The Rice team, drawn from the departments of Chemical and Biomolecular Engineering, Chemistry, and Materials Science and Nanoengineering, along with the Rice Advanced Materials Institute, has developed a relatively soft and flexible porous material that selectively transports lithium ions. The work, published in the peer-reviewed journal Chemical Science, introduces a new design philosophy for porous solid electrolytes, one that treats mechanical adaptability not as an afterthought but as a core performance criterion, sitting alongside ionic conductivity and chemical stability in the designer’s toolkit.

The material at the heart of the study is called ZnBTCA, and it belongs to a family of compounds known as metal-organic frameworks, or MOFs. These are porous crystalline materials built from metal atoms connected by organic molecules into repeating, lattice-like structures. What makes MOFs so compelling for battery applications is their architecture: they contain tiny, ordered channels whose size and chemistry can be tailored to move or hold specific molecules and ions. In principle, a MOF can act as a molecular sieve for charge, allowing lithium ions to stream through its pores while blocking everything else. Many MOFs previously explored as battery electrolytes, however, have been constructed from relatively rigid aromatic linkers — the organic struts that connect the metal nodes — which produce frameworks that are stiff and mechanically unyielding.

ZnBTCA takes a fundamentally different route. Instead of an aromatic linker, it employs a flexible aliphatic linker, an organic molecule with a pliable carbon-chain backbone. That single design choice ripples through the entire material: the framework itself becomes softer and more mechanically adaptable, capable of deforming and conforming rather than resisting. The researchers also emphasized practicality in their selection of building blocks. The material is assembled from relatively abundant, low-cost components, including zinc — one of the most accessible metals in the battery chemist’s repertoire — and an inexpensive aliphatic linker, an important consideration if such materials are ever to move from the laboratory bench toward real-world manufacturing.

First author Zina Deriche, a graduate student in chemical and biomolecular engineering at Rice, framed the shift in thinking that underpins the work. MOF electrolytes, she noted, are often designed primarily around how effectively they transport ions. The Rice study demonstrates that the mechanical properties of the framework can also be an important part of the design equation. It is a deceptively simple reframing, but one with significant consequences: a material that can flex with the battery as it charges and discharges may hold up better over thousands of cycles than one that cracks or delaminates under mechanical stress.

The dual functionality of ZnBTCA is what excites the team most. The material’s flexible molecular building blocks make it relatively soft, while its negatively charged framework structure favors the movement of positively charged lithium ions through its channels. Stavroula Alina Kampouri, a corresponding author of the study and assistant professor across chemical and biomolecular engineering, chemistry, and materials science and nanoengineering, described the convergence of these properties as the most thrilling aspect of the result. The same material, she explained, brings together flexible building blocks that make the framework softer and more adaptable with a negatively charged structure that promotes selective lithium-ion transport. Bringing those properties together, she said, offers a new way to design MOF electrolytes by considering how ions move and how the material behaves mechanically at the same time.

The experimental results lend strong support to the design concept. The researchers first exchanged the sodium ions residing within the framework for lithium ions, a process known as ion exchange that is common in MOF chemistry. After the exchange, they found that roughly 95 percent of the mobile charge-balancing ions in the material were lithium, while the framework itself remained structurally intact — a critical demonstration that the delicate crystalline lattice could survive the swapping of its guest ions without collapsing. Selectivity matters enormously in an electrolyte: a material that carries only lithium ions, rather than a mixed crowd of mobile species, delivers cleaner, more efficient battery operation and helps prevent parasitic reactions that degrade performance over time.

Further electrochemical testing quantified just how selective the material is. In measurements of the current flowing through the material, lithium ions accounted for approximately 79 percent of the total ionic current — a high lithium-ion transference number that indicates the framework’s channels and charged surfaces are indeed preferentially shepherding lithium rather than the counterions left behind by the exchange process. For electrolyte designers, that figure represents a meaningful benchmark, because a high transference number reduces the concentration gradients that build up inside a cell during fast charging and discharging, which in turn improves power delivery and reduces energy losses.

Perhaps the most demanding test the team ran addressed one of the most feared failure modes in next-generation batteries: short circuiting. In lithium-metal batteries, where a strip of pure lithium metal serves as one electrode, repeated charging and discharging can cause lithium to grow unevenly, forming needle-like structures that eventually pierce the electrolyte and create an unwanted electrical pathway — a short circuit that can disable or even ignite a cell. To probe how ZnBTCA would cope with this challenge, the researchers incorporated the material into a solid electrolyte membrane and sandwiched it between two lithium-metal electrodes. The cells then operated stably for nearly 300 hours as the researchers progressively increased the current, with no evidence of short-circuiting under the conditions tested. While laboratory cycling tests are far from the final word on commercial viability, the result suggests that a softer, more conformable electrolyte membrane may help suppress the mechanical pathways through which dendrites typically propagate.

The broader significance of the work lies in the design principle it establishes. To the team’s knowledge, ZnBTCA is the first MOF electrolyte built from an aliphatic linker with a flexible carbon-chain backbone, a distinction that opens an essentially unexplored region of chemical space for electrolyte development. Sibani Lisa Biswal, chair of chemical and biomolecular engineering and the William M. McCardell Professor in Chemical Engineering at Rice, as well as a professor of materials science and nanoengineering, summarized the lesson: researchers can think beyond just the chemistry and pore structure of these materials, because mechanical flexibility can also become a design tool for developing MOF electrolytes for solid-state batteries. The results suggest that mechanical flexibility could take its place alongside chemical composition, pore structure, and framework charge as a tunable parameter in the rational design of battery electrolytes.

The research was supported by a GEM Fellowship as well as the National Science Foundation under grant No. 2404376, and portions of the work used resources of the Shared Equipment Authority at Rice. As solid-state batteries edge closer to commercial reality in electric vehicles, consumer electronics, and grid storage, the challenge is no longer simply finding materials that conduct lithium ions well — it is finding materials that can survive the mechanical realities of a battery’s life. By proving that a MOF can be simultaneously soft, selective, and structurally robust, the Rice team has added a promising new entry to the electrolyte design playbook, and perhaps a hint that the future of the solid-state battery will be built not on rigid crystals, but on frameworks that know how to flex.

Subject of Research: Flexible metal-organic framework electrolytes for selective lithium-ion conduction in solid-state batteries

Article Title: A softer porous material could help improve solid-state batteries

Article References: A softer porous material could help improve solid-state batteries. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: solid-state batteries, metal-organic frameworks, lithium-ion transport, ZnBTCA, electrolytes, mechanical flexibility, Rice University, Chemical Science, aliphatic linker, lithium-metal anodes, ion selectivity, energy storage

The case is striking not only because of the dramatic recovery but also because of the therapeutic history that preceded it. The patient had been treated with a proton pump inhibitor, a class of acid-suppressing drugs used by millions of people worldwide for reflux and ulcer disease, as well as with metformin, the first-line medication for type 2 diabetes. Both drugs are increasingly recognized as causes of magnesium depletion. Proton pump inhibitors interfere with intestinal magnesium absorption, while metformin has been linked to urinary magnesium losses in patients with diabetes. The Belgian authors note that their patient had also experienced unexplained transient neurological deficits in the past, episodes that now appear, in retrospect, to have been heralds of the same underlying electrolyte disturbance.

Magnesium is the fourth most abundant cation in the body and the second most abundant inside cells, and its physiological reach is enormous. It serves as a cofactor for hundreds of enzymatic reactions, stabilizes the electrical gradients of excitable membranes, regulates calcium channels, and modulates the release of neurotransmitters at both central and peripheral synapses. Because roughly two thirds of body magnesium is stored in bone and only a tiny fraction circulates in the plasma, the serum level is an imperfect window into total body stores. This biochemical reality helps explain why hypomagnesemia is frequently overlooked: clinicians may assume that a normal-looking electrolyte panel excludes deficiency, or simply never order the measurement at all during the chaotic early minutes of a suspected stroke.

The neurological consequences of magnesium depletion are well documented in other contexts. Low magnesium lowers the seizure threshold, produces tremor, muscle cramps, and tetany, and can cause characteristic eye-movement abnormalities and ataxia. Severe cases have even produced reversible cerebellar syndrome with brain imaging showing swelling of the cerebellum, a phenomenon described in several case reports over the past decade. What has been far rarer in the literature is the attribution of focal deficits, weakness or speech disturbance confined to one side of the body in a pattern that strongly suggests a localized brain lesion, to magnesium deficiency alone. The Belgian team argues that this rarity is at least partly an artifact of under-testing rather than a true biological rarity.

The mechanisms by which magnesium depletion could produce focal symptoms are biologically plausible. Experimental work on cerebral penetrating arterioles has shown that magnesium promotes vasodilation in these small vessels, so its absence may impair the fine regulation of blood flow in deep brain regions. Magnesium also acts as a natural antagonist of the NMDA receptor, damping excitotoxic cascades that amplify injury during ischemia, and clinical studies in patients with aneurysmal subarachnoid hemorrhage have explored magnesium infusions as a protective agent during episodes of critical perfusion. In a deficient state, the loss of these protective functions could render vulnerable brain circuits transiently dysfunctional, producing deficits that mimic vascular occlusion but reverse when the electrolyte balance is restored.

Stroke mimics are far from a marginal concern in emergency neurology. Large registries of patients referred to stroke services have found that a substantial proportion of code-stroke activations ultimately receive a non-ischemic diagnosis, ranging from seizures and migraines to metabolic derangements and functional disorders. Distinguishing true stroke from mimic is one of the hardest tasks in acute medicine, because the cost of missing a real infarct is measured in dead brain tissue, while the cost of treating a mimic with thrombolytics includes potentially catastrophic bleeding. Imaging tools such as CT perfusion have improved this discrimination, but they depend on recognizing which metabolic conditions can distort the pictures they produce. A severe electrolyte abnormality that resolves with repletion belongs firmly on that list.

The literature review accompanying the Belgian case assembles the scattered evidence that severe hypomagnesemia can indeed present as an acute stroke. Earlier reports have described patients in emergency settings and in rehabilitation units whose hemiparesis and aphasia cleared after magnesium replacement, and a 2025 review in Neurological Sciences explicitly asked whether hypomagnesemia is a rare but overlooked stroke mimic. The pattern across these reports is consistent: the patients often carry risk factors for magnesium wasting, such as proton pump inhibitor use, diuretics, alcohol use, diabetes, or diarrhea, and their deficits improve as the serum level normalizes, a trajectory that would be unusual for a completed infarct. The authors of the new report emphasize that despite this documentation, magnesium levels are rarely measured during the acute phase of a suspected stroke.

Why does this gap persist? Part of the answer lies in the historical organization of emergency testing. A classic study from 1990 found that when serum magnesium was measured routinely rather than only when specifically requested, previously unrecognized hypomagnesemia was detected at a meaningful rate, suggesting that clinicians systematically underestimate its prevalence. Magnesium also interacts closely with potassium homeostasis, so refractory hypokalemia is a well-known clue to coexisting magnesium depletion, yet this teaching point competes with dozens of others in crowded emergency protocols. In the hyperacute stroke pathway, where every additional test is weighed against the clock, magnesium has simply never earned a place on the mandatory panel. The Belgian authors argue that the case they present, together with the accumulating literature, makes a compelling case for changing that practice.

The clinical implications are considerable. For patients on proton pump inhibitors and metformin, the combination seen in this case, the possibility of drug-induced magnesium wasting deserves explicit attention whenever neurological symptoms appear, particularly if the deficits fluctuate or if vascular imaging fails to show an occlusion. For stroke teams, the report suggests that a serum magnesium level, a cheap and rapid assay available in virtually every hospital laboratory, could be added to the initial workup of suspected stroke without meaningfully delaying treatment. Identifying a reversible metabolic mimic before administering thrombolytic drugs would spare patients an unnecessary bleeding risk, and correcting the deficiency could resolve deficits that would otherwise trigger days of admission, monitoring, and secondary stroke workup aimed at a vascular cause that does not exist.

The Belgian case report, approved by the ethics committee of the Hopital Civil Marie Curie in Charleroi and published with the patient’s informed consent, is unlikely to overturn stroke care on its own. But it adds a vivid data point to a quiet shift in neurological thinking: that the electrolyte milieu of the brain is not a background variable but an active determinant of focal function. As the authors and their predecessors in the literature suggest, some of the deficits long assumed to be vascular may instead be the nervous system’s distress signal when a critical mineral runs out. For a condition whose treatment is as simple as magnesium repletion, overlooking it is a luxury that emergency medicine can no longer afford.

Subject of Research: Severe hypomagnesemia as a reversible stroke mimic causing focal neurological deficits

Article Title: Stroke mimic caused by severe hypomagnesemia: a case report and literature review

Article References: Vidal Bankier, D., Renard, F., & Manto, M. (2026). Stroke mimic caused by severe hypomagnesemia: a case report and literature review. Journal of Neurology, 273(10), Article 632. https://doi.org/10.1007/s00415-026-14155-8

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14155-8

Keywords: hypomagnesemia, stroke mimic, magnesium, neurology, proton pump inhibitors, metformin, aphasia, hemiparesis, emergency medicine, electrolytes, case report, Journal of Neurology

The research, published in the Journal of Eating Disorders, analysed the medical records of 62 adults hospitalised for anorexia nervosa at a single French centre between January 2020 and December 2025. To be included, patients had to stay in hospital for more than five days, carry a diagnosis of anorexia nervosa, and have no pre-existing liver disease that could confound the interpretation of their liver enzymes. The overwhelming majority of participants, 95 percent, were female, with a mean age of 28.9 years. Their mean body mass index at admission was 13.47 kilograms per square metre, a figure that places most patients deep within the range of extreme, organ-threatening malnutrition.

The findings on the prevalence of starvation hepatitis were striking in their own right. Half of all patients, 31 of the 62 admitted, already had the condition when they arrived, defined by elevated transaminases, the enzymes alanine aminotransferase and aspartate aminotransferase, in the context of advanced protein-energy malnutrition. This liver injury is thought to arise because prolonged starvation depletes the liver of the substrates it needs to maintain its own cells, triggering autophagy, steatosis-like changes and hepatocyte dysfunction. Clinically, it matters because the liver sits at the centre of the metabolic machinery that must be revived when feeding resumes, and because severe transaminase elevations can themselves signal hepatic compromise that requires careful management.

Refeeding syndrome, the study’s second focus, is defined as a clinically significant decrease in plasma phosphate, potassium and/or magnesium within five days of resuming caloric intake. The mechanism is deceptively simple but physiologically profound. In starvation, insulin secretion falls to minimal levels and the body switches to burning fatty acids and ketone bodies, while intracellular stores of phosphate, potassium and magnesium are progressively depleted even though serum levels may look deceptively normal. When carbohydrates are suddenly reintroduced, insulin surges, driving glucose and electrolytes into cells and causing serum phosphate and other electrolytes to plummet. The consequences can include cardiac arrhythmia, respiratory failure, seizures, and death. Clinical guidelines from bodies such as NICE and the French Haute Autorité de Santé already recommend low initial caloric intake and close electrolyte monitoring in high-risk patients, but identifying who is truly high risk remains an inexact science.

In the new cohort, refeeding syndrome occurred in 19 patients, or 30.6 percent overall, within the first five days of nutritional rehabilitation. But the distribution was far from even. Among patients who had starvation hepatitis at admission, 48.4 percent went on to develop refeeding syndrome, compared with just 12.9 percent of those whose liver enzymes were normal when they arrived. That difference, statistically significant with a p-value of 0.006, hinted at a strong biological relationship between the two conditions, one that clinicians had long suspected but that had never before been formally tested in this population.

To confirm that the association was real and not merely a reflection of patients being sicker in general, the researchers performed a multivariate logistic regression, adjusting for body mass index and initial caloric intake, two factors that are themselves known to influence refeeding risk. The result held. Starvation hepatitis at admission was the only independent predictor of refeeding syndrome, with an adjusted odds ratio of 3.6 and a 95 percent confidence interval of 1.08 to 12.2, and a p-value of 0.038. In other words, even after accounting for how thin the patient was and how aggressively calories were reintroduced, the presence of liver injury on arrival more than tripled the odds of the electrolyte crisis that follows.

The biological plausibility of this link is a subject worth unpacking. The liver is the primary site of glycogen storage, gluconeogenesis, and phosphate-dependent ATP production, all of which are central to the pathophysiology of refeeding. A malnourished liver that is already injured may have diminished capacity to handle the sudden metabolic load of renewed nutrition, including the surge in insulin-driven intracellular uptake of phosphate that characterises refeeding syndrome. Hepatic depletion of adenosine triphosphate and phosphorus stores may leave patients with less reserve to buffer the shift, so serum phosphate falls faster and further. The authors propose that starvation hepatitis may therefore serve not just as a marker of extreme malnutrition but as a window into the functional state of the metabolic machinery that refeeding will stress.

What makes the finding clinically compelling is the simplicity of the marker involved. Transaminase measurements are a routine part of the standard blood panel drawn from virtually every hospitalised patient, requiring no specialised equipment, no imaging, and no additional cost. If a team knows that elevated liver enzymes on the admission screen flag a patient as being at especially high risk of refeeding syndrome, they can act on that information immediately: starting calories at a lower level, advancing the feeding plan more cautiously, and scheduling more frequent checks of phosphate, potassium and magnesium over the critical first five days. Electrolyte abnormalities detected early can often be corrected with supplementation before they progress to arrhythmia or organ failure, which is precisely the kind of preventable deterioration that structured monitoring is designed to avert.

The study does have limitations inherent to its design. It was retrospective, meaning the researchers relied on data collected in the course of routine clinical care rather than on a prospectively standardised protocol, and it was conducted at a single centre with a modest sample size of 62 patients. The confidence interval around the odds ratio, while excluding the null value, is wide, reflecting the small number of events involved. Residual confounding cannot be excluded, and the study population, overwhelmingly female and severely underweight, may not fully represent the broader spectrum of eating disorders seen in clinical practice. Nevertheless, the consistency of the effect with known physiology, and the fact that it persisted after adjustment for the most obvious confounders, lend the finding credibility and make it a candidate for confirmation in larger, prospective cohorts.

For clinicians working in eating disorder units, medical wards and intensive care settings, the study offers a practical rule of thumb grounded in evidence: check the liver panel before the first meal. For researchers, it opens a new line of inquiry into the hepatic mechanisms that connect starvation and refeeding, potentially identifying further biomarkers of metabolic reserve. And for patients and families, it is a reminder that recovery from severe anorexia nervosa is a medically intricate process, one in which the pace of renourishment can matter as much as the nutrition itself. A simple blood test, the study suggests, may help clinicians get that pace right for the patients who need caution most.

Subject of Research: The association between starvation hepatitis and refeeding syndrome in hospitalised patients with severe anorexia nervosa

Article Title: Starvation hepatitis and refeeding syndrome in anorexia nervosa: a retrospective cohort study

Article References: DEGLAIRE, J., ROULAND, A., GERSTER, T., BRINDISI, M.-C., BOUILLET, B., LUU, M., & MOUILLOT, T. (2026). Starvation hepatitis and refeeding syndrome in anorexia nervosa: a retrospective cohort study. Journal of Eating Disorders. https://doi.org/10.1186/s40337-026-01779-4

Image Credits: AI Generated

DOI: 10.1186/s40337-026-01779-4

Keywords: anorexia nervosa, starvation hepatitis, refeeding syndrome, malnutrition, liver enzymes, electrolytes, nutritional rehabilitation, hypophosphatemia, eating disorders, hospitalisation, risk prediction, transaminases

The commercial case for sodium-ion technology is strengthening rapidly. Recent industrial breakthroughs between 2025 and the present have seen announced manufacturing capacities exceed hundreds of gigawatt-hours, with particularly aggressive deployment in China. Commercial electric-vehicle packs now achieve energy densities above 170 watt-hours per kilogram, while the technology’s inherent safety, thermal stability and strong low-temperature performance make it ideal for grid-scale energy storage, where cost per kilowatt-hour matters more than compact size. High-power sodium chemistries are also attracting interest for drones, portable electronics and robotic platforms that demand rapid charging at low cost.

Yet sodium-ion batteries face stubborn material-level bottlenecks rooted in the element’s physics. The sodium ion is significantly larger than its lithium counterpart, with an ionic radius of 1.02 angstroms compared with 0.76 angstroms, leading to sluggish diffusion kinetics and substantial structural strain during repeated charging cycles. Layered oxide cathodes can undergo phase transitions and transition-metal migration that cause voltage fade, while hard carbon anodes suffer from low initial Coulombic efficiency and unstable interface formation. Alloy-type anodes based on tin, antimony and phosphorus promise theoretical capacities exceeding 1000 milliampere-hours per gram, but swell by more than 300 percent during sodiation, pulverizing particles and draining capacity. At high charging rates, non-uniform sodium deposition, dendrite growth and electrolyte decomposition compound these durability problems.

The review identifies four critical knowledge gaps that have kept traditional research methods from closing the performance gap with lithium. First, the solid electrolyte interphase, the protective film that forms on the anode, remains a molecular black box in sodium systems because sodium-based interphase components dissolve far more readily than their lithium analogues, driving continuous electrolyte consumption. Second, predicting high-voltage degradation and phase transitions with density functional theory calculations is computationally prohibitive. Third, high-quality standardized datasets for sodium materials lag dramatically behind the lithium literature. Fourth, many machine learning models function as inscrutable black boxes, offering predictions without the physical insight researchers need to understand structure-property relationships.

To overcome these obstacles, Hu argues, researchers are deploying an increasingly sophisticated AI toolbox. Graph neural networks represent crystal structures directly as atomic graphs, learning relationships between connectivity and properties without manual feature engineering, and have proven especially powerful for predicting formation energies, phase stability and electrode voltages in layered oxides. Deep neural networks serve as fast surrogate models for electrochemical performance metrics, while generative models such as generative adversarial networks, variational autoencoders and diffusion models can propose entirely new crystal structures that exist in no database. Inverse design flips the traditional workflow on its head: instead of making a material and measuring its properties, researchers specify desired performance targets and let the algorithm identify candidates likely to meet them.

Each cathode family demands a different AI strategy. For layered transition-metal oxides, machine learning models trained on density functional theory data distinguish P2 from O3 structures using descriptors like sodium concentration and mixing entropy, and AI-guided screening has identified phase-stable iron-nickel-manganese-titanium high-entropy oxides with improved cycling retention and air stability. Polyanionic cathodes of the sodium superionic conductor type benefit from active-learning frameworks that slash computational costs while exploring thousands of compositions, with generative models coupled to quantum simulations finding candidates with volume changes below 4 percent. Prussian blue analogues, whose modular open frameworks are plagued by vacancy defects and structural water, are ideally suited to variational autoencoders and Gaussian process regression that optimize composition and suppress degradation mechanisms.

On the anode side, machine learning is equally transformative. Ensemble methods such as XGBoost and random forests trained on biomass-derived datasets now predict reversible capacity and initial Coulombic efficiency from synthesis parameters, revealing how carbonization temperature tunes porosity and graphitic ordering in hard carbon. For alloy anodes, crystal-structure search algorithms combined with machine-learning-accelerated annealing identify thermodynamically stable sodium-alloy phases, while high-entropy alloy concepts redistribute sodiation-induced stress. Anode-free designs, in which sodium plates directly onto a bare current collector for maximum energy density, rely on AI to engineer artificial interphases and electrolyte additives that suppress dendrites, with engineered sodium-metal interfaces reported to reach Coulombic efficiencies approaching 99.96 percent under favorable conditions.

Electrolyte development may be where AI delivers its greatest leverage, because the design space of solvent, salt and additive combinations is astronomically large. Descriptor-based screening using HOMO-LUMO energies, dielectric constants and solvation energies allows algorithms to shortlist solvents that optimize sodium-ion coordination, while deep learning correlates additive chemistry with the formation of inorganic-rich protective layers composed of sodium fluoride, sodium oxide and sodium carbonate. Bayesian optimization combined with quantum calculations rapidly identifies additives that outperform conventional fluoroethylene carbonate, and physics-informed machine learning embeds thermodynamic constraints into models to accelerate prediction of ionic conductivity while preserving physical consistency. The same frameworks are being used to design localized high-concentration and fluorine-free electrolytes that suppress dendrite growth through favorable solvation structures.

The review’s most ambitious vision is the fully autonomous, closed-loop discovery platform, in which AI prediction, robotic synthesis and automated electrochemical testing feed results back into ever-improving models without human intervention. Active learning continuously selects the most informative experiments, Bayesian optimization explores sparse chemical spaces efficiently, and emerging agentic materials-science frameworks can plan, execute and refine entire development workflows. But Hu is candid about the remaining hurdles: sodium-specific datasets remain too small and scattered, dynamic interfacial phenomena span too many length and time scales for current models, and explainable AI methods will be essential to convert black-box predictions into actionable scientific insight. If those challenges can be met, self-driving laboratories could compress the journey from computational concept to validated battery material from years to months, accelerating sodium-ion batteries toward their role as the sustainable backbone of grid storage, electric mobility and beyond.

Subject of Research: AI-driven discovery of cathode, anode and electrolyte materials for sodium-ion batteries

Article Title: Review of AI-driven discovery of cathode anode and electrolyte materials for advancing sodium-ion batteries

Article References: Hu, Z. (2026). Review of AI-driven discovery of cathode anode and electrolyte materials for advancing sodium-ion batteries. Discover Industrial Chemistry and Materials, 1(1), Article 21. https://doi.org/10.1007/s44508-026-00022-x

Image Credits: AI Generated

DOI: 10.1007/s44508-026-00022-x

Keywords: sodium-ion batteries, artificial intelligence, machine learning, graph neural networks, generative AI, cathode materials, anode materials, electrolytes, solid electrolyte interphase, inverse design, high-throughput screening, autonomous laboratories

The core insight of the new study is deceptively simple. Different anions interact with surrounding materials in markedly different ways, and one of the most important determinants of that behaviour is how tightly each atom within the anion holds on to its electrons. An anion that readily shares electron density with a neighbouring electrode material will behave very differently in a battery than one that clings stubbornly to its charge. Until now, quantifying this so-called electron-donating tendency for a wide range of anions has required laborious, expensive laboratory experiments, one chemical at a time, which has slowed the pace of materials discovery considerably.

The University of East London team, working alongside their co-authors, attacked the problem with a two-pronged strategy combining experiment and computation. On the experimental side, they turned to X-ray photoelectron spectroscopy, or XPS, a technique that works by firing X-rays at a material and carefully measuring the energies of the electrons that are knocked loose in the process. Because each element and each chemical bonding environment produces a characteristic signature, XPS allows researchers to identify precisely which atoms are present in a sample and, crucially, how those atoms are bonded to their neighbours. By examining the photoelectron spectra of anions interacting with battery-relevant surfaces, the team could read off directly how strongly specific atoms within each anion were gripping their electrons.

That measurement alone would have been a useful incremental advance, but the researchers went a step further. They then built computer models capable of recreating those spectroscopic measurements virtually, calibrating the simulations against the experimental XPS data until the two agreed. Once the computational framework was validated, it could be used to predict the electron-donating behaviour of anions that had never been measured in the laboratory, effectively generating a fast, inexpensive screening tool. In practical terms, this means chemists could evaluate thousands of candidate anions on a computer and shortlist only the most promising ones for physical synthesis and testing, dramatically reducing the time and cost that typically separate a good idea from a working material.

The team describes the resulting quantities as element-specific donor numbers for anions, a metric that captures, atom by atom, the willingness of a negatively charged species to share electron density with its environment. Donor numbers have a long history in chemistry as a way of ranking how strongly solvents and other species bind to central atoms, but extending the concept to individual atoms within anions, and grounding it in hard spectroscopic data rather than indirect inference, is what distinguishes the new work. The approach gives researchers a common, quantitative language for comparing anions that previously had to be assessed through scattered, experiment-specific observations.

The immediate application is battery chemistry, where the choice of anion in an electrolyte influences everything from ionic conductivity to the stability of the electrode-electrolyte interface and, ultimately, how many charge and discharge cycles a cell can survive before degrading. Fluorinated anions, for example, are prized in lithium-ion and next-generation lithium metal batteries precisely because of the way their electron-withdrawing behaviour shapes protective interphases on electrode surfaces. With a validated predictive model in hand, researchers designing such electrolytes can now reason about candidate anions in terms of measured and modelled electron-donation properties rather than relying on trial-and-error, focusing their efforts on the materials that show the most promise.

Dr Richard Matthews, Senior Lecturer of Physical and Computational Chemistry at the University of East London and co-author of the study, emphasised the wider significance of the advance. We now have a much clearer picture of how anions interact with other materials, and that opens up some exciting possibilities, he said. By being able to predict these interactions, we can focus our efforts on the materials that show the most promise for better batteries and beyond. His framing captures the essential economic argument behind the work: in a field where every new candidate material must traditionally pass through synthesis, characterisation and testing, any method that filters out dead ends on a computer can save months of effort and substantial research funding.

The implications, however, stretch well beyond the electrochemistry laboratory. The researchers note that the measurement-and-modelling workflow could eventually be scaled up into a comprehensive database spanning many different elements and bonding environments. Such a dataset would be a valuable raw material for machine learning and artificial intelligence systems, which thrive on large volumes of high-quality, consistently defined measurements. Trainable models fed with element-specific donor numbers could surface unexpected patterns in chemical behaviour, propose novel anion structures for synthesis, and generally compress the discovery cycle for a wide range of chemical processes, not only those relevant to energy storage but also catalysis, corrosion science and medicinal chemistry, where anion interactions play equally decisive roles.

The study also illustrates a broader trend in modern materials research, in which the boundary between measurement and prediction is becoming increasingly porous. Spectroscopy provides the ground truth that anchors computational models to reality, while the models extend the reach of a limited number of experiments to a vast space of unmeasured chemistry. This virtuous loop, in which each new experimental data point improves the accuracy of predictions across an entire chemical family, is quickly becoming the standard playbook for labs hoping to compete in the race toward improved batteries, including the solid-state systems and sodium-based chemistries that many companies and national programmes are pursuing as alternatives to today’s lithium-ion technology.

For now, the University of East London team and their collaborators have laid the foundation, publishing a peer-reviewed method that others can adopt, replicate and extend. The paper, titled Element-Specific Donor Numbers for Anions, appeared in the Journal of the American Chemical Society under the DOI 10.1021/jacs.6c06567. If the framework spreads as widely as its authors hope, the humble anion, long treated as an afterthought in popular discussions of battery technology, may finally receive the systematic, quantitative attention it deserves, and consumers could one day reap the benefits in devices that hold their charge longer, degrade more slowly and cost less to develop.

Subject of Research: Element-specific donor numbers for anions as a predictive tool for battery material discovery

Article Title: Scientists discover method that could lead to longer-lasting batteries

Article References: Scientists discover method that could lead to longer-lasting batteries. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: batteries, anions, X-ray photoelectron spectroscopy, University of East London, electrolytes, computational chemistry, machine learning, energy storage, materials science, donor numbers, chemical discovery, electrochemistry