Most of what gets written about sodium cells is chemistry — cathode families, electrolyte salts, the anode question. A published application that appeared in the July 16, 2026 US drop, assigned to CONTEMPORARY AMPEREX TECHNOLOGY CO., LIMITED, is not about any of that. US20260204669A1, titled “BATTERY MANUFACTURING PROCESS, BATTERY TREATMENT DEVICE, BATTERY, AND ELECTRIC DEVICE,” is a filing about a step on the line. That placement is the signal. When a cell maker files on formation sequencing rather than on materials, the problem it is describing is one that shows up in a plant, in a takt time, in a piece of equipment somebody has to buy.

The problem, as the record states it, is water. Residual moisture inside a finished cell is a nuisance in any chemistry, but the description spells out why it bites harder here: The reduced sodium metal has high activity and is prone to violent reactions with water in the battery, generating gases such as H2, thereby causing significant capacity loss. Once sodium metal deposits at the negative electrode, any water still in the cell has something extremely eager to react with. The description further notes that leftover water hinders sodium deposition, and that uneven sodium deposition easily causes local dendrites, leading to short circuits in a cell. So the water has to be gone before the sodium arrives. The question is how you get it out.

What the record says the industry does now

The application does not compare itself to a dry room, ambient humidity control, or moisture management on the plant floor. None of those appear in the record. Its stated baseline is a single, specific process step:

In the prior art, water is generally removed from batteries through high-temperature baking, but in this water removal method, the water inside the battery cannot be effectively removed.— BATTERY MANUFACTURING PROCESS, BATTERY TREATMENT DEVICE, BATTERY, AND ELECTRIC DEVICE, US20260204669A1

Baking is a real line item. It is an oven, a dwell, energy, and floor space between two other operations, and the record's contention is that it does not fully reach the water trapped inside the assembled cell. The application also names a case where it says baking is particularly difficult — for a cell on which water removal through “high-temperature baking” is difficult, such as a cell with a Nasicon-type phosphate positive electrode material system. That is a narrow, concrete callout, and it is the closest the record comes to saying which cells this process is for.

The disclosed approach inverts the order. Rather than trying to drive water out before the electrolyte goes in, the process injects electrolyte first, deliberately, so that water trapped in the electrode sheets and separator has somewhere to diffuse to — into the electrolyte, where it can reach the negative electrode. Then the cell is charged, but only partway. Claim 1 is spare: a battery manufacturing process comprising acquiring a battery to be treated, injecting an electrolyte, and charging the battery before sodium in the battery is reduced at a negative electrode. Worth noting for anyone reading the claim itself: the phrase “sodium-ion” does not appear in claim 1. Sodium enters through the negative-electrode language and through the CPC classification — H01M 10/054, non-lithium secondary cells, alongside H01M 10/446 and H01M 10/058. “Sodium cell” is a fair reading of the record; “claim 1 is sodium-ion-specific” is not what the claim says.

The mechanism lives in claim 2, which specifies charging at a first charge voltage higher than the reduction potential of H+ and lower than the reduction potential of sodium. That is a window. Inside it, water — or hydrogen ions from water decomposing in the electrolyte — takes electrons at the negative electrode and leaves as hydrogen gas. Sodium, whose reduction potential sits above the ceiling of the window, stays put. The abstract describes the sequence: in the process of charging the battery, water may receive electrons at the negative electrode and undergo a reduction reaction, producing hydrogen gas and hydroxide; or water may decompose in the electrolyte to generate hydrogen ions, which are then reduced at the negative electrode to hydrogen gas. Only after the water is consumed does claim 9 take the cell past sodium's reduction potential, at which point the sodium metal that deposits never meets water at all. The description writes the reactions as 2H2O+4e−→H2+2OH− and 2H2O+NaPF6+4e−→NaPO2F2+2H2+4F−; the first is printed unbalanced in the record, and is reproduced here as written rather than corrected.

The dependent claims are where a process engineer would look, and they read like a recipe someone has actually run. Claim 3 puts the first charge voltage at 1.0 V to 2.9 V, preferably 1.7 V to 2.9 V, more preferably 2.5 V to 2.7 V. Claims 4–5 heat the cell during that step — 25° C. to 70° C., preferably 45° C. to 60° C. — which speeds water's diffusion out of the cell material and into the electrolyte. Claims 6–7 pull a vacuum during the same step, at or above −80 kPa and below 0 kPa, preferably −30 kPa to −10 kPa, which draws off hydrogen as it forms and pushes the equilibrium toward removing more water. Claim 8 sets the dwell at 4 h to 48 h, preferably 6 h to 18 h. Claim 11 closes the loop: after the second-voltage charge, more electrolyte goes in, the cell is evacuated, protective gas is injected, and it is sealed. The description offers one efficiency statement — It can be understood that at 12 h, the water removal efficiency is close to 90%. That is the description talking about the process, not a validated cell spec, and it should be read that way. These are claimed preferred ranges in a pending filing, not disclosed production settings — but their shape is informative on its own: heat, vacuum, and a multi-hour hold are formation-stage parameters. The filing also covers the apparatus — claim 12 recites a battery treatment device with a battery acquisition unit and a charge unit configured to charge before sodium is reduced — the tell that somebody is thinking about the box on the line, not only the recipe running in it.

Reading the cohort

Five other applications carrying Contemporary Amperex assignee strings published in the same July 16 drop, and they sit almost entirely on the hardware and equipment side. US20260204759A1 describes a cell housing where both tabs exit the same end surface, one running to the terminal and the opposite-polarity one straight to the housing. US20260204612A1 covers a winding mandrel whose secondary members retract inward so the assembly narrows for jelly-roll release, and US20260204611A1 a driven pressing plate that reciprocates to bend a cell tab onto the terminal post during assembly. US20260201509A1 is a materials filing — an aluminum alloy for battery boxes formulated to suppress the Al2Cu phase — and US20260200307A1, titled simply “VEHICLE,” describes a pack with a raised platform tucked under the seat cushion to add volume without intruding on passengers. The observation available here is a modest one, and only about these six records: the hero is the only one that touches sodium, and the only one that is a formation-process filing rather than a hardware, equipment, structure, or materials filing. Six records published on one date are not a strategy, and this cohort should not be read as one. What it does show is that when a sodium-specific record surfaced in this drop, it surfaced as a manufacturing process with a device claim attached — the form a problem takes once it has moved from the chemistry question to the line question. US20260204669A1 is a published application, pending, not granted and not a product; the record describes what the process does, and is silent on whether anyone is running it.