Use AI to find rare minerals
(Nanowerk News) Machine learning models can predict the location of minerals on Earth – and possibly other planets – by leveraging patterns in mineral associations.
Science and industry seek deposits of minerals to better understand our planet’s history and extract them for use in technologies such as rechargeable batteries.
Shaunna Morrison, Anirudh Prabhu and colleagues set out to create a tool to find the occurrences of certain minerals, a task that has long been both an art and a science, relying on individual experience, along with a healthy dose of luck.
The team created a machine learning model that used data from the Mineral Evolution Database, which includes 295,583 mineral sites from 5,478 mineral species, to predict the occurrence of previously unknown minerals based on association rules.
The authors tested their model by exploring the Tecopa basin in the Mojave Desert, the famous Martian analogous environment. The model can also predict the location of geologically important minerals, including alteration of uraninite, rutherfordine, andersonite, and schröckingerite, bayleyite, and zippeite.
In addition, the model finds areas of promise for critical rare earth elements and lithium minerals, including monazite-(Ce), and allanite-(Ce), and spodumene.
Mineral association analysis can be a powerful predictive tool for mineralogists, petrologists, economic geologists, and planetary scientists, according to the authors.
This research was published in PNAS Nexus (“Predicting new mineral occurrences and planetary analogue environments through analysis of mineral associations”).
