Determination of Elemental Composition and Content in Stream Sediments by Laser-Induced Breakdown Spectroscopy

The stream sediment (SS) records evolution information of the water system structure and sedimentary environment in specific regions during different geological periods, which is of great significance for studying the ancient planetary environment and the law of water system changes. Based on the SS of different geographical environments on Earth, remote laser-induced breakdown spectroscopy (remote-LIBS) technology combined with the multidimensional scaling-back propagation neural network (MDS-BPNN) algorithm was used to conduct an in-depth analysis of remote qualitative and quantitative detection of the elemental composition and content of SS. The results show that the detection system based on remote LIBS combined with an artificial neural network algorithm can achieve an ideal quantitative analysis of major and trace elements. The coefficients of determination (R-2) of the test set for major elements is greater than 0.9996, and the root mean square error (RMSE) is less than 0.7325. The coefficients of determination (R-2) of the test set for trace elements is greater than 0.9837, and the root mean square error is less than 42.21. In addition, for the application scenario of exploring extraterrestrial life, biominerals represented by stromatolite phosphorite (SP) are easy to form sand and enter into SS under weathering. Therefore, this paper discusses the feasibility of using remote-LIBS technology to detect and identify such minerals under the disappearance of SPs' macro- and micro-characteristics. From our research, we can find that remote-LIBS technology is the preferred candidate for discovering dust-covered biominerals. In geological environments rich in water system sedimentary rocks, such as Mars' ancient riverbeds, LIBS technology is crucial for deciphering the "life signals" hidden in the Martian sand.