The Model

New Technologies Make Exploration in Rocks Far More Effective

Our exploration model applies to all energy metals, but let’s focus on lithium for this discussion.

The demand for energy metals such as lithium is explosive, and much of that demand will be met by new sources in rock deposits such as claystones, pegmatites and volcanogenics.

Lessons From American Lithium

As co-founder and CEO of American Lithium, Mike and his team prospected and validated one of the largest lithium deposits in North America — all in claystone rocks. Throughout that process, Mike saw how the conventional approach to evaluating lithium in rocks is very costly and time consuming. It requires taking large number of samples in the field, gathering and storing them in batches, repeatedly sending the batches to a distant lab for analysis, and waiting days or weeks for the results. Under this model, a simple exploration project can easily take 8-12 weeks and cost of tens thousands of dollars.

Even then, this model’s time and cost structure typically imposes limits on the number of samples taken and analyzed, which in turn limits a full understanding of where a deposit’s highest grades really are. If a project moves to the extraction phase, this hazy understanding leads to excess material being extracted from a deposit and sent through processing. Processing is expensive, requiring resources such as sulfuric acid, soda ash, lime, and limestone, along with water, power, transportation, labor, environmental compliance and more. The result is a growing pile of low or no-grade material that cost as much to process as the valuable higher grade that’s kept.

A New Model

Mike and his team began developing a new, tech-driven model for exploration and extraction that can eliminate many of the inefficiencies of the traditional approach. Work began at American Lithium, and Mike continued it here at Global Subsurface. In 2021, Global Subsurface, American Lithium and Lawrence Berkeley National Laboratory were jointly awarded an Office of Science grant from the United States Department of Energy to refine the work.

The model starts with mobile spectrometry units that an operator can take into the field and instantly analyze a sample, with on-screen feedback showing exactly what minerals are present. The ease and speed of analysis means that one operator in a rented pickup truck can take hundreds if not thousands of samples in a matter of days.

With far more sample data to work with, the other component of our model is a machine learning materials database that finds patterns and trends in an ocean of data—things that can easily go undetected under the conventional approach where a geologist inspects columns on a spreadsheet. The AI-driven database is a tool, though, not a savant. Our decades of experience guides it to the lessons it needs to learn, and its deep analysis points out trends we couldn’t see otherwise.

What’s more, since this AI-driven analysis can take place on a laptop in the field, it directs a far better-informed exploration strategy that can be fine-tuned by the day. It also means that far less time is needed on a property to fully analyze its deposit, so leases for exploration rights can be dramatically shortened, allowing for more properties to be prospected along with properties that are not available on longer-term leases.

And, of course, if promising grades are found and long-term leases secured, prospectors can move into extraction with a clearer blueprint of where the highest grades are, resulting in smaller waste piles, less resources consumed, less environmental impact, and significant cost savings.

In the Field

We’re already applying this model to several exploration projects in claystones, pegmatites and volcanogenics. We’re excited by the results we’re seeing, and expect this model to help unlock significant new sources of energy metals in the future.