The need for critical minerals (CM) and rare earth elements (REEs), which are essential materials in tech such as solar panels, smart phones, and computers, is high and likely to grow.
Rare…but not rare:
-
The crux of the problem is in the name, “rare,” which is a bit of a misnomer in terms of raw relative abundance but more related to the economic accumulation of CMs and REEs in one location that can be profitably extracted.
-
Locating and extracting the materials are also a huge part of what makes REE and CM procurement rather expensive.
Driving the news: A partnership of the National Energy Technology Laboratory (NETL), Ramaco Resources, and Weir International seeks to transform and improve the efficiency of these processes. In short, they want to make REEs and CMs easy to find and quantify.
Background: The partnership began in 2018 and is funded by the DOE’s Technology and Commercialization Fund.
How will they do it? Bright lights and machine learning. Gabriel Creason, a geo-data scientist at NETL gave a high-level overview of the process: “We are combining our proven method of predicting geologic deposits with handheld X-ray fluorescence (XRF) and machine learning for near-real-time characterization of a site’s critical mineral potential.”
The machine-learning technology was developed by NETL’s Geoscience Artificial Intelligence & Assessment (GAIA) research group.
From data to deposits: The method is a classic data-driven model informed by physics—specifically geologic and geospatial knowledge-data guided by REE accumulation mechanisms—that systematically assess and identify areas of higher enrichment.
How it works:
-
Gather and inventory available knowledge and data.
-
Group regions together based on shared geologic characteristics.
-
Calculate the potential enrichment score, which describes and quantifies areas where how favorable past conditions were for REE accumulation.
-
Perform spatial integration or predictions and observations.
The latest move forward in the project gives a framework that can guide and inform the data acquisition process.
Get it fast: The technology breaks down traditional barriers to resource characterization like limited field data and slow, expensive lab analyses by enabling real-time resource detection and quantification.
Power to the people: The project’s lead researcher summed up the motivations behind the work saying, “Putting the power of rapid detection and resource prediction in the hands of the commercial sector helps unlock new sources of CMs.”
To read the release about the work from NETL, go here.