Skip to main content

Experimental constraints on reactive flow in the lunar mantle and high-Ti magma petrogenesis

Basaltic lunar mare volcanism has erupted a variety of melt compositions that far exceeds that of terrestrial basalts. Most notably, TiO2 contents of lunar basalts vary by almost two orders of magnitude and can reach up to 18 wt.% in red/black primitive pyroclastic glasses [1]. Such high Ti contents point to a role for ilmenite-bearing cumulates (IBC), which are predicted to form near the end of solidification of the lunar magma ocean [2]. No experimental study investigating the melting of pure IBC or mixtures of IBC and peridotite [e.g., 3] has, however, succeeded in reproducing the composition of high-Ti lunar basalts.

We propose that lunar high-Ti basalts originate as IBC partial melts, but have their composition modified through reactive flow in the lunar mantle. We simulate this process by experimentally reacting an IBC partial melt composition [3] with high-Mg# olivine and orthopyroxene crystals at 1400-1500 °C and 1.5 GPa in a piston cylinder press. This leads to a peritectic reaction where orthopyroxene undergoes dissolution-reprecipitation to form a more Ca- and Ti-rich rim, in addition to Fe-Mg exchange between IBC melt and minerals. The composition of the reacted melt shifts to lower Mg# and overlaps with that of high-Ti basalts. No growth of ilmenite or clinopyroxene is observed, thus satisfying the condition that high-Ti melts lack these phases at high-pressure multiple saturation.

[1] Delano (1986) JGR 87 A171–A181; [2] Ringwood & Kesson (1976) LPSC Proc. 1697–1722; [3] van Orman & Grove (2000) MAPS 35 783–794.


Martijn Klaver1, Tim Elliott2, Stephan Klemme3
1WWU Münster, Germany;University of Bristol, United Kingdom; 2University of Bristol, United Kingdom; 3WWU Münster, Germany
GeoMinKöln 2022