Browse Items (7684 total)
Sort by:
-
Acid functionalized Arachis hypogaea skin based carbon nanosphere as efficacious material for enhanced energy storage
The present introduces a single step approach for enhancing supercapacitor performance by utilizing acid-functionalized porous carbon derived from the inner skin of Arachis hypogaea as a sustainable biomass precursor. Through pyrolysis at 800 C in a nitrogen atmosphere, the resulting porous carbon material demonstrates unique structural and electrochemical behavior as confirmed by FTIR, XRD, Raman spectroscopy, FE-SEM, HR-TEM,EDS,BET analyses. The acid functionalized variant (FAH8) significantly outperformed the non-functionalized carbon (AH8), showing a fourfold increase in specific capacitance. Electrochemical evaluations revealed that FAH8 achieved a high specific capacitance of 273 Fg?1 at 0.25 Ag?1 in 3 M KOH, with an energy density of 22.5 Wh kg?1 and a power density of 125 W kg?1 in a three-electrode setup. The symmetrical CR2032 device of FAH8 exhibited a maximum capacitance of 98 Fg?1 and displayed excellent stability, with 98.5 % efficiency and 97.4 % capacitance retention after 7500 cycles. Notably, the device also delivered a high energy density of 23.17 Wh kg?1 and power density of 325.0 W kg?1. The enhanced performance attributed by the simple acid functionalization highlights the potential of this material in energy storage. Thus, the study not only emphasizes the effective use of low-cost biomass precursors but also provides a straightforward functionalization strategy to boost energy storage capabilities, paving the way for sustainable high-performance supercapacitors. 2025 Elsevier Ltd -
Synergistic g-c3n4/v2o5/pani composite for electrochemical energy storage
This work illustrates the synthesis of a ternary hybrid composite (g-C3N4/V2O5/PANI) from graphitic carbon nitride, vanadium pentoxide, and Polyaniline via hydrothermal method followed by in-situ polymerization. Morphological analysis confirms the integration of vanadium pentoxide (V2O5) and polyaniline (PANI) within the interlayer spaces of graphitic materials. The resultant hybrid composite structure facilitates rapid diffusion and ion movement at the electrode-electrolyte interface. Additionally, incorporating V2O5 within a polymer matrix alongside graphitic material generates diverse electrical profiles, enhancing electrochemical performance. The electrochemical characteristics of g-C3N4/V2O5/PANI composites were examined by Cyclic voltammetry (CV), Galvanostatic charge-discharge (GCD), and Electrochemical impedance spectroscopy (EIS). The GCD analysis shows that the g-C3N4/V2O5/PANI composite exhibits a specific capacitance of 880 Fg?1 at a current density of 1 Ag?1, retaining 78 % of its initial capacitance after executing 2000 cycles at 3 Ag?1. Furthermore, a symmetric supercapacitor was constructed using g-C3N4/V2O5/PANI composite material as the electrode, showing a capacitance of 246 Fg?1 when measured at an input current density of 1 Ag?1. This study demonstrates g-C3N4/V2O5/PANI is a potential electrode material for supercapacitor application. 2024 -
A robust explainable machine learning pipeline for transformer health index prediction addressing data pathologies and redundancy
Power transformers are critical infrastructure assets where unexpected failures incur severe technical and economic penalties. This study proposes a robust, explainable machine-learning (ML) pipeline for predicting the transformer Health Index (HI) using routinely collected dissolved gas analysis (DGA) and dielectric measurements. To ensure model reliability, the pipeline specifically addresses data pathologiesnamely extreme skewness and heavy tailsusing YeoJohnson transformations, while mitigating multicollinearity through hierarchical correlation clustering (|r| ? 0.85) followed by a Variance Inflation Factor (VIF) screening (VIF ? 5). Four high-performance ensemblesRandom Forest, XGBoost, LightGBM, and CatBoostwere optimized via randomized cross-validation. Experimental results on a dataset of 470 records demonstrate consistent generalization across all models (RMSE ? 0.022), with Random Forest providing superior accuracy (MAPE ? 1.24%). A Taylor diagram confirmed consistent generalization (correlation ? 0.730.78 and matched variance), while residual analysis showed minimal bias. SHAP explanations indicated that dibenzyl disulfide (DBDS) and interfacial tension (Interfacial V) were the most influential positive drivers of HI; water content tended to depress HI; and several gases (e.g., methane, hydrogen, acetylene, CO) contributed positively at higher concentrations. The proposed workflow was robust to skew/heavy tails and multicollinearity, required no feature scaling, and produced transparent, practitioner-ready insights that support condition-based maintenance at fleet scale. 2026 Elsevier B.V. -
AI-driven load forecasting and energy management in smart grids using hybrid deep models
Modern power systems are becoming more complex, and integrating renewable energy sources (RES) calls for sophisticated solutions for accurate load forecasting and efficient energy management. To improve forecast accuracy and operational efficiency in smart grids, the research suggests a hybrid deep learning (DL) structure that blends convolutional neural networks (CNN) with long short-term memory (LSTM) systems. The LSTM element records sequential connections within historical energy usage, while the CNN element extracts geographical features from environmental variables such as temperature, humidity, and solar radiation. A comprehensive preprocessing pipeline comprising data cleaning, normalization, and feature selection ensures high-quality inputs for model training. The proposed LSTM-bCNN model is evaluated using a publicly available dataset, and its performance is benchmarked against traditional and contemporary models including ARIMA, SVM, RF, and standalone LSTM. According to findings from experiments, the mixture model obtains the highest R-squared (R) value, the lowest Mean Absolute Error (MAE), and the Root Mean Squared Error (RMSE), confirming its robustness in capturing complex patterns in energy consumption. This research highlights the possible of hybrid DL models in enabling intelligent, adaptive, and resilient energy management systems (EMS) within next-generation smart grids. 2026 Elsevier B.V. -
High-speed portrait video segmentation using the hybrid combination of deep-learning models and boundary movement adjustment
As global warming intensifies, the development of energy-efficient Artificial Intelligence (AI) technologies has become crucial. Additionally, the growing demand for on-device AI in smartphones, extended reality devices, and autonomous vehicles necessitates AI systems that can function effectively on low-performance hardware. To address these needs, this study proposes hybrid methods in the field of Portrait Video Segmentation (PVS). Our proposed hybrid models leverage Deep-learning based Segmentation Models (DSMs) and a novel Boundary Movement Adjustment (BMA) process to achieve speed and accuracy balance. The Hybrid Serial Model (HSM) not only accelerates PVS but also improves energy efficiency while maintaining a similar level of accuracy. On the other hand, the Hybrid Parallel Model (HPM) enables high-performance PVS even on low-performance devices, ensuring no video frames are lost during high-speed segmentation processing. Tests conducted on Jetson Nano, Raspberry Pi, and a desktop PC demonstrate the effectiveness of these models, showing improvements in PVS speed while maintaining accuracy close to that of traditional DSMs. HSM increased PVS speed from 15.2 Frames Per Second (FPS) to 25.1 FPS on a desktop PC with a 0.5 % accuracy loss, and from 6.3 FPS to 16.5 FPS on a Jetson Nano with a 1 % loss. HPM reached 30 FPS on a desktop PC with a 0.05 % loss, and 29.7 FPS on a Jetson Nano with a 1 % loss. On the Raspberry Pi, the HPM method improved speed from 2.9 FPS to 29.8 FPS, demonstrating its adaptability for low-performance devices. 2025 Elsevier Ltd -
Putting sustainable human resource management and workplace eudaimonic well-being into cross-cultural context
This study examines how sustainable human resource management (HRM) impacts employee work engagement and eudaimonic well-being across cultural contexts that differ on individualism-collectivism dimension. Theoretically, the study draws from Self-Determination Theory (SDT; Ryan & Deci, 2017) and the model of culture fit (Aycan et al., 1999). Using data from 14,502 employees nested in 54 countries working in a variety of positions across different sectors, we found support for our hypothesized modelthat is, sustainable HRM was positively related to employee eudaimonic well-being via enhanced work engagement. The study found that one moderating effectthe relationship between work engagement and eudaimonic well-beingwas stronger in countries that are more individualistic rather than collectivistic. The findings provide support for the universality of the SDT-based approach to understanding employee experiences based on sustainable HRM and cultural variations that inform work-related eudaimonic well-being. Our study advances existing cross-cultural research on sustainable HRM and employee well-being. 2026 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies. -
Stepwise hydrothermally synthesized gold nanoparticles supported copper metal-organic frameworks as an impedimetric immunosensor for the ultrasensitive detection of pancreatic cancer
Carbohydrate antigen (CA199) is a frequently used biomarker for detecting and prognosis of pancreatic cancer. Early detection of pancreatic cancer remains a challenge in routine clinical analyses, including imaging techniques such as magnetic resonance imaging, ultrasonography, and computed tomography. There is an urgent urge to develop robust sensors like electrochemical immunosensors that provide low-priced and sensitive biomarker detection. A potential electrochemical immunosensor comprising Au nanoparticles supported on Cu MOF, HKUST-1 (Au@HKUST-1) on screen-printed carbon electrodes (SPCE) was developed via a one-pot stepwise hydrothermal method for the ultralow level detection of CA 199 in human serum using electrochemical impedance spectroscopy (EIS). CA 199 could be detected in a wide range of concentrations, including 0.01 U mL-1 to 35 U mL-1. Au@HKUST-1/SPCE sensor displayed ultralow levels of detection of CA 199 with a limit of detection (LOD) in PBS as 0.17 U mL-1 and serum as 0.08 U mL-1. The limit of quantification of the immunosensor is 0.53 U mL-1 and 0.26 U mL-1 in PBS and serum, respectively. This work also emphasizes the ultrasensitive detection of pancreatic cancer using CA 199 biomarker in serum samples, and achieved a reliable analytical platform for the detection of CA 199 and other biomarkers. 2025 Elsevier Ltd. -
Ruthenium phosphate-embedded poly-(3,4 diaminobenzoic acid)-based electrode for enhanced sensing of 2,4-dichlorophenol in water samples
2,4-Dichlorophenol (2,4-DCP) is found to have a prevalent application in synthesizing many industrial materials, meanwhile leading to toxicological effects on human health and aquatic life. This work demonstrates the construction of a highly responsive electrochemical sensing platform for 2,4-DCP, based on ruthenium phosphate electrodeposited over poly-(3,4 diaminobenzoic acid)-loaded carbon fiber paper (Ru-Pi/PDABA/CFP). Surface modification of the conducting polymer with Ru-Pi improves electrocatalytic performance by enhancing available electrocatalytic sites and rapid charge transmission channels. The developed electrode was characterized using XRD, XPS, and SEM studies to substantiate the formation of Ru-Pi/PDABA/CFP hybrid material, and electrochemical studies further evidence the improved electrochemical performance upon electrode modification. Cyclic voltammetric studies showcased 2-fold enhanced catalytic activity of Ru-Pi/PDABA/CFP compared to the bare CFP. Differential pulse voltammetric outcome corroborated outstanding electroanalytical metrics towards 2,4-DCP, unveiling an appreciably minimal limit of detection (LOD) of 1.47 nM and a low quantification boundary (LOQ) of 4.37 nM in a wide concentration-response linearity of 5 450 nM. The interferences from foreign substances produced only negligible signal modulations (<4.6 %) on the current amplitude of 2,4-DCP, validating the sensor's excellent selectivity towards the target analyte. Further, the application of Ru-Pi/PDABA/CFP was extended for the 2,4-DCP assay in actual tap and lake water samples. 2025 -
Harnessing immobilized copper salophen complex for the electrochemical synthesis of phenazine
This study introduces a novel approach utilizing a modified electrode, denoted as Cu-PDABA-CFP, as a pivotal catalyst in the electrochemical synthesis of phenazine. Through bulk electrolysis of o-phenylene diamine in an acetonitrile medium, facilitated by lithium perchlorate as the supporting electrolyte, the electrode serves as a fundamental component in this synthetic endeavor. The modification process entails the immobilization of a copper-salophen complex, synthesized in accordance with established literature protocols, onto the electrode surface. Surface characterization of the modified electrode was meticulously conducted to get critical insights into the structural morphology and topographical features of the electrode surface, pivotal for understanding its electrochemical behavior. Concurrently, electrochemical characterization studies were undertaken to evaluate the inherent activity of the modified electrode. To elucidate the intricate electrochemical reaction mechanism underlying the synthetic transformation, an exhaustive screening of reaction conditions was meticulously undertaken. The findings presented herein contribute not only to advancing our fundamental understanding of electrochemical processes but also hold promise for the development of novel electrochemical methodologies with broader applicability in synthetic chemistry and materials science. 2025 Elsevier Ltd -
Novel biogenic CNS@AgNPs hybrid nanostructures for electrochemical detection of sucralose: Experimental and in silico strategies
Carbon, the most abundant and versatile element, has played a significant role in scientific innovations, forming the backbone of material science and nanotechnology. This study presents the first reported simultaneous photogenic synthesis of carbon nanospheres (CNS) integrated with silver nanoparticles (CNS@AgNPs) using Coriander sativum seed extract for sucralose detection. The CNS@AgNPs formation, mediated by oleic acid from the extract, was confirmed with GCMS analysis. The morphology of the CNS@AgNPs was characterized using SEM, TEM, XPS, XRD, Raman, BET, Diffuse Reflectance Spectroscopy (DRS), and Thermogravimetric analysis (TGA). The fabricated GE/Nafion/CNS@AgNPs electrode demonstrated an intense oxidation peak current at +0.7V, with Differential Pulse Voltammetry (DPV) showing a linear response from 2.0 to 14?M, with a LOD and LOQ of 0.2?M and 0.62?M (R2=0.998), respectively. The Density Functional Theory (DFT) studies revealed key mechanistic insights, including the methanol cleavage energy (~3.135נ103eV) and HOMO-LUMO differences between neutral sucralose and its cationic form. Monte Carlo (MC) simulations confirmed favourable adsorption energy (?52.739kcal/mol) with specific binding interactions (3.3578.653 influencing electron transfer pathways. This eco-friendly approach presents the potential of sustainable materials for developing efficient electrochemical sensors for detecting artificial sweeteners in real samples. 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. -
Biosynthesized carbon quantum dots/g-C3N4/Co3O4 composites for effective methylene blue dye degradation and DFT study
In this study, we aimed to develop a new, efficient photocatalyst, graphitic carbon nitride/carbon quantum dots/cobalt oxide (g-C3N4/CQDs/Co3O4 (CCC)), via a hydrothermal route. The composite was synthesized through a simple hydrothermal method, with the Co3O4 nanoparticles (NPs) systematically varied to 3, 5, and 10 %. The resulting samples are comprehensively characterized using various techniques, including X-ray diffraction (XRD), Raman spectroscopy, Fourier-transform infrared (FT-IR) spectroscopy, X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), transmission electron microscopy (TEM), BrunauerEmmettTeller (BET) surface area analysis, vibrating sample magnetometry (VSM), thermogravimetric analysis (TGA), and ultraviolet-visible (UVVis) spectroscopy. Photocatalytic activity was evaluated using methylene blue (MB) dye under UV light. Among the prepared samples, the 3 % Co3O4 NPs loaded CCC catalyst has shown superior photocatalytic efficiency of 94.5 % within 120 min, which is higher than that of the 5 and 10 % Co3O4 NPs loaded CCC composite and better than that of the pristine materials. The results are obtained for optimized conditions at a concentration of 5 ppm, 0.05 g and pH 10. The 3 % CCC composite has exhibited excellent reusability and stability upto five cycles. Furthermore, Density Functional Theory (DFT) was used to understand the crystal structure and electronic properties of the prepared composite. The results have demonstrate that the novel CCC composite is a promising catalyst for the degradation of MB dye in aqueous solutions and environmental remediation. 2025 Elsevier B.V. -
Exploring thermal and entropic behaviors in nanofluid stagnation point flow with nonlinear dynamics
This study investigates the optimization of heat and mass transfer in nanofluid stagnation point flow by analyzing entropy generation and its underlying physical mechanisms. Nanofluid technology, widely applied in thermal energy storage, and heat exchangers represents a significant advancement in modern thermal systems. While nanofluids enhance heat transfer rates, optimizing thermal conductivity through nanoparticle dispersion remains a key challenge. This work also incorporates the effects of a nonlinear chemical reaction to evaluate its impact on coupled heat and mass transport. The governing nonlinear partial differential equations, including momentum, energy, and concentration expressions, are reduced to a system of coupled ordinary differential equations using local similarity transformations. These equations are solved numerically using a Runge-Kutta scheme in MATLAB. The results, presented through tables and graphs, demonstrate how velocity, temperature, and concentration profiles vary with key physical parameters. Entropy generation is shown to increase with higher porosity, while reductions in slip and Williamson fluid parameters decrease it. Furthermore, the skin friction coefficient increases by approximately 7 % when the magnetic parameter M increases from 0 to 0.5, whereas the Nusselt number decreases by nearly 28.6 % as M increases from 0 to 1. Additionally, the local Sherwood number decreases by approximately 16.7 % when the permeability parameter Kp increases from 0 to 0.3. These findings provide practical insights into enhancing nanofluid based heat and mass transfer systems for engineering applications. 2025 The Authors. -
Theoretical insights into the role of nanoclusters in anticancer drug delivery systems
The present investigation focuses on the adsorption of Deguelin and Raloxifene, two potential anticancer drugs, on beryllium oxide (Be12O12) and magnesium oxide (Mg12O12) nanoclusters using advanced computational approaches. The pristine nanoclusters are first optimized for their structural and electronic properties, revealing unique geometric and electronic characteristics that influence their interaction with the drug molecules. Various initial configurations are explored to identify the most stable adsorption sites, with adsorption energies (Eads) indicating significant interactions between the drug molecules and the nanoclusters. The adsorption on Mg12O12 nanoclusters displays chemisorption behavior with higher Eads values, whereas Be12O12 nanoclusters exhibit physisorption, implying a weaker yet stable interaction. Furthermore, electronic structure analysis, including density of states (DOS) and HOMO-LUMO gap evaluations, indicated that drug adsorption alters the electronic properties of the nanoclusters, particularly for Be12O12, where a notable reduction in the band gap is observed. These findings suggest that Be12O12 and Mg12O12 nanoclusters hold promise as effective carriers for Deguelin and Raloxifene, offering insights into their potential application in targeted drug delivery systems. 2025 Elsevier B.V. -
Electrospun PAN/TEMPO nanofiber electrode: Dual charge storage mechanism for supercapacitor applications
An advanced electrode material for asymmetric supercapacitors was created by electrospinning a polyacrylonitrile (PAN)/2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) composite. Strong interfacial interactions between the PAN chains and TEMPO nitroxide radicals were confirmed by Fourier transform infrared spectroscopy, which showed partial suppression of the PAN nitrile (C tbnd N) vibrations. X-ray diffraction revealed increased short-range molecular ordering of PAN caused by TEMPO via dipoledipole interactions, without changing the semicrystalline structure. In morphological studies, the incorporation of TEMPO showed decrease in the fiber diameter and enhanced surface roughness as compared to PAN fibers, resulting in the interconnected nanofibrous network with enhanced electrolyte accessibility. A mesoporous architecture with a quantifiable surface area and pore volume was characterized by BET analysis. A higher D-to-G band intensity ratio was found by Raman spectroscopy, which quantitatively indicated the formation of defects and improved electrochemically active sites in the PAN/TEMPO composite. The PAN/TEMPO electrode facilitates a dual charge storage mechanism that combines electrical double-layer capacitance from the nanofibrous PAN matrix and pseudocapacitance from reversible TEMPO redox activity due to these synergistic structural and chemical modifications. The assembled asymmetric supercapacitor exhibits a stable energy density of 7.71 Wh kg?1 and a power density of 365.33 W kg?1, and the composite electrode provides improved capacitive performance in acidic electrolyte. Additionally, Raman and EIS studies were performed for the PAN/TEMPO electrode after performing 5000 galvanostatic charge/discharge cycles to check the stability of the material. Overall, this work provides a novel approach to design supercapacitor electrodes with a structuredefectredox synergy in TEMPO-modified electrospun PAN nanofibers. 2026 The Authors -
Early detection of mental health disorders using machine learning models: An analysis based on behavioral and voice data
Mental illnesses are to be detected promptly and correctly to intervene effectively and in time. In this paper, a multi-stage NeuroVibeNet model of early mental disorders detection based on multimodal behavioral and voice data is proposed. It starts with the preprocessing of data that is high-quality and consistent, such as mean imputation, min-max normalization, outlier detection, noise reduction, and short-time energy extraction. The majority of the advanced methods employed in extracting temporal, spectral, and complex features include multiscale entropy, soft dynamic time warping, spectral contrast analysis, formant frequency analysis, and a one-dimensional convolutional neural network autoencoder. The feature selection is done via a sparse autoencoder that is used to maximize relevance and minimize redundancy. The chosen features are fed into the NeuroVibeNet architecture, where TabNet is used to process behavioral data, and Capsule Networks are used to process voice data to allow learning representations with attention and hierarchy. Lastly, a voting-based ensemble classifier uses the two modalities to combine the predictions to make strong classification decisions. The structure is coded in Python and tested on three benchmark datasets with the accuracy of 0.9839, 0.9856, and 0.9855, which is better than the current approaches. Copyright 2026. Published by Elsevier Ltd. -
A reliable inter-domain routing framework for autonomous systems using hybrid Blockchain
Inter-domain networks face several routing challenges, such as security, scalability, and reliability concerns in existing BGP-based systems. These challenges are exacerbated by the increasing number of interconnected networks and the lack of a standardized approach for routing data between them. Hybrid Blockchain-based framework has proposed for inter-domain routing in autonomous systems in this research. The framework combines the use of traditional routing protocols with the distributed ledger technology of Blockchain. It leverages the salient features of both to create a more secure and efficient routing framework. The Blockchain component provides a decentralized and tamper-proof ledger for storing routing information, while the traditional routing protocols handle the actual exchange of data between autonomous systems. The framework is designed to enhance the security of inter-domain routing by incorporating the use of digital signatures and information sharing among participating autonomous systems. Each participating system maintains a copy of the distributed ledger and can verify the authenticity of routing information using digital signatures. It ensures that only legitimate and authorized data is transmitted between autonomous systems, mitigating the risk of malicious attacks or illegitimate routing. The proposed framework obtained 87.73 % Route calculation Speed, 90.41 % Route filtering, 93.77 % Fault tolerance, 94.10 % Load balancing, 95.54 % Hop count, 95.13 % bandwidth consumption, 93.94 % Security Management and 96.29 % Convergence time. The framework employs a consensus mechanism for updating and validating the routing information, ensuring consistency and accuracy in the routing decisions. It also reduces the reliance on a single central authority and distributes the decision-making process among participating systems. 2024 Elsevier Ltd -
Stroke disease classification from computed tomography images using Inception Harmonic LeNet and wavelet- symmetrically weighted local gradient pattern features
Stroke is a leading cause of mortality, making prompt and precise diagnosis essential for effective treatment. Computed Tomography (CT) screening is crucial in identifying stroke types, particularly ischemic and hemorrhagic strokes. Existing automated methods lack the accuracy and consistency required for reliable stroke diagnosis. Therefore, a novel Inception Harmonic LeNet (InHLeNet) approach is devised for stroke disease classification. Initially, CT scans are collected and subjected to preprocessing, which is done using guided filtering and an improved Non-Subsampled Shearlet Transform (NSST) threshold. The filtered images are then segmented using the Dimension fusion U-Net (D-UNet). Subsequently, augmentation is performed by local augmentation and self-augmentation, where local augmentation introduces localized variations within each CT image, and self-augmentation generates feature-guided transformations of lesion regions. Further, the wavelet transforms with Symmetrically Weighted Local Gradient Pattern (Wavelet-SWLGP) features are extracted. Lastly, stroke disease is classified using InHLeNet, which merges InceptionV3Net, LeNet, and Harmonic analysis. The performance of InHLeNet is assessed using several evaluation metrics, including accuracy, True Positive Rate (TPR), True Negative Rate (TNR), and Matthews Correlation Coefficient (MCC). The results attained using the InHLeNet model is accuracy of 96.888%, TNR of 96.381%, MCC of 96.777%, and TPR of 97.988%, with image size, highlighting its effectiveness. 2026 Elsevier Ltd -
NorBlueNet: Hyperspectral imaging-based hybrid CNN-transformer model for non-destructive SSC analysis in Norwegian wild blueberries
Soluble solids content (SSC) is a vital parameter in blueberries, reflecting the concentration of dissolved sugars (primarily fructose and glucose) and directly influencing the fruit's sweetness, flavour, and ripeness. As part of this study, Norwegian wild blueberries were carefully hand-picked from a forest in Norway and subsequently imaged using a hyperspectral camera to capture their detailed spectral characteristics. This study introduces NorBlueNet, a hybrid CNN-transformer architecture, for accurately predicting SSC in wild blueberries through hyperspectral imaging and deep learning. This hybrid architecture combines CNN layers for local feature extraction and spatial hierarchy representation, followed by transformer layers that capture global relationships and long-range dependencies. The hybrid approach combines the computational advantages of CNNs with the advanced attention mechanisms of transformers, achieving enhanced accuracy while maintaining computational efficiency. A comprehensive evaluation is conducted by comparing the proposed model with two additional deep learning models on the custom dataset. The results indicate that the NorBlueNet achieves the highest prediction accuracy, with an R2 = 0.98, RMSE = 0.0136, and RPD = 9.3759 thereby demonstrating its superior performance. To foster community engagement, collaboration and facilitate re-implementation of our work, we have made our code available at:https://github.com/NorBlueNet. 2025 -
Dual purpose behavior of Ni-PTC MOF for high performance supercapacitor and water splitting applications
Metal-organic frameworks (MOFs) have elicited significant interest as next-generation materials for storing and converting energy, owing to their structural versatility and tunable physicochemical properties. In the present work, a nickel-based MOF, referred to as Ni-PTC, was synthesized via a straightforward method and explored for its dual functionality as a supercapacitor electrode and an electrocatalyst for overall water splitting. Structural and morphological analyses confirmed the materials high surface area, hierarchical porosity, and excellent crystallinity. As a supercapacitor electrode, Ni-PTC delivered a high specific capacitance of 953.86 F g?1 at 1 A g?1 and demonstrated superior cycling durability, retaining 92 % of its initial capacitance after 5000 cycles. Its electrocatalytic performance was assessed for both hydrogen (HER) and oxygen evolution reactions (OER), exhibiting overpotentials of 241 mV and 400 mV, respectively, at a current density of 10 mA cm?2. The catalyst also showed excellent operational stability, underscoring its potential in energy-related applications. 2026 Elsevier B.V. -
Comprehensive and comparative analysis of sorbitan ester (Span) niosomes as emerging vesicular drug delivery platform: Fabrication, characterization, release dynamics, biocompatibility profiling and toxicological implications
To overcome the limitations and related adverse side effects of conventional drug delivery, niosomes, aka non-ionic surfactant vesicles, have emerged as an effective vesicular drug delivery system (VDDS) for the past few years. This study represents a comparative analysis of physico-chemical characteristics, in vitro and in vivo biocompatibility of synthesized sorbitan ester (Span) niosomal vesicles. In brief, Span 20, Span 40, Span 60 and Span 80 surfactants, along with equimolar concentration of cholesterol, were used to synthesize blank, Biochanin A (model hydrophobic drug) and Crocin (model hydrophobic drug) loaded niosomes. Characterization techniques unveiled that all niosomes were polydispersed sphericles with hydrodynamic diameter of 300 nm to 650 nm and PDI < 0.550. Fourier transform infrared spectroscopy (FTIR) and UVvisible spectroscopy (UV-Vis) analysis of drug loaded niosomes showed respective characteristic peaks of Biochanin A and Crocin, indicating effective drug encapsulation with EE% varying from 58.975 to 90.050. Among all formulations, Span 60 and Span 40 niosomes sketched satisfactory yield, drug encapsulation (EE%), loading efficacy (LD%), drug release and stability. Results obtained from in vitro biocompatibility study depicted that all niosomes had marked drug delivery efficacy with minimum cytotoxicity (<25 %) and haemolysis (<27 %) at 500 g/mL concentration. After a consecutive 14-day exposure to blank niosomes (100 mg/kg body weight) by intraparetonial injection, treated swiss albino mice exhibited little to no significant changes in body weight, organ weight, haematological and biochemical parameters, with normal hepato-renal histological characteristics. Thus, the study portrayed a mechanistic and comparative evaluation in vitro and in vivo applicability of niosomes with detailed sub-acute toxicological profiling. 2025 Elsevier B.V.