Experimental evidence for a shallow cumulate remelting origin of lunar high-titanium mare basalts - Dr. Yanhao Lin

Lunar high-Ti mare basalts are among the first lunar samples returned from human space exploration missions. Understanding the origin of lunar mare basalts is crucial for advancing our knowledge of the Moon’s formation, evolution, and the relationship between the Earth-Moon system. However, key details regarding the formation and origin of lunar high-Ti basalts remain unclear. Recently, Dr. Yanhao Lin’s research team at HPSTAR (the Center for High Pressure Science and Technology Advanced Research) conducted new research using experimental petrology methods to investigate the origin of lunar high-Ti mare basalts. Their findings propose that these basalts originated from an ilmenite-bearing cumulate layer (IBCL). This study, titled "Experimental evidence for a shallow cumulate remelting origin of lunar high-titanium mare basalts," was published in Communications Earth & Environment. The first author of the study is Shuchang Gao, a joint master student at HPSTAR and the China University of Geosciences (Beijing).

Lunar mare basalt samples, primarily returned from the Apollo and Chang’e missions, show significant differences from the  Earth basalts in both composition and formation age. The most striking contrast lies in their titanium dioxide (TiO₂) content (Fig. 1). Earth’s mid-ocean ridge basalts (MORB) typically have a narrow range of TiO₂ content, averaging about 1.6 wt.%. In contrast, lunar mare basalts as well as ultramafic glass beads exhibit a wide range of TiO₂ contents, which is used as a basis for subdivision into three major groups (i.e.,<1 6="" very="" low-ti="" or="" vlt="">6 wt.% = high-Ti basalts). The very high abundances of titanium in the high-Ti mare basalts require involvement of one or more titanium-rich subsurface mineral phases. Ilmenite, armalcolite, and rutile appear to be the only lunar minerals sufficiently rich in Ti to provide a viable source for the Ti in these samples. The formation, chemical composition, and dynamic behavior of lunar subsurface reservoirs containing these TiO₂-rich minerals in the context of the LMO model have been studied extensively over the past decades. Nevertheless, key aspects of the formation of the high-Ti mare basalts, including the composition of their source, their depth(s) of origin, as well as the process(es) that caused partial melting in the first place, have not been resolved.

Fig. 1 . The TiO₂ versus Mg# (molar Mg/(Mg + Fe)) of lunar basalts and average mid-ocean ridge basalts (MORB).

Here, we show the results of a series of partial melting experiments at low pressure (~0.4 GPa) and a range of temperatures (1050–1350 °C) on three bulk compositions. All three are based on LMO crystallization step LBS10 from previous experimental study. This is the first step that yielded ilmenite and thus an IBCL. Varying amounts of plagioclase were added to the LBS10 composition to assess the effect of inefficient plagioclase extraction during the late stages of LMO crystallization. Experimental results were augmented with thermodynamic modeling of partial melting of the second ICBL formed during later LMO crystallization step LBS11. Our results are compared with the major and trace element compositions, as well as isotopic systematics, of high-Ti mare basalt samples to further assess the source depth and formation mechanism of these enigmatic rocks.

The representative TiO₂ content ranges (9–13 wt.%) of Apollo mare basalts are reached at 1150–1300 °C. At experimental temperatures of approximately 1220–1260 °C, corresponding to partial melt proportions of ~35–55 per-cent, abundances of TiO2, MgO and FeO in melts, and melt Mg#, from the three experimental series are consistent with the compositions in high-Ti mare basalts (Fig. 2b–d). Partial melts generated from the starting composition of the IBCL mixed with 5 wt.% plagioclase best fit the compositional characteristics of high-Ti mare basalts. In contrast, the melts have SiO₂ contents higher than those of high-Ti mare basalts (Fig. 2a), whereas the CaO contents are lower (Fig. 2e). We hypothesize that it may be possible to eliminate the gap in Fig. 2a, e and achieve the composition of high-Ti mare basalts by remelting LBS11, or a mixture of LBS10 and LBS11, or a combination of LBS10, LBS11, plagioclase, and additional clinopyroxene (cpx) to account for the low CaO content observed in our experiments at shallow depths. In order to test this hypothesis, melting simulations were conducted by combining the Monte Carlo method with alphaMELTs software. In summary, remelting experiments and simulations support the idea that high-Ti mare basalt compositions can be reproduced by remelting of a shallow-depth hybrid lunar mantle, composed of IBCL+plagioclase+clinopyroxene.

Fig. 2. Comparison among major element abundances in Apollo 11 and 17 high-Ti mare basalts (pink and blue crosses), in experimental melts from this study (green symbols) and in Monte Carlo-alphaMELTS simulated melts (gray and yellow circles). a SiO₂ versus MgO. b Al₂ O₃ versus MgO. c TiO₂ versus MgO. d FeO versus MgO. e CaO versus MgO.

To complement major element abundance considerations, we assess if a shallow remelting mechanism is consistent with the observed trace element patterns and isotope systematics of the high-Ti basalts. We focus on the rare earth elements (REE), and on Sm-Nd-Lu-Hf isotopic systematics. Taken together, REE abundances, Lu-Hf-Sm-Nd isotopic data, as well as major element considerations all support the hypothesis that the lunar high-Ti mare basalts were formed by low-pressure, high-degree partial melting of an ilmenite-bearing source. This source contained small percentages of imperfectly segregated plagioclase, TIRL, and a small but important amount of urKREEP component.

Fig. 3. CI-normalized REE patterns of high-Ti mare basalt parental melts and modeled lunar mantle cumulate melts. The horizontal green line refers to the initial bulk LMO trace element concentration.

Partial melting experiments of late-stage lunar cumulate compositions with variable proportions of plagioclase formed during the crystallization of the lunar magma ocean at low pressure were conducted. These experiments indicate that the lunar high-TiO₂ mare basalts were formed by large-degree partial melting of a shallow source. A deep source suggested by studies of multiple saturation points is not required. Our hypothesis is supported by major and trace element observations as well as isotopic considerations. This formation process is consistent with an impact-induced triggering mechanism for lunar volcanism. Large impacts may have had an important role in the history of volcanism on the Moon.

月球月海高钛玄武岩是人类开展月球探测任务以来返回的第一批月球样品之一，研究月球玄武岩的起源对我们理解月球的形成、演化乃至地月系统之间的联系都具有重要意义。然而，有关月海高钛玄武岩形成和起源仍不明确。近日，北京高压科学研究中心林彦蒿研究员课题组使用实验岩石学方法对月海高钛玄武岩源区进行探究，提出月海高钛玄武岩起源于浅部月幔含钛铁矿源区，相关工作以“Experimental evidence for a shallow cumulate remelting origin of lunar high-titanium mare basalts”为题发表于Nature子刊《Communication Earth & Environment》，论文第一作者为北京高压科学研究中心和中国地质大学（北京）联合培养硕士生高书畅。