Browse
Log in / Register
Share
A laboratory push to turn coal ash into a strategic mineral source
March 17, 2025 — Deep inside Worcester Polytechnic Institute’s research facilities, a team of engineers, geochemists, and machine-learning specialists is testing a radical idea: using AI-designed biomolecules to extract rare earth elements and high-purity silica from the mountains of coal ash, red mud, and mine tailings scattered across the American landscape. The project, funded by a $3.3 million grant from the National Science Foundation, aims to reinvent industrial waste as a domestic supply chain for critical minerals.
The approach mirrors natural processes. Microscopic diatoms build glass-like shells at ambient temperatures. Sea sponges polymerize silica in seawater. By studying these biological mechanisms, the WPI team hopes to replicate them at industrial scale—replacing thousand-degree furnaces and harsh chemical baths with low-energy proteins and catalysts.
U.S. coal-fired power plants still generate tens of millions of tons of ash every year, and decades of legacy disposal have left behind roughly two billion tons of coal ash in ponds and landfills, according to industry inventories compiled by the American Coal Ash Association. Alongside it, aluminum refineries pump out caustic red mud, and abandoned hardrock mines have left thousands of acres of tailings that leach heavy metals into nearby rivers.
All three streams contain materials America currently imports at great cost. Rare earth elements—essential to electric vehicles, wind turbines, defense electronics, and medical imaging—sit locked in these sludges in concentrations that traditional extraction methods can’t economically recover. The same is true of silica, the starting point for glass, semiconductors, silicones, and concrete admixtures.
Import dependency frames the urgency. USGS Mineral Commodity Summaries continue to rank rare earths among the most import-reliant minerals, with the United States depending on overseas suppliers—predominantly China—for the vast majority of its REE needs. Clean energy tax credits and defense stockpile requirements have pushed federal agencies to seek domestic alternatives. New mining, however, carries its own environmental, permitting, and community costs, which is why waste streams look increasingly attractive.
The core science is biosilicification—the same process diatoms use to condense dissolved silica into ornate cell walls. In nature, enzymes guide silicon chemistry without high temperatures or extreme acidity. The research team is hunting for synthetic biomolecules that can do the same thing to industrial sludge: selectively bind to silicon and critical metals, pull them out of complex mixtures, and release them for reprocessing.
The payoff would be significant if it works. Instead of importing silicon metal refined in overseas arc furnaces, a refinery could digest its own waste into feedstock for solar panels, battery components, or construction materials. A coal ash landfill could become a functioning raw-materials depot.
Leading the effort is Mingjiang Tao, an associate professor in civil and environmental engineering at WPI, who directs the project’s core research on bio-enabled metallurgy and biosilicification. Alongside him are Carrick Eggleston, a geochemist who specializes in silicate dissolution and repolymerization chemistry, and Yan Wang, a materials scientist widely known for lithium-ion battery recycling who brings experience in closed-loop mineral processing.
The five-year grant comes from the National Science Foundation’s Growing Convergence Research program, which deliberately funds projects that cross conventional scientific boundaries. The institutional roster extends well beyond Worcester: researchers from George Mason University, UC San Diego, UMass Amherst, and the University at Buffalo are contributing expertise in molecular engineering, computational biology, AI modeling, and isotope geochemistry.
Finding a biomolecule that efficiently attacks a specific silicate matrix is no small lab chore. The search space is immense, and running physical experiments on every candidate could consume the entire grant period. Instead, the team is using AI to model how bioengineered molecules bind to silica-rich waste surfaces, allowing them to shortlist the most promising targets before touching a test tube.
Those AI predictions feed into geochemical maps of the dissolution and repolymerization pathways—essentially a step-by-step picture of how silicon and rare earths move from a solid, waste-bound state into soluble forms that can be recovered. The pairing of predictive modeling with thermodynamic mapping is what gives the project its convergence character: biology, geochemistry, and computation under one roof.
The team is also borrowing proven playbooks from rechargeable battery recycling. Wang’s earlier work on extracting precious metals from spent lithium-ion batteries has already shown how selective leaching and precipitation can separate tightly interwoven compounds. Scaling up those methods, or adapting them to coal ash chemistry, could shave years off development timelines.
This WPI project is not running in isolation. The U.S. Department of Energy has channeled substantial funding through its Unlocking Critical Minerals from Coal and Coal Byproducts program, led by the National Energy Technology Laboratory (NETL). That effort has already produced pilot-scale extraction systems working on real coal ash and acid mine drainage sites, mostly based on chemical precipitation and ion exchange.
What is new about the WPI approach is its biological layer. Where NETL’s chemistry-based recovery trains rely on reagent consumption and energy input, bio-directed pathways are designed to operate at ambient conditions with minimal waste discharge. If WPI’s biomolecules perform as modeled, they could offer a complementary—and considerably greener—recovery route, potentially integrated with the extraction infrastructure NETL has already validated.
An estimate cited by the research team puts the scale of opportunity in eye-catching terms: U.S. coal ash landfills alone hold roughly 11 million tons of rare earth elements, valued at about $8.4 billion. That figure represents roughly eight times the nation’s entire reported domestic reserve of rare earth minerals, according to the researchers—meaning a substantial “mine” already sits above ground, close to roads, grid power, and existing industrial infrastructure.
There is strategy beyond commerce here. Rare earths recovered from waste could feed Department of Defense stockpiles, permanent magnets, guidance systems, and advanced aircraft sensors. A domestic supply line that starts with a toxic landfill and ends with defense-grade magnets is the kind of circular-economy win that policymakers on both sides of the aisle have been seeking.
WPI’s longer-term goal is even more ambitious: a “whole-material” system that leaves no residue behind. Residual silica would flow into concrete admixtures, glass manufacturing, or silicone chemistry, while recovered critical minerals move into high-tech supply chains. Waste becomes a complete resource stream instead of a perpetual containment problem.
Whether these biomolecules prove viable at industrial scale remains to be tested; moving from beaker to holding pond is always the hardest part of any process breakthrough. But the direction is clear: the United States is growing weary of both overseas dependency and the environmental burden of its own industrial past.
If WPI’s bio-AI approach succeeds—or even if it simply advances the science enough for others to build on—coal ash ponds that once symbolized industrial pollution will increasingly be viewed as strategic assets. The next generation of domestic critical minerals may not emerge from a new mine shaft but from a biology-inspired recipe cooked in a Worcester laboratory, guided by algorithms and powered by the waste itself.
