The research, led by Qindong Yang of Tsinghua University with colleagues from the National Institute for Environmental Studies in Japan, the University of Southern Denmark and Tianjin and Nankai Universities, is the first to couple province-resolved fleet projections for China with explicit battery degradation modelling and dynamic material flow analysis. That combination matters, because most previous assessments of shared mobility’s resource benefits assumed batteries simply last a fixed number of years. In reality, a shared electric car can be on the road for many more hours a day than a privately owned one, racking up charge cycles and deep discharges that age its cells far faster than the average owner’s vehicle would.

The team’s modelling shows just how consequential that distinction is. For vehicles equipped with lithium nickel manganese cobalt oxide batteries, the high-energy chemistry known as NMC811 that many manufacturers favour for long-range driving, shared mobility was expected to save between 5.3 and 8.1 million tonnes of cumulative lithium demand by 2060 under static lifetime assumptions. When degradation was factored in realistically, those savings shrank to between 2.6 and 4.7 million tonnes. In other words, the faster turnover of degraded shared-fleet batteries could consume a large share of the raw material reductions that car sharing was supposed to deliver.

Not every chemistry suffers equally, however. The study finds that more durable batteries, notably lithium iron phosphate, the chemistry that dominates China’s mass-market electric fleet, and lithium nickel cobalt aluminium oxide cells, can preserve or even amplify the material savings of shared mobility. This creates what the authors describe as a chemistry-dependent dividend: the environmental case for sharing vehicles is only as strong as the longevity of the batteries inside them. A shared fleet built on robust, degradation-resistant packs delivers genuine resource relief, while one built on energy-dense but fragile cells may simply burn through minerals at nearly the same pace as private ownership.

Behind these numbers lies a subtle temporal trap that the researchers call a trade-off in timing. Intensive use accelerates battery retirement, which sounds bad but has a hidden benefit: retired packs feed the recycling stream sooner, boosting the supply of secondary materials recovered from end-of-life batteries. At the same time, many circular-economy strategies route retired electric vehicle batteries into second-life applications, typically stationary grid storage, before they are finally recycled. The study shows that this detour delays recycling, locks secondary material supply to vehicle manufacturing timelines, and increases near-term demand for newly mined lithium and other critical minerals. The very strategy designed to squeeze extra value from spent batteries can, paradoxically, deepen short-term dependence on primary resources.

The scale of the analysis is one of its distinguishing strengths. Rather than treating China as a single homogeneous market, the team resolved their projections at the provincial level, capturing the enormous variation in climate, driving conditions and use intensity across the country. Batteries in hot southern provinces degrade faster than those in cooler northern regions, and the duty cycles of ride-hailing fleets in megacities differ sharply from those of private cars in smaller urban areas. By overlaying these regional degradation patterns on dynamic material flow models, the researchers could track how lithium, cobalt and nickel circulate through the Chinese fleet year by year, distinguishing between material locked in operating vehicles, material awaiting recycling and material returning from the scrap stream.

Dynamic material flow analysis, the methodological backbone of the study, treats the vehicle fleet as a kind of reservoir of embodied metals. Batteries enter that reservoir at manufacturing and exit at end of life, where they can either be recycled back into the production loop or diverted into second-life storage before eventual recovery. The timing of these flows determines whether a closed-loop system, in which recycled material fully satisfies new battery demand, arrives in time to relieve mining pressure. The study maps the conditions under which China’s battery system achieves that closed loop, and finds that degradation assumptions shift the timing substantially. Shared mobility with fast-degrading NMC811 packs floods the recycling pipeline earlier but with a much larger total volume of batteries that must be replaced, while durable chemistries slow the flow but reduce cumulative demand overall.

The findings carry immediate relevance for policy and industry strategy, both in China and beyond. China’s Ministry of Industry and Information Technology has already established standards for the comprehensive utilisation of waste electric vehicle batteries, and the country’s dominance in lithium-ion battery recycling gives it a head start on circular supply chains. But this study suggests that policy cannot treat recycling, second-life deployment, battery chemistry selection and mobility models as independent levers. A fleet-planning decision as mundane as whether ride-hailing vehicles use lithium iron phosphate or NMC811 cells propagates through decades of material demand, recycling capacity needs and import dependence. The authors argue that accounting for battery degradation is therefore critical to aligning shared mobility, battery choice and circular strategies with transport decarbonization goals.

There is also a broader lesson about how the clean energy transition is modelled. Life cycle assessments and material demand projections frequently rely on static assumptions about product lifetimes, an approach that is computationally simpler but can misrepresent systems in which usage intensity fundamentally alters durability. The battery community has made rapid progress on the science of degradation, from physics-based ageing models to machine-learning approaches that can predict cycle life from early discharge data, and this study demonstrates how powerful it is to embed that understanding directly into resource assessments. The same logic likely applies to other heavily used clean technologies, from grid-scale storage batteries to hydrogen fuel cells in commercial fleets, where duty cycles differ dramatically from laboratory baselines.

For consumers and investors watching the electric vehicle market, the takeaways are tangible. Durable chemistry choices, thoughtful thermal management and realistic usage projections all shape how much mining the mobility transition ultimately requires. Shared mobility remains a valuable tool for cutting emissions and materials demand, but the new analysis makes clear that it is not automatically a resource panacea. Whether car sharing delivers its promised material dividend depends on batteries that can endure the punishing schedules of shared fleets, on recycling systems timed to collect retired packs without unnecessary detours, and on planners who treat degradation not as an afterthought but as a first-order variable in the circular economy of transport.

Subject of Research: Material demand and circularity implications of battery degradation in shared electric mobility in China

Article Title: Material circularity implications of battery degradation in shared electric mobility

Article References: Yang, Q., Xu, G., Chen, W., Tong, D., & Cao, Z. (2026). Material circularity implications of battery degradation in shared electric mobility. Nature Sustainability. https://doi.org/10.1038/s41893-026-01969-7

Image Credits: AI Generated

DOI: 10.1038/s41893-026-01969-7

Keywords: shared mobility, battery degradation, lithium, NMC811, lithium iron phosphate, recycling, second-life batteries, material flow analysis, critical minerals, China, electric vehicles, circular economy

In a study published in the journal Ionics, Zhuo Wang, Xu Zhang, Long Peng, Jingwen Zhu, and Qiquan Li developed an analytical model that captures the lithiation-induced stress inside bilayered silicon/graphite electrodes, the kind of layered architectures that battery manufacturers are increasingly exploring as a pragmatic compromise between silicon’s enormous capacity and graphite’s mechanical reliability. Their work, published on 13 October 2026, provides one of the clearest mechanical pictures yet of why layer arrangement matters, and it delivers concrete design rules that electrode engineers can apply directly.

The core of the study is a mechanoelectrochemical coupling model, a mathematical framework that links the electrochemical state of charge to the mechanical stress it generates. As lithium ions flood into an electrode during charging, the active materials expand; as they are withdrawn during discharge, the materials contract. Because silicon expands far more than graphite, and because the two materials possess different elastic moduli, a bilayer electrode becomes a mechanical mismatch problem: the silicon-rich layer wants to swell while the graphite-rich layer resists, and the thin metallic current collector underneath is dragged along for the ride. The researchers’ model tracks how these competing tendencies translate into stress distributions and bending deformations across the entire electrode stack.

What the analysis revealed is that the electrode stress originates primarily from the volumetric expansion associated with the increasing state of charge, meaning that the deeper a battery is charged, the more mechanical punishment its anode endures. This is a crucial insight for battery designers, because it implies that the most damaging stresses accumulate precisely during the high-capacity operation that silicon electrodes are meant to deliver. Managing stress is therefore not an optional refinement but a fundamental requirement if silicon-based anodes are ever to reach industrial maturity, a point the authors emphasize when they describe mechanoelectrochemical coupling issues as bottlenecks hindering the industrial development of silicon-based lithium-ion batteries.

The team systematically examined how two architectural variables, the relative position of the graphite- and silicon-based active layers and the thickness ratio between them, influence the stress landscape. For single-sided bilayer electrodes, they compared configurations in which the silicon-based layer sits above the graphite-based layer versus the reverse arrangement. The results were striking. Placing the silicon-based layer on top, a configuration the researchers designate Architecture II, produces increased bending deformation of the whole electrode. Somewhat counterintuitively, this configuration slightly lowers the compressive stress within the active layers themselves, but it comes at a hidden cost: a significant increase in stress within the current collector, the thin metallic foil that carries electrons in and out of the electrode.

That last finding may prove to be the study’s most consequential contribution. Current collectors are rarely the focus of electrode design conversations, yet they are the structural backbone of the entire assembly. If a particular layer arrangement offloads stress from the active materials only to concentrate it in the collector, the electrode may fail in a different and equally destructive way, through collector wrinkling, delamination, or fracture. The researchers’ model makes these hidden trade-offs visible, allowing designers to see that the mechanical budget of an electrode cannot be optimized layer by layer in isolation; the whole stack, including the collector, must be considered as a coupled system.

The picture changes for symmetric double-sided bilayer electrodes, in which active layers coat both faces of the current collector. Here, the researchers found that the sequence of the active layers has no significant effect on the electrode stress, a result that simplifies design choices considerably for double-sided architectures. However, a second lever proved powerful: increasing the thickness ratio of the graphite-based layer to the silicon-based layer effectively mitigates stress, particularly within the current collector. In other words, giving graphite the lion’s share of the thickness acts as a mechanical buffer, absorbing and distributing the deformation that the silicon layer would otherwise impose on the fragile collector beneath.

These conclusions arrive at a moment when layered and gradient silicon/graphite electrodes are attracting intense attention across the battery research community. Recent work has explored dual-layer hybrid graphite anodes for fast charging, gradient silicon distributions for energy- and power-dense cells, and porosity-engineered layered composites, all converging on the same idea: spatially organizing silicon within the electrode can dramatically improve performance. What the Zhongyuan team adds is a rigorous analytical treatment of the mechanical consequences, complementing electrochemical studies that focus on capacity, rate capability, and cycling life. By grounding architecture decisions in stress analysis, their model helps explain why some layered designs succeed in the laboratory while others fail prematurely.

The study also builds on a rich mechanistic literature. Prior experimental work has measured strains in composite battery electrodes during cycling and documented the interplay between electrochemical reactions and mechanical responses in silicon-graphite anodes, including its impact on degradation. First-principles studies have shown that the elastic properties of lithiated silicon and graphite themselves evolve with lithium concentration, with silicon softening as it lithiates and graphite stiffening during intercalation. The new analytical model incorporates this mechanoelectrochemical reality, evaluating how stress evolves as the state of charge climbs, rather than treating the electrode as a static structure. That dynamic perspective is essential for a component that may swell and relax hundreds of times over its service life.

From a mechanical perspective, the researchers distilled their numerical evaluation of the analytical solution into several electrode design strategies, offering manufacturers a practical playbook. For single-sided coatings, the message is to think carefully before stacking silicon on top, since the apparent relief in active-layer stress is offset by collector stress amplification and greater bending. For double-sided coatings, the message is more encouraging: layer order matters little, but deliberately weighting the graphite layer’s thickness relative to the silicon layer can meaningfully reduce the stresses that threaten the collector. As the industry races toward higher-energy-density cells, such rules of thumb, grounded in quantitative stress analysis rather than trial and error, could shorten the path from laboratory concept to factory floor. The work was supported by the Key Projects of Science and Technology of Henan Province and the Young Backbone Teacher program of Zhongyuan University of Technology, and it underscores a broader truth about the energy transition: sometimes the breakthrough is not a new material, but a smarter way to arrange the ones we already have.

Subject of Research: Mechanoelectrochemical stress modeling of bilayered silicon/graphite electrodes in lithium-ion batteries

Article Title: Effects of bilayered architecture design on the mechanical stability of silicon/graphite-based electrodes for lithium-ion batteries

Article References: Wang, Z., Zhang, X., Peng, L., Zhu, J., & Li, Q. (2026). Effects of bilayered architecture design on the mechanical stability of silicon/graphite-based electrodes for lithium-ion batteries. Ionics. https://doi.org/10.1007/s11581-026-07582-0

Image Credits: AI Generated

DOI: 10.1007/s11581-026-07582-0

Keywords: lithium-ion batteries, silicon anode, graphite, bilayer electrode, lithiation-induced stress, mechanoelectrochemistry, current collector, electrode architecture, volumetric expansion, battery degradation, analytical modeling, energy storage

The team, led by Prof. Chen Guohua, Chair Professor of the Department of Chemical and Biological Engineering at HKUST, developed a cartilage-inspired polymer nanocoating that suppresses the degradation and wear of battery cathode materials where it begins: at the internal grain boundaries of individual cathode particles. The work, carried out with Dr. Liu Qiang, a Research Assistant Professor in the same department, and Dr. Liu Yutong, a postdoctoral fellow who began the research during her PhD at The Hong Kong Polytechnic University, demonstrates that a coating only about fourteen nanometers thick can act as a dynamic shock absorber inside battery particles, holding capacity at exceptionally high levels after thousands of charge-discharge cycles.

The problem the coating addresses is known as intergranular cracking, and it is the central obstacle standing between laboratories and the commercial deployment of next-generation nickel-rich layered cathodes. These nickel-rich materials are attractive because they pack more energy into the same volume, which is exactly what manufacturers of electric vehicles and consumer electronics want. But during cycling, the lithium ions that shuttle in and out of the cathode cause repeated volume changes in the material. Those changes generate severe dynamic internal stress, and because layered cathode particles are polycrystalline aggregates built from many smaller primary particles, the stress concentrates along the boundaries between those grains. Cracks propagate, the electrical and ionic pathways inside the particle are severed, isolated fragments stop contributing to capacity, and the cathode degrades cycle by cycle.

The HKUST team’s solution came from an unlikely teacher: the human knee. Cartilage in human joints performs a remarkable mechanical function, deforming under load and dissipating external stress so that the brittle bones beneath it are protected from damage. Replicating that biological strategy at the nanoscale, the researchers designed a crosslinked shape-memory polymer with the chemical designation p(BA-V4D4). The material is engineered to balance two properties that normally pull against each other: deformability, which lets it absorb and redistribute mechanical stress, and strength, which keeps it from simply being squeezed out of the way. When placed between adjacent primary particles inside a cathode, the optimized copolymer behaves, in the team’s own description, like cartilage between bones.

Getting a polymer into the interior of a battery particle is not straightforward, and this is where the team’s process innovation matters as much as the material itself. The researchers used initiated Chemical Vapor Deposition, or iCVD, a gas-phase technique previously developed in Chen’s laboratory. In iCVD, monomer vapors are delivered to the substrate and polymerize in situ on its surface, requiring no liquid solvents. Because the reactants travel as vapor, they can penetrate spaces that liquid coatings cannot reach. In this work, the monomer vapor penetrated deep into the electrode’s secondary particles, reaching the vulnerable internal boundaries between primary particles through nanosized channels just ten to twenty nanometers wide and roughly four thousand nanometers deep, and polymerized there to form a resilient, highly uniform stress-absorbing layer roughly fourteen nanometers thick.

Dr. Liu Qiang, a co-corresponding author of the study, explained the mechanism in the university’s announcement. By deploying a shape-memory polymer as an adaptive stress dissipator, he said, the team can suppress intergranular cracking through a solid stress delocalization mechanism, and the polymer structure was engineered to achieve an optimal balance between deformability and strength. He credited the laboratory’s earlier CVD technology for enabling the coating to be integrated into grain boundaries through those incredibly narrow nanosized channels. Dr. Liu Yutong, a co-first author, added that traditional methods cannot penetrate the grain boundaries of cathode materials or allow researchers to tune the mechanical properties of the polymer easily, whereas the iCVD method transports reactants into those boundaries to form a protective coating via in-situ polymerization. She noted that the team has successfully applied the technology to nickel-rich NCM cathodes and to commercial LiFePO4 materials, demonstrating its universal applicability across different cathode chemistries.

The shape-memory character of the polymer is what elevates it beyond a passive binder. A conventional coating might cushion a particle once, but under the repeated cyclic loading of thousands of charge-discharge cycles, a passive layer would fatigue and fail. A shape-memory polymer, by contrast, can deform under stress and recover its configuration, delocalizing stress dynamically rather than merely absorbing a single impact. In effect, the coating converts the sharp, localized stress concentrations that would otherwise nucleate cracks into distributed, manageable deformation across the particle’s interior. That is precisely the role cartilage plays in a joint, and it is why the researchers describe their material as biomimetic rather than simply protective.

For Prof. Chen, the achievement is the culmination of a research thread that began two decades ago. He recalled that the idea of applying CVD polymer coatings to energy storage materials took shape during his sabbatical at MIT in 2005, where CVD polymers were being applied to pharmaceutical particles for controlled drug release. His laboratory has since been, in his words, the first to successfully apply the technology to lithium-ion battery materials, and it previously demonstrated that CVD can coat conductive polymers in research published in Nature Energy in 2019. The new p(BA-V4D4) work extends that platform to a broader range of functional polymeric materials, positioning iCVD as a general tool for engineering the mechanical environments inside energy storage devices rather than only their surfaces.

The implications reach beyond conventional liquid-electrolyte cells. Prof. Chen said the team is now exploring the application of the technology to solid-state batteries, where mechanical stress at interfaces is an even more acute problem and where the coating could help address critical challenges related to both high-energy density and safety. Solid-state designs replace the flammable liquid electrolyte of conventional cells with a solid medium, but maintaining stable contact and managing stress between rigid solid components remains one of the field’s hardest problems, making an adaptive stress-dissipating interlayer an appealing candidate. Meanwhile, the team is scaling up the iCVD process to produce coated cathode particles at the 1,000-kilogram scale, enough to manufacture large batteries for industrial applications spanning consumer electronics, electric vehicles, and electric vertical takeoff and landing aircraft, a category of aircraft whose viability depends heavily on battery energy density and cycle life.

The research was a collaborative effort involving experts from universities and institutions across Hong Kong and the Chinese Mainland, contributing materials synthesis, characterization, and data science. Co-first authors alongside Dr. Liu Yutong are Prof. Zuo Wenhua of Xiamen University and Dr. Wang Wei, a PhD graduate of The Hong Kong Polytechnic University. Key contributors include Dr. Yu Hailong and Prof. Huang Xuejie of the Institute of Physics at the Chinese Academy of Sciences, Prof. Liu Xiang of Beihang University, Prof. Ruan Haihui and Dr. Lin Cong of PolyU, Dr. Zhang Kai of the Institute of High Energy Physics at the Chinese Academy of Sciences, Prof. Du Ke of Central South University, and Prof. Pan Feng of Peking University Shenzhen Graduate School. The study, titled Dynamic delocalization of stress in brittle battery positive electrode active materials by shape-memory polymer nanocoating, was published in Nature Nanotechnology on 23 July 2026. If the scale-up succeeds, the humble engineering lesson embedded in human joints, that resilient, deformable layers protect brittle structures from repeated stress, may soon be running inside the batteries of phones, cars, and aircraft, quietly extending the working life of the devices that depend on them.

Subject of Research: A biomimetic shape-memory polymer nanocoating that suppresses intergranular cracking in lithium-ion battery cathode materials.

Article Title: HKUST develops novel cartilage-inspired polymer nanocoating to substantially extend battery lifespan

Article References: HKUST develops novel cartilage-inspired polymer nanocoating to substantially extend battery lifespan. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: lithium-ion batteries, cathode materials, nanocoating, shape-memory polymer, initiated chemical vapor deposition, nickel-rich cathodes, intergranular cracking, biomimetic materials, solid-state batteries, HKUST, Nature Nanotechnology, battery degradation

To appreciate why this matters, it helps to understand what formation actually is. When a lithium-ion cell is assembled, it is not yet a functional battery. It must first be charged and discharged in carefully controlled cycles, a process called formation, during which the solid electrolyte interphase, a critical protective layer on the anode, is established. This stage is among the most expensive and time-consuming steps in battery manufacturing, and manufacturers around the world use different formation protocols, varying the charging currents, voltages, and cycle patterns, in pursuit of faster production or better performance. The trouble is that these protocol differences leave distinct fingerprints on the cell’s long-term degradation behavior. A prediction model trained on batteries formed one way may fail badly when applied to batteries formed another way, which is precisely the problem the research team set out to solve.

Traditional approaches to remaining useful life prediction, the researchers note, often suffer from poor accuracy and weak robustness when batteries have undergone different formation protocols. The core difficulty is that formation-induced differences create systematic biases in the early-cycle data that prediction models rely on. Features extracted from charge-discharge curves may look deceptively similar across cells while hiding fundamentally different aging trajectories. The new method confronts this head-on with a two-pronged strategy: an unscented transformation tailored for variable formation protocols, and a hybrid framework combining an Unscented Particle Filter with a Gated Recurrent Unit neural network, complete with a protocol-sensitivity compensation module designed to mitigate the bias that protocol differences inject into predictions.

The first component of the framework is an exponential empirical degradation model, fitted using an Unscented Particle Filter, that captures the underlying aging mechanism of the battery. Empirical degradation models describe how capacity fades over cycles using mathematical functions, and the particle filter is a Bayesian inference technique that continuously updates the model’s parameters as new measurements arrive. The unscented variant is particularly well suited to this task because it propagates uncertainty through nonlinear transformations without requiring the model to be linearized, a step that can introduce serious errors in battery systems where degradation is anything but linear. By anchoring the prediction in a physically motivated degradation model, the framework avoids the worst failure mode of purely data-driven approaches: extrapolating confidently into regions where the training data offers no guidance.

The second component is where the deep learning enters. The researchers constructed a Gated Recurrent Unit neural network, a streamlined relative of the long short-term memory architecture that has become a workhorse of time-series prediction. The network takes key features from constant-current and constant-voltage charge-discharge cycles as its inputs, treats the battery’s state of health as a hidden state that evolves over time, and outputs the remaining useful life under varying formation protocols. Gated recurrent units are prized for their ability to capture temporal dependencies in sequential data while remaining computationally lighter than full LSTM networks, making them attractive for applications where predictions must be updated continuously as new cycling data streams in.

What makes the architecture genuinely novel is the way these two components collaborate. Rather than letting the neural network operate alone, the framework integrates empirical modeling with data-driven prediction, allowing the particle-filter-based degradation model to refine the network’s RUL estimates while the protocol-sensitivity compensation module adapts dynamically to variations in formation conditions. In effect, the empirical model provides a physically grounded skeleton, the neural network fleshes out the complex, protocol-dependent details, and the compensation module corrects for the systematic distortions that different formation histories introduce. This model-data fusion philosophy reflects a broader trend in battery prognostics, where researchers increasingly recognize that neither pure physics nor pure machine learning is sufficient on its own.

Validation was carried out on public data from the University of Michigan Battery Lab, specifically the fast formation study dataset published through the university’s Deep Blue Data repository. This dataset is a valuable testing ground precisely because it contains batteries cycled under different formation protocols, allowing the researchers to systematically quantify and model the impact of protocol differences on long-term degradation and remaining useful life prediction. The results were striking: the proposed method achieved an R-squared value of 99.41 percent, meaning it explained nearly all of the variance in the observed degradation trajectories, and a root mean square error of 2.23 percent, indicating tight agreement between predicted and actual values. The team reports that the method significantly outperformed comparison methods in both state of health and remaining useful life prediction accuracy, demonstrating superior cross-protocol generalization and robustness compared with existing hybrid approaches.

The practical implications of this work extend well beyond the laboratory. Accurate remaining useful life prediction is essential for determining appropriate battery replacement times, a decision with enormous consequences for safety, cost, and sustainability. In electric vehicles, overestimating a battery’s life risks stranding drivers with a depleted pack, while underestimating it wastes money by replacing cells that still have serviceable capacity. In aerospace and space applications, where some of the co-authors’ institutions operate, the stakes are even higher, since battery failure can compromise entire missions. A prediction framework that remains accurate regardless of how the cells were formed could allow manufacturers to mix batteries from different production lines or formation batches into the same fleet management system without sacrificing prognostic reliability, something current methods struggle to guarantee.

There is also a manufacturing angle that could prove commercially significant. Formation is one of the most energy-intensive and capital-heavy stages of cell production, and the industry has been actively exploring faster formation protocols to cut costs and increase throughput. One reason manufacturers hesitate to adopt aggressive fast-formation schemes is uncertainty about how those choices will affect long-term life and whether existing quality-control and prognostic tools will still work. A prediction method that explicitly compensates for protocol sensitivity could lower that barrier, giving manufacturers the confidence to experiment with formation strategies that shorten production times without losing the ability to forecast field performance. The research team, which includes authors affiliated with CATL, the world’s largest battery manufacturer, is well positioned to understand these industrial pressures.

The study, led by Gaige Chen and Haitao Meng, who contributed equally, with Yahong Yang, Xiaoshen Zhang, Wei Deng, and Jinsong Liu, was supported by the National Natural Science Foundation of China and builds on the group’s earlier work on transfer learning for state of health prediction under cross-formation protocols. As batteries proliferate through every corner of the energy economy, tools like this hybrid UPF-GRU framework point toward a future in which the lifespan of every cell can be forecast accurately from its earliest cycles, regardless of the manufacturing choices that shaped it. That kind of predictive certainty could reshape how batteries are designed, produced, warranted, and retired, turning one of the industry’s most persistent blind spots into a solved problem.

Subject of Research: Machine learning-based remaining useful life prediction of lithium-ion batteries across different formation protocols

Article Title: Remaining useful life prediction of lithium-ion batteries across formation protocols

Article References: Chen, G., Meng, H., Yang, Y., Zhang, X., Deng, W., & Liu, J. (2026). Remaining useful life prediction of lithium-ion batteries across formation protocols. Ionics. https://doi.org/10.1007/s11581-026-07472-5

Image Credits: AI Generated

DOI: 10.1007/s11581-026-07472-5

Keywords: lithium-ion batteries, remaining useful life, formation protocols, unscented particle filter, gated recurrent unit, state of health, battery degradation, machine learning, battery manufacturing, prognostics, CATL, University of Michigan Battery Lab

Predictive maintenance rests on a deceptively simple bargain. Algorithms trained on operational data learn to estimate a battery’s state of charge, state of health and remaining useful life, allowing operators to schedule interventions before failures occur rather than after. But the models behind these estimates are typically trained under a set of assumptions: that the battery will perform a defined duty, that environmental conditions will resemble those seen during training, and that the monitoring pipeline feeding data to the algorithm will stay consistent. The Perspective’s central claim is that all three assumptions are increasingly false for modern storage assets, and that the resulting mismatch, which the authors call operational drift, is a structural failure mode rather than an occasional nuisance.

The sources of drift are more varied than they might first appear. Application requirements evolve as grid markets change and new revenue streams emerge, so a system installed for peak shaving may later be dispatched for frequency regulation with very different charge and discharge profiles. Duty cycles shift accordingly, and research has shown that dynamic cycling can actually enhance battery lifetime, meaning that degradation trajectories can bend in either direction when usage patterns change. Environmental exposure varies too, from extreme cold climates that demand heating and cooling loads to be factored into system sizing, to temperature differences that strongly influence which degradation mechanisms dominate. Even the monitoring pipeline itself can drift, as sensor biases, random errors, low sampling rates and intermittent data unavailability alter the statistical character of the very signals the algorithms depend on.

To bring order to this complexity, the authors classify drift by its origin in evolving asset roles and then organize its effects into three layered functional mismatches. The first is a data-layer mismatch: the distribution of inputs seen in deployment no longer matches the training distribution, a phenomenon statisticians call covariate shift. The second is a model-layer mismatch, where the relationship the algorithm learned between features and outcomes no longer holds, for example when a new duty cycle activates degradation pathways such as lithium plating that were rare during training. The third is a temporal-layer mismatch, arising because degradation is not stationary in time; knee points on capacity-fade curves mark transitions from gradual to accelerated decline, and a model calibrated on early-life behavior can be badly wrong after such a transition.

Each mismatch layer, the authors argue, calls for a targeted transfer strategy rather than a one-size-fits-all fix. At the data layer, domain adaptation techniques can align source and target distributions, using statistical distances such as maximum mean discrepancy or Wasserstein distance to quantify and correct the shift. Adversarial variants go further, learning features that a domain classifier cannot distinguish, so that a model trained on laboratory cells or on one fleet can generalize to another. At the model layer, transfer learning approaches range from freezing a backbone network and attaching small trainable adapter layers, to physics-informed schemes that embed electrochemical knowledge, such as Arrhenius temperature dependence, directly into activation functions. Residual learning offers another elegant option: instead of learning the full mapping, the model learns only the difference between a physics-based reference prediction and observed behavior, which keeps adaptations small and interpretable.

The temporal layer demands perhaps the most delicate machinery, because adapting a model to new conditions can destroy what it already knows. The Perspective explicitly assesses candidate strategies against four failure risks: negative transfer, where adaptation to a new domain actually degrades performance; catastrophic forgetting, where learning new behavior erases old competence; poorly calibrated uncertainty, where the model’s confidence no longer reflects its true reliability; and low sample efficiency, where adaptation demands more target data than operators can realistically collect. Continual learning methods, including regularization-based approaches and bilevel-optimized schemes, have been shown to preserve earlier knowledge while absorbing new operating conditions, and meta-learning can adapt to a fresh battery with only a handful of target measurements.

What makes the argument timely is the sheer scale of the stakes. Grid-scale storage is expanding rapidly as renewable generation grows, and batteries increasingly provide multiple services simultaneously, from arbitrage and peak shaving to regulation and reliability support. Studies of vehicle-to-grid operation show that the dominant degradation mechanism determines whether extra energy throughput is worth the days of life lost, which means maintenance decisions are inseparable from market strategy. Meanwhile, retired electric vehicle batteries are flowing into second-life deployments where their histories are heterogeneous and often poorly documented, and where a model trained on first-life data may face a target domain it has never seen. The authors position transferable predictive maintenance as a life-cycle-spanning capability that connects manufacturing, first-life operation and second-life deployment into one continuous chain of health information.

The framework also carries implications that reach back to the factory floor. Research on transfer learning for battery cell manufacturing suggests that models trained during formation and early testing can predict lifetime outcomes, and that low-state-of-charge resistance measured immediately after manufacturing is a rapid predictor of how formation protocols will shape long-term durability. If such early-life signals could be carried forward through a battery’s passports and reused when the cell changes roles, the cold-start problem of second-life health estimation would shrink dramatically. Recent work on history-free prediction of second-life degradation trajectories from a single partial cycle, using interpretable mixture-of-experts models, hints at how much can be recovered even from sparse information, but the authors contend that systematic transferability would make such feats routine rather than exceptional.

None of this can be built on algorithms alone, and the Perspective is unusually concrete about the infrastructure required. The authors call for data infrastructures that preserve context across the life cycle, foundation models for batteries that encode broad electrochemical and operational knowledge and can be cheaply adapted to new tasks, and open benchmarks that test transferability under realistic drift rather than stationary laboratory conditions. Initiatives toward unified battery data descriptions and large-scale real-world datasets of retired cells point in this direction, as does work on federated learning, which allows fleets owned by different parties to collaborate on health estimation without sharing raw data. Privacy-preserving collaboration matters because the data that would make transferable maintenance most powerful, spanning manufacturers, fleet operators and recyclers, is exactly the data that commercial interests are most reluctant to pool.

The authors close with regulatory recommendations, arguing that standards should require calibrated uncertainty alongside point predictions and should encourage the audit dimensions their framework defines, so that operators can verify not just whether a model is accurate but whether it remains trustworthy as conditions drift. If they are right, the payoff extends well fewer failures and longer asset lives: batteries that can document their own health across changing roles would become easier to resell, repurpose and recycle, tightening the loop of a circular battery economy. In a decade when storage assets are expected to outlive the markets they were built for, the ability of their maintenance algorithms to move with them may prove as important as the chemistry inside the cells. Transferable predictive maintenance, the authors suggest, is the missing link between the batteries we build and the many lives we ask them to live.

Subject of Research: Transferable predictive maintenance for battery energy storage systems under operational drift

Article Title: Transferable predictive maintenance for battery energy storage systems

Article References: Tao, S., Liang, C., Tavallaey, S. S., Thulin, N., Wik, T., Zio, E., Pecht, M. G., Mo, H., & Zou, C. (2026). Transferable predictive maintenance for battery energy storage systems. Nature Reviews Electrical Engineering. https://doi.org/10.1038/s44287-026-00337-z

Image Credits: AI Generated

DOI: 10.1038/s44287-026-00337-z

Keywords: battery energy storage, predictive maintenance, transfer learning, operational drift, domain adaptation, continual learning, state of health, second-life batteries, uncertainty quantification, foundation models, grid storage, battery degradation

Electrochemical impedance spectroscopy, or EIS, works by applying a small sinusoidal voltage or current perturbation across a range of frequencies and measuring the cell’s response. At high frequencies, the technique reveals the resistance that ions and electrons encounter as they move through electrolytes and across interfaces. At low frequencies, it exposes the slower processes of diffusion, the way lithium ions wander through electrode materials to find their storage sites. By fitting the resulting spectra to equivalent circuit models, researchers can separate these contributions and assign numerical values to them. In this study, the two most telling parameters were the charge transfer resistance, denoted R-CT, which describes the energetic barrier that lithium ions must overcome to cross the electrode-electrolyte interface, and the Warburg impedance, W, which captures the resistance to solid-state diffusion deep within the electrodes.

The team systematically varied the state of charge, or SOC, of polymer lithium-ion cells and recorded impedance spectra at each point, paying particular attention to what happens when the cells are driven into the over-discharge regime. The results revealed a striking asymmetry between the high and low ends of the voltage window. In the high state-of-charge region, spanning voltages from 4.2 volts down to 3.7 volts, the charge transfer resistance increased steadily while the Warburg impedance remained relatively steady. This makes physical sense: when a cathode is nearly full of lithium or an anode is nearly empty, the interfacial kinetics change, but the diffusion pathways through the bulk electrode material remain largely intact.

The picture changed dramatically in the low state-of-charge and over-discharge zone, covering voltages from 3.6 volts down to 2.0 volts. Here the researchers observed a significant rise in the Warburg impedance alongside a drop in the charge transfer resistance. This combination signals that the dominant limitation shifts from the interface to the bulk: lithium ions find it increasingly difficult to diffuse through the electrode material, while the interfacial reaction kinetics are paradoxically altered in a way that lowers the apparent charge transfer barrier. The authors interpret this as evidence of enhanced diffusion restrictions and altered interfacial kinetics, a signature that over discharge reshapes the interior landscape of the cell rather than merely fouling its surfaces.

One of the most important findings concerns the subtlety of the damage. When the researchers compared spectra taken at equivalent voltages and equivalent cycle numbers, they found that over discharge does not produce dramatic spectral distortion. Instead, it induces progressive shifts in the interfacial resistance and in the low-frequency response of the cell. This is a crucial insight for battery diagnostics, because it means that a catastrophic-looking failure may be preceded by gradual, measurable changes that only careful impedance monitoring would catch. A battery management system that tracks these parameters over time could, in principle, flag a cell that has been over-discharged long before its capacity loss becomes obvious to the user.

The study also examined how these impedance parameters evolve with repeated cycling. Both the charge transfer resistance and the Warburg impedance increased gradually as the cells were cycled, with the tenth cycle exhibiting significantly higher values than the first. This progressive growth suggests cumulative electrode deterioration and a steady loss of ionic transport efficiency. In practical terms, each abuse event, each excursion into the over-discharge zone, leaves a residue of damage that accumulates. The electrodes lose their ability to shuttle lithium ions efficiently, and the battery’s internal resistance climbs, which in turn increases heat generation, reduces power output, and accelerates further degradation in a self-reinforcing loop.

The implications for battery management systems are significant. Modern BMS technology increasingly relies on impedance-based methods to estimate the state of charge and state of health of cells, and a rich body of literature has developed around using EIS for these purposes. Previous studies have shown that impedance measurements can support state-of-charge estimation even for aging batteries, and machine learning approaches have been layered on top of impedance data to improve prediction accuracy. However, most of this work assumes the battery operates within its normal voltage window. The Daegu University study adds a missing piece: a quantitative description of how the key impedance parameters behave when that assumption is violated, which is exactly the situation that occurs in real-world abuse scenarios such as deep discharge during vehicle breakdowns, faulty electronics, or user error.

The findings also connect to a broader research effort on over-discharge failure mechanisms. Earlier work on commercial lithium iron phosphate cells showed that over discharge causes structural and chemical changes in electrodes, including copper dissolution from the anode current collector at extreme depths of discharge, which can lead to internal short circuits. Other studies have documented effects on performance and thermal stability. What distinguishes the new study is its focus on polymer lithium-ion batteries, a format favored for its flexibility and safety, and its systematic separation of the impedance response into interfacial and diffusional components across the full discharge range, including the abusive zone below the normal cutoff voltage.

For consumers, the research underscores a familiar message with new precision: the damage from draining a battery too far is real, measurable, and cumulative. For engineers, it provides a diagnostic roadmap. Because the impedance changes induced by over discharge are systematic rather than chaotic, they can be incorporated into the models that battery management systems use to monitor cell health. A vehicle or device that periodically measures its cells’ impedance could detect the characteristic rise in Warburg impedance and the shifting interfacial resistance that betray a history of deep discharge, then adjust charging strategies or flag the cell for inspection. The authors emphasize that impedance monitoring of this kind is important for early degradation detection and battery management in polymer lithium-ion batteries.

As the world electrifies transportation and deploys ever-larger stationary storage systems, the economics of battery longevity become staggering in scale. Extending the useful life of a battery pack by even a few percentage points translates into enormous savings of cost, raw materials, and environmental impact. Studies like this one, which translate the invisible electrochemistry of degradation into measurable electrical signatures, are the foundation of that effort. The over-discharged battery, it turns out, does tell the story of its abuse; it simply tells it in the language of impedance, at frequencies and magnitudes that only careful spectroscopy can hear. Learning to listen, the Daegu University work suggests, may be one of the most practical steps toward smarter, longer-lived, and safer energy storage.

Subject of Research: Electrochemical impedance analysis of polymer lithium-ion batteries under over-discharge conditions

Article Title: Impedance analysis of polymer lithium-ion batteries under over discharge

Article References: Subedi, M., & Jeong, H. K. (2026). Impedance analysis of polymer lithium-ion batteries under over discharge. Ionics. https://doi.org/10.1007/s11581-026-07501-3

Image Credits: AI Generated

DOI: 10.1007/s11581-026-07501-3

Keywords: lithium-ion battery, over discharge, electrochemical impedance spectroscopy, charge transfer resistance, Warburg impedance, state of charge, battery degradation, polymer battery, battery management system, ionic transport, electrode deterioration, energy storage

The team’s starting point was a rich dataset of 72 real-driving trips performed with a BMW i3, a compact electric vehicle whose battery and heating circuit data were recorded under genuine operational conditions. This kind of data is gold for battery researchers. Synthetic drive cycles, such as the standardized profiles used in certification tests, smooth away the stop-and-go chaos, temperature swings, and driver behavior that dominate everyday motoring. Previous comparative studies have shown that battery degradation patterns on synthetic cycles can differ markedly from those observed in the wild, which means control strategies tuned in the laboratory often underperform on the street. By anchoring their models to 72 authentic journeys, Chen and Zhou ensured that every prediction and optimization they made reflected the messy reality of daily driving.

The first technical step was validation. The researchers constructed detailed vehicle and heating circuit models and checked them against the recorded trip data, confirming that their simulations reproduced the thermal and electrical behavior of the actual car. With trustworthy models in hand, they then developed strategies for a clever energy trick: using regenerative braking energy to heat the cabin. In an electric vehicle, braking normally wastes kinetic energy as heat at the friction pads, but regenerative systems recover part of it as electricity. Routing that recovered energy into cabin heating rather than drawing it from the traction battery eases the load on the cells, and the study reports that such strategies improved both the vehicle’s driving range and battery life. It is a double win: passengers stay warm, and the battery endures fewer stressful charge-discharge excursions.

From there, the study pivoted to machine learning. The researchers framed a prediction problem in which environmental variables, vehicle variables, battery variables, and thermal system variables serve as inputs, while the target outputs are the battery’s final State of Charge and the distance traveled in kilometers. In plain terms, they wanted an algorithm that could look at the conditions of a trip and accurately forecast how much energy would remain in the battery and how far the car would go. Four candidate models were systematically compared: XGBoost, AdaBoost, CatBoost, and K-Nearest Neighbor. Each was assessed with a rigorous protocol, an 80/20 train-test split combined with five-fold cross-validation, which guards against the common pitfall of an algorithm that merely memorizes its training data.

Among the four contenders, CatBoost, a gradient-boosting method developed originally to handle categorical features gracefully, emerged as the strongest baseline performer. On the held-out test set it achieved a coefficient of determination, or R², of 0.7260, a root mean square error of 0.0734, and a mean absolute error of 0.0592, while also posting the highest cross-validation accuracy of the group with an R² of 0.802. For readers unfamiliar with these metrics, R² measures the fraction of variance in the outcomes that the model explains, with 1.0 being perfect, while the error metrics quantify the typical size of the model’s mistakes. A baseline R² above 0.72 on real-world driving data is respectable, but the authors were not satisfied.

The study’s most distinctive contribution lies in what came next: metaheuristic optimization. Metaheuristics are search algorithms inspired by natural processes, from the flocking of birds to the foraging of insects, that explore vast parameter spaces far more efficiently than brute-force grid searches. Chen and Zhou deployed three such algorithms to fine-tune the CatBoost model’s hyperparameters, the internal settings that govern how aggressively the model learns and how it balances bias against variance. The winner was the Horned Lizard Optimization algorithm, a relatively new nature-inspired method, which when coupled with CatBoost produced the study’s best-performing framework, dubbed HLO-CatBoost.

The improvements were substantial. The optimized HLO-CatBoost framework achieved a test R² of 0.8408, with the root mean square error falling to 0.0559 and the mean absolute error to 0.0434. In other words, hyperparameter tuning by the horned lizard algorithm lifted the explained variance by more than eleven percentage points while cutting the typical prediction error by roughly a quarter compared with the untuned baseline. That gap matters in practice: a battery management system that predicts remaining charge and range more accurately can make smarter decisions about when to heat the cabin, when to limit power, and how to keep cells inside their optimal temperature window, directly translating into extra kilometers per charge and slower long-term degradation.

Crucially, the authors did not stop at headline accuracy numbers. They subjected their framework to a battery of further analyses designed to probe its robustness and trustworthiness. Environmental scenario testing examined how the model behaves across different ambient conditions, a critical check for a system intended to operate from freezing winters to hot summers. Uncertainty analysis quantified the confidence one can place in individual predictions. Computational efficiency assessments confirmed that the framework is fast enough for practical deployment rather than being a laboratory curiosity. A Williams plot, a standard diagnostic from chemometric modeling, was used to identify outliers and define the model’s applicability domain, flagging any predictions that fall outside the territory the model learned from. Finally, feature importance analysis opened the black box, revealing which input variables most strongly drive the predictions of final state of charge and travel distance.

This emphasis on interpretability and validation distinguishes the work from much of the machine learning literature applied to batteries, where a single impressive accuracy figure is often the whole story. Battery thermal management systems, or BTMSs, are safety-critical components: they must prevent thermal runaway, a catastrophic failure mode in which a cell overheats and triggers a chain reaction, while simultaneously squeezing out every possible increment of efficiency. A model that cannot explain itself, or that fails unpredictably at the edges of its training distribution, is a liability in that context. By combining rigorous cross-validation, outlier diagnostics, scenario analysis, and feature attribution, Chen and Zhou have assembled the kind of evidence package that engineers and regulators need before trusting an intelligent management system with a live vehicle.

The broader implications reach across the electric vehicle industry. Range anxiety remains one of the most cited barriers to EV adoption, and a meaningful share of that anxiety stems from the unpredictability of range in extreme weather, when heating and cooling loads can slash advertised figures. Machine learning frameworks trained on genuine driving data, and sharpened by metaheuristic optimization, offer a path toward management systems that anticipate those losses and compensate for them, whether by pre-heating the battery while the car is still plugged in, by harvesting braking energy for the cabin, or by modulating thermal loads in real time. The study’s authors, who collected and analyzed the data and evaluated and edited the manuscript, report no competing interests, and their work was supported by research programs at Chongqing institutions along with technical support from the marketing department of Chongqing Changan New Energy Automobile. As electric fleets grow and software-defined vehicles become the norm, the marriage of real-world data, gradient boosting, and nature-inspired optimization may well become a standard ingredient in the batteries that move us, turning every braking event and every degree of ambient temperature into information that helps a car go farther on the same charge.

Subject of Research: Machine learning optimization of electric vehicle battery thermal management using real-world driving data

Article Title: Optimizing automotive battery thermal regulation strategies for enhanced efficiency in real driving cycles

Article References: Chen, B., & Zhou, C. (2026). Optimizing automotive battery thermal regulation strategies for enhanced efficiency in real driving cycles. Ionics. https://doi.org/10.1007/s11581-026-07474-3

Image Credits: AI Generated

DOI: 10.1007/s11581-026-07474-3

Keywords: electric vehicles, battery thermal management, real driving cycles, CatBoost, Horned Lizard Optimization, machine learning, state of charge, regenerative braking, BMW i3, hyperparameter tuning, range prediction, battery degradation

At the heart of the review is a deceptively simple observation: the overall efficiency and degradation of a battery are determined by the contributions of its individual components, and the most consequential of those components are often the thinnest. When a lithium-ion cell is charged for the first time, the electrolyte decomposes on the surface of the negative electrode, forming a nanometer-scale film known as the solid electrolyte interphase, or SEI. A counterpart film, the cathode electrolyte interphase or CEI, forms on the positive electrode. These films are paradoxical creatures: they are a necessary evil, passivating the electrode and preventing further electrolyte breakdown, but they also add resistance that restricts the movement of lithium ions during electrochemical reactions. As the review emphasizes, the formation of the SEI, the CEI, and lithium dendrites creates major resistance at the electrode materials, and this interfacial drag is a central cause of capacity fade, impedance growth, and eventual battery failure.

The SEI has been studied since the early days of rechargeable lithium technology, but its chemistry remains fiendishly complex. It is a heterogeneous mosaic of inorganic compounds such as lithium fluoride, lithium oxide, and lithium carbonate, embedded in organic species derived from solvent decomposition. The review highlights recent work showing that the functional groups present at the surface of silicon anodes, one of the most promising high-capacity negative electrode materials, directly shape which SEI components form and how stable the film turns out to be. Researchers are increasingly designing electrolyte formulations specifically to regulate the SEI on lithium metal anodes, tailoring the balance between organic-rich and inorganic-rich interphase compositions to achieve both fast ion transport and long-term mechanical resilience. Quantitative frameworks for anti-corrosive passivation design, aimed at extending the calendar life of lithium metal batteries, represent the newest frontier in this effort.

On the positive side of the cell, the review surveys the evolution of cathode chemistry from the layered oxides that dominate today’s market to the high-voltage spinels and lithium-rich compositions that promise greater energy density. Layered lithium nickel manganese cobalt oxides remain the industry standard, but increasing the nickel content raises capacity at the cost of structural instability, surface reactivity, and transition metal dissolution. High-voltage spinel cathodes such as lithium manganese nickel oxide offer operating voltages approaching five volts, but they suffer from electrolyte oxidation and oxygen loss at the electrode surface. The review discusses surface modification, doping, electrolyte optimization, and the deliberate introduction of oxygen deficiencies as complementary strategies to stabilize these demanding materials. Recent work on cobalt-free lithium nickel oxide electrodes for sulfide-based all-solid-state batteries shows how far the push toward sustainable, high-voltage cathodes has progressed.

Lithium-rich layered oxides deserve special attention because they deliver extraordinary capacities by activating anionic redox, meaning that oxygen in the crystal lattice participates in charge storage alongside the transition metals. The catch is voltage fade: with repeated cycling, the operating voltage drifts downward, eroding energy density. Recent research has traced this phenomenon to trapped molecular oxygen within the cathode structure, a finding that finally gives designers a concrete target. The review also covers structural stabilization of lithium-rich cathodes through composite electrolytes that induce a rigid yet adaptive interphase at the cathode surface, protecting the material while preserving its exceptional capacity. Surface modification of these lithium-rich cathodes, the review notes, is one of the most active areas in current battery research.

The most tantalizing prize in the field remains the lithium metal anode. Replacing graphite with pure lithium could dramatically increase energy density, because lithium offers the highest theoretical specific capacity of any anode material and the lowest electrochemical potential. But lithium metal is notoriously unruly. During charging, lithium ions plate onto the anode surface, and if the deposition is not uniform, needle-like structures called dendrites grow across the cell. These can pierce separators, cause internal short circuits, and in the worst cases trigger fires. Even short of catastrophic failure, dendrites create dead lithium that is electrically disconnected from the electrode, permanently draining the cell’s capacity. The kinetics of the lithium electrodeposition process, including the role of exchange current density in determining whether lithium deposits smoothly or erupts into dendrites, are now understood in far greater detail than a decade ago.

Against this backdrop, the review devotes substantial space to artificial interphase engineering, one of the most promising strategies for making lithium metal anodes practical. Instead of relying on the spontaneously formed, often fragile native SEI, researchers build engineered protective layers on the lithium surface before the cell ever operates. These artificial interphases can be designed to conduct lithium ions efficiently while blocking electrons and mechanically suppressing dendrite initiation. Three-dimensional host structures that guide bottom-up lithium deposition represent a related approach, physically confining where lithium is allowed to plate. Recent demonstrations of vertically oriented one-dimensional titania nanoflake architectures show how carefully designed scaffolds can stabilize lithium deposition. Even more striking, recent experiments have achieved high plating currents without dendrite formation at the interface between a lithium anode and a solid electrolyte, challenging long-held assumptions about the limits of fast charging in lithium metal systems.

Another emerging strategy highlighted in the review is the use of multicomponent lithium metal alloys as negative electrodes. Rather than plating pure lithium, these electrodes exploit solid-solution alloys, in which lithium is stored within a matrix of other metals. Recent work on multicomponent solid-solution alloy anodes and on lithium-aluminum alloy electrodes for all-solid-state batteries demonstrates that alloying can smooth the electrochemical potential landscape, moderate volume changes, and suppress dendrite growth. Computational tools are accelerating this search: neuroevolution potentials, a machine learning technique, have recently been used to screen ion transport in lithium-rich alloys and predict which compositions will conduct lithium fastest, compressing years of trial-and-error experimentation into computational screening.

The electrolyte itself is undergoing a quiet revolution. The review discusses electrolyte diluents engineered with large electrostatic potential differences to enable fast charging and slow discharging in lithium metal batteries, as well as anion-enrichment interfacial strategies that enable high-voltage, anode-free configurations. All-solid-state batteries, in which a solid electrolyte replaces the flammable liquid, appear throughout the review as both a solution and a new set of challenges. Solid electrolytes promise improved safety and the possibility of pairing lithium metal anodes with high-voltage cathodes, but they introduce their own interfacial problems, including contact loss and space-charge layers. Adaptive interphase designs that work under pressure-free conditions are among the recent breakthroughs bringing all-solid-state cells closer to commercialization.

What emerges from the Ionics review is a picture of a field converging on a single insight: the interface is everything. Whether the challenge is dendrite formation on lithium metal, oxygen loss from high-voltage cathodes, or transition metal crossover poisoning the anode, the decisive battles are fought in layers of material only nanometers thick. The stabilization strategies surveyed, from artificial interphases and alloy anodes to surface-modified cathodes and rationally designed electrolytes, all aim to make those invisible walls stronger, thinner, and more ion-conductive. If the pace of recent progress continues, the review suggests, the next generation of batteries may deliver the combination of high energy density, long cycling life, and safety that has eluded the industry for decades, and it will do so by mastering chemistry at a scale far smaller than the eye can see.

Subject of Research: Recent advances in electrode materials, interfacial chemistry, and stabilization strategies for lithium-ion and lithium-metal batteries

Article Title: Recent advances in lithium-ion and lithium-metal batteries: Electrode materials, interfacial chemistry, and stabilization strategies for enhanced performance

Article References: Singh, V., Sharma, Y., Kumar, A., Singh, D., Anand, V., & Heera, P. (2026). Recent advances in lithium-ion and lithium-metal batteries: Electrode materials, interfacial chemistry, and stabilization strategies for enhanced performance. Ionics. https://doi.org/10.1007/s11581-026-07547-3

Image Credits: AI Generated

DOI: 10.1007/s11581-026-07547-3

Keywords: lithium-ion batteries, lithium-metal anodes, solid electrolyte interphase, cathode electrolyte interphase, dendrite suppression, artificial interphase engineering, high-voltage cathodes, lithium-rich layered oxides, multicomponent lithium alloys, all-solid-state batteries, electrolyte design, battery degradation

The new work, published in the open-access journal iScience, describes a low-frequency bidirectional pulsed current heating method that operates at frequencies of one hertz or below. The central innovation is adaptivity: rather than applying fixed pulse parameters, the system continuously adjusts the pulse currents according to the battery’s real-time state, ensuring that the anode potential never crosses the lithium-plating threshold throughout the entire heating process. The result, according to the authors, is the fastest possible heating that a given battery chemistry can safely tolerate, a genuine non-destructive approach validated on both lithium iron phosphate and nickel-cobalt-manganese cells.

To understand why this matters, it helps to consider the landscape of existing solutions. External heating methods, which rely on thermal fluids, heat pipes, or phase change materials placed outside the cell, tend to produce uneven temperature distributions and accelerate aging. Internal heating, by contrast, exploits the battery’s own internal resistance to generate Joule heat directly where it is needed, minimizing losses. Alternating current heating and mutual pulse heating, in which battery modules alternately charge and discharge one another, have both been explored extensively. Bidirectional pulsed current strategies, in which positive and negative pulses alternate to keep the state of charge nearly constant, have shown particular promise, but earlier high-frequency implementations demanded costly power electronics that limited real-world adoption.

The research team began by systematically mapping how pulse parameters influence heat generation. Using 2 ampere-hour LFP soft-pack cells placed in a temperature chamber, they varied the ratio of positive to negative pulse amplitudes from 1:1 up to 1:5 while holding the pulse period fixed at six seconds. Across both low and high states of charge, a larger amplitude ratio produced a greater temperature rise, a consequence of the fundamental relationship that heat generation scales with the square of current. They then varied the pulse period from fractions of a second up to 36 seconds and found that longer periods heated the battery more effectively, although the benefit plateaued beyond roughly ten seconds. That plateau defined their operating point: a ten-second pulse period, well within the low-frequency regime that avoids expensive high-frequency excitation hardware.

The truly distinctive element of the study is its use of three-electrode cells to police the lithium-plating boundary in real time. The researchers modified commercial cells by implanting a lithium titanate reference electrode at the center of the electrode stack, allowing them to measure the anode potential directly during operation, something impossible with conventional two-terminal cells. Through hybrid pulse power characterization tests conducted at temperatures from minus ten to twenty-five degrees Celsius and across a range of states of charge, they constructed a three-dimensional map of the maximum charging current the battery could tolerate at each condition without lithium deposition. This boundary current map became the adaptive brain of the heating algorithm, dictating the positive pulse amplitude as temperature and state of charge evolved.

The preheating experiments themselves were conducted under conditions mimicking a real electric vehicle arriving at a charger on a frigid winter day. With cells adjusted to ten percent state of charge and stabilized at minus fifteen degrees Celsius, the bidirectional pulses were applied. For the LFP cell, heating from minus fifteen to ten degrees took 2,068 seconds, an average temperature rise of 0.73 degrees Celsius per minute. Crucially, the measured anode potential, corrected against the reference electrode’s equilibrium potential of 1.565 volts versus lithium, remained above zero volts throughout the entire process, the electrochemical signature that no metallic lithium was plating onto the graphite anode at any moment.

Durability is where many preheating schemes have quietly failed, so the team subjected cells to one hundred repeated heating cycles at fixed state of charge. Because the bidirectional pulses deliver equal charge and discharge capacity within each cycle, the state of charge never drifts, allowing every cycle to begin from identical conditions. The authors note that with typical weekly charging and a three-month winter, one hundred heating cycles correspond to roughly seven years of real-world service, comparable to the lifetime of a traction battery. Capacity measurements taken before the test, after fifty cycles, and after one hundred cycles showed essentially no degradation. Post-mortem analysis sealed the case: transmission electron microscopy revealed an intact, uncracked SEI film, scanning electron microscopy showed an orderly lamellar electrode structure with no dendritic deposits, and energy-dispersive spectroscopy confirmed the elemental composition was unchanged.

To test whether the framework generalizes beyond one chemistry, the researchers repeated the entire procedure with a 1.75 ampere-hour commercial NCM811 pouch cell, independently recalibrating its lithium-plating boundary rather than simply copying LFP parameters. The results were striking. The NCM811 cell heated from minus fifteen to ten degrees in just 865.8 seconds, an average rise of 1.75 degrees Celsius per minute, nearly double the LFP rate. The reason lies in the interplay of resistance and allowable current: the NCM cell’s internal resistance is roughly half that of the LFP cell, but its lithium-plating boundary current is nearly twice as large, so its heating power, proportional to current squared times resistance, is consistently higher across all temperatures. The NCM cell likewise survived one hundred heating cycles with no measurable capacity loss and no lithium plating detected.

The physics underlying these results rewards aggressive current within safe limits. Because heat generation is proportional to the square of current, the alternating charge and discharge pulses of the bidirectional strategy allow larger amplitudes than constant-current heating would, since the alternating polarity alleviates anode polarization and prevents the anode potential from collapsing below zero volts. The authors point out an encouraging trend for the future: as battery technology improves fast-charging capability, the allowable lithium-plating boundary current grows, which in turn permits even larger pulse currents and faster heating, even as falling internal resistance reduces the Joule heating contribution per ampere.

The study is candid about its limitations. The pulse period was optimized at zero degrees and extrapolated to minus fifteen, an assumption supported but not exhaustively characterized across the full temperature range. The long-term stability of the implanted reference electrodes under repeated pulsing remains an open question, the durability test did not combine heating with subsequent fast charging, and the hardware costs and system-level energy efficiency of the required bidirectional power electronics were not quantified. Even so, the demonstration that a low-frequency, adaptive, chemistry-agnostic pulse strategy can warm frozen batteries at up to 1.75 degrees per minute without any detectable damage represents a meaningful step toward electric vehicles and stationary storage that shrug off winter. For the millions of drivers who watch their range evaporate on cold mornings, the prospect of a battery that heats itself safely and swiftly may be the most welcome news of the season.

Subject of Research: Non-destructive low-temperature pulse preheating of lithium-ion batteries

Article Title: Low-frequency non-destructive adaptive fastest pulse preheating for lithium-ion batteries

Article References: Li, C., Mao, S., Gao, M., He, M., Sun, Y., Wang, D., Han, X., Xia, C.-J., Wang, X., Lu, Y., & Gao, W. (2026). Low-frequency non-destructive adaptive fastest pulse preheating for lithium-ion batteries. iScience, 29(10), Article 117715. https://doi.org/10.1016/j.isci.2026.117715

Image Credits: AI Generated

DOI: 10.1016/j.isci.2026.117715

Keywords: lithium-ion batteries, preheating, bidirectional pulse current, lithium plating, cold climate, electric vehicles, anode potential, three-electrode cells, LFP, NCM811, fast charging, battery degradation

The researchers’ approach centers on electrochemical impedance spectroscopy, a technique that probes a battery by applying small alternating-current signals across a wide range of frequencies. Because different physical processes inside the cell respond at different frequencies, the resulting impedance spectrum acts like a fingerprint of battery aging. Slow degradation mechanisms such as the growth of the solid electrolyte interphase, loss of active material and charge transfer slowdowns each leave distinctive traces in the spectrum, which is why the method has long been favored for state of health estimation. The challenge is that these processes overlap in the frequency domain and that real-world measurements are contaminated by noise, both of which undermine straightforward regression models that try to map raw spectra directly onto battery health.

To overcome this entanglement, Song and Zhao turned to the distribution of relaxation times, or DRT, a mathematical deconvolution technique that transforms an impedance spectrum from the frequency domain into the time constant domain. Instead of overlapping frequency responses, the DRT produces a set of peaks, each corresponding to a distinct relaxation process within the cell. As a battery degrades, the positions, heights and shapes of these peaks shift in physically meaningful ways, effectively giving the model a cleaner and more interpretable set of inputs. The DRT framework has a long pedigree in electrochemistry, and recent work has refined its regularization and discretization, but its integration with modern deep learning architectures for battery diagnostics remains an active frontier.

The new model, named DRT-MSFCNN, is a physics-feature-enhanced multi-scale convolutional neural network. Multi-scale convolutional branches allow the network to examine features of the DRT profile at several resolutions simultaneously, capturing both broad peak structures and fine morphological details that may signal specific aging mechanisms. In parallel, feature-wise linear modulation, or FiLM, modules inject statistical physical features derived from the profiles directly into the network’s internal processing. Rather than forcing the network to rediscover electrochemical structure from scratch, the architecture lets measured physics guide the learning process, a strategy that reduces sensitivity to noise and improves generalization to cells the model has never seen.

Attention modules complete the design, helping the network decide which portions of the DRT profile deserve the greatest weight when estimating health. Ablation experiments, in which components of the model are systematically removed, confirmed that multi-scale convolution, FiLM conditioning and attention each contribute complementary improvements, indicating that the architecture’s strength comes from the combination rather than any single trick.

What sets the evaluation apart is its rigor. The experiments used a publicly available impedance spectroscopy dataset and a battery-level, independent protocol called Leave-One-Battery-Out, or LOBO, in which the model is trained on all batteries except one and then tested on the withheld cell. This simulates the hardest and most realistic deployment scenario, estimating health for a battery whose aging history is entirely unknown. Many published methods are validated by splitting data randomly within the same batteries, a practice that inflates apparent accuracy, so the LOBO protocol provides a more honest benchmark of real-world capability.

The results were striking. DRT-MSFCNN achieved a macro-average root mean square error of 0.74 percent and a mean absolute error of 0.60 percent on the test batteries, meaning that on average its health estimates landed within roughly two-thirds of a percentage point of the true value. It outperformed both traditional machine learning baselines, such as methods built on support vector regression and random forests, and competing deep learning models that worked directly on impedance spectra or simpler feature sets. In a field where fractions of a percent translate into significant differences in how aggressively a battery can be used or when it should be retired, this margin matters.

The implications extend across the battery life cycle. For electric vehicles, accurate health tracking supports more reliable range prediction and safer warranty assessment. For second-life applications, where retired electric vehicle batteries are repurposed for stationary storage, fast and trustworthy health estimation determines whether a cell is worth reusing. Because impedance spectroscopy can in principle be performed quickly and even during operation, as recent operando studies have demonstrated, a robust estimation pipeline built on impedance data could eventually run continuously inside battery management systems rather than requiring lengthy laboratory characterization.

The study also illustrates a broader trend in artificial intelligence for the physical sciences: rather than replacing physical understanding with brute-force learning, the most effective models embed physics into the learning pipeline. By first decoupling the impedance spectrum into interpretable relaxation processes and then feeding those physics-rich profiles to a network designed to exploit their structure, the researchers obtained a system that is both more accurate and arguably more trustworthy than purely data-driven alternatives. The work was supported by the National Natural Science Foundation of China, and the underlying dataset is publicly available, allowing other groups to benchmark against it and accelerate progress.

Challenges remain before such methods reach commercial battery management systems, including the computational cost of DRT deconvolution and the need for impedance hardware on board vehicles, but the direction is clear. As battery fleets grow into the billions of cells, the ability to read a battery’s inner condition through a physics-informed neural lens could reshape how society maintains, reuses and ultimately trusts its energy storage infrastructure.

Subject of Research: Physics-enhanced state of health estimation of lithium-ion batteries using distribution of relaxation times and multi-scale deep learning on electrochemical impedance spectroscopy.

Article Title: Lithium-Ion battery SOH estimation via a physics-feature-enhanced method integrating DRT mechanism and multi-scale deep learning

Article References: Song, C., & Zhao, G. (2026). Lithium-Ion battery SOH estimation via a physics-feature-enhanced method integrating DRT mechanism and multi-scale deep learning. Ionics. https://doi.org/10.1007/s11581-026-07519-7

Image Credits: AI Generated

DOI: 10.1007/s11581-026-07519-7

Keywords: lithium-ion battery, state of health, electrochemical impedance spectroscopy, distribution of relaxation times, deep learning, convolutional neural network, FiLM, battery diagnostics, energy storage, machine learning, battery degradation, Guangxi University