Bepi-Colombo

The project

Fig. 1. Spectral radiance of Mercury at different temperatures. The maximum values are located in the NIR range . Fig. 2. Sample spectrum of the feldspars albite and anorthite for different grain sizes in the range from 6 to 18 microns

MERTIS is a combined TIR Spectrometer with an radiometer instrument proposed for the orbiter of Bepi Colombo. From Earth-based telescopic measurements of hermean spectra, which are mostly disk-integrated and often affected by problems resulting from the planet’s proximity to the Sun and/or from telluric absorption's we know that early measurements in the NIR (0.65 < l < 2.5 µm) suggested pyroxene with 5.5 % FeO due to a weak Fe2+-absorption at 0.89 µm (McCord & Clark 1979). Later measurements at low air mass did show no absorption feature that cannot be attributed to incomplete removal of telluric absorption bands (Vilas 1985). The lack of any feature between 0.5 and 1.0 µm does not preclude the presence of orthopyroxenes (opx); it only suggests that any opx would be an endmember enstatite (MgSiO3). The spectral slope suggests that agglutinitic glasses could be present in the regolith (Vilas 1985). The spectral slopes and the absence of any feature suggest anorthite as main component, and from the albedo value it was concluded that at most 3 wt-% FeO and 1 wt-% TiO2 can be present in the regolith (Blewett et al. 1997). Such a low FeO content would be highly exceptional and needs confirmation.

Only very few measurements in the spectral TIR range from 7.5 to 13 µm (Sprague and Roush 1998 and refs. therein, Sprague et al. 2000, Cooper et al. 2001) are known. Interpretation of the data is complicated by the need of both thermal modeling and atmospheric correction. There is evidence for spectral variation with the observed hermean locality, and indication for features common to feldspars and mixtures of feldspar and pyroxene, respectively. As a result we conclude that only i n the TIR-range feldspars can be detected and specified by means of their diagnostic spectral signatures: Christiansen frequency, reststrahlen band, and transparency feature (see figure 2). Pyroxenes and most other minerals can be detected and specified in this spectral range, too. A suitable experiment to investigate the chemical composition and to identify the rock-forming minerals must use the TIR range.

Main Scientific Objectives

• Study of Mercury‘s surface composition

• Identification of rock-forming minerals

• Mapping of the surface mineralogy

• Study of surface temperature and thermal inertia

The instrument

The NIR, SWIR and TIR spectroscopy are powerful in characterizing the mineralogy of solid planetary surfaces. Most of the diagnostic absorption bands of the rock forming minerals appear in these spectral ranges.

For wavelengths longer than about 2 µm (and even shorter near perihelion and at the equator) the thermal flux of planets like Mercury exceeds the flux reflected from its surface.

In the NIR-range pure feldspars (expected to belong to Mercury’s main constituents) have no specific spectral signature! On Earth, feldspars sometimes may be detected by weak features which result from accessory impurities. Nevertheless, also feldspar spectra without any feature in this range were found (see, e.g., Hunt and Salisbury 1970). For a reliable detection of any feldspar the NIR-range is highly questionable. A specification of the type and composition of a feldspar (i.e., concerning its position within the mixture series Albite - Anorthite, and Albite - Orthoclase, respectively) is quite impossible in the NIR.

In the TIR-range, however, feldspars can be detected and specified by means of their diagnostic spectral signatures: Christiansen frequency, reststrahlen band, and transpareny feature. Further, also pyroxens and most other minerals can be detected and specified in this spectral range. Therefore, to investigate feldspar and pyroxene spectra (e.g. on Mercury) it is essential to have spectral measurements in the TIR range.

Detectors for the three spectral ranges: 0.4-1 m m (VIS/NIR-CCD), SWIR (0.8-2.5 m m ) and TIR exist in Europe and were used at DLR already for other experiments. The only candidate for the TIR detector is the Microbolometer technology (standard chips with 320x240 pixel, no cooling is required). The best performances for this technology have detectors offered by Raytheon, Santa Barbara. These Bolometer types were applied in the THEMIS Instrument for the Mars 2001 Orbiter (Workshop on Bolometers, ESTEC June 2002). A European version is available from SOFRADIR and was used in the FUEGO project, where the DLR delivered the main camera based on the BIRD technology. Currently the Microbolometers are used by the DLR in terrestrial projects.

In the case of Mercury, the spectral radiance at day site shows that the thermal emission starts to dominate the all over radiance already at wavelength larger than 1.2 µm (at 725°K) depending on the surface albedo (see figure 1). The range between 0.8 and 2.8 µm is a transition region characterized by the overlapping of the reflected solar radiation and the thermal emission. However, Mercury’s thermal flux exceeds the flux reflected from its surface. This makes near-infrared spectroscopic techniques very difficult, but enables emittance spectroscopy in the TIR range. This spectral range has high potential for mineral identification because it is in that region that the major rock-forming minerals have their fundamental vibration bands. Therefore, a TIR channel is a powerful option in order to provide more valuable information about Mercury’s mineralogy.

Table 1: Basis parameter for a TIR Spectrometer
Focal length f	50 mm
f – number f no	2.0
Optical efficiency h opt	0.54
Micro bolometer matrix Raytheon Santa Barbara	320 x 80 @ 50 µm
Spectral channel width d l	90 nm
Spectral bandwidth D l	450 nm
Spectral range	7 – 14 m m
Detectivity D*	10 9 cm Hz 1/2 W –1
Resolution on ground	M × 200 m
Envisaged resolution on ground (400 km orbit)	200 - 600 m (n = 1 - 3)
Swath width	64 km
Optic module	160x130x50 mm³

 

Contributions of the team

Scientific contributions

Laboratory measurements in the TIR; data analysis and interpretation, thermal modeling.

Hardware contributions

IR detector plus read out electronic, optical design, opto mechanic, structure, system integration, calibration and verification. Overall instrument management, system engineering and system integration.

Internal structure, composition, and evolution of Mercury implications for design of instruments for space craft

Modeling the thermal evolution history of Mercury, studying the chemistry of the rocks on the surface of Mercury, observed either from ground or from a space-craft like Messenger or Bepi Colombo, may hold clues to the formation of the Mercury. Large amounts of mantle forming silicates (like olivine and pyroxenes), if observed, will support the differential accretion model of Weidenschilling (1978). This scenario assumes the accretion of dense, iron rich core materials and the sorting of planetesimal building blocks by gas drag in the very early phase of the formation of the solar system. From analyzing the polar moment of inertia and the density additional evidence for Fe/Si ratios can be found. The giant impact model (Wetherill 1988, Benz et al. 1988) in which most of the crust is removed by an impactor of perhaps about 20% the size of the planet could possibly be supported by the detection and identification of mafic to ultra-mafic silicate rock and other rock material typically of assumed in the mantle of terrestrial planets. In this case Ca, Al, and alkali will be depleted without the enrichment of refractory elements and the FeO abundance will depend on the oxidation state of the protoplanetary material. An enrichment of refractory material and depletion of alkalis and FeO, on the other hand, may indicate a hot formation (Cameron 1985, Fegley and Cameron 1987) of the planet. In this model, the Sun in an early T-Tauri phase would have had a substantially enhanced energy output, resulting in surface temperatures of Mercury of ~2500-3000K. Other formation scenarios allow a volatile-rich composition. For instance, it has recently been noted that the density of Mercury is close to that of an eutectic Fe-FeS alloy (Harder and Schubert, 2001). They propose that Mercury may basically consist of a core with a thin silicate shell surrounding it. A sulfur-rich composition has also been proposed by Sprague et al. (1995, 1996). In this case there may be polar deposits of sulfur and it is very likely that sulfur bearing minerals may be detected on the surface. As a consequence the identification and detection of mineral phases expected on the yet unknown surface is vital.

The structure of Mercury will depend on its composition. If the composition is refractory as is consistent with some of the formation scenarios above, then the core will be mostly iron and the silicate shell may be about 600 km thick. The core most likely is partly solid but a fluid outer core shell is required to explain the magnetic field. The gravity measurements combined with observations of the libration rate and obliquity planned for Messenger and Bepi Colombo will allow improved models of the interior. Among the expected results are constraints on the structure of the core, and the topography of the core/mantle boundary and the crust/mantle boundary (e.g., Spohn et al. 2001).

The NIR, SWIR and TIR spectroscopy are powerful in characterizing the mineralogy of solid planetary surfaces. Most of the diagnostic absorption bands of the rock forming minerals appear in these spectral ranges. For wavelengths longer than about 2 µm (and even shorter near perihelion and at the equator) the thermal flux of planets like Mercury exceeds the flux reflected from its surface.

In the NIR-range pure feldspars (expected to belong to Mercury’s main constituents) have no specific spectral signature. On Earth, feldspars sometimes may be detected by weak features which result from accessory impurities. But, nevertheless, also feldspar spectra without any feature in this range were found (see, e.g., Hunt and Salisbury 1970). For a reliable detection of any feldspar the NIR-range is highly questionable. A specification of the type and composition of feldspar (i.e., concerning its position within the mixture series Albite - Anorthite, and Albite - Orthoclase, respectively) is quite impossible in the NIR.

In the TIR-range, however, feldspars can be detected and specified by means of their diagnostic spectral signatures: Christiansen frequency, reststrahlen band, and transparency feature. Further, also pyroxenes and most other minerals can be detected and specified in this spectral range. Therefore, to investigate feldspar and pyroxene spectra on Mercury it is essential to have spectral measurements in the TIR range.

Conclusions

In the case of Mercury, the spectral radiance at day site shows that the thermal emission starts to dominate the all over radiance already at wavelength larger than 1.2 µm (at 725°K) depending on the surface albedo. The range between 0.8 and 2.8 µm is a transition region characterized by the overlapping of the reflected solar radiation and the thermal emission. However, Mercury’s thermal flux exceeds the flux reflected from its surface. This makes near-infrared spectroscopic techniques very difficult, but enables emittance spectroscopy in the TIR range. This spectral range has high potential for mineral identification because it is in that region that the major rock-forming minerals have their fundamental vibration bands. Therefore, a TIR channel is a powerful option in order to provide more valuable information about Mercury’s mineralogy.