Waiting for the sodium battery? 2027 is a bet on the lithium price

Salt is everywhere. Sodium is the sixth most abundant element in the earth's crust, the raw material costs a few hundred dollars a tonne, it is not toxic and it shrugs off the cold. You can build a battery out of it.

So you go and buy one. You can't.

The video explains why in ten minutes, with the b-roll attached. This article is the dossier underneath it: the tables, the measurement dates, the two price quotes that contradict each other, the four trading days on the Guangzhou exchange, and the methods section of the paper everyone points at without opening it. If you never watch the video, you miss nothing here.

One thing up front, because it shapes how you read the rest: this is not about whether sodium is better. It is about why a technology that demonstrably works is not on the shelf.

Key takeaways

• The year 2027 comes from trade press, not from research. The only peer-reviewed study that actually models the question says "in the 2030s" — with a caveat attached.
• "Parity" in that study means one chance in five. The authors define Price Parity as the point where sodium has a 20% probability of being cheaper than LFP. Only above 80% does it become Price Advantage.
• The raw material is dirt cheap and the cell is not. Sodium carbonate costs hundreds of dollars a tonne, lithium carbonate tens of thousands. Yet sodium manufacturer HiNa quoted a higher price per watt-hour for sodium than for lithium in March 2026.
• The factory is not the barrier. Sodium-ion is treated in the field as a drop-in technology for existing lithium-ion lines. The machines stay; the materials and the suppliers change. It is not happening because it does not pay, not because it cannot be done.
• Two companies tried and no longer exist. BlueSky Energy, maker of the GreenRock saltwater battery best known in the Netherlands, went bankrupt in September 2022. Natron Energy raised $363 million and filed a WARN notice on 3 September 2025.
• Sodium gets cheaper than lithium when lithium gets more expensive. On 9 August 2025 the mining permit for CATL's Jianxiawo project in Jiangxi expired. Lithium on the Guangzhou exchange rose almost a fifth in four trading days, and it was precisely in those weeks that the headlines appeared saying sodium was "cheaper again". Nothing about sodium had changed.
• China is not holding sodium back — it is pushing. China holds 8% of lithium reserves and imported roughly 60% of the lithium it processed in 2024. Its power sits in refining, and that position does not change when the metal changes.

"Salt battery" means three different things

Before a single figure: the word is bad, and that is part of the problem.

In the Netherlands, most people searching for zoutbatterij are looking for heat storage — a tank of salt that holds warmth for your heating system. Different product, different market, different physics from what this article covers. The most-watched Dutch video on that keyword is about heating your house.

What this article covers is electrical storage, and within that there are again three families: sodium-ion, solid-state sodium chloride, and saltwater. They share an element and almost nothing else.

Within sodium-ion — the family closest to a home battery — there is yet another split, and it determines what a cell can do:

CATL's layered oxide dominates the EV segment; Prussian blue targets stationary storage. Natron Energy, which returns below, was a Prussian blue company.

The raw material is dirt cheap. The cell is not.

This is the heart of the misconception, and it is a reasonable one: if the stuff you build it from costs almost nothing, surely the product gets cheap?

The raw-material side does check out. Sodium carbonate — soda ash, what actually enters the factory — costs hundreds of dollars a tonne. Battery-grade lithium carbonate costs tens of thousands.

Then the cell. This is the figure the argument breaks on:

That is the company selling sodium, stating that sodium costs more.

Now the honesty a video cannot carry: our own knowledge base holds a different figure.

Same unit, different answer. The likely explanation is that the first is a market-wide band from a smaller player and the second is product-specific quoting from two large manufacturers, with different reference dates. What matters: both point the same way. Sodium is at parity or more expensive. Neither is the figure, and presenting either as the figure does something the sources do not support.

For scale on the other side: LFP fell to a record low in 2025, averaging $108 per kilowatt-hour at pack level, with lowest observed prices of $36/kWh at cell level and $50/kWh at pack level for stationary storage. Cause: overcapacity and price competition. That is the target sodium has to duck under, and the target keeps moving down.

The factory is not the barrier

The usual explanation is that the industry would have to retool, and that this is too expensive. It does not hold.

Sodium-ion is treated in the field as a drop-in technology for existing lithium-ion lines. Machine Design puts it plainly, on the authority of the IDTechEx report Sodium-ion Batteries 2023-2033: "Na-ion is a drop-in technology for the current production lines of Li-ion batteries."

What changes are the materials — hard carbon instead of graphite, a different cathode — and the suppliers. The machines largely stay where they are.

That inverts the argument. If sodium were cheaper, the switch could happen fast. It is not happening because it does not pay.

Two companies tried and no longer exist

This is the part that rarely makes an overview, because it is not a technical matter.

BlueSky Energy (Austria), maker of the GreenRock saltwater battery best known in the Netherlands, was founded in 2013 and set up production in Frankenburg in 2020. In September 2022 it filed for insolvency, citing overindebtedness, after persistent supply problems dating back to Covid and despite seven-figure funding. In July 2023 the remaining batteries, production equipment and inventory were auctioned off. Part of the stock was defective. Anyone who owns one has had no factory support since.

Natron Energy (United States) raised $363 million from investors including Khosla Ventures, Chevron and United Airlines, plus $55.4 million in April 2025. On 3 September 2025 came a WARN notice: 95 layoffs, both sites closed, and the planned $1.4 billion gigafactory in Edgecombe County, North Carolina — 24 GWh a year, 1,000 jobs — cancelled.

So money was not the problem. Nor was the technology — both companies had working products. What was missing was a market willing to pay more for a cell that stores less energy per kilogram than an alternative getting cheaper every year.

The evidence: one permit, four trading days

Our knowledge base has carried the line for years that sodium only wins structurally when lithium is expensive — a commodity market, not a technology curve. Until last year that claim had no event underneath it.

On 9 August 2025 the mining permit for CATL's Jianxiawo project in Yichun, Jiangxi province, expired. Without renewal the mine closed. Benchmark Mineral Intelligence estimated that a prolonged shutdown affects roughly 60,000 tonnes of lithium carbonate equivalent in Jiangxi.

What the exchange did, day by day:

That is +19.7% in four trading days. And it was precisely in those weeks that the headlines appeared everywhere saying sodium was "cheaper again" than lithium.

Nothing about sodium had changed. No breakthrough, no new factory, no better cathode. One permit ran out.

The denominator decides the percentage

Four different percentages circulate for how big Jianxiawo was. They do not contradict each other — they count different things:

Hearing "3%" and reading "10%" looks like a contradiction. It is four different slices of four different pies. So name the denominator the moment the figure appears on screen, or every "but it's 6%" reply becomes unanswerable.

What the only serious study actually says

The year 2027 appears in trade blogs, in comparison pieces and in virtually every Dutch overview. "By 2027, Sodium ion Battery Costs Will Fully Drop Below LiFePO4." "Sodium Batteries to Undercut LFP Costs by 2027."

Yao, Benson and Chueh (Stanford) modelled 6,048 scenario combinations for Nature Energy, using componential learning curves bounded by mineral prices and technical design floors. Their result, in the published version: there are sodium pathways that could become cost-competitive with low-cost lithium-ion variants in the 2030s — and even then only "assuming substantial progress can be made along technology roadmaps via targeted R&D".

That is five years of difference and a caveat. But the methods section carries something stronger, and it is the reason this article exists:

Read "sodium reaches parity with LFP" and you think: same price. What the only serious model says is: one chance in five that it is cheaper. And even that sits in the 2030s.

The closing line of the abstract leaves little room: "being price advantageous against low-cost lithium-ion variants in the near term is challenging."

BNEF cut its forecast in 2025 and still expects growth in 2026

On 18 June 2025 BloombergNEF wrote, in its own publication:

"Expectations for sodium-ion batteries among manufacturers have cooled as LFP prices continue their downward trend, leading to BNEF's lowered expectations for sodium-ion technology to scale."

In May 2026 BNEF expects sodium to gain share in stationary storage after all, while calling it "generally more expensive than lithium-ion batteries today due to limited scale". Analyst Isshu Kikuma cites 60 GWh for CATL–HyperStrong and 5 GWh for Peak Energy–Jupiter Power.

Those two do not contradict each other. "Lowered" is not "negative": in 2025 BNEF tempered how fast sodium scales, and it still expects growth. The point is the direction of the revision and the reason behind it — LFP kept falling.

The scale problem in one line: sodium-ion was less than 1% of lithium-ion production in 2025. Global sodium capacity sat below 5 GWh; the forecast for 2027 is 100 GWh, with CATL and BYD together holding roughly 60%.

Is China holding sodium back?

This is a thought many viewers arrive at themselves: China sits on the battery chain, so it protects its lithium position. Tested, it flips.

And the strongest counter-indication: Chen Liquan, founding father of China's battery industry and awarded the National Highest Science and Technology Award in July 2026, publicly called for an accelerated pivot to sodium. His argument was industrial security, not price.

The economics say the same. A Chinese lithium salt producer, to Fastmarkets, after the lithium price nearly halved:

"now that lithium price fell a lot, it may not be worth switching to sodium. The move… is more strategic than economical."

Sodium is a strategic technology, not an economic one. That is why China pushes it, why almost nobody else does, and why it is not on your wall.

Where sodium does win

Stopping here would be unfair, because on two counts sodium does something lithium cannot.

Thermal behaviour. Ferdigg and Mair measured sodium (NVPF), LFP and NMC under identical conditions:

NMC is worst on both axes, by a factor of four on temperature. Against NMC, sodium is demonstrably cooler in a thermal runaway.

Zero volts. A sodium cell can go fully flat, to 0 V, and be shipped that way. International transport rules require lithium to ship at no more than 30% state of charge. The cause is cell chemistry: lithium uses copper foil on the anode side, which dissolves below roughly 1.5 V, while sodium uses aluminium on both sides. For storage, shipping and long idle periods that is a real advantage.

And the technology does exist, at grid scale. CATL announced TENER Sodium in Munich on 22 June 2026:

The "half the cost of LFP" headline attached to this product comes from a YouTube video and is carried by no primary or trade source. Worse, that video mixes two products. Its description covers CATL TENER Sodium, its spoken content covers Unigrid NaCrO₂, and the title glues them together. In the spoken version the cost claim is also conditional and attributed: "Tan says at equal scale, sodium ion should cost about half what LFP does to produce." That condition — at equal scale — is exactly what does not exist, because sodium was less than one percent of lithium production last year.

What is usable from the same corner, and is not a manufacturer figure: Syntropic Power had Unigrid's cells tested at the ISO 17025-accredited battery lab of the Rochester Institute of Technology. The 210 Ah cells delivered 216–218 Ah, cell efficiency of 97.9% or better, and after 1,004 cycles at 100% depth of discharge still held above 99% capacity. Independent lab, so a different category from a datasheet.

What this means if you are deciding now

Three things to weigh together:

1. The scenario where sodium wins is the scenario where lithium gets expensive. Good news for sodium, bad news for your wallet, because in that scenario every battery costs more. You are not waiting for a discount; you are waiting for a price shock at the competitor.
2. The Dutch net-metering scheme ends on 1 January 2027. That affects when you buy far more directly than which chemistry sits in the cell. It is a separate calculation, with your own consumption in it, and it stands apart from this dossier.
3. What you buy today is LFP, and that is not the bad option. Its pack price fell to a record low, and it is the target sodium has to get under. Our dossier puts LFP at 81.2% of installations in December 2025; that figure carries no opened primary source on our side, so read it as an order of magnitude rather than a measurement.

Whoever waits is waiting on a one-in-five chance, in a decade that has not started yet. That is a defensible choice, as long as you know that is the choice you are making.

What is deliberately not in here

This section belongs, because the omitted claims say as much as the included ones.

• "Eight thousand cycles against six thousand." Cut. 8,000 falls within the LFP range of 6,000 to 10,000 cycles to 70–80% remaining capacity at 80–100% depth of discharge. Without those conditions the numbers are not comparable at all.
• "The next generation sits about seventeen percent higher in energy density." Retracted. 175 Wh/kg is a cell figure (CATL Naxtra), 205 Wh/kg is a pack figure (Shenxing PLUS). Different units. Cell for cell, mainstream LFP sits around 160–162 Wh/kg, so sodium at 175 is actually above it.
• What a line conversion costs in euros. Not found in this round, so no figure is quoted. The drop-in point carries the reasoning without it.
• A total figure for global LFP production in GWh. Not found, so not quoted. What does exist is a share — and that one appears above with its limitation attached, rather than as a hard number.

Frequently asked questions

The video

The accessible version of this story is on Jay's Desk: why the sodium battery is not on your wall, in ten minutes, with the jar of salt on the desk.

This article is the complete version: the methods section of the paper, the daily quotes, the two price statements that contradict each other, and the sources attached.

Transparency

No sponsorship, no affiliate links, no brand preference. Nothing was lent, given or paid for this video or this article. No product I sell appears in it, and no manufacturer saw this piece in advance.

Where a figure comes from a manufacturer, it says so. Where two sources disagree, both appear. Where a figure needs a reference date to be true, that date is stated.

Sources