along the cooling margins, with little to no contribution from crystal settling.