Data for 'Solar wind neon storage in vesicles in space weathered lunar samples: Implications for neon behavior in planetary materials'

<p>Solar wind neon is incorporated into lunar regolith grains and other planetary materials via solar wind implantation. Understanding the abundance of neon relative to other solar wind gases and its trapping and storage within planetary materials can inform on both parent body processing and volatile cycling. Microstructural defects in lunar regolith grains, including vesicles formed by space weathering processes, have previously been identified to store other solar wind-derived volatiles such as hydrogen, water, and helium. Here, we use transmission electron microscopy and electron energy loss spectroscopy to identify the presence of solar wind neon and quantify its abundance in vesicles within a space weathered lunar regolith grain. The direct observation of solar wind neon trapped in vesicles offers a new understanding of the space weathering history of lunar regolith grains and other planetary materials rich in solar wind gases. The storage of solar wind neon in nanoscale vesicles also has implications for its fractionation, diffusivity, and retention at high temperatures which may affect interpretations of the exposure and processing histories of lunar and other planetary materials as derived from noble gas analyses.</p>

<p><strong><span style="font-size:11pt"><span style="line-height:107%"><span style="font-family:Aptos,sans-serif"><span style="font-size:12.0pt"><span style="line-height:107%"><span new="" roman="" style="font-family:" times="">TEM and EELS Analyses</span></span></span></span></span></span></strong></p>

<p>The microstructure and chemistry of the agglutinate grain was studied via TEM imaging, high-angle annular dark-field (HAADF) scanning TEM (STEM) imaging, and energy-dispersive X-ray spectroscopy (EDS) using a 200 keV Thermo Scientific Talos 200X TEM in the PEMC, and a 300 keV Thermo Scientific Themis Z S/TEM in the PEMC and at the Center for Electron Microscopy and Analysis at the Ohio State University (OSU) (Kling et al., 2025). Electron energy loss spectroscopy (EELS) analyses of the agglutinate FIB sections were performed on the Themis Z S/TEMs at both Purdue University and OSU to measure the presence of implanted solar wind-derived gases such as He, Ne, H, and water. EELS analyses at Purdue were collected as linescans and EELS analyses at OSU were collected as spectrum images. Further details on acquisition parameters are presented in Kling et al., (2025).</p>

<p>Electron energy loss spectroscopy analyses of the Ne<sup>+</sup>-irradiated SCO were performed on the Themis Z at Purdue operated in STEM mode with an accelerating voltage of 300 keV using a monochromator to maximize energy resolution in the low-loss energy region. We performed these measurements using a monochromator excitation of 0.8 with a camera length of 145 mm and current ranging between 0.12 and 0.16 nA. EELS data were acquired with a Gatan<sup>TM</sup> Quantum 965 EELS detector and the Gatan<sup>TM</sup> DigitalMicrograph<sup>TM</sup> acquisition software. These measurements were collected as linescans and spectrum images (SIs). HAADF images or intensity profiles were acquired simultaneously with the EELS spectra. Spectrum images were also acquired using drift correction. Pixel sizes ranged from 0.2 to 15 nm depending on the sizes of analyzed vesicles.</p>

<p>EELS analyses were performed in Dual EELS mode, enabling the simultaneous acquisition of two EELS signals. These measurements were performed in two ways. The first method was meant to capture the zero-loss peak (ZLP) in the “Low-Loss” signal and the energy region containing the Ne valence excitation at ~16.85 eV in the “High-Loss” signal. These measurements were performed with an energy dispersion of 0.025 eV/ch, an imaging aperture of 2.5 mm, and an energy resolution of ~0.2 eV as measured by the full-width half-maximum of the ZLP. The Low-Loss signal energy and time were set to 0 eV and 0.005 s, respectively, and the High-Loss signal energy and time were set to 6 eV and 0.5-1 s, respectively, with pixel times matching those of the high-loss exposure times of either 0.5 or 1 s. Binning of acquired data varied between 1x1, 1x5, 1x65, and 1x130.</p>

<p>    The second method for Dual EELS acquisition was to capture the ZLP and the low-loss energy region including the Ne valence excitation in the Low-Loss signal and the core-loss energy region surrounding the Ne K-edge (Ne-K) at 867 eV in the High-Loss signal. To maximize the counts for identification of the Ne-K while maintaining the energy resolution necessary to simultaneously identify the Ne valence excitation, we used a 5 mm imaging aperture and an energy dispersion of 0.05 and 0.1 eV/ch. The Low-Loss signal was acquired with an energy and exposure time of 0 eV and 0.005 s, respectively, and High-Loss signal with an energy between 760-820 eV and exposure time of 0.5s. These measurements were acquired with the 'UFAST' setting, optimized for high-speed data acquisition and having a binning of 1x130.</p>

<p>    Following EELS analyses, we also performed EDS analyses on the SCO sample for additional verification of the presence of neon using the Super-X quad-silicon drift EDS detector on the Themis Z. EDS mapping was performed on a region of the SCO including both irradiated and unirradiated areas as well as areas previously mapping using EELS.</p>

<p style="margin-bottom:11px"><strong><span style="font-size:12pt"><span style="line-height:115%"><span style="font-family:Aptos,sans-serif">EELS data processing</span></span></span></strong></p>

<p>The processing and analysis of electron energy-loss spectroscopy (EELS) data were performed using the Gatan<sup>TM</sup> DigitalMicrograph<sup>TM</sup> software and the HyperSpy open-source Python library. The initial processing of the EELS data (including zero-loss peak alignment, high intensity pixel removal, and background removal) and identification of the Ne valence excitation via principal component analysis and non-negative matrix factorization follow the previous methodology from Kling et al. (2025).</p>

<p>Here, we describe the steps taken to extract the Ne valence excitation peak and identify its peak center. After the initial processing of the data, the spectra were cropped to the region surrounding the Ne valence excitation: ~14-20 eV. To reduce contributions of noise in the identification of the peak maximum, we in some cases performed denoising of the spectra prior to isolation of the Ne peak using PCA. Similar to the methods of Kling et al. (2025), we performed a single-variable decomposition of the data and then reconstructed the data with a reduced number of components. These components describe nearly all the variability within the dataset, and the excluded components constitute noise. hus, the reconstructed data are "denoised." A baseline was fit to the background of the data, excluding the energy region contained within the Ne valence excitation. This baseline fit was performed by masking the Ne peak and fitting a polynomial function to the background outside the mask. After fitting the baseline, it was subtracted from the data to isolate the Ne valence excitation. We then fit a Gaussian curve to the residual Ne valence excitation and found the maximum of the Gaussian. Using this maximum value, we calculated the atomic density of Ne in the vesicle using Equation 1. We used a similar procedure to isolate and extract the He-K edge and quantify the He atomic density in vesicles where it was present.</p>