The lunar surface, lacking the protective shield of an atmosphere and magnetic field, is continuously altered by space environments through a variety of processes, including the bombardment of (micro)meteorites, solar wind, and cosmic rays. These alteration processes, known as space weathering, occur widely on airless bodies such as asteroids, Mercury and the Moon.

Metallic iron nanoparticles (npFe⁰) are ubiquitously produced by space weathering, changing the optical spectral of airless bodies over time and complicating the interpretation of their surface properties. The optical effects largely depend on the particle size: smaller npFe⁰ redden the reflectance spectra, whereas larger npFe⁰ cause darkening. Despite decades of research, the origins of npFe⁰ of different sizes remain enigmatic, particularly regarding the special roles of the two key agents: micrometeorite impacts versus solar wind irradiation. Generally, lunar soils are exposed to both impacts and irradiation environments concurrently (Fig. 1), making it challenging to separate the respective effects of micrometeorites and solar wind on npFe⁰ formation. This significantly hampers our ability to understand and predict color changes of lunar regions or other airless bodies exposed to different space environments.

Our paper begins with a study of diverse glassy materials in the Chang’e-5 lunar soils. Observing lunar particles one by one, we interestingly found that, besides glass beads with smooth surfaces, there are also numerous glass beads featuring bulges on their extremities (Fig. 2). The bulges were then identified as metallic iron. Subsequent precise observations unveiled that a single such glass bead can preserve npFe⁰ of varying sizes and different distributions (Fig. 3), including surface-correlated small npFe⁰ with sizes of approximately several nanometers (SnpFe⁰), volume-correlated large npFe⁰ of roughly tens of nanometers (LnpFe⁰) and extremity-correlated ultralarge npFe⁰ reaching up to approximately hundreds of nanometers (ULnpFe⁰).

Typically, distinguishing npFe⁰ of different origins coexisting in a single experiment, such as in a transmission electron microscopy (TEM) image, poses a significant challenge. Consequently, many studies tend to prioritize investigating a certain mechanism of npFe⁰ formation. In this study, we comprehensively considered the observed npFe⁰ within individual glass beads, and conducted a combined analysis of the sources of npFe⁰ of different sizes, aiming to clarify the relative contributions of micrometeorite impacts and solar wind irradiation to the generation of different-sized npFe⁰.

We found that these glass beads with rotational shapes can help distinguish npFe⁰ formed before and after the solidification of the host glass beads. These beads are products of rapid cooling of hypervelocity impact-ejected melted droplets, and their unique shapes result from non-axisymmetric rotations of viscous droplets during flight. The centrifugal forces generated by rotational motion can elongate spherical viscous melts into oblate spheroids, ellipsoids or dumbbells. The observed concentration of ULnpFe⁰ and LnpFe⁰ towards impact glass extremities is attributed to the migration of denser Fe⁰ in the impact-generated melts driven by centrifugal forces during rotations, indicating their formation prior to the solidification of the melts to form glass beads. Namely, impact-derived shock and thermal effects can melt and transform Fe-bearing grains into Fe⁰, which can then quickly nucleate and grow up into LnpFe⁰ within the melts, potentially coalescing into ULnpFe⁰. Upon solidification of the melts, the formed (U)LnpFe⁰ are preserved within the glass beads.

Additionally, we realized that the impact-derived (U)LnpFe⁰ can help identify solar wind-irradiated regions on the surfaces of the glass beads. Solar wind irradiation can easily damage the metallic matrix, resulting in vesicular textures on the near-surface LnpFe⁰ and surface-exposed ULnpFe⁰. Vesicles gradually decrease in size and eventually disappear with increasing depth, marking the irradiated regions. Within the irradiated depth, abundant SnpFe⁰, densely filling the the surfaces of the glass beads, were identified. Similar to the distribution of vesicles in the metallic matrix, SnpFe⁰ in the glass matrix gradually decrease in size and abundance, and finally disappear beyond the penetration depth of solar wind. These results indicate that after formation, the exposed glass beads underwent long-term irradiation environments, and the implantation of solar wind ions produced abundant SnpFe⁰, resulting in SnpFe⁰-rich rims along the entire surfaces of the glass beads. The observed gradient distribution of SnpFe⁰ and vesicles aligns with the reported decrease in solar wind-derived OH/H₂O content with depth, corresponding to the decreasing amount of implanted solar wind ions along the depth direction.

We further examined different types of Chang’e-5 mineral grains and observed that the conclusions drawn from glass beads apply to lunar minerals. Small npFe⁰ were commonly found in the solar wind-irradiated rims of Chang’e-5 lunar grains, but were absent in the outmost vapor-deposited layers. This suggests that the prevailing impact-induced vapor deposition mechanism for producing small npFe⁰ is not applicable to the observed CE-5 samples. Our findings highlight that solar wind irradiation, rather than vapor deposition, is the primary agent for the formation of small npFe⁰, meaning that solar wind is responsible for the reddening of optical spectral. This is consistent with a series of solar wind flux-dependent weathering effects in spectroscopic observations.

Our combined study on the origins of npFe⁰ of different sizes reveals that small and large npFe⁰ with different optical effects are the respective products of solar wind irradiation and micrometeorite impacts, highlighting that solar wind and micrometeorites both play important yet distinct roles in space weathering. These findings provide valuable insights for understanding and predicting space weathering behavior under complex space environments. For instance, they can explain the unique weathering behavior of lunar swirls, where reduced irradiation and normal impact environments finally result in significantly fewer small npFe⁰ but similar large npFe⁰ compared to surrounding regions. However, whether the revealed effects of solar wind irradiation and micrometeorite impacts are universal or variable needs to be verified by samples weathered in diverse space environments. The Chang’e-6 lunar soils will present an excellent opportunity to address this question and further enhance our comprehension of space weathering.