By Kyle Proffitt
January 22, 2025 | One topic of interest at the 2025 Advanced Automotive Battery Conference, held December in Las Vegas, was the significant advances being made with lithium-sulfur batteries. Speakers from Lyten, Coherent, and Fraunhofer IWS discussed specific chemistries, architectures, challenges, and successes working with this chemistry, culminating in some predictions that we will see Li-S cells in (some) automotives this decade.
The battery chemistry in a Li-S system is somewhat unique. Unlike NMC, LFP, graphite, or many other options in which the lithium ions intercalate into available space, Li-S is a conversion system. In the cathode, sulfur as S8 covalently bonds with lithium ions through a series of reactions to create Li2S—two lithium ions for each sulfur atom—and this creates some unique outcomes.
All three of the speakers in this session sang the praises of Li-S. Advantages for sulfur include low cost, wide availability, and high energy density. The chemistry also does away with contentious metals such as cobalt, nickel, and manganese. Celina Mikolajczak, Chief Battery Technology Officer at Lyten, said that with Li-S, “the upside is huge.” She started with cost. “When we look at the price that Li-S cells can achieve, it drops below LFP.” She also highlighted the benefits of sulfur abundance. “You can pretty much make Li-S cells on any continent with a local supply chain,” she said.
Rob Murano, Senior Director of Product Development and Commercialization at Coherent, went even further, saying that “sulfur is the ideal cathode material…it’s abundant, it’s dispersed around the globe, it’s easily minable… for many industries it’s actually a waste product.” He showed a comparison of sulfur active material costs of 7 cents per kWh, compared with $30-40/kWh for NMC or $10-15/kWh for LFP; that adds up to pack-level costs for Li-S one half to one third those of NMC or LFP.
Susanne Doerfler, Group Manager at Fraunhofer Institute for Material and Beam Technology, was focused largely on the high energy density. “Aviation is the most promising application, because we need lightweight batteries, and every kilogram counts,” she said. Li-S can fill that bill. With traditional NMC-based Li-ion batteries capping out at about 300 Wh/kg, she presented a semi-solid Li-S system projected to reach 700 Wh/kg.
So the promise is evident; Li-S is poised to double battery energy density at a fraction of the cost, using local supply chains. However, no commercial Li-S batteries are available yet, due to a number of challenges. One primary factor is that the conversion of S8 to Li2S does not occur in a single clean step but instead proceeds through intermediate polysulfide species Li2SX (where X = 6-8), and these polysulfides are highly soluble in many common liquid electrolytes. The result is referred to as the polysulfide shuttle, and it contributes to rapid battery degradation, such that Li-S cells don’t usually survive many cycles. Additionally, sulfur is an insulator, requiring carbon or other additives to maintain electrical conductivity. “If you can solve these problems, you’d pretty much have the holy grail of electrochemical energy storage,” Murano said.
The speakers each described their different approaches to overcoming these challenges and moving Li-S toward commercialization.
Lyten: 3D Graphene, 3D Anodes, Unique Use Cases
Mikolajczak did not go to lengths to explain all of their underlying technology, but Lyten uses a proprietary three-dimensional graphene cathode material in a composite with sulfur to mitigate the polysulfide shuttle and improve conductivity. They acquire the 3D graphene as a fortuitous byproduct of a methane capture process.
She showed a roadmap of Lyten’s progress in energy density and cycle life with Li-S cells, starting from about 30 cycles and 185 Wh/kg in 2021 and improving to 250-300 cycles and 310 Wh/kg in 2024. She said this represents nearly a 100% annualized cycle life improvement and expects the trend to continue because the newer chemistry provides several avenues for unexplored technological advances. “In a couple years, can we achieve 1,000 cycles? We think so,” she said.
She highlighted that factors including charge rate, temperature, and depth of discharge (DoD) have dramatic effects on battery performance metrics with their system. For instance, their P1 cell developed in May 2024 lasts about 250 cycles at a full DoD before falling below 70% capacity, but operating at just 20% DoD, it lasted 2,000 cycles (still going) with no appreciable loss of capacity. This protocol is suitable for low earth orbit satellites, Mikolajczak said. Alternatively, higher temperatures and much slower charge/discharge rates greatly increase energy density, to the extent that C/50 charge/discharge at 35 °C translates to 350 Wh/kg for the same cell that released 248 Wh/kg at 25 °C and C/3. “This is a conversion chemistry; that temperature really matters,” she explained.
Lyten is also developing their own 3d anodes using lithium composites. Mikolajczak showed results where the 3d anodes nearly doubled the cycle life for their cells. These 3d anodes had an additional advantage in safety, as she showed that they increased thermal runaway temperatures from a baseline of 125 °C up to 236 °C.
Their latest cell reaches 313 Wh/kg at C/3 and 362 Wh/kg at C/10. Mikolajczak explained that “if you’re looking at 8-10 hours of life for the battery… this battery is actually going to give you a lot more usable energy than the C/3 discharge would suggest. This suddenly makes this cell really interesting for a whole range of applications where this is going to outperform lithium-ion cells.”
Mikolajczak also showed a nice demonstration of their cells in action. She said they get a lot of questions about the weird voltage profile for Li-S, operating around two volts and having a couple of plateaus during discharge. She showed a video of a small drone operating with a 6.5 Ah Li-S battery pack at 14.7 V. A comparable lithium-polymer-powered drone was heavier with only 5.2 Ah of energy. Ultimately, she says, Li-S makes practical and usable cells.
She also shared some good news about calendar life. “The prevailing wisdom is that Li-S is terrible in calendar life,” she said. However, they discovered that cells experienced their worst irreversible capacity loss within the first month, up to 13% if held at 30% state of charge (SoC), but then held remaining capacity quite well. At other SoC levels, the outlook was much better. “If you want to take cells shipped at 100% SOC and sit them in a container, they will sit for months, and then you can recover the cell,” Mikolajczak said. In other words, those fully charged cells do lose about 25% of their charge after sitting for 5 months, but it’s only short term.
Finally, she shared some of Lyten’s future plans, including a 100 MWh production line in 2025 and a 6 GWh facility in 2027. She expects customers from this gigafactory to include last-mile delivery vehicles, heavy equipment, and limited production automotives.
Coherent: Immobilize Sulfur in Electrophilic Carbon
Murano was exuberant about Coherent’s advances around Li-S. “We think we’ve solved sulfur, with chemistry,” he said. The key advance he discussed is in the area of chalcogen (group 16 element) immobilization, stemming from years working with selenium and now sulfur chemistry.
Electrophilic carbon is the key, he explained. “We take what is a carbon host material, and we chemically process it… to make what we define as electrophilic carbon, or electron-deficient carbon.” He said this electrophilic carbon can then form high activation energy bonds with sulfur, preventing the formation of polysulfide species, “even in liquid electrolytes that are not tailored to prevent polysulfide shuttle… they don’t need to be because you suppress the formation.”
Murano showed data where the use of their immobilization strategy created a “single long, slow discharge plateau that would be consistent with the presence of no polysulfides.” He said Argonne National Lab and others have validated that no polysulfide species are present.
Additionally, sulfur immobilization solves the conductivity problem. In a composite “with 64% of the material being sulfur—conductivity of rubber—you would expect your conductivity would not be very good,” he said, “but we effectively double the conductivity of the composite over that of the carbon by itself.” The carbon was already a good conductor, so this is a vast improvement over the electrical conductivity of sulfur.
Murano said they believe in a “belt and suspenders” approach. In addition to immobilizing sulfur, they have created a “hydrophilic gate” that is permeable to lithium ions and electrons but impermeable to polysulfide species. They cover cathode particles with this “near zero mass” material and further limit any potential polysulfide leak.
The result for Coherent is 2.2 Ah pouch cells with specific energy density up to 205 Wh/kg. Working at 80% depth of discharge, C/10 charge rate, C/3 discharge, 25 °C, and < 1 MPa of external pressure, he said “right now, we’re at about 270 cycles with around 92% capacity remaining”. A projection showed this cell reaching around 500 cycles with 80% capacity retention, and Murano said the goal is to reach 800+ cycles in 2025. He said their platform has the potential to reach 265-400 Wh/kg.
Their cells also showed good performance at low temperatures, with over 80% capacity retention at 0 °C. “The reason for that is largely our electrolyte system,” Murano said without providing details. He closed saying that he expects the first cell products to be on the market within the next 12 months and added a small teaser for other future directions: “we have belief and some evidence that this would work for oxygen as the cathode working material as well.”
Fraunhofer IWS: Improving Charge Rate, Solid-State and Semi-Solid-State Li-S
Doerfler started with the goals she hears from OEMs regarding prototype cells: >400 Wh/kg, >1C charge/discharge rate, and at least 200 cycles. However, “it’s quite challenging to go over 0.1C” with Li-S, she said.
She discussed optimization of liquid solvents as an avenue for improving this charge rate, finding that LiFSI in a DME/hydrofluoroether-based electrolyte enabled some success in pouch cells. These could cycle at least 35 times while releasing about 800 mAh per gram of sulfur. She emphasized the importance of moving experiments to pouch cells, saying that to get to a target 450 Wh/kg, the electrolyte to solvent ratio needs to be under 1.5, which can’t be accomplished with coin cells. “This is where we separate the sheep from the goats,” she explained.
Another approach for solving the sulfide shuttle problem is to move to an all-solid-state system. Typically, however, high stack pressures are needed, and the insulating properties of sulfur are exacerbated. High energy ball milling (HBM) of cathode material can be used to decrease particle sizes and homogenize the mixture with more triple grain boundaries—places where sulfur, electrolyte, and carbon meet. However, HBM is an hours-long batch process. Doerfler referred to work published June of 2024 in which they reported success performing low energy ball milling for 15 minutes. She said the type of carbon makes a big difference when low energy milling is performed, and a mesoporous carbon is favored.
Combining low-energy milling with a dry transfer electrode coating process developed at Fraunhofer and the use of a reinforced separator, they created all-solid-state pouch cells with energy density of 1500 mAh per gram of sulfur and cycled them 20 times, something Doerfler said was not successfully reported prior without including indium in the anode. The study is under revision for publication.
They also returned to some of the lessons with optimized electrolytes and created a semi-solid system where the cathode includes argyrodite solid electrolyte and no liquid, whereas the anode includes some liquid electrolyte. She showed these pouch cells successfully cycling 25 times under low (0.6 MPa) stack pressure and said that the semi-solid system really enables increased energy densities, projecting a stack-level density up to 700 Wh/kg.
How Soon in EVs?
Following this series of presentations, the question was raised: how soon might we see these in vehicles? Mikolajczak began listing some of the technical challenges of scaling up for this kind of operation, but she pointed to an advantage for Lyten in that their process works with standard lithium-ion equipment, meaning a Li-ion gigafactory could be repurposed. Ultimately, she predicted that small-volume shipments could reach automotive manufacturers as early as 2027, but we’d generally start seeing Li-S in EVs “toward the end of the decade.” Murano added that there’s a model for this transition in LFP. He predicted that one day, “the OEMs suddenly [will] get a quote where it’s dollars per kWh over every other spec, as long as it’s good enough, and you’re going to see a lot of forecasting agencies very surprised by the uptake of sulfur. I think, in ‘28, ‘29, you’re going to start seeing vehicles running around with sulfur in them.” Pressed on the optimism of this outlook, he added, “I didn’t say high volume.”
