Secondary nucleation at crystal surfaces as a potential mechanism for in situ nucleation and growth of crystals at magma chamber margins

There is a large body of field, textural and chemical evidence from mafic layered intrusions indicating that crystals predominantly nucleate and grow in situ, i.e., at solidification fronts along the margins of evolving magma chambers. Campbell was the first to provide an explanation for this fundamental mode of crystallization in which new nuclei appear on existing crystals along chamber margins, a process he referred to as ‘heterogeneous self-nucleation’. Holness et al. (2023) have recently argued, however, that this nucleation mechanism is implausible because continued growth of existing crystals is more kinetically favourable than nucleation of new grains of the same phases. As an alternative explanation, Holness et al. (2023) have returned to the classical settling model in which crystals nucleate and grow throughout the entire magma chamber. Massive chromitites in the Bushveld Complex then arise due to gravity-induced settling of chromite clusters from the convecting resident melt on the chamber floor, with formation of the clusters being due to ‘synneusis’ – random collision of suspended grains leading to adherence. This mechanism is, however, inconsistent with existing field, textural and chemical observations from massive chromitites and other igneous rocks of the Bushveld Complex. Most of these observations are indicative of the rock forming via some form of repeated nucleation and growth of new crystals on existing ones at the chamber floor. A challenge is thus to identify the mechanism of in situ nucleation and growth of crystals along the margins of layered intrusions that would be reconcilable with the ground-truth observations. In this study we suggest that the most likely candidate is secondary nucleation caused by seed crystal surfaces - a process that has never been invoked in igneous petrology - but plays a key role in the formation of crystals in industrial crystallizers. During this process, Van Der Waal’s attractive forces in the vicinity of a crystal face can stabilize new nuclei of the same phase in proximity of the original crystal. We present experimental evidence of growth of chromite clusters from chromite-saturated mafic magma in support of this argument. Since secondary nucleation is induced by pre-existing parent crystals acting as catalysts for further nucleation, one may logically expect it to actively operate along the crystal-rich solidification fronts at the floors, walls, and roofs of evolving magma chambers.