A 2025 solid-state lithium–sulfur study hit about 505 Wh/kg at the cell level, and a Tsinghua quasi-solid-state battery reached 604 Wh/kg, more than double many conventional lithium-ion cells. If either chemistry survives the jump from lab to mass-produced EV packs, it could mean the same range with a much lighter battery, or far longer range without the extra weight.

The first thing to say about a 604 Wh/kg battery cell is that it is not a 604 Wh/kg electric vehicle pack.

That distinction sounds technical, but it is the difference between an interesting laboratory result and a car that can be bought, insured, fast-charged, parked in summer heat and driven for years. Cells are the basic electrochemical units. Packs include casing, cooling, wiring, structural supports, sensors, safety systems and the electronics that keep the cells within operating limits. Pack-level energy density is therefore lower than cell-level energy density, sometimes much lower.

Still, the cell numbers matter. In 2025, two battery papers pushed into territory that explains why researchers keep returning to chemistries beyond conventional lithium-ion. One reported a solid-state lithium-sulfur route with about 505 Wh/kg at the cell level. Another, a Tsinghua University-led paper in Nature, reported quasi-solid-state lithium pouch cells at 604 Wh/kg and 1,027 Wh/l. Those values sit far above many lithium-ion cells now used as practical engineering benchmarks.

The finding is worth taking seriously, but it should not be read as the final word. These are research results, not production vehicle packs. The useful question is narrower and more important: what would happen if batteries in this energy-density range could survive the climb from carefully made cells to mass-produced automotive systems?

What Wh/kg actually changes

Watt-hours per kilogram is a measure of gravimetric energy density. It tells you how much energy a battery stores for each kilogram of battery mass. For electric vehicles, that number matters because battery weight is not passive. A heavier pack needs more structure, affects handling, changes tyre loads, and can reduce some of the efficiency gained by carrying more stored energy in the first place.

If a chemistry can store the same energy in less mass, a vehicle can carry the same range with a lighter battery. If the vehicle keeps the same pack mass but stores more energy inside it, range can increase without simply making the pack larger. Those are the two futures implied by the headline numbers: lighter batteries for the same range, or longer range without adding the same weight penalty.

That does not mean range doubles just because a cell number doubles. A vehicle is not a cell test fixture. Engineers must still account for pack architecture, thermal control, safety margins, charging behaviour, ageing, manufacturing yield and cost. But a higher cell-level ceiling gives designers more room to trade between range, weight, durability and price.

The lithium-sulfur attraction

Lithium-sulfur batteries have been attractive for a simple reason: sulfur is light, relatively abundant and theoretically capable of storing a large amount of charge. The chemistry offers a route to high gravimetric energy density without relying on the same mix of transition metals that shape many lithium-ion supply chains.

The difficulty is that sulfur is not an easy electrode material to make behave inside a practical battery. In liquid electrolyte lithium-sulfur cells, intermediate sulfur compounds can dissolve and migrate, hurting efficiency and cycle life. In all-solid-state versions, one of the central problems is contact. The reaction has to proceed where sulfur, electronic pathways and ion-conducting solid material all meet. If too much sulfur sits away from those active interfaces, the cell carries material that is theoretically useful but practically underused.

A January 2025 paper in Nature Materials addressed that interface problem directly. Daiwei Wang, Donghai Wang and colleagues described all-solid-state lithium-sulfur batteries using mixed ionic-electronic conductors in sulfur cathodes. The aim was to move the conversion reaction beyond the narrow traditional three-phase boundary and allow more of the sulfur to participate.

In the paper’s abstract, the team reported active sulfur ratios up to 87.3 percent, conversion degrees above 94 percent, discharge capacity above 1,450 mAh/g and cycle life beyond 1,000 cycles in their tested all-solid-state lithium-sulfur cells. The broader reported cell-level figure of about 505 Wh/kg is why the work is being discussed alongside the highest-density lithium-metal and quasi-solid-state results of the year.

The caution is built into the achievement. High utilisation in a research cell does not settle how the chemistry behaves in large-format packs, across temperature ranges, under fast charging, after years of vibration, or through the quality-control demands of an automotive factory. It does show that one of the old lithium-sulfur bottlenecks can be attacked at the level of electrode architecture rather than treated as a fixed penalty.

The Tsinghua quasi-solid-state result

The 604 Wh/kg number comes from a different battery family. In the September 2025 Nature paper, Xue-Yan Huang, Chen-Zi Zhao, Wei-Jin Kong, Qiang Zhang and colleagues reported a quasi-solid-state polymer electrolyte paired with lithium-rich manganese-based layered oxide cathodes and an anode-free cell design.

The abstract describes an in-built fluoropolyether-based polymer electrolyte designed to form fluorine-rich interfacial layers on both cathode and anode sides. The point was not only to store more energy, but to reduce harmful interfacial reactions that limit cycling and safety. In pouch cells, the authors reported 604 Wh/kg and 1,027 Wh/l, along with nail-penetration safety performance in a fully charged condition. In coin cells, they reported more than 500 cycles at 25 degrees Celsius.

That combination is why the paper is notable. Energy density alone can be misleading if it comes at the cost of cycle life, safety or manufacturability. A high number produced by an unusually delicate configuration is less useful than a slightly lower number that can be produced consistently and operated under real conditions. The Tsinghua-led work tries to address the interface and safety problems that often decide whether a high-energy lithium-metal design remains a paper result or becomes a candidate for engineering scale-up.

Even so, quasi-solid-state does not mean problem-free. Polymer electrolytes have to move ions quickly enough, tolerate high voltages, maintain contact with changing electrodes, resist dendrite growth at the lithium side, and perform across the temperatures expected of a vehicle. Anode-free designs can save mass, but they also place severe demands on lithium plating and stripping efficiency over many cycles.

Why the pack is the harder test

Battery announcements often drift from cell numbers to vehicle promises too quickly. The hard part is that an EV pack is a managed system, not a box of ideal cells. Every kilogram saved at the cell level can be partly given back by the hardware needed to keep those cells safe and durable.

A pack also has to survive abuse cases. It must handle crash loads, charging faults, manufacturing variation, cold starts, hot parking lots, repeated fast charging and years of partial charging patterns that rarely resemble laboratory cycling protocols. A chemistry that looks excellent at small scale can fail commercially if it is expensive, sensitive to moisture, hard to manufacture at yield, or dependent on processes that are slow and fragile.

There is another reason to be careful with the comparison to conventional lithium-ion. Commercial lithium-ion is not one chemistry. It includes nickel-rich cells, lithium iron phosphate cells, different formats, different safety margins and different pack designs. Some are chosen for energy density, others for cost, durability or safety. A cell-level comparison captures one important axis, not the whole engineering decision.

That is why the two 2025 results are best understood as signals rather than forecasts. They show that researchers are still finding credible routes above the energy-density range that shaped the first major era of electric vehicles. They do not show that the next generation of EV packs will automatically be twice as energy-dense, or that lithium-ion is about to be displaced on a fixed timetable.

The real promise is optionality

If either chemistry survives scale-up, the most important change may not be a single dramatic range number. It may be optionality. A city vehicle could keep ordinary range while using a smaller, lighter battery. A long-distance vehicle could carry more energy without becoming heavier in the same proportion. Commercial vehicles could trade saved mass for payload. Aviation, where every kilogram matters, would watch closely even if the first automotive use came earlier.

For now, the honest reading is restrained. The 505 Wh/kg and 604 Wh/kg figures are cell-level results from research environments. They are not pack-level guarantees, production costs, warranty lifetimes or charging-network solutions. But they are also not trivial. They point to a battery problem being worked at its root: how much energy can be stored before weight, safety and degradation take the gains back.

That is the quiet importance of the 2025 studies. They do not give the EV industry a finished battery. They give it a higher ceiling to test against, and a reminder that range is not only a software problem, a charging problem or an aerodynamics problem. It is still, fundamentally, a chemistry problem.