Textural and Geochemical Evidence for Magnetite Production upon Antigorite Breakdown During Subduction

Joana F Vieira Duarte,Francesca Piccoli,Thomas Pettke,Jörg Hermann

doi:10.1093/petrology/egab053

Abstract

AbstractMagnetite stability in ultramafic systems undergoing subduction plays a major role in controlling redox states of the fluids liberated upon dehydration reactions, as well as of residual rocks. Despite their relevance for the evaluation of the redox conditions, the systematics and geochemistry of oxide minerals have remained poorly constrained in subducted ultramafic rocks. Here, we present a detailed petrological and geochemical study of magnetite in hydrous ultramafic rocks from Cerro del Almirez (Spain). Our results indicate that prograde to peak magnetite, ilmenite–hematite solid solution minerals, and sulfides coexist in both antigorite-serpentinite and chlorite-harzburgite at c. 670 °C and 1·6 GPa, displaying successive crystallization stages, each characterized by specific mineral compositions. In antigorite-serpentinite, magnetite inherited from seafloor hydration and recrystallized during subduction has moderate Cr (Cr2O3 &lt; 10 wt%) and low Al and V concentrations. In chlorite-harzburgite, polygonal magnetite is in textural equilibrium with olivine, orthopyroxene, chlorite, pentlandite, and ilmenite–hematite solid solution minerals. The Cr2O3 contents of this magnetite are up to 19 wt%, higher than any magnetite data obtained for antigorite-serpentinite, along with higher Al and V, derived from antigorite breakdown, and lower Mn concentrations. This polygonal magnetite displays conspicuous core to rim zoning as recognized on elemental maps. Cr–V–Al–Fe3+ mass-balance calculations, assuming conservative behavior of total Fe3+ and Al, were employed to model magnetite compositions and modes in the partially dehydrated product chlorite-harzburgite starting from antigorite-serpentinite, as well as in the serpentinite protolith starting from the chlorite-harzburgite. The model results disagree with measured Cr and V compositions in magnetite from antigorite-serpentinites and chlorite-harzburgites. This indicates that these two rock types had different initial bulk compositions and thus cannot be directly compared. Our mass-balance analysis also reveals that new magnetite formation is required across the antigorite-breakdown reaction to account for the mass conservation of fluid-immobile elements such as Cr–V–Al–Fe3+. Complete recrystallization and formation of new magnetite in equilibrium with peak olivine (Mg# 89–91), chlorite (Mg# ∼95), orthopyroxene (Mg# 90–91), and pentlandite buffer the released fluid to redox conditions of ∼1 log unit above the quartz–fayalite–magnetite buffer. This is consistent with the observation that the Fe–Ti solid solution minerals (hemo-ilmenite and ilmeno-hematite) crystallized as homogeneous phases and exsolved upon exhumation and cooling. We conclude that antigorite-dehydration reaction fluids carry only a moderate redox budget and therefore may not be the only reason why the magmas are comparatively oxidized.

Highlights

Magnetite is a very common accessory mineral forming upon hydrothermal alteration of peridotite
A crucial point that so far has never been addressed in subducted abyssal serpentinites is that, depending on the protolith, magnetite abundance inherited from oceanic alteration can vary widely, ranging from more than 6 wt% to nearly zero (Whittaker & Wicks, 1970; Seyfried et al, 2007; Evans, 2008; Beard et al, 2009; Andreani et al, 2013; Klein et al, 2014; Bonnemains et al, 2016)
We investigated in great detail the petrological and geochemical systematics of magnetite and ilmenite–hematite solid solution formed upon ocean-floor hydration and prograde subduction metamorphism, which discloses important insight into mineral growth during continued metamorphism

Summary

Introduction

Magnetite is a very common accessory mineral forming upon hydrothermal alteration of peridotite (i.e. serpentinization; Cannat et al, 1992; Cannat, 1993). A crucial point that so far has never been addressed in subducted abyssal serpentinites is that, depending on the protolith, magnetite abundance inherited from oceanic alteration can vary widely, ranging from more than 6 wt% to nearly zero (Whittaker & Wicks, 1970; Seyfried et al, 2007; Evans, 2008; Beard et al, 2009; Andreani et al, 2013; Klein et al, 2014; Bonnemains et al, 2016). It is impossible to know a priori the initial magnetite content of metamorphic serpentinites and it is impossible to state if magnetite is consumed by relying only on its abundances in rocks of different metamorphic degrees from various localities Another possibility that has not been explored yet is if magnetite could be formed during dehydration reactions involving antigorite and chlorite that hosts some ferric iron Another possibility that has not been explored yet is if magnetite could be formed during dehydration reactions involving antigorite and chlorite that hosts some ferric iron (e.g. Evans et al, 2012; Debret et al, 2015; Evans & Frost, 2021)

Methods

Results

Conclusion