Editorial Article
Four Hundred and Thirty-Five Gigawatt-Hours Is a Capacity Number
There are two figures in circulation for how many sodium-ion cells the world made in 2025. One says 8.1 GWh. The other says 47.2 GWh, up more than 150 per cent year on year, and it arrives with a table, a year, and an air of authority. The gap between them is 5.8 times. One of them is wrong, and the useful thing is that you do not need to know which institution to trust to find out which one.
A gigawatt-hour of sodium-ion is not an opinion. It is a quantity of powder. Cathode active material delivers roughly 130 mAh/g at about 3.1 volts on the polyanion route and 150 mAh/g at 3.6 volts on layered oxide, which puts the requirement at 2,430 to 2,470 tonnes of cathode per gigawatt-hour — call it 2,450 — no matter whose factory assembles the cell. That constant is a property of the chemistry. It does not care about the forecast, the press release, or the market report.
So test the numbers against the material that shipped. Reported sodium-ion cathode shipments for 2025 were 20,000 tonnes. Twenty thousand divided by 2,450 is 8.16 gigawatt-hours. Test the prior year as well: 9,000 tonnes of cathode against reported shipments of 3.7 GWh gives 2,432 tonnes per gigawatt-hour — within 0.7 per cent of the constant. Two independent years, two independent measurements, one answer.
Now run the other figure through the same gate. 47.2 GWh would have required 115,640 tonnes of cathode in 2025. That is very nearly the 120,000 tonnes the same compilation reports for 2026.
That is the resolution, and it is not a compromise between two institutions. 47.2 gigawatt-hours and 120,000 tonnes belong to the same year, and the year is not 2025. Somebody moved a number one column to the left, and a base year moved with it.
Why a base year is worth an article
Because the same mistake, made at the top of the curve, is where the interesting money is. Published forecasts put sodium-ion cell output at 435 GWh by 2030. This market’s own path — 8.1 GWh in 2025 compounding at 67.25 per cent, which is the volume line published in the sodium-ion materials report — arrives at roughly 106 GWh in 2030. The two are 4.1 times apart, and the entire disagreement sits inside the next four years, which is the only part of any forecast a supplier can actually plan against.
Ask what 435 GWh would require, in tonnes, by 2030:
| Input for 435 GWh by 2030 | Tonnes |
| Cathode active material (2,450 t/GWh) | 1,065,750 |
| Hard carbon anode (1,129 t/GWh) | 491,115 |
| Total active material | 1,556,865 |
| Memo: cathode shipped in 2025 | 20,000 |
| Implied growth in cathode tonnage, 2025 to 2030 | 53.3 times |
Fifty-three times in five years. Not impossible — this is a chemistry with a 54.86 per cent value CAGR and a 67.25 per cent volume CAGR written into its own base case, so the industry is not short of ambition. But 53 times is a statement about kilns, calciners, precursor lines and permitting, and the reason to be careful is that every one of the 435 gigawatt-hours has already been announced by somebody. Capacity announcements are the cheapest document in this industry to produce. Cells that shipped are the only number that had a purchase order behind it.
The wedge between tonnes and dollars
Here is the part that a bigger cell number does not fix. In this market, tonnage compounds at 67.25 per cent and value at 54.86 per cent — a gap of 12.4 percentage points, sustained for eight years. The arithmetic consequence is that revenue per tonne falls from RMB 39,799 to RMB 21,482, or 46.02 per cent over the window, 7.42 per cent a year.
That is not a forecast about competitive intensity. It is the industry’s own commercial claim read backwards. Every published price line in this chain is falling on purpose, because the entire pitch for sodium is that it gets cheap enough to matter. A supplier who plans capacity against a gigawatt-hour number and a revenue line against a price that falls 7.42 per cent a year is planning two different businesses.
And the material mix inside the cell is not stable either. Hard carbon anode was USD 72.7 million of the USD 162.6 million market in 2025 — 44.70 per cent of the value from 31.55 per cent of the tonnage — because at RMB 56,390 a tonne it is the most expensive material in a sodium-ion cell and the only one whose price nobody can independently confirm. A gigawatt-hour of sodium-ion is 3,579 tonnes of active material: 2,450 of cathode and 1,129 of hard carbon. The equivalent LFP gigawatt-hour is about 3,400 tonnes — 2,551 of cathode and roughly 850 of graphite. Sodium is a more materials-intensive chemistry, and its most expensive material is the one with no published index.
Scale, stated in the only unit that cannot be flattered
The whole of this market was USD 162.6 million in 2025. The LFP cathode market it is designed to displace was USD 8.30 billion. Sodium-ion materials are therefore 1.96 per cent of the thing they are meant to replace.
That ratio is the honest frame for every gigawatt-hour claim in the sector. If sodium-ion reaches 435 GWh in 2030 it will not be a niche; if it reaches 106 it will still be a real business with a real supply chain and a real price war. But the two outcomes are different industrial systems, and the difference between them is a factor of four, which is roughly the width of a forecast nobody can yet falsify with a shipment.
What would change this
Two falsifiers, both dated and both checkable.
One: if reported cathode shipments for 2026 land near 120,000 tonnes, then 47.2 GWh is the correct 2026 figure, the base year was simply mislabelled in the compilation, and the 2030 path should be re-underwritten upward rather than argued about. That would be a genuine correction to this analysis and it would show up in a shipment line, not a press release.
Two: if hard carbon pricing becomes independently quoted — an index, a published contract band, anything a buyer can point at — the 44.70 per cent value share becomes contestable, and the anode’s position as the most expensive tonne in the cell starts to behave like a normal commodity instead of a negotiated one. Until then, the most valuable material in the cell is the one with the least evidence behind its price.
Neither test requires trusting a forecaster. Both require counting the powder. That is the whole method, and it is why the base year was worth the argument: 2,450 tonnes of cathode per gigawatt-hour is the only number in this market that everybody has to agree with.
This analysis draws on the sodium-ion battery materials market report, which settles the 2025 base year at 8.1 GWh, and on the sodium-ion electrolyte and NaPF6 report, which supplies the second chemical constant in the chain — 1,300 tonnes of electrolyte per gigawatt-hour of sodium cells.
Frequently Asked Questions
How large is the sodium-ion battery materials market?
USD 162.6 million in 2025, projected to USD 5,376.5 million by 2033 at a 54.86% baseline CAGR. That is 1.96% of the USD 8.30 billion LFP cathode market it is designed to displace. Volume grows faster than value — 67.25% against 54.86% — so revenue per tonne falls 46.02% over the window, or 7.42% a year.
Why does 47.2 GWh for 2025 sodium-ion output appear in industry compilations?
Because the figure is internally inconsistent with its own sourcing. At 2,450 tonnes of cathode per gigawatt-hour, 47.2 GWh requires 115,640 tonnes of cathode. The same compilation reports 20,000 tonnes of cathode shipped in 2025 and 120,000 tonnes for 2026 — so 47.2 GWh and 120,000 tonnes describe the same year, and it is not 2025. Two independent years of cathode data put 2025 at 8.1 to 8.2 GWh.
What would 435 GWh of sodium-ion cells in 2030 require in materials?
1,065,750 tonnes of cathode active material and 491,115 tonnes of hard carbon — 1.56 million tonnes of active material in total. Against 20,000 tonnes of cathode shipped in 2025, that is 53.3 times the tonnage in five years, and it implies a plant base most of which has so far been announced rather than built.
Which material captures the most value in a sodium-ion cell?
Hard carbon anode. It was USD 72.7 million of the USD 162.6 million 2025 market — 44.70% of the value from 31.55% of the tonnage — because at RMB 56,390 a tonne it is the most expensive material in the cell, and it is also the only major material whose price has no independent public quote.