Industries / Mining and materials / Mining and quarrying / Salt and mineral extraction / Lithium mining
Direct extraction
Direct lithium extraction (DLE) pulls lithium out of brine in hours with a sorbent, an ion-exchange material, a solvent or a membrane, instead of waiting months for the sun to evaporate the water.
4
Milestones achieved, from 2015 to 2025.
2
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
2015
Simbol's Salton Sea collapse
Simbol, founded in 2008, licensed technology from Lawrence Livermore National Laboratory to take lithium, manganese and zinc from the hot brine that geothermal power plants pump up near California's Salton Sea, and in 2010 won US$3 million from the US Department of Energy toward a US$9.6 million project. Its former chief executive, John Burba, had earlier pioneered selective lithium extraction at FMC Lithium. By 2013 the company said it had made battery-grade lithium carbonate from geothermal brine and run its demonstration plant for more than 9,000 hours, and in January 2015 it planned full-scale production for 2018. Within weeks it laid off many of its workers, and the Los Angeles Times reported in 2020 that it collapsed in 2015, shortly after Tesla offered to buy it for US$325 million.
LessonA process that runs in a demonstration plant still fails if the company cannot pay for the full plant.
Credit
Simbol Materials; Lawrence Livermore National Laboratory (technology)
2021
Sorbents for magnesium-rich lakes
Lithium can be precipitated from brine with soda ash at a reasonable cost only where there is less than about six times as much magnesium as lithium, and many Chinese salt-lake brines hold far more, by a 2026 review in Precision Chemistry. In August 2021 Qinghai Salt Lake Fozhao Lanke Lithium Industry in Qinghai province reported 10,000 t a year of lithium carbonate in industrial production from a plant of adsorption towers followed by nanofiltration and reverse osmosis membranes, with a 20,000 t a year battery-grade line in trial production. A January 2025 survey by Centro de Energía UC in Chile lists the adsorption plants of Lanke Lithium and Zangge Lithium in China among the commercial DLE operations.
Credit
Qinghai Salt Lake Fozhao Lanke Lithium Industry; Zangge Lithium
2023
Chile chooses direct extraction
Chile's national lithium strategy, launched in April 2023, commits the country to move from evaporation ponds to DLE and to protect at least 30% of its salt-lake ecosystems, by Argus Media's account. Chile was the world's second-largest lithium producer, with an estimated 261,000 t of lithium carbonate equivalent (LCE, the weight of lithium carbonate that a given amount of lithium would make) in 2024, by US Geological Survey figures that Argus cites.
Credit
The Government of Chile
2025
A royalty for brine lithium
On 28 May 2025 the Arkansas Oil and Gas Commission voted unanimously to set a royalty of 2.5% on lithium extracted from brine in the Reynolds Unit, in Columbia and Lafayette counties, for the venture of Standard Lithium and Equinor. It was the state's first lithium royalty rate. State law requires a "fair and equitable" royalty, and the venture needed one to seek financing for a project estimated at more than US$1 billion. The commission had turned down an earlier application, which offered 1.82%, after at least 100 people and companies objected.
LessonBefore a DLE plant on oilfield brine can be financed, the state has to settle what the owners of the brine are paid.
Credit
Arkansas Oil and Gas Commission
Watch for
Rincon's large DLE plant
Rio Tinto says it began building a DLE plant at Rincon in Salta, Argentina, in 2025, sized for 60,000 t a year of battery-grade lithium carbonate (a 3,000 t starter plant and a 57,000 t expansion) at a capital cost of US$2.5 billion, with first production expected in 2028 and three years to reach full output. In 2026 it secured US$1.175 billion of loans from the International Finance Corporation, IDB Invest, Export Finance Australia and the Japan Bank for International Cooperation.
Achieved whenthe full plant ships battery-grade lithium carbonate at a reported rate, confirmed in audited filings or by an outside party.
Claimed by
Rio Tinto (reported by Mining Weekly, April 2026)
DLE tests on Andean brine
ENAMI, Chile's state mining company, says eight international laboratories tested DLE on brine samples from its Altoandinos project. By its account they used on average 36 m³ of water per tonne of LCE, 55 times less than evaporation ponds, and recovered 92% of the lithium against 42% in ponds, and a 75,000 t a year DLE plant would need 10 ha against 1,020 ha for ponds.
Achieved whenthe test methods and results are published in full, or a pilot plant on Altoandinos brine repeats the recovery and water figures under outside review.
Claimed by
ENAMI, June 2025 (reported by Argus Media)
Open
01
DLE promises, tracked
A public ledger of DLE projects worldwide that records each first-production date a company has announced, with the date it announced it, the date the first product actually shipped, and the latest reported output against nameplate capacity, updated every quarter.
The briefCredit · unclaimed02
One test case for reinjection
A shared test problem for pumping lithium-stripped brine back into a salt flat's aquifer, of the kind groundwater modelers use to check their programs against each other: a stated salar shape, brine densities, pumping and injection rates, run in at least two free programs with and without density effects, reporting how fast the lithium grade falls at the production wells.
The briefCredit · unclaimed
Open 01 · the brief
DLE promises, tracked
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public ledger of DLE projects worldwide that records each first-production date a company has announced, with the date it announced it, the date the first product actually shipped, and the latest reported output against nameplate capacity, updated every quarter.
Why it is still open
Dates move and projects stop: Simbol planned full production for 2018 and collapsed in 2015. The lists that exist are snapshots by stage. A January 2025 survey by Centro de Energía UC lists DLE projects by stage and calls its list not exhaustive, and an October 2025 report by the Federal Reserve Bank of Dallas counted 66 US lithium projects, 21 of them planning DLE, and found 38 still exploring. Searches for this brief found paid project databases but no free ledger of promised against actual dates.
Built with what exists
Company releases and filings on EDGAR, SEDAR+ and the Australian Securities Exchange; the project lists in the Centro de Energía UC and Dallas Fed reports; a large language model such as Claude to pull dates from releases, with a person checking each one (see the Note "Seven more flagged"); a spreadsheet kept on GitHub or Zenodo so every change is dated.
First step
Take the Centro de Energía UC project list and, for ten projects, record the first start date each company announced and the current one.
Done looks like
At least 40 DLE projects in at least six countries, each with every announced start date and its source, and a published median delay in months for projects announced before 2023.
Open 02 · the brief
One test case for reinjection
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A shared test problem for pumping lithium-stripped brine back into a salt flat's aquifer, of the kind groundwater modelers use to check their programs against each other: a stated salar shape, brine densities, pumping and injection rates, run in at least two free programs with and without density effects, reporting how fast the lithium grade falls at the production wells.
Why it is still open
DLE with aluminum-based sorbents had not yet been run at industrial scale with the spent brine returned underground, because it could dilute the brine the plant feeds on; Eramet sends its spent brine to evaporation ponds while it studies reinjection (Centro de Energía UC, January 2025). The public models disagree in method. A 2021 study of the Salar de Atacama, published by Jade Cove Partners and Zelandez, used a model that treats all water as the same density and calls itself a starting point; a 2026 abstract by a team at Spain's CSIC and two Barcelona universities reports that responses with density included differ markedly from constant-density predictions. Searches for this brief found single-site models but no shared test case.
s3.amazonaws.comstatic1.squarespace.commeetingorganizer.copernicus.org
Built with what exists
MODFLOW 6, the US Geological Survey's free groundwater program, with its buoyancy package for density-driven flow and its groundwater transport model; FloPy, its Python interface; SEAWAT, an older USGS program for variable-density flow, as the second code; the hydraulic values published in the 2021 Atacama study.
First step
Rebuild the 2021 Atacama model in MODFLOW 6 from the values in the report, reproduce its main result, then switch on the buoyancy package.
Done looks like
The test case and input files published under an open license, results from two programs that agree on lithium grade at the wells within 5% after 20 simulated years, and the gap between runs with and without density effects reported.