The effect of chlorine on the liquidus of basalt: First results and implications for basalt genesis on Mars and Earth

Chlorine (Cl) is abundant, to percent levels, on Mars, in Mars basalts, and in some Earth basalts. Yet, little is known about the effects of Cl on basalt phase equilibria, which provide crucial constraints on melting temperatures, melt compositions, and melt production. To explore the effects of Cl on basalt equilibria, we located the liquidus (at high P and T) for a Martian basalt composition (the rock Humphrey, Gusev Crater, Mars) with 0.7% Cl-added, and obtained mineral and melt compositions from the experimental charges. Addition of Cl produces an unexpectedly large change in the liquidus—it is shifted to lower temperature by ~ 50 °C, and the field of pigeonite stability is enlarged so that the point of liquidus cosaturation in olivine & pigeonite is shifted down pressure by ~ 4 kbar (from 12.5 kbar and 1355 °C in the Cl-free composition to 8.5 kbar and 1305 °C). This temperature shift is comparable to that produced by addition of ~ 0.8% H2O; so, on a molar basis, Cl is twice as effective as H2O in reducing this basalt's liquidus. Conceptually, the shift in liquidus temperature and enlargement of the field of pigeonite + melt is consistent with formation of complexes between Cl and network-modifying cations (e.g., Mg, Fe, Ca), which would make the latter unavailable to modify (depolymerize) the silicate network. However, the actual effect of Cl is larger than predicted by this simple model, even allowing up to 8 cations complexed per Cl. For Mars, these results suggest that Cl may play a crucial role in its basalts' generation and evolution. While, some aspects of Martian basalt petrogenesis have been consistent with experimental results on water-rich systems (e.g., to ~ 2% H2O in basalt magma); the work here suggests that similar experimental results would be obtained for Cl-rich, H2O-poor systems, and that Martian basalts would contain little water. For Earth, the work here suggests that Cl must be considered explicitly in the petrogenesis of Cl-rich magmas, like those of subduction zones.