THE ORIGIN OF CHROMITE SEAMS IN THE RHUM COMPLEX, SCOTLAND: A PETROGRAPHIC AND PHASE DIAGRAM INTERPRETATION

The Rhum Complex, a well-known layered mafic igneous complex, contains chromite seams; a potential source of PPGEs. We present a petrographic analysis of a troctolite, chromitite, and anorthosite layers from Rhum – captured in a thin section. Olivine (OL) orthocumulate troctolite consists of seriate, subhedral-anhedral, cumulus Ol (<5mm) in a loose framework of touching grains, sparse cumulus oxide, and intercumulus plagioclase (Pl). The orthocumulate chromitite layer, with intercumulus Pl, is sinusoidal and of variable thickness (1.25-1.68mm). Along the contact, Ol are tabular, long axes parallel to the chromitite contact. Ol in contact with intercumulus chromitite Pl is extensively scalloped and embayed with cumulus chromite nestled in embayments. The opposing side of the chromitite is an anorthosite with three subzones: At the contact with chromitite, adcumulate tabular Pl crystals are imbricated, transitioning to a planar fabric, and finally a Pl mesocumulate with intercumulus CPX. In the ternary system Ol-Di-An the troctolite crystallization sequence Ol→Cr→Pl (no CPX) constrains the isopleth to the Fo-Liq field, likewise the anorthosite Pl→Sp→CPX (no Ol) isopleth is constrain to the An-Liq field, requiring two distinct parent magmas. We suggest injection of a magma containing liquidus Pl dissolved the cumulate Ol creating a “karsted” surface. Addition of “Ol” component into the adjacent melt shifted its bulk composition into the Spinel-Liq field resulting in crystallization of cumulus chromite. Imbrication of cumulus Pl above this contact is due to flow, with thicker accumulation of placer chromite in troughs rather than crests of the “unconformable surface”. Planar Pl above this contact may record flow or compaction, with compaction decreasing away from the contact. Debate on the formation of chromite due to replenishments pivots on the origin of the anorthosite either due to reaction between troctolite and primitive picrite (Kaufmann et al., 2020) or a primary replenishment (Latypov et al., 2020). Our interpretations are consistent with a locally dynamic interface between an older dissolving troctolite and a younger flowing high-alumina melt with liquidus plagioclase resulting in the formation of thin chromite seams on the thermal-chemically eroded “unconformable” chamber floor.