We present the results of high pressure, high temperature multiple saturation experiments at variable oxygen fugacity $(f_{O_2})$ conditions (IW+1.5 and IW−2.1) on three lunar high titanium ultramafic glasses: the Apollo 17 Orange glass (A17O, 9.1 wt% TiO₂), the Apollo 15 Red glass (A15R, 13.8 wt% TiO₂), and the Apollo 14 Black glass (A14B, 16.4 wt% TiO₂). We performed experiments in graphite ($f_{O_2}$ = IW+1.5) and iron ($f_{O_2}$ = IW−2.1) capsules. The experimentally determined multiple saturation points (MSPs) in graphite capsules are 2.5 GPa and ∼1530 °C (A17O), 1.3 GPa and ∼1350 °C (A15R), and 1.55 GPa and ∼1425 °C (A14B). In iron, we found MSPs of 3.3 GPa and ∼1565 °C (A17O), 2.8 GPa and ∼1490 °C (A15R), and 4.0 GPa and ∼1540 °C (A14B). These results, when combined with previous experiments on the lunar ultramafic glasses, indicate that the increase in the pressure of multiple saturation is linearly proportional to the TiO₂ content of the melt ($\mathrm{\Delta }MSP\left(GPa\right)=0.14\times Ti{O}_2(wt\%)-0.15$, R² = 0.93, RMSE = 0.2 GPa). The high depths of melting correlated with the lowest $f_{O_2}$ conditions are hard to reconcile with buoyancy constraints on these iron and titanium rich magmas. In addition, measurements of $f_{O_2}$ on the orange glass as well as the presence of iron blebs in the glasses suggest that the glasses were reduced during eruption. To reconcile buoyancy constraints with $f_{O_2}$ estimates, we present a model in which the high titanium magmas experienced higher $f_{O_2}$ conditions at their source, but underwent subsequent reduction at shallow depths (4–52 km) just prior to their eruption. In this model, we can then further bracket the depth of melting to be from the minimum multiple saturation pressure in graphite to the deepest depth at which the magmas are buoyant: assuming the Hess and Parmentier (1995) post overturn cumulate mantle, the depths of melting range from ∼550–770 km for the A17O glass, ∼260–490 km for the A15R glass, and ∼320–350 km for the A14B glass.

我们展示了可变氧逸度下的高压、高温多重饱和实验的结果$(F_{{○}_2})$三种月球高钛超镁铁质玻璃的条件（IW+1.5 和 IW-2.1）：Apollo 17 橙色玻璃（A17O，9.1 wt% TiO ₂）、Apollo 15 红色玻璃（A15R，13.8 wt% TiO ₂）和Apollo 14 黑色玻璃（A14B，16.4 wt% TiO ₂）。我们在石墨中进行了实验（$F_{{○}_2}$ = IW+1.5) 和铁 ($F_{{○}_2}$ = IW-2.1) 胶囊。实验确定的石墨胶囊中的多个饱和点 (MSP) 为 2.5 GPa 和 ∼1530 °C (A17O)、1.3 GPa 和 ∼1350 °C (A15R) 以及 1.55 GPa 和 ∼1425 °C (A14B)。在铁中，我们发现 MSP 为 3.3 GPa 和 ∼1565 °C (A17O)、2.8 GPa 和 ∼1490 °C (A15R)，以及 4.0 GPa 和 ∼1540 °C (A14B)。这些结果与之前对月球超镁铁玻璃的实验相结合，表明多重饱和压力的增加与熔体中的 TiO ₂含量成线性比例（$\mathrm{\Delta }米\mathrm{小号}磷\left(G磷\mathrm{一个}\right)=0.14\times 吨\mathrm{一世}{○}_2(w吨\%)-0.15$, R ²  = 0.93, RMSE = 0.2 GPa)。熔化深度与最低熔化深度相关$F_{{○}_2}$这些条件很难与这些富含铁和钛的岩浆的浮力限制相协调。此外，测量$F_{{○}_2}$橙色玻璃上以及玻璃中铁泡的存在表明玻璃在喷发期间减少了。协调浮力约束与$F_{{○}_2}$估计，我们提出了一个模型，其中高钛岩浆经历了更高的$F_{{○}_2}$在它们的源头条件下，但在它们喷发之前在浅层深度（4-52公里）进行了随后的减少。在这个模型中，我们可以进一步将熔融深度划分为从石墨中的最小多重饱和压力到岩浆漂浮的最深深度：假设 Hess 和 Parmentier（1995）倾覆后的累积地幔，深度A17O 玻璃的熔化范围约为 550-770 公里，A15R 玻璃的熔化范围约为 260-490 公里，A14B 玻璃的熔化范围约为 320-350 公里。