The relationship between the Fe3+/FeT of MORB and the Fe3+/FeT of the solid, convecting, upper mantle is unknown yet critically constrains geophysical observations and mantle fO2 as a function of depth, temperature, lithology, and time. Natural observations and experimental determinations of Fe3+ partitioning between basalts and peridotite residues suggest that as melting proceeds in the spinel stability field, mineral chemistry and mode in the peridotite may evolve such that the system maintains approximately constant melt Fe3+/FeT, and hence constant residue fO2 [1]. To accurately model the evolution of the rock-melt system and project the composition of peridotite back along the melting column to infinitesimal melt fractions will require quantification of mineral chemistries (including Fe3+/FeT) as a function fO2, P, and T. We equilibrated silicate melts and mineral assemblages of olivine, orthopyroxene, and spinel, ± clinopyroxene, over a range of fO2 and spinel Cr# at 1 atmosphere and 1.5GPa. We rigorously quantify fO2 – and check for system equilibrium - using multiple barometers (glass, mineral, alloy). Spinel oxybarometry and glass Fe3+/FeT both record the measured furnace (or calculated FePt capsule) fO2. increases as a function of spinel Fe2O3 [2] and decreases as a function of temperature [3]. New 1 atmosphere experiments show that increases by a factor of 2 to 2.5 as spinel Cr# increases from approximately 0.18 to about 0.65. Because average spinel Cr# increases as a function of mantle potential temperature, we predict that spinel Cr# and Tp will exert competing effects on during MORB generation and modulate observed variations in MORB Fe3+/FeT as a function of extent of melting and Tp [4].

[1] Birner et al., EPSL (2021) EPSL 566: 116951.

[2] Davis and Cottrell, (2018) Am Min 103.7: 1056-1067.

[3] Davis and Cottrell, (2021) CtMP 176.9: 67.

[4] Cottrell and Kelley, (2011) EPSL 305.3-4: 270-282.