Multivariate Comparison of Lunar Soil Simulants

Abstract

Since 1960, when the first probes landed on the Moon, multiple specimens of lunar soil have been available for research. The very limited volume of lunar soil samples means that they can be utilized only for the most important research programs. Therefore, multiple artificial lunar soil materials, called lunar soil simulants (LSSs), have been developed over the last 55 years. Lunar soil simulants are used to enable large-scale lunar research programs. They are supposed to be precise copycats of lunar soil in chemical composition, mineralogy, grading, and mechanical characteristics. Nevertheless, the vast majority of artificial and commercially available simulants have not met scientific expectations and proved unsuitable for specific lunar engineering purposes. There is a pressing need for a complex but at the same time efficient, reliable, and fast method for the assessment of lunar soil simulants. In the authors’ opinion, an analytical approach based on multivariate statistics involving multiple parameters of a given simulant would be the most feasible solution to this problem. Principal component analysis, which is a special case of factor analysis, was chosen as the most suitable for the task. Moreover, based on chemical composition data from real lunar samples and the results of multivariate calculations, we proposed to use the ilmenite sand as the base for LSSs that could be easily obtained, separated, and standardized in both terrestrial and extraterrestrial locations.