TLDR. Part 1 covered the seven-step flowsheet, part 2 covered why brine chemistry makes each DLE technology a site-specific bet. This issue is the proof: the one DLE plant that has actually run commercially since 1997, what its three decades did and did not establish, and the question that actually matters once you strip away the headline.
Let's get into it.
Submit a battery industry whisper…
The one DLE plant that has been running since Clinton was president
Here is the part that should reset how you read every DLE headline: the first commercial DLE facility opened at Salar del Hombre Muerto, Catamarca Province, Argentina, in the late 1990s. Remote, high-altitude salar, production wells, a selective adsorption plant pulling lithium out of a soup of sodium, potassium, magnesium, calcium, and boron. Rio Tinto, the current owner, states that Fénix has made lithium carbonate from Hombre Muerto brine since 1997 using proprietary selective-adsorption DLE technology.
So when a company frames DLE as next-generation, remember an early commercial version has run in Argentina for nearly three decades. The genuine innovation on the table now is not inventing selective lithium extraction, it's deploying DLE across a wider range of brines, at larger scale, lower concentrations, higher recovery, and lower water and chemical use, at a cost that actually competes with conventional production, all at once. Whether that combination is achievable is still an open question.
The timeline:

At Fénix, brine gets pumped from Hombre Muerto, fed to a proprietary selective adsorption plant, and comes out as a lithium-rich, lower-impurity solution, which then gets concentrated through both solar evaporation and mechanical vapor-recompression evaporation before feeding lithium carbonate and lithium chloride production. Notice what that means: Fénix is a hybrid, DLE plus ponds, not DLE instead of ponds. That should retire the assumption that "real DLE" has to eliminate evaporation to count.
Rio Tinto states the first phase of a 10,000-tonne expansion completed in May 2024, with current operating capacity listed at 32,000 tonnes per year, including 4,000 tonnes LCE (lithium carbonate equivalent) of lithium chloride capacity through Güemes. Fénix is not just old, it's an old DLE operation still being expanded, on a hybrid flowsheet that contradicts half the DLE marketing out there right now.
What Fénix actually proved (and the much longer list of what it didn't)
What it proved: DLE can work at industrial scale. Selective adsorption can run for decades commercially. DLE can improve separation ahead of final concentration. DLE and evaporation are not mutually exclusive. A hybrid flowsheet can beat a technically "pure" one. Brine-specific process knowledge is a real, durable operating advantage.
What it did not prove: That the same sorbent works on every brine. That adsorption beats every other DLE family. That DLE can eliminate ponds and still pencil out economically. That DLE universally cuts fresh-water use. That high recovery guarantees low cost. That reinjection always works. That DLE hands you battery-grade product automatically. That a lab result scales into a reliable plant. That a modern geothermal or oilfield-brine project shares Fénix's economics.
Fénix proved DLE can work. It did not prove every DLE technology works on every brine, and it definitely did not prove every application pencils out. All eyes are now on Eramet at Centenario-Ratones to find out if lightning strikes twice in the same mountains.
Why everyone is suddenly obsessed with DLE
The drivers are real: demand for geographically diversified lithium supply, pressure to improve recovery from existing brine assets, a desire to cut dependence on giant pond systems, government money chasing domestic critical minerals, genuine advances in sorbents, membranes, and process control, and interest in geothermal and oilfield brines outside the traditional Lithium Triangle. Add investor appetite for anything promising a smaller footprint and higher recovery, and you have a full hype cycle.
Two things worth separating from the hype directly. Shorter cycles are not automatically cheaper: a shorter extraction residence time can cut work-in-process inventory, but once a conventional pond system reaches steady state, faster residence time does not by itself translate into lower unit operating cost. Nontraditional brines are not automatically a win either: geothermal and oilfield brines could be a real strategic supply source, but that depends on the whole list above, well flow, impurities, temperature, pumping cost, reinjection, water balance, infrastructure, not just concentration and recovery.
DLE is a flowsheet, not a machine
DLE's pitch is genuinely attractive: faster recovery, smaller footprint, higher recovery, wider resource applicability. What financiers, customers, and regulators actually need before they believe it is evidence: long-duration continuous operation, stable recovery under changing feed conditions, realistic water and reagent consumption, sorbent or membrane life, reinjection performance, and total capital and operating cost, plus a financeable return on what is usually a multibillion-dollar facility. So ask the question most companies will not answer directly, because it ruins the thesis: what is the total cost from brine to qualified lithium product, including pretreatment, extraction, water, reagents, concentration, conversion, waste management, and reinjection?
Direct lithium extraction could turn out to be one of the more consequential shifts in brine-based lithium production. It may genuinely enable faster recovery, smaller footprints, and access to brines that ponds handle badly. None of that makes it a shortcut around geology, hydrogeology, water management, separation chemistry, corrosion, scaling, chemical logistics, downstream conversion, or commercial qualification.
Success is not a laboratory recovery percentage. It is years of stable operation, defensible mass and water balances, manageable reagent consumption, consistent product quality, reliable reinjection, and a total cost that actually competes.
"Does DLE work?" is not the interesting question anymore. Several forms of selective lithium extraction clearly work, Fénix has receipts. The question worth asking is: which DLE process works on this particular brine, at this flow rate, with this water and reagent balance, for enough cycles, to produce a qualified lithium chemical at a competitive total cost? The answer looks a lot less exciting than the headline, and a lot more like an actual engineering project.
Some Resources We Love:
Rio Tinto's Fénix operations page, for the operator's own description of the Hombre Muerto flowsheet and current capacity.
Eramet's Centenario-Ratones project updates, the newest large-scale adsorption reference point, still ramping.
