The growing demand for sustainable energy solutions highlights the need to extend the use of lithium-ion batteries (LIBs) in first-life applications (e.g., electric vehicles) or repurpose them for second-life uses like energy storage. However, most existing research primarily focuses on first-life applications, with limited attention to the unique challenges of second-life batteries, where accurate estimation of the State of Health (SOH) at low levels (<80%) becomes increasingly difficult due to nonlinear degradation mechanisms. This study addresses this gap by introducing a machine learning approach leveraging Nonlinear Frequency Response Analysis (NFRA) data for SOH estimation in second-life applications down to 60%. NFRA outperformed traditional Electrochemical Impedance Spectroscopy (EIS), achieving >98% prediction accuracy for second-life batteries, even when trained on first-life data alone, and >96.7% using published datasets. NFRA captures nonlinear responses, such as energy losses linked to Li+ transport and solid-electrolyte interface dynamics, which EIS fails to detect. Two predictive models, Long Short-Term Memory (LSTM) networks and Nonlinear Autoregressive with External Input (NARX), were tested, with LSTM reducing root mean square error (RMSE) by up to 30% compared to NARX. NFRA consistently reduced RMSE by over 39% relative to EIS in second-life phases. These findings establish NFRA as a reliable tool for enhancing SOH predictions, enabling safer, more efficient battery repurposing and extended lifetimes.

A sulfonic-functionalized Zr-type metal–organic framework (UiO-66-SO3H) was synthesized, and UiO-66-SO3Li was obtained via subsequent lithiation treatment. The UiO-66-SO3H framework was incorporated into a polymer matrix as a functional filler to obtain high-performance solid composite electrolytes (SCEs). Structural and morphological analyses, as well as various electrochemical tests, were performed. Owing to well-defined porous channels, abundant lithium-ion transport sites, and effective regulation of polymer crystallinity by SO 3 − moieties at 60°C, the optimized SCE-2 (10wt

Anion Exchange Membrane Water Electrolysers (AEM) use a membrane capable of hydroxide ion transport to enable water splitting under alkaline conditions. This requires the use of alkaline electrolyte (0.1 – 1M) in particular at the anode to both hydrate the membrane and also ensure the high activity of the non-precious metal catalysts. However, this opens up several possible design options for the architecture of AEM catalyst layers considering that there will be a mixed ionic conductivity of both ionomer and electrolyte. Furthermore, under oxidising conditions at the anode, nickel oxide containing species are formed and have a low electrical conductivity which may interfere with the utilisation and performance of the catalyst layer. To investigate how the micro-structural design of the catalyst layer affects the performance of the anode, computational modelling was used to reconstruct an equivalent 3D nano-to-micro structure based on parameters extracted from SEM images of spherical nickel nano-powders [1]. Effective electrical and ionic conductivity with ionomer and bubble fraction was computed for the entire three-phase design space. Using an image-based two-phase flow method, capillary pressure saturation and relative permeability curves were extracted from the reconstructed catalyst layers. This process is shown by Figure 1. Using this microstructural information, a volume averaged 1D electrochemical (ionic, electronic and kinetic charge transport) and two-phase flow model (implicit pressure explicit saturation (IMPES)) was created to model the performance of the catalyst layer under different conditions. This approach showed that the anode catalyst layer has different dominating overpotentials at different material and electrolyte compositions. The effective electronic conductivity is limiting in the anode, which shifts the current distribution towards the side closest to the porous transport layer. This type of model was then used to understand the thickness limitations of catalyst coated membrane layers compared to catalyst coated substrates. [1] Rossini, Matteo, et al. "Rational design of membrane electrode assembly for anion exchange membrane water electrolysis systems." Journal of Power Sources 614 (2024): 235062. Figure 1

Ionomers are essential components in proton (PEM) and anion (AEM) exchange membrane electrolysers, facilitating ion transport and influencing catalyst layer microstructure and overpotentials.[1] Ion-conducting functional groups, such as quaternary ammoniums in Sustanion and sulfonates in Nafion, are known to poison catalyst surfaces, reducing the electrochemically active area and overall performance. The mechanisms underlying these interactions remain poorly understood, as studies often use clean metal surfaces that do not accurately represent industrial catalysts.[2] This project investigates the effects of a range of organic functional groups on the catalytic activity of platinum nanoparticles for alkaline water electrolysis. Researchers have attempted to compare the structure-activity relationship of catalysts with various ionomers, however these studies are complicated by variations in ionomer architecture and molecular weight which affect ion conductivity and mass transport and thus the apparent catalyst activity. Here, the interaction of organic functional groups with the catalyst surface has been decoupled from macroscopic physicochemical effects by generating model platinum nanoparticle catalysts immobilized on graphite. The graphite support was then functionalised with a series of ligands tethered to pyrene which can anchor itself to graphite through π-π stacking. Successful immobilization and surface coverage of the pyrene ligands was confirmed and quantified using UV-vis, Raman and 13 C MAS NMR spectroscopy. The resulting pyrene-modified catalysts were evaluated electrochemically to isolate structure–function relationships. Catalysts will be characterised using XPS, in-situ Raman, and MAS NMR spectroscopy to identify key surface descriptors governing catalyst activity and to inform the rational design of improved ionomer–catalyst interfaces for next-generation electrolysers. [1] ACS Electrochem. 2025, 1, 2, 239–248 [2] ACS Catal. 2014, 4, 10, 3772–3778 Figure 1

Protic ionic liquids (PILs), prepared from a stoichiometric acid-base mixture, can function as electrolytes in proton exchange membrane fuel cells (PEMFCs) under low humidity. However, using platinum (Pt) microelectrode studies, we showed that in stoichiometric PILs, oxygen reduction (ORR) and hydrogen oxidation (HOR) kinetics are sluggish due to the lack of free protons for proton-transfer reactions. This can be addressed by adding excess acid to form non-stoichiometric PILs. In this work, we introduce new PILs: butyl pyrrolidinium triflate [Bu-Pyrr][TfO] and ethyl piperidinium triflate [Et-Pip][TfO], featuring thermally stable cations (boiling points: 155 degrees C and 131 degrees C). Adding 200 mM triflic acid enhanced Pt activity, yielding ORR currents of -7.17 and -3.08 mA/cm2 at 0 V vs. RHE, and HOR currents of 2.31 and 1.43 mA/cm2 at 0.75 V vs. RHE for [Bu-Pyrr][TfO] and [Et-Pip][TfO], respectively. ORR was limited by O2 solubility beyond 300 and 400 mM added acid for [Bu-Pyrr] [TfO] and [Et-Pip][TfO], respectively, while HOR currents remained linearly dependent on added acid up to 500 mM. This study concludes that stoichiometric PILs are unsuitable for PEMFCs. Although non-stoichiometric PILs support electrochemical activity, cation adsorption on Pt increases overpotentials and reduces limiting currents; [Bu-Pyrr] raised ORR and HOR overpotentials by -10% and 6%, respectively.

Based on the confinement of the ionic liquid (IL) within the HKUST-1 metal–organic framework (MOF), this work successfully developed a novel gel polymer electrolyte (HKU-5), for room-temperature (RT) quasi-solid lithium metal batteries. X-ray diffraction, scanning electron microscopy, N2 adsorption-desorption isotherm analysis and different electrochemical characterizations were carried out. In Li|Li symmetric cells, HKU-5 enables stable lithium plating/stripping for over 800 h (0.1 mA·cm−2) and effectively suppresses lithium dendrite growth. When assembled Li|LiFePO4 full cells using HKU-5 as the separator, it exhibits excellent cycling stability and interfacial compatibility, retaining a discharge specific capacity of 151.2 mA·h·g−1 for 100 cycles at 0.5 C and RT. This study presents a promising IL@MOF-based strategy for developing next-generation electrolytes toward practical lithium metal batteries.

Electrochemical CO2 reduction to formate (HCOO-) presents a sustainable strategy for carbon valorisation but remains challenging due to low selectivity and competing side reactions. In this work, we report a Cu3Sn gas diffusion electrode (GDE) synthesized via a scalable electrochemical spontaneous deposition (ESD) method, enabling highly selective CO2-to-formate conversion. The optimised CuSn35min electrode, corresponding to the Cu3Sn alloy phase, achieves a Faradaic efficiency of similar to 90% for HCOO- at -0.85 V vs RHE in half-cell testing, while effectively suppressing hydrogen evolution and C2 product formation. Mechanistic and electrokinetic analyses, supported by in situ Raman spectroscopy, reveal a shift from *CO dimerization to an *OCHO-mediated pathway, driven by the synergistic interaction between Cu and Sn. Post-electrolysis structural characterization indicates the formation of surface SnO2 and Cu2O species on the Cu3Sn framework, suggesting a dynamically reconstructed alloy-oxide interface as the catalytically active phase. Full-cell evaluation in a Cu3Sn GDE parallel to Pt mesh configuration under continuous CO2 flow delivers similar to 82% formate Faradaic efficiency and similar to 73% energy efficiency at 180 mA cm- 2, with formate Faradaic efficiency retained above 76% over 60 h of continuous operation. This work demonstrates a cost-effective and scalable route for designing alloy-based GDEs for efficient CO2 electroreduction.

Semiconductor photocatalysis harnesses solar energy to achieve H2 production and the oxidative degradation of antibiotic contaminants, offering viable solutions to energy scarcity and environmental pollution. However, developing visible-light-driven and dual-functional heterojunctions remains a considerable challenge. In this study, we synthesized 2D/0D Co3Se4/ZnSe nanocomposite photocatalysts using an in situ hydrothermal method, evaluating their efficacy for photocatalytic H2 generation and tetracycline (TC) degradation. The composites demonstrated significantly enhanced photocatalytic performance over individual ZnSe and Co3Se4. Notably, the CZ-0.10 composite, incorporating 10 mol% Co3Se4, achieved the highest H2 evolution rate of 1766 μmol·g⁻1·h⁻1 under visible light, a value roughly 17-fold that of ZnSe. Additionally, the TC degradation efficiency of CZ-0.10 reached 84%, significantly outperforming Co3Se4 (6%) and ZnSe (49%). Reusability tests indicated excellent stability for CZ-0.10 across multiple photocatalytic cycles. Furthermore, degradation pathway analysis and mung bean sprout toxicity assays demonstrated that TC was transformed into low-toxicity small molecules, leading to a substantial decrease in solution toxicity. Combined experimental results and density functional theory (DFT) calculations confirmed the formation of a Z-scheme heterojunction at the Co3Se4/ZnSe interface, which effectively facilitated charge carrier separation and enhanced redox capabilities, resulting in improved photocatalytic efficiency. This work presents an effective method for constructing metal selenide-based Z-scheme photocatalysts, demonstrating their promising potential for H2 evolution and antibiotic degradation.

This study reports oxidation-resistant composite anion exchange membranes (AEMs) based on quaternised styrene-ethylene-propylene-styrene (QSEPS) ionomers reinforced with porous PTFE (1 mu m pore size, 70 % porosity) and stabilised using cerium oxide (CeO2) nanoparticles as regenerative radical scavengers. Ex-situ Fenton tests confirmed that CeO2 effectively mitigates oxidative degradation, with the QSEPS/PTFE composite containing 6 wt% CeO2 retaining 87 % of its weight after 30 h at room temperature, compared with the complete degradation of the CeO2-free QSEPS membrane under the same conditions. The improved stability was accompanied by a modest reduction in hydroxide conductivity in deionised (DI) water as QSEPS/CeO2(6 wt %)/PTFE only reached 0.035 S cm-1 at 60 degrees C compared with 0.048 S cm-1 for QSEPS/PTFE. However, this conductivity penalty was largely mitigated using 0.1 M KOH supporting electrolyte, where the area specific resistance (ASR) decreased from 0.339 to 0.148 Omega cm2 for QSEPS/CeO2(6 wt%)/PTFE and from 0.222 to 0.115 Omega cm2 for QSEPS/PTFE. Long-term durability tests under DI water circulation at 2 V and 60 degrees C showed markedly improved performance retention with QSEPS/CeO2(6 wt%)/PTFE exhibiting a 34.9 % current density loss over 900 h compared with 59.1 % for its CeO2-free counterpart, QSEPS/PTFE, corresponding to degradation rates of 147 and 314 mu A cm-2 h-1, respectively. The results demonstrate that CeO2 nanoparticles can effectively suppress hydroxyl-radical-induced degradation, thereby extending AEM lifetime under water electrolyser conditions.

Transition metal perovskites and derived compositions have long been suggested as electrocatalysts for the oxygen electrochemical reactions in alkaline media. This paper presents a study of the alkaline stability of LaBO3 perovskites (B = Co, Ni, Mn, and Fe) by exposing powdered samples to a 2M NaOH solution with pH > 14 for variable periods of time. The elemental analysis of the supernatant test solution reveals the presence of Fe only after 48 h reaction time, whereas Co is apparent within the first 12 h. Ni and Mn are detected after 24 h in quantities like those obtained for Co after 12 h. These data suggest that the initial dissolution is primarily determined by the nature of the transition metal, in excellent agreement with calculated Pourbaix equilibrium diagrams. The analysis of the powders by electron microscopy combined with energy-dispersive spectroscopy and X-ray photoelectron spectroscopy confirms the underlying compositional changes, which are particularly important on the surface of the particles, where the transition metal cations tend to be depleted, thereby forming lanthanum-enriched regions. It is concluded from these tests that the chemical stability increases in the series Co < Ni < Mn < Fe. A degradation of kinetic parameters for oxygen reduction and evolution reactions (ORR and OER), assessed by linear scanning voltammetry and time-dependent galvanostatic measurements, is observed with variable magnitude depending on the transition metal, following the same compositional trend of the chemical stability series. The effect of the transition metal on ORR and OER performance is well explained by the e(g) orbital occupation model. The dissolution of these types of materials in strong alkaline media, potentially aggravated in compositions where lanthanum is partly substituted by alkaline earths, underlies complex compositional changes that may determine the performance and stability of the incorporating device, e.g., fuel cells, metal-air batteries, electrolyzers, or supercapacitors.

This study investigates the oxygen evolution reaction (OER) performance and degradation mechanisms of CoFe2O4 (CFO) and its composite with carbon nitride (CFO/CN). The CFO/CN electrocatalyst exhibits superior initial catalytic activity, achieving a higher current density (42 mA cm-2 at 1.58 V vs. RHE), lower overpotential (359 mV at 10 mA cm-2), and a smaller Tafel slope (45.2 mV/dec) due to enhanced charge transfer, oxygen vacancy stabilization, and pseudocapacitive contributions from CN. However, prolonged cycling leads to structural degradation, oxygen vacancy depletion, and surface amorphization, resulting in performance decline. Despite this, CFO/CN maintains better stability at 10 mA cm-2, sustaining a lower operating voltage than CFO over 24 hours. Post-OER XRD and XPS confirm that CN mitigates severe catalyst degradation, preserving active sites and conductivity. These findings show that while higher activity can be achieved, maintaining long-term stability remains a challenge, highlighting the importance of developing strategies to preserve oxygen vacancies and strengthen the catalyst’s structure for sustained OER performance.

Understanding the alkaline stability of quaternary ammonium (QA) cations tethered to polymer backbones in anion-exchange membranes (AEMs) is crucial to advance the long-term performance of anion-exchange polyelectrolyte-based fuel cells and electrolyzers. A library of model QA cations with N-phenyl and N-benzyl tethers has been synthesized, and comparative alkaline degradation studies revealed that the former are much less stable toward hydroxide attack than their benzylic counterparts. Density functional theory (DFT) studies support the relative stability of the QA cations and demonstrate the critical effect of hydroxide solvation on alkaline stability as well as the degradation pathway. The 3-benzyl-3,6-diazaspiro[5.5]-undecane-6-ium (N-benzyl-ASU, 8) cation was found to be the most stable QA group, with a half-life of 2,595 h at 80 °C and 14,363 h at 60 °C in 3 M NaOD at a hydration number of 4.8, despite its N-phenyl-ASU counterpart (5) having a higher energy lowest unoccupied molecular orbital (LUMO); this suggests that the LUMO energy alone may not be an accurate indicator of alkaline stability. This study highlights the importance of considering the method of tethering the QA group to the polymer backbone and controlling the level of hydroxide hydration when developing QA cations for use in AEM-based devices. The structure-stability correlations arising from this work will inform the design of heteroatom donor-containing QA-based head groups with improved stability profiles.

Correction for ‘Efficient ethane production via SnCl4 Lewis acid-enhanced CO2 electroreduction in a flow cell electrolyser’ by Sankeerthana Bellamkonda et al., J. Mater. Chem. A, 2025, https://doi.org/10.1039/D5TA00176E.

Intensification of water electrolysers is essential to lower the cost of green hydrogen. In this study we design and test a novel alkaline water electrolyser operating without a separator diaphragm (membraneless), using instead a convective flow barrier between electrodes, saving material and energy cost. For the first time, high-speed imaging has confirmed conditions required for a clear electrode gap, with bubble separation up high current density >4 A cm-2. This design was enabled through computational fluid dynamics simulations (OpenFOAM) developed for coupling ionic and kinetic charge transport in the electrolyte and electrodes with the insulating effect of bubble two-phase flow. A pseudo-2D, 3D-printed transparent electrolyser cells validated the fluid flow vectors using particle image velocimetry of bubbles, as well as tracking bubble transport inside the cell. The simulated current density distribution on the electrodes, along with the two-phase flow simulations showed membraneless separation of bubbles was possible and later proven in the experiments using an inactive porous barrier layer with pore diameter of < 51 mu m. Modelling at larger scale detailed that two-phase flow of bubbles effects charge-transfer instead of ohmic transport in this system. Theoretical single-phase flow transport models predict at 50 cm scale it can operate at low current density with low crossover (<% 1 with minimum Reynolds number of 80) but experimental measurements indicated precise control of electrolyte mixing is required to achieve this in practice. Scenarios investigated in this study de-risk the next steps reducing the electrode gap and growth of catalyst, enabling predictions of 1 A cm-2 at 1.8-2.5 V.

There has been significant interest in the development of oxygen evolution reaction (OER) catalysts without the use of precious metals, in order to reduce the cost of electrolyzers for green hydrogen production. Herein, different weights (up to 20%) of CeO2 added to lanthanum manganese perovskite oxide were synthesized by using the sol-gel method, followed by calcination at 900 °C in air. The phase purity of the prepared CeO2-lanthanum manganese perovskite oxide electrocatalysts were investigated by using X-ray diffraction (XRD), followed by Rietveld analysis. The results confirmed that only perovskite and CeO2 phases were present without extra impurity phases. An iodometric titration technique was employed to determine the chemical formula and the average oxidation state of Mn in the prepared electrocatalysts. The optimized electrocatalyst containing ∼10 wt % of CeO2 content with lanthanum manganese perovskite (LCM-0.1) showed improved OER activity, achieving a greater than 22-fold increase in generated current density at 1.9 V versus RHE (reversible hydrogen electrode) in 0.1 M KOH compared to pure LaMnO3. The electrocatalysts were tracked via in-operando Raman spectroscopy and ex-situ XPS spectroscopy, which evidenced reconstruction of the catalyst surface. The seen electrocatalytic activity improvement has been attributed to restructuring of the catalyst surface to form the MnOOH structure. A water electrolyzer was fabricated using the optimized LCM-0.1 electrocatalyst, and the device performance was evaluated with different loadings of the LCM-0.1 catalyst in the anode. The results suggest that the incorporation of cerium oxide in perovskite-based catalysts can be utilized as a method to promote OER electrocatalytic activity.

Composite anion exchange membranes (AEMs) based on poly(terphenylene piperidinium) (PTPiQA) and impregnated with varying loadings of quaternized graphene oxide (QGO) as filler were developed, and their properties as anion exchange membranes for use in water electrolysis (AEMWEs) and fuel cells (AEMFCs) were explored. This study investigates the trade-off between mechanical robustness, ionic conductivity, and alkaline stability in QGO-reinforced twisted polymer backbones. QGO synthesized by functionalization with ethylenediamine (EDA), followed by quaternization with glycidyl trimethylammonium chloride (GTMAC), was used as a filler for PTPiQA, and the properties of the resulting composites PTPiQA-QGO-X investigated as a function of QGO loading for X between 0.1 and 0.7 wt%. Among all compositions, PTPiQA-QGO-0.3% exhibited the highest OH− conductivity of 71.56 mS cm−1 at room temperature, attributed to enhanced ionic connectivity and water uptake. However, this increase in conductivity was accompanied by a slight decrease in ion exchange capacity (IEC) retention (91.8%) during an alkaline stability test in 1 M KOH at 60 °C for 336 h due to localized cation degradation. Mechanical testing revealed that PTPiQA-QGO-0.3% offered optimal dry and wet tensile strength (dry TS of 42.77 MPa and wet TS of 30.20 MPa), whereas higher QGO loadings yielded low mechanical strength. These findings highlight that 0.3 wt% QGO balances ion transport efficiency and mechanical strength, while higher loadings improve alkaline durability, compromising mechanical durability and guiding the rational design of AEMs for AEMWEs and AEMFCs.

As a portable energy storage device, lithium-ion batteries are widely used in various industries. However, with the increasing number of consumers, the number of accidents caused by the instability of aging lithium-ion batteries, such as spontaneous combustion due to thermal runaway, has also increased, raising significant safety concerns among consumers and manufacturers. Battery aging is fundamentally the result of structural changes within the cell, which are presented as phenomena such as swelling of the battery. These internal structural alterations lead to an inhomogeneous current distribution during operation, further exacerbating the instability of the battery. To address the above issue, we propose an innovative magnetic sensor-based battery state-of-health prediction method that differs from traditional state-of-health prediction approaches, such as data-driven models and equivalent circuit-based methods. Instead of relying on indirect data analysis, our approach employs a magnetic field sensing device to capture real-time variations in the battery magnetic field during charging and discharging. A traditional partial least squares regression(PLSR) model and a convolutional neural network (CNN) are trained to capture the evolving characteristics of the battery’s surface magnetic field, which grows increasingly inhomogeneous with aging. By directly utilizing the intrinsic properties of the battery instead of relying on secondary data analysis to infer lifespan, our method achieves a high R2 value of 0.92, providing a non-destructive, more precise, and reliable approach to predicting battery state-of-health.

In the domain of battery energy storage systems for Electric Vehicles (EVs) applications and beyond, the adoption of machine learning techniques has surfaced as a notable strategy for battery modeling. Machine learning models are primarily utilized to forecast the state of batteries, specifically focusing on analyzing the state of charge (SOC). A crucial barrier to the adoption of machine learning algorithms in SOC estimation is their lack of explainability. Current ML SOC estimators researched, are highly focused on model performance while neglecting model explainability. This lack of explainability is crucial for stakeholders as they require trust and assurance in models' predictions, particularly applications like EVs, where erroneous SOC estimation can cause potential safety hazards. This work focuses on the explainability of SOC estimation using the Tsetlin Machine. The Tsetlin machine (TM) is a novel intrinsically explainable machine learning algorithm that offers competitive accuracy across various applications with simple Boolean-based logic computation. This paper aims to bridge the gap of model explainability by also comparing the explainable AI technique SHAP with the novel TM algorithm. The results present the TM as a highly competitive algorithm for accurate SOC estimation while offering a precise quantification of the influence of individual attributes to the final prediction without the need for post-hoc techniques.