Influence of Particle Size in Laboratory Emissivity of Mercury Analogue Material Mixtures under Simulated Mercury Conditions

The Mercury Radiometer and Thermal Infrared Spectrometer (MERTIS) is part of the payload on the joint ESA/JAXA mission BepiColombo to Mercury. The onboard spectrometer will map the thermal emissivity of Mercury in the 7-14 µm wavelength range [1]. Understanding how Mercury's surface properties influence remote sensing data is essential for interpreting the data of MERTIS. Therefore, to support the interpretation of MERTIS data, various emissivity measurements already have been performed at the Planetary Spectroscopy Laboratory (PSL) at the German Aerospace Center (DLR) in Berlin [2]. This study aims to extend the knowledge acquired by systematically investigating the effect of grain size on emissivity and bidirectional reflectance spectra of Mercury analogue mixtures. 

Two endmembers, an enstatite and an anorthoclase, were prepared in three grain size fractions: 0–25 µm, 25–63 µm, and 125–250 µm. In addition to the pure endmembers, intimate mixtures of the two endmembers with ratios of 25:75, 50:50, and 75:25 were created, including both homogeneous and heterogeneous grain size combinations, resulting in a total of 33 samples.

Mid-infrared emissivity spectra were acquired at 150 °C, 250 °C, 350 °C, and 450 °C under simulated Mercury pressure condition (0.7 mbar) in the DLR-built emissivity chamber at PSL. The sample spectra were calibrated using emissivity spectra of a graphite slab serving as approximation of a blackbody, and additional hemispherical reflectance spectra of the samples, which were taken under ambient temperature conditions (Fig. 1).

Figure 1: Sample measurements (thick lines) with approximated blackbody target measurements (color gradient) for data calibration.

In addition to the emissivity measurements, bidirectional reflectance spectra of the same samples were acquired across a wide spectral range (VIS to MIR). Spectra were taken before and after heating in the emissivity chamber to investigate the influence of heating upon the samples. Next to the chamber-heated samples, fresh sample of all mixtures were measured in a vacuum-oven under the same temperature and lower pressure condition as in the emissivity chamber. The oven-heated samples were used to additionally investigate the influence of sample preparation and sample handling on the spectra. Those samples are in fact directly heated in the vacuum oven inside their sample holders, while in the case of the sample measured in bidirectional reflectance after the emissivity measurements, those powders are scooped into the reflectance sample holder from the emissivity sample cup, this process altering the original layering inside the sample because of the thermal gradient inside the powdered samples heated in vacuum. All bidirectional reflectance spectra were merged, creating one full wavelength-range spectrum for each sample (Fig. 2).

Figure 2: Bidirectional reflectance spectra from UV to MIR before (left) and after (right) calibration.

The results show that grain size systematically influences spectral features, particularly the peak width of the Christiansen Feature and the occurrence of Transparency Feature in the emissivity measurements, as well as their maximum emissivity value. Spectral behavior is influenced by clinging fines, scattering regimes, and thermal gradients. In addition, both the emissivity and reflectance data show that spectral properties and trends are strongly related to relative grain size, and especially in the mixtures, the appearance of features is linked to the endmember with the smaller grain size (Fig. 3). Furthermore, oven-heated samples show more consistent results than chamber-heated samples, highlighting the importance of sample preparation and handling.

This study demonstrates systematic grain size effects and identifies robust grain size indicators in intimate mixtures. The findings of this study highlight the need to consider these grain size effects when interpreting remote sensing spectra of Mercury to contribute to a more reliable interpretation of spectral data from MERTIS and other spectral suites onboard the BepiColombo mission. Additionally, the results highlight the need for consistent sample handling in laboratory set-ups.

References:

[1]       H. Hiesinger and J. Helbert, “The Mercury Radiometer and Thermal Infrared Spectrometer (MERTIS) for the BepiColombo mission,” Planet. Space Sci., vol. 58, no. 1–2, pp. 144–165, Jan. 2010, doi: 10.1016/j.pss.2008.09.019.

[2]       A. Maturilli, J. Helbert, I. Varatharajan, and H. Hiesinger, “Emissivity Spectra of Analogue Materials at Mercury P-T Conditions,” presented at the 48th Annual Lunar and Planetary Science Conference, Mar. 2017, p. 1427. Accessed: Sep. 24, 2025.