The Smackover Limestone Formation represents a vital resource for both geothermal energy and lithium (Li) recovery, especially as the demand for critical minerals and renewable energy continues to rise. This study leverages advanced artificial intelligence (AI) and machine learning (ML) techniques to optimize the identification, extraction, and co-production of Li from geothermal brines. The Smackover Formation has played a pivotal role in the U.S. energy sector for over a century, contributing significantly to the country’s conventional energy economy. Since the 1950s, the region’s brines have been commercially tapped for bromine, which is found in abundance within the deep sedimentary basins of the formation. These brines, however, also contain a complex array of valuable components, presenting new opportunities for the extraction of critical minerals such as lithium, potassium, boron, and iodine. Recent reports have identified lithium concentrations as high as 1,700 mg/L in these deep-basinal oilfield brines of the Gulf Coast. Building upon extensive multi-dimensional datasets from the Smackover region—encompassing geological, geochemical, and geothermal data—this study employs LANL’s nonnegative matrix factorization with k-means clustering (NMFk) to uncover hidden signatures within the data. This enables the identification of lithium-rich zones and the development of optimized extraction strategies. By analyzing key attributes such as brine composition, temperature, and pressure, the AI/ML models predict the factors that drive lithium mobilization and extraction efficiency. This AI-driven approach refines geothermal brine management, enhancing both lithium recovery and geothermal energy production. The data-driven methodology offers a more efficient and sustainable means of resource identification while reducing the environmental impact of traditional methods. Our research advances the integration of AI and ML in resource assessment, presenting a novel approach to the co-production of geothermal energy and critical minerals.

Press the Back button in your browser, or search again.

Copyright 2025, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.