No abstract

The solar wind continuously impacts on rocky bodies in space, eroding their surface, thereby contributing significantly to the exosphere formations. The BepiColombo mission to Mercury will investigate the Hermean exosphere, which makes an understanding of the precise formation processes crucial for evaluation of the acquired data. We therefore developed an experimental setup with two microbalances that allows us to compare the sputter behavior of deposited thin solid layers with that of real mineral samples in the form of pressed powder. In addition, this technique is used to study the angular distribution of the sputtered particles. Using 4 keV He+ and 2 keV Ar+ ions, the sputter behavior of pellets of the minerals enstatite (MgSiO3) and wollastonite (CaSiO3) is studied, because these minerals represent analogs for the surface of the planet Mercury or the Moon. Pellets of powdered enstatite show significantly lower sputter yields than thin amorphous enstatite films prepared by pulsed laser deposition. 3D simulations of sputtering based on surface topography data from atomic force microscopy show that the observed reduction can be explained by the much rougher pellet surface alone. We therefore conclude that sputter yields from amorphous thin films can be applied to surfaces of celestial bodies exposed to ion irradiation, provided the effects of surface roughness, as encountered in realistic materials in space, are adequately accounted for. This also implies that taking surface roughness into account is important for modeling of the interaction of the solar wind with the surface of Mercury.

Rocky planets and moons experiencing solar wind sputtering are continuously supplying their enveloping exosphere with ejected neutral atoms. To understand the quantity and properties of the ejecta, well-established binary collision approximation Monte Carlo codes like TRIM with default settings are used predominantly. Improved models such as SDTrimSP have come forward, and together with new experimental data, the underlying assumptions have been challenged. We introduce a hybrid model, combining the previous surface binding approach with a new bulk binding model akin to Hofsäss & Stegmaier. In addition, we expand the model implementation by distinguishing between free and bound components sourced from mineral compounds such as oxides or sulfides. The use of oxides and sulfides also enables the correct setting of the mass densities of minerals, which was previously limited to the manual setting of individual atomic densities of elements. All of the energies and densities used are thereby based on tabulated data, so that only minimal user input and no fitting of parameters are required. We found unprecedented agreement between the newly implemented hybrid model and previously published sputter yields for incidence angles up to 45° from surface normal. Good agreement is found for the angular distribution of mass sputtered from enstatite MgSiO3 compared to the latest experimental data. Energy distributions recreate trends of experimental data of oxidized metals. Similar trends are to be expected from future mineral experimental data. The model thus serves its purpose of widespread applicability and ease of use for modelers of rocky body exospheres.

No abstract

<p>Rocky bodies in space without a protective atmosphere like Mercury are subject to harsh conditions, including the bombardment by solar wind ions. This leads to the ejection of particles along with alteration of the surface properties like composition and morphology. These ion-induced sputter processes contribute to the formation of the Hermean exosphere [1]. Therefore, understanding the involved fundamental ion-solid interaction processes is important for the modelling of exosphere creation. From a computational standpoint, those impacts are usually investigated using simulation codes based on the Binary Collision Approximation (BCA) like SRIM [2] or SDTrimSP [3]. However, past studies revealed that especially for compound targets relevant as Hermean regolith analogues, simulations have to be adapted by means of modified input parameters to reproduce experimental results [4]. Thus, laboratory measurements are strongly needed to ensure that valid inputs enter the exosphere modelling.</p> <p align="justify">A well-proven method for such investigations of sputtering is the Quartz Crystal Microbalance (QCM). It allows to measure sputter yields of thin films deposited onto a quartz resonator with high precision in real time and<em> in situ</em><em> </em>[5]. Expanding on this technique, we use a setup in which we place a second QCM in the vicinity of the irradiated target, facing the centre of particle emission [6]. Ejecta that stick to its surface result in a mass accumulation, which is resolved through the piezoelectric properties of the quartz resonator. By moving this second catcher-QCM in an arc around the target, the angular emission characteristic of sputtered particles can be probed. The experimental realisation is sketched in figure 1. The setup also allows for experiments with targets that cannot be deposited onto a resonator or whose surface properties change during the deposition process. Particularly, the above-mentioned films used in, e.g. [4], are vitreous and smooth. We therefore extended our studies to pellets from ground and pressed mineral specimens that retain some roughness for a more realistic regolith analogue [7]. Through comparison of these measured angular distributions with reference samples, also total sputter yields can be determined for targets whose yields cannot be measured with a single QCM.</p> <p align="justify"><img src="" alt="" /></p> <p align="justify">Figure 1: Sketch of the setup. A target is irradiated with an ion beam under an angle of incidence α. The angular distribution of ejecta (blue shaded area) is probed with the catcher-QCM by varying the angle α<sub>C</sub>.</p> <p align="justify">We initially used a 2 keV Ar<sup>+</sup> ion beam to irradiate both an enstatite (MgSiO<sub>3</sub>) film and a pellet under 60° and 45° incidence with respect to the sample surface normal. This choice of projectile has a high sputter yield and produces sufficient signal at the catcher-QCM to allow for a proof of principle. After initial hardships, modified sample preparation routines made reproducible quantification of the obtained data possible. We attribute differences in shape and magnitude of the sputtered particle angular distributions between the sample types to the different roughnesses of the sample configurations. Geometric considerations alone are sufficient to describe the qualitative behaviour of our results [8]. Simulation results to visualise the impact of surface roughness on the sputter yield as a function of incidence angle are given in figure 2. For a solar wind relevant projectile species, the same measurements were also carried out with a 4 keV He<sup>+</sup> ion beam. This presentation will include the results from both the He and the Ar irradiation experiments.</p> <p align="justify"><img src="" alt="" /></p> <p align="left">Figure 2: Sputter yields Y for a perfectly smooth surface (as simulated by SDTrimSP, blue) and for the surface of an enstatite (MgSiO<sub>3</sub>) pellet used in this study in orange. The latter are obtained using a home-made code based on geometric considerations [8]<em>.</em></p> <p align="left"> </p> <p align="left"><strong>References</strong></p> <p align="left">[1] P. Wurz et al.: Planet. Space Sci. <strong>58</strong>, 1599, 2010 <br />[2] J.F. Ziegler et al.: Nucl. Instrum. Methods Phys. Res. B: Beam Interact. Mater. At., <strong>268</strong>, 1818, 2010 <br />[3] A. Mutzke et al.: SDTrimSP Version 6.00. Max-Planck-Institut für Plasmaphysik, 2019 <br />[4] P.S. Szabo et al.: Astrophys. J., <strong>891</strong>, 100, 2020 <br />[5] G. Hayderer et al.: Rev. Sci. Instrum., <strong>70</strong>, 3696, 1999 <br />[6] H. Biber et al.: EPSC2021, online, EPSC2021-526, 2021 <br />[7] N. Jäggi et al.: Icarus, <strong>365</strong>, 114492, 2021<br />[8] C. Cupak et al.: Appl. Surf. Sci., <strong>570</strong>, 151204, 2021</p>