The Pore Picks Lithium
How a crosslinking-regulated membrane separated lithium from magnesium in natural salt-lake brine—and how much engineering still remains
The Resource Is Dissolved Inside a Crowd
Finding lithium is not the same as producing usable lithium. In salt-lake brines, the desired ions share the liquid with much larger quantities of magnesium and other dissolved minerals. A process has to distinguish the valuable material from that chemical crowd before the resource can become a useful supply.
That makes lithium recovery a selectivity problem, not simply a filtration problem. Lithium and magnesium can be difficult for a membrane to distinguish because their hydrated sizes are close enough that an ordinary pore may admit both. Open the path too widely and competing ions pass through. Make it too restrictive and useful lithium transport can slow.
The challenge is especially relevant as demand grows. The International Energy Agency’s 2026 outlook projects lithium demand to rise more than threefold by 2040 and says projected supply deficits persist through 2035. Those figures do not validate any particular extraction technology, but they explain why better separation methods matter.
The new result is about one possible tool for that job: a membrane whose internal structure is controlled during manufacture so that it can favor lithium over magnesium under electrodialysis.
Crosslinking Turns a Membrane Into a Molecular Gate
In a Nature Communications study published on October 31, 2026, a multinational team led by Sun Yat-sen University regulated how tightly a polymer membrane crosslinked as it formed. Crosslinking joins polymer chains into a network. Changing its degree changes the distribution of subnanometre pores in that network.
That manufacturing step is the central idea. The membrane is not presented as a magnetic collector or a complete extraction plant. It is a precisely formed barrier whose pore architecture can be tuned into a narrow window for lithium-magnesium separation.
Under one defined electrodialysis condition, with a magnesium-to-lithium mass ratio of 140, the researchers reported a lithium-to-magnesium separation factor as high as 6,454. A separation factor measures how strongly the process favors one ion relative to another under the tested conditions.
It is not a recovery percentage, a purity percentage or an industrial production rate. The number therefore shows strong selectivity in that experiment, but it does not say how much saleable lithium a full plant would produce.
The Natural-Brine Test Is the More Revealing Result
Idealized mixtures can hide the complexity of a real resource, so the team’s untreated natural salt-lake brine test matters. That brine contained multiple competing ions and had a magnesium-to-lithium mass ratio of 87. In that test, the membrane maintained a separation factor above 1,200.
This is evidence that the crosslinking-regulated pore architecture retained substantial selectivity in a chemically crowded natural sample. It does not make the membrane a commercial solution, but it goes beyond a result obtained only in a simple two-salt liquid.
The distinction between the two numbers is important. The peak factor of 6,454 belongs to the defined high-magnesium laboratory condition. The value above 1,200 belongs to untreated natural brine with multiple competing ions. Combining them into a single performance claim would erase the conditions that give each result meaning.
Neither result establishes total lithium yield, final battery-grade purity, energy consumption, water demand, chemical inputs, capital cost or environmental impact. Those questions require measurements across an integrated process, not one selectivity metric.
A Bigger Sheet Is a Milestone, Not a Factory
Membrane performance can change when a material moves from a small laboratory coupon to a larger manufactured area. The researchers fabricated a sheet measuring 25 by 35 centimetres and reported stable performance through 15 days of repeated operation.
Those are meaningful scale-up signals. A larger sheet tests whether the pore structure can be made consistently across more material, while repeated operation begins to test whether the performance persists beyond a short demonstration.
But the limits are just as important. Fifteen days is not a multi-year service life, and one sheet is not a commercial module or a full extraction facility. The study does not establish sustained throughput across industrial membrane stacks, resistance to months of fouling, durability through repeated cleaning, or consistent performance as natural brine composition changes.
The next steps must connect material science to process engineering. Manufacturers would need to reproduce the pore distribution across larger areas, assemble the membranes into durable electrodialysis modules, and determine how operation, cleaning and replacement affect performance over time.
What Would Turn the Result Into a Process?
Several measurements would clarify the membrane’s practical value. Engineers need lithium throughput per unit area, total recovery across all process stages, the composition and purity of the output, electricity and chemical requirements, water balance, cleaning frequency, membrane lifetime and cost.
They also need extended tests with variable natural brines. Minerals can accumulate on membrane surfaces, and the chemistry of a resource can shift. A system that works for days under a controlled protocol may behave differently after months of repeated concentration and cleaning.
The environmental and economic case likewise remains open. Strong selectivity could be useful, but selectivity alone cannot establish that a process is low-cost, low-energy, low-water or environmentally preferable. Those conclusions need full process data and comparative assessment.
The honest conclusion is narrower and more useful than a breakthrough slogan. The researchers showed that controlling polymer crosslinking can tune subnanometre pores for strong lithium-magnesium separation, including in untreated natural salt-lake brine. They also moved from a tiny material concept to a larger sheet with 15 days of repeated operation.
That is a credible step toward a process component. The next test is whether the same precision can survive manufacturing scale, durable modules and the economics of continuous operation. The membrane has crossed an important laboratory threshold. It has not yet crossed from a paper to a mine.
Stay ahead of the curve! Subscribe for more insights on the latest breakthroughs and innovations.


