An application assigned to TOYOTA JIDOSHA KABUSHIKI KAISHA published on July 16, 2026 under the number US20260204733A1, titled “BATTERY AND METHOD FOR MANUFACTURING BATTERY.” It is a pending application, not a granted patent, and the distinction matters for how much weight the document can carry. What it does carry is a fairly specific description of where the engineering difficulty in a solid-state cell is being addressed — and in this filing, that place is not the cathode. It is the anode, and more precisely the interior composition of the anode across its own thickness.

Claim 1 sets the architecture before it sets the chemistry. It recites a first current collector layer with a pair of first electrode layers disposed on one surface and the other surface of that collector, then a pair of solid electrolyte layers, a pair of second electrode layers, and a pair of second current collector layers, each mirroring outward. That is a symmetric, double-sided stack built around one central collector. It is worth flagging that the application’s abstract describes the arrangement more loosely — as first electrode layers disposed on the first current collector layer, without the double-sided requirement. The bipolar limitation exists in the claims. The abstract, as descriptive background, is broader than what is claimed, and only the claims define scope.

The point of novelty arrives at the end of claim 1. The second electrode layers each contain an electrode active material and a solid electrolyte, and the ratio between them is deliberately non-uniform: richer in solid electrolyte at the face meeting the solid electrolyte layer, poorer at the face meeting the outer current collector. Claim 1 says nothing about which electrode is which. It speaks only in terms of “first” and “second.” The role assignment does not appear until dependent claim 3, and that claim is unambiguous about it.

the first current collector layer is a cathode current collector layer; the first electrode layers are cathode layers; the second electrode layers are anode layers; and the second current collector layers are anode current collector layers.— BATTERY AND METHOD FOR MANUFACTURING BATTERY, US20260204733A1

Read together, claims 1 and 3 describe a cell whose anode is compositionally graded through its depth — electrolyte-rich where it faces the separator, active-material-rich where it faces the current collector. Claim 4 then narrows once more: the anode layers each contain silicon as the electrode active material. That is a dependent limitation sitting two levels down the chain, from claim 1 to claim 3 to claim 4, so silicon is not claimed at the independent level. But its presence in the chain is the clearest tell in the document about the problem being addressed. Silicon anodes are attractive on an energy-density basis and difficult on a mechanical one, because the material changes volume as it cycles. In a liquid cell an electrolyte can flow back into contact. In a solid-state cell it cannot. Grading the electrolyte content through the anode is a structural answer to a structural problem, and the claim set is organized around it.

The manufacturing claim is the commercial tell

For a reader tracking where capital gets committed rather than where research gets published, claim 5 is the more informative half of this application. It is a method claim, and it does not describe grading as a continuous gradient laid down in one pass. It describes building the graded anode by transfer, in discrete steps: press a laminate carrying first electrode layers on both faces of the first current collector; transfer a solid electrolyte layer onto each; transfer a first partial second electrode layer onto each of those; then transfer a second partial second electrode layer on top, with the electrolyte content of the first partial layer higher than that of the second.

Claim 2 covers the resulting structure by the same logic — at least one second electrode layer made of a plurality of partial second electrode layers laminated in the thickness direction, with electrolyte content stepping down from the separator side outward. The grading, in other words, is quantized into sub-layers rather than smoothly varied. That is a meaningful distinction for anyone modeling process cost. A stepped, transfer-based build is described in terms of discrete coated films and a lamination sequence, each partial layer carried on its own substrate. It implies a process line organized around transfer materials and stacking accuracy rather than around a single deposition step with a controlled composition ramp.

None of that is a statement about what Toyota will manufacture. An application discloses an approach the applicant thought worth protecting; it is not a production plan, a capex commitment, or a schedule. What it does indicate is the shape of the process the applicant was describing when the application was drafted, and that shape is closer to conventional electrode-film handling than to a novel deposition regime.

Where it sits in the cohort

The application does not stand alone in the July 16 publication set. Records assigned to the same applicant published alongside it across the energy-storage stack, including US20260204709A1 (SECONDARY BATTERY), US20260204665A1 (POWER STORAGE DEVICE), and US20260204617A1 (METHOD OF MANUFACTURING POWER STORAGE DEVICE AND POWER STORAGE DEVICE). Adjacent to the cell itself, US20260204753A1 (BATTERY PACK AND BATTERY PACK MANUFACTURING METHOD) and US20260204670A1 (METHOD FOR DETECTING ABNORMALITY IN STORAGE BATTERY) address pack assembly and fault detection. The spread suggests attention distributed across cell internals, pack construction, and diagnostics in the same window rather than concentrated on a single layer of the problem.

The classification is consistent with that reading. The application is classified under H01M 10/0525 and H01M 10/0585, covering lithium secondary cells and thin-layer stacked constructions, alongside H01M 4/366 and H01M 4/386, which reach electrode active materials including silicon, and several H01M 50 subclasses covering separators and their manufacture. Named inventors are Tetsuya Waseda, Hideyuki Tokioka, Yuki Sato, Takuya Matsuyama, Kenichi Kakishita, and Takuya Kimura.

The honest summary is narrow. Toyota filed on a solid-state cell architecture whose distinguishing feature is a graded anode, described as buildable by sequential transfer of two partial layers of differing electrolyte content, with silicon named as the anode active material in a dependent claim. It published July 16 and remains pending. It tells you what problem the applicant was working on and what construction route it described for solving it. It does not tell you what will be built, when, or at what cost — and any figure attached to those questions would have to come from somewhere other than this document.