The Potential of AI in Discovering New Minerals on Earth and Beyond

Rocks and minerals play a crucial role in the development of civilizations, providing essential raw materials. In the context of a technological society, minerals, especially those containing rare elements, are highly sought after. Traditionally, mineral discoveries have been made through a combination of perseverance and luck.

Over the past two centuries, scientists have come to realize that minerals are not randomly distributed throughout the Earth. Instead, they are often found in specific combinations known as paragenesis. A paragenesis is a collection of minerals that form under particular physico-chemical circumstances, such as a specific chemical composition in the host rock or the presence of specific temperature and pressure conditions.

In an exciting new study, researchers have explored the potential of Artificial Intelligence (AI) to predict the most likely locations of minerals by uncovering the governing rules behind these mineral combinations.

The research team fed an AI with data from the Mineral Evolution Database, which contains detailed information on 295,583 mineral localities and 5,478 mineral species. They employed a powerful method called Mineral Association Analysis (MAA) to identify relationships between different minerals and establish rules based on these associations.

To validate their model, the researchers examined the Tecopa formation in the Mojave Desert, California. This geologic formation consists of a series of lake and river sediments deposited millions of years ago in a tectonic basin, combined with hot fluids from the underground, resulting in the formation of rare minerals.

By analyzing the mineral composition of the Tecopa formation, the AI model successfully predicted the presence of exotic uranium minerals like rutherfordine, andersonite, schröckingerite, bayleyite, and zippeite. Moreover, the model identified promising areas for critical rare earth elements and lithium minerals, including monazite, allanite, and spodumene.

The authors believe that mineral association analysis has the potential to uncover unknown mineral deposits and accurately predict the mineral inventory of any given locality, provided sufficient data is available. Furthermore, this model can be applied not only to Earth but also to other rocky planetary bodies.

NASA’s Reconnaissance Orbiter has been mapping minerals on Mars for the past 16 years but has primarily focused on the most common rock-forming minerals. By utilizing Mineral Association Analysis, scientists may be able to discover more exotic minerals that have eluded detection thus far.

The study titled “Predicting new mineral occurrences and planetary analog environments via mineral association analysis” was recently published in the prestigious journal PNAS Nexus (2023) and has been made possible with the collaboration of PNAS Nexus.

+

Denial of responsibility! TechCodex is an automatic aggregator of the all world’s media. In each content, the hyperlink to the primary source is specified. All trademarks belong to their rightful owners, and all materials to their authors. For any complaint, please reach us at – [email protected]. We will take necessary action within 24 hours.