EFFECT OF SURFACE ROUGHNESS ON THE THERMAL INFRARED SPECTRAL CHARACTERISTICS OF MARS-RELEVANT SILICA-RICH MATERIALS

Thermal emission spectrometers in orbit and on the surface of Mars have located several silica-rich deposits, both in outcrop and as loose regolith material. Spectra indicate the presence of opaline silica and more crystalline silica phases such as quartz. Given that bona fide (i.e., carbonaceous) microbial biosignatures are preserved in ancient silica-rich materials on Earth, and that silica-rich materials can originate in a variety of ways, we would like to know whether highly siliceous cherts might produce diagnostic spectral characteristics indicative of their mode of origin. In this regard, we are investigating how mineral textures, grain size, sample density, and surface roughness of silica-rich materials affect their thermal infrared spectra. This study focused on quantifying the effects of surface roughness on the spectral character of the manufactured samples that were studied.

To quantify how surface roughness affects thermal infrared spectral characteristics, we roughened with known grit sizes different types of sample materials (i.e., fused quartz, glass, and silicon wafers) and measured surface roughness with the use of a stylus profilometer and a scanning electron microscope. This approach allowed us to assess how different degrees of roughening in materials with different types of structural ordering could affect the infrared spectra of silica-rich materials roughened in a reproducible manner. Stylus profilometry data were quantified using a scale-dependent root-mean-square roughness parameter. Scales of observation ranged from 10 μm to 1 cm. Fractal dimensions were also estimated to characterize surface roughness. We found that changes in spectral contrast could be linked with the magnitude and type of surface roughness, and that these factors may confound estimations of relative amounts of silica phases in natural samples. Our findings demonstrate the utility of a protocol that results in reproducibly roughened sample surfaces for spectral analysis. Such a protocol will be particularly helpful as spectral datasets continue to grow with the inclusion of data from powdered, sawn, and weathered surfaces of geological samples.