Editorial Article
The Salt That Tripled and Took the Cell With It
The most expensive thing in the electrolyte is not made by an electrolyte company
Start with the mass balance, because it settles what kind of market this is. Every lithium-ion cell carries roughly 200 to 400 grams of fluorine, most of it in the electrolyte salt. That fluorine began, six processing steps upstream, as calcium fluoride in a fluorspar mine.
Global fluorspar production across all grades is approximately 8 to 9 million tonnes a year. China accounts for roughly 60 per cent of it, with some estimates placing the share at 69 per cent. About 60 per cent of the total is acid-grade fluorspar — CaF₂ at 97 per cent or better — and even that is not what the battery chain buys.
Battery-grade hydrogen fluoride demands tighter impurity caps at the premium tier: arsenic below 5 ppm and phosphorus below 100 ppm. Standard acidspar (CaF₂ ≥97%, SiO₂ ≤1%, CaCO₃ ≤1.5%) feeds the general HF industry perfectly well. The material at ≥97.5% CaF₂ with the lower impurity profile is a smaller subset again.
Three sequential bottlenecks therefore sit between a mine and a salt: geographic concentration, a premium-grade subset inside it, and purity-limited HF conversion capacity that sits mostly in one country. Any one of the three propagates directly to the salt, and the salt is the single largest line in the electrolyte.
The price is set two markets away, and one of them is sulphur
LiPF₆ tripled in the second half of 2025. It quoted RMB 168,500 a tonne in China in January 2026 — against RMB 79,500 for the sodium salt that is meant to replace it, a salt that is 52.82 per cent cheaper by weight and 61.87 per cent cheaper per mole of charge carrier.
That inversion is not a lithium story, or not only one. LiPF₆ is made from PCl₅, anhydrous hydrogen fluoride and a lithium source. So is every other hexafluorophosphate, including the sodium one. When the lithium salt’s price tripled, the shared fluorine and shared phosphorus went where the price was. A sodium salt producer and a lithium salt producer are drawing on the same balance.
Then the acid underneath the fluorine. HF is fluorspar plus sulphuric acid, so the anion in every one of these salts is exposed to a sulphur market that has nothing to do with batteries. On 28 February 2026 the Middle East war closed the Strait of Hormuz, through which roughly half of seaborne sulphur transits. The report records the resulting sulphur crisis as a live input risk to lithium salt and battery-metal processing, with China protecting domestic supply.
So the electrolyte is priced by four things: a lithium contract, a fluorspar grade, a sulphur chokepoint, and an HF impurity specification. Not one of them is a battery market.
And yet the price went bearish in the same window
In July 2026 battery-grade lithium carbonate spot traded at RMB 150,000 to 154,000 a tonne, down 5.0 per cent week on week on incremental supply and cautious demand. LiPF₆ production costs were modestly raised in the first quarter of 2026 — and the price index stayed bearish across North America, Europe and Asia, because ample inventories and steady cell-maker operating rates capped the upside.
That is the shape of this market and it is the most important thing to understand about it. The cost base moves on commodity shocks that the seller does not control. The selling price cannot recover them, because there is too much capacity standing behind it.
Tinci halted a RMB 2.65 billion domestic electrolyte project amid overcapacity and fierce competition, even as its offshore projects proceeded. Its operating rates ran at 90 per cent. A highly consolidated structure with six tier-one players and high capital intensity, in which the marginal tonne is still priced by whoever is willing to run at 90 per cent — that is not a contradiction. It is what a market with a fixed-cost plant and a specification product looks like when the specification stops being scarce.
The specification just changed, and it is the strongest demand signal here
China’s GB 38031-2025 took effect in July 2026 and requires no fire and no explosion in thermal runaway. That is not a fuel-economy target. It is a materials mandate.
Meeting it means reformulating with flame-retardant additives and novel lithium salts. That raises value per tonne, rewards performance-differentiated product over commodity blend, and consolidates supply toward whoever can absorb the compliance cost. It is the one demand driver in this market that does not depend on how many cars are sold.
From the other direction, and inside the same forecast window, the European Chemicals Agency’s Committee for Risk Assessment adopted an opinion in March 2026 on a broad PFAS restriction under REACH, covering many of the fluorinated compounds used in battery electrolytes. The final scope and transition periods remain uncertain. A phase-out, or stringent use limits, would force reformulation of most current electrolyte products sold in Europe.
Two regulators pulling in opposite directions: one mandating more performance chemistry into the cell, one threatening the chemistry family the cell currently runs on. Both raise the value of a formulation file. Neither is a volume assumption.
The substitution is low-risk on paper, and the reason is instructive
Samsung SDI targets mass production of all-solid-state batteries in the second half of 2027, with samples receiving positive customer feedback. EcoPro BM operates a 40-tonne-per-year sulfide-based solid electrolyte pilot plant, has passed qualification tests, and targets mass production in 2027.
Forty tonnes a year. Against a market of USD 11.49 billion in 2025 growing to USD 41.08 billion by 2033 at a 17.27 per cent baseline CAGR — 19.77 per cent optimistic, 14.77 per cent conservative — with the peak annual growth rate of 19.18 per cent arriving in 2029.
That is why the report puts substitution risk at low and overall market risk at high with geopolitical exposure high. The threat to this forecast is not solid state. It is the cost base, and it is the capacity.
Geography has become the product
Green Energy Origin acquired Mitsubishi Chemical’s electrolyte assets in Memphis, Tennessee and in the United Kingdom, alongside an existing 200,000-tonne-a-year plant in the Czech Republic. Feon Energy signed an MOU with Orbia for US-based next-generation electrolyte manufacturing.
Neither is cheaper than China on any operating line that matters. They are capital spent to buy market access in a decade when access is being rationed by rule rather than by price — and for a European or American cell maker, a supplier inside the customs perimeter is now worth a premium that has nothing to do with its cost of production.
The operating risk does not disappear when the plant moves. At CATL’s Debrecen facility in Hungary, a government office reported in March 2026 that hazardous materials were stored improperly and without permits in the semi-finished section; unauthorised use of the building was prohibited and an environmental protection procedure initiated. In June 2026 approximately one litre of diethyl carbonate leaked from a storage container during a transfer near production line 4, controlled within five minutes with no injuries.
A litre of solvent, five minutes, no injuries — and for a supplier, the consequences run to contract renegotiation, stricter storage protocols and higher insurance. That is what a hazardous-materials licence is worth, and it is a cost of the electrolyte that never appears in a salt price.
What the market is actually underwriting
Three exposures, and they do not offset each other.
One: a cost base nobody controls. Fluorspar grade, sulphur freight, lithium carbonate. The 2026 sulphur shock demonstrated that by-product feedstock dependence transmits disruption faster than contract terms can reprice it.
Two: a price that cannot hold a shock. Costs rose in the first quarter of 2026 and the index went bearish anyway. Any model that assumes cost pass-through in this market has to explain why it did not happen in the year it was most needed.
Three: a specification that is being rewritten twice at once. GB 38031-2025 pushes toward performance chemistry; the PFAS opinion pushes against the fluorine family itself. A formulator can win the first and lose the second, in the same eight years.
What is not on the list is demand. EV and stationary storage growth, capacity localisation and long-term supply agreements are the reasons this market reaches USD 41.08 billion by 2033. The volume is the easy part. The tonne is the problem — because the tonne is priced by three suppliers and only one of them makes batteries.
Frequently Asked Questions
How much fluorine is in a lithium-ion cell, and where does it come from?
Roughly 200 to 400 grams per cell, most of it in the electrolyte salt LiPF6. That fluorine begins six processing steps upstream as calcium fluoride in a fluorspar mine. Global fluorspar output is about 8 to 9 million tonnes a year across all grades, China accounts for roughly 60 per cent (some estimates 69 per cent), and only about 60 per cent of that is acid-grade material at CaF2 97 per cent or better. The battery chain needs the premium end of that subset, with arsenic below 5 ppm and phosphorus below 100 ppm.
Why did LiPF6 prices rise sharply and then turn bearish?
Because the cost and the price are set in different places. HF is fluorspar plus sulphuric acid, so the salt is exposed to a sulphur market that has nothing to do with batteries; on 28 February 2026 the Strait of Hormuz closure disrupted a chokepoint that carries roughly half of seaborne sulphur. LiPF6 is also made from PCl5, anhydrous HF and a lithium source, so it is exposed to lithium carbonate, which traded at RMB 150,000 to 154,000 a tonne in July 2026. In the first quarter of 2026 production costs were modestly raised and the price index still went bearish, because ample inventories and steady cell-maker rates capped the upside.
What are the two regulatory changes acting on battery electrolytes?
China's GB 38031-2025 took effect in July 2026 and requires no fire and no explosion in thermal runaway. Meeting it compels reformulation with flame-retardant additives and novel lithium salts, which raises value per tonne and favours suppliers able to absorb compliance cost. In the opposite direction, ECHA's Committee for Risk Assessment adopted an opinion in March 2026 on a broad PFAS restriction under REACH covering many fluorinated battery electrolyte compounds, with the final scope and transition periods still uncertain.
How serious is the solid-state threat to liquid electrolyte demand?
Low. Samsung SDI targets all-solid-state mass production in the second half of 2027 and EcoPro BM runs a 40-tonne-a-year sulfide solid electrolyte pilot targeting 2027 — set against a market of USD 11.49 billion in 2025 reaching USD 41.08 billion by 2033 at a 17.27 per cent baseline CAGR. The higher risks are cost and capacity: Tinci halted a RMB 2.65 billion domestic electrolyte project amid overcapacity while running at 90 per cent operating rates, and geopolitical exposure is high because China controls the fluorine chain from fluorspar through HF to LiPF6 conversion.