The Solid-State Battery Gap: Strong Lab Numbers, Unproven Factories

8 min read

Key Claim: The central question in solid-state battery development is not whether the cells work in a laboratory — they do — but whether any company can produce them at the yields, volumes, and costs required for automotive deployment; as of 2026, none has.

A modified Mercedes-Benz EQS drove from Stuttgart to Malmö in September 2025 — 1,205 km — on a single charge of Factorial Energy’s lithium-metal solid-state cells, arriving with 137 km of range still remaining. QuantumScape’s QSE-5 cell claims 844 Wh/L at the cell level and retains more than 95% capacity after 1,000 cycles in laboratory testing. CATL says its 20 Ah solid-state cells hit 500 Wh/kg in trial production, though the figure has not been independently verified — which, as it happens, is precisely the kind of lab-to-production gap this article is about.

These are real numbers — not speculative projections. The problem is that none of them have been demonstrated at production scale. In 2026, the central question in battery technology is not whether solid-state cells can outperform lithium-ion on a laboratory bench. They can, and the data are convincing. The question is whether any company can build them in the volumes, at the yields, and at the cost required to put them in vehicles. On that measure, the gap between what has been shown and what has been shipped remains wide.

What “Production Scale” Actually Means

The distinction between cell-level performance and manufacturing capability matters more in battery technology than in almost any other sector, because the path from a single working cell to a gigawatt-hour factory involves engineering problems that are qualitatively different from the chemistry problems that produced the cell. Consider QuantumScape’s QSE-5: a cell that achieves 844 Wh/L in laboratory testing and retains 95% capacity after 1,000 cycles — yet as of early 2026, QuantumScape has not disclosed how many of those cells its Eagle Line can produce per shift, or what fraction pass quality inspection.

Solid-state batteries replace the liquid electrolyte in conventional lithium-ion cells with a solid material — typically a ceramic, glass, or polymer. The solid electrolyte must be deposited in layers that are simultaneously extremely thin (to minimise resistance), mechanically robust (to survive the volume changes as the battery charges and discharges), and essentially defect-free (because a single pinhole creates a short-circuit pathway). At the laboratory scale, researchers can produce a handful of cells with exacting control. At the gigafactory scale, yield — the percentage of cells that leave the line without a defect — becomes the governing economic variable.

No company has publicly disclosed solid-state electrolyte yield data at industrial scale, because no company is yet producing at industrial scale. What exists in 2026 is a tiered spectrum: lab cells demonstrating peak performance, prototype vehicles integrating those cells for range and durability tests, and pilot lines that are beginning to translate lab processes into something closer to manufacturing equipment. The step from pilot line to high-volume production is the one that has consistently slipped.

QuantumScape: Furthest Along in the West, Still Pre-Commercial

QuantumScape is the most closely watched Western entrant, partly because of its Volkswagen Group partnership and partly because it is publicly traded, making its milestones and setbacks unusually transparent.

In October 2025, QuantumScape began shipping B1 samples of its QSE-5 cell — a 5 Ah, anode-free lithium-metal cell with a ceramic (garnet-type) separator produced using its proprietary Cobra process. The B1 designation indicates these are second-generation engineering samples, intended for rigorous testing by OEM customers to support future vehicle programmes, not for installation in vehicles today.

In February 2026, the company inaugurated its Eagle Line pilot facility in San Jose, California — the first highly automated production line for QSE-5 cells. The Eagle Line is designed to produce cells for OEM sampling and to validate the Cobra separator process at a scale greater than the company’s previous lab equipment, but it is not a commercial manufacturing facility. QuantumScape has not disclosed Eagle Line throughput in cells per day or GWh equivalent.

The Volkswagen Group relationship suggests the industry is managing risk rather than committing to volume — a structure that could be read as hedging. In July 2024, Volkswagen’s battery subsidiary PowerCo signed a non-exclusive licence to manufacture up to 40 GWh/year using QuantumScape’s technology. In July 2025, that agreement was expanded, with PowerCo committing up to $131 million over two years tied to scale-up milestones and gaining rights to produce an additional 5 GWh/year of QSE-5-based cells. The structure is worth noting: Volkswagen is paying for technology rights and milestone-based development rather than placing a purchase order for cells. Commercial mass production from this partnership is targeted for 2027–28.

Toyota: The Longest Timeline, Revised Again

Toyota’s solid-state programme is the most discussed and most repeatedly delayed in the industry. The company initially targeted solid-state batteries for production vehicles by 2020, then 2023, then 2026. The current target is 2027–28 for initial limited production in Lexus flagship models, with mass production scheduled later in the decade.

What distinguishes Toyota’s approach from most competitors is its focus on the material supply chain before the cell line. In early 2025, Idemitsu Kosan — a Japanese energy company — announced it is building a production plant for lithium sulphide, the key precursor for sulfide-type solid electrolytes, with a capacity of 1,000 metric tonnes per year. The plant is targeted for operational readiness in June 2027. Toyota is collaborating with Idemitsu and Sumitomo Metal Mining on both the material and cell development programmes.

The choice of sulphide electrolytes is significant. Sulphide materials are softer and more amenable to the deformation required during battery assembly than the harder oxide ceramics (such as the garnet electrolytes used by QuantumScape), which makes them better suited to high-volume manufacturing. The tradeoff is chemical stability: sulphide electrolytes react with moisture and require carefully controlled manufacturing environments, adding cost and complexity. Toyota’s investment in a dedicated material supply chain suggests the company is treating the electrolyte precursor as a strategic input rather than a commodity — a rational position if sulphide becomes the dominant route to mass production.

The credibility question remains. Toyota has made these announcements before and postponed. There is currently no independent third-party verification of its manufacturing readiness, and no public yield data from its pilot lines.

Solid Power: A Revealing Pivot

Solid Power’s trajectory offers a different kind of signal. The Colorado-based company was founded as a solid-state cell developer, partnered with BMW and Samsung SDI, and went public via SPAC in 2021. In 2025, the company’s full-year revenue was $21.7 million — and the growth was driven not by selling cells but by a line-installation agreement with SK On, the South Korean battery manufacturer, to install Solid Power’s solid-state electrolyte production technology at SK On’s facility.

The pivot from cell developer to electrolyte technology licensor is significant. It reflects the economic logic of the current moment: large battery manufacturers have the capital and manufacturing infrastructure to build cells at scale; what they lack is a validated solid-state electrolyte process. Solid Power’s customers are paying to acquire that process, not to buy finished cells. For 2026, the company plans to commission a continuous electrolyte production line and is evaluating a joint venture for electrolyte manufacturing in Korea at 500 metric tonnes per year capacity.

BMW has road-tested a BMW i7 equipped with Solid Power cells. The company’s trajectory — from cell developer to electrolyte licensor, with revenue driven by process installation rather than product sales — may be the most honest indicator of where the industry actually stands: the valuable intellectual property is in manufacturing the electrolyte, not assembling the cell.

Other Programmes in Brief

Samsung SDI, under a trilateral agreement with Solid Power and BMW, will supply all-solid-state cells to BMW for evaluation vehicles expected in late 2026, with Samsung SDI targeting mass production in 2027. Separately, Samsung Electro-Mechanics has indicated that solid-state cells may appear in consumer wearables such as the Galaxy Ring as early as 2026 — a low-volume application that would generate real-world data without requiring automotive-scale manufacturing.

CATL, meanwhile, has its 20 Ah solid-state cells at a technology readiness level (TRL) of approximately 4 out of 9, targeting level 7–8 by 2027, equivalent to small-batch production. The company’s short-term goal is to progress from 20 Ah samples to 60 Ah automotive-grade prototypes — another illustration of the distance between a working cell and a production-ready one.

The Science Is Still Generating Surprises

The production challenge is not merely engineering execution. A study published in Nature in early 2026, “Electrochemical corrosion accompanies dendrite growth in solid electrolytes,” identified a failure mode that the prevailing mechanical stress model had not predicted.

Using operando birefringence microscopy to observe dendrites propagating through garnet-type solid electrolyte (Li₆.₆La₃Zr₁.₆Ta₀.₄O₁₂), the researchers found that dendrites can penetrate the electrolyte at stresses up to 75% lower than would be predicted by the fracture stress of the material under mechanical load alone. The explanation: at higher current densities, electrochemical corrosion at the dendrite tip decomposes the electrolyte locally, inducing phase transitions that create a net molar volume contraction. This removes material from around the advancing dendrite front — a self-reinforcing mechanism that bypasses the mechanical barrier the solid electrolyte was supposed to provide.

The practical implication is that cell designs optimised against mechanical failure criteria may still fail through a different pathway at operating current densities. This does not mean solid-state batteries cannot be made to work; it means the engineering parameters for reliable operation are not yet fully characterised, even in well-studied garnet electrolytes. For sulphide-type electrolytes — the preferred commercial route for Toyota, Idemitsu, and most automotive-scale programmes — equivalent characterisation remains ongoing, and the same corrosion-driven failure mode has not yet been ruled out in those chemistries.

The Cost Gap That Matters More Than Energy Density

Performance per kilogram is not the only variable that matters. IDTechEx estimates current solid-state battery production costs at $400–800/kWh — compared with approximately $115/kWh for lithium-ion in 2024. Even at the low end of that range, solid-state cells cost roughly 3.5 times more per unit of energy stored.

For context: lithium-ion battery costs fell from over $1,000/kWh in 2010 to $115/kWh by 2024 through a combination of chemistry optimisation, manufacturing scale, and supply chain maturation developed over more than a decade of high-volume production. Solid-state batteries are at the beginning of that curve — except the curve for solid-state will not start bending until production volumes justify capital investment in specialised deposition and assembly equipment, which in turn requires OEM commitments that depend on cost parity that does not yet exist.

The first commercial solid-state cells will almost certainly enter high-value, low-volume applications: premium EVs, aerospace, defence, and potentially consumer wearables. These volumes will generate data and operational experience but will not produce the manufacturing learning curve that drives cost reduction at the pace the automotive market requires. This dynamic is not unique to batteries — it mirrors the cost trajectory that defined early silicon quantum computing investments, where premium deployments preceded any mass-market cost curve.

What to Watch

The signals worth monitoring are not the range records or energy density announcements. Those will keep coming. The indicators that matter are:

Electrolyte yield and throughput data. The first time a company publishes or discloses verified throughput figures for a solid-state electrolyte production line — cells per shift, defect rates — that will be the leading indicator that the manufacturing gap is closing, not widening. Given current pilot-line activity, any such disclosure is most likely in the 2026–27 window; its absence beyond that point would be informative in itself.

OEM purchase commitments. A licensing agreement or a milestone-based development contract is not a purchase order. When an OEM places a firm order for solid-state cells at a defined price and volume for a named vehicle programme, that changes the economics of the entire supply chain. The earliest credible window for such a commitment is 2027, contingent on Toyota or Samsung SDI delivering on their near-term milestones.

Toyota’s 2027 delivery. Toyota has the most systematic material supply chain build-out of any programme currently visible. If Idemitsu’s lithium sulphide plant comes online on schedule in June 2027 and Toyota delivers even a limited-run Lexus with solid-state cells in 2027–28, it will be the first credible data point that the production transition is tractable. If the timeline slips again, the industry will face a harder conversation about whether 2030 mass production targets are realistic.

Cost trajectory announcements. Any manufacturer that begins disclosing a cost per kWh — even preliminary — is signalling manufacturing maturity. The continued absence of such figures from any major programme is itself a data point.

This article was produced with AI assistance and reviewed by the editorial team.