The integration of renewable energy (RE) sources and the increasing complexity of energy management (EM) systems have resulted in considerable advances in energy usage optimization in smart renewable energy communities (RECs). Existing methods rely on forecasting techniques that necessitate accurate projections of future energy prices, RE generation, user comfort, and behavior. While these strategies are effective in controlled contexts, they have limitations in dynamic, real-time scenarios where system inputs fluctuate unexpectedly. Using prior knowledge or forecasts might result in computational cost in real-time energy optimization. This study proposes a novel approach to real-time adaptive EM in RECs that overcomes the need for prior knowledge and the overhead of forecasting future buildings' EM systems. To tackle the uncertainties in system input dynamics (i.e., RE generation process, battery storage, load arrival processes, demand, and dynamic pricing), this study presents a one-slot-look-ahead queue-based Lyapunov optimization framework. This approach allows for real-time EM systems and reduces user discomfort in smart buildings that are linked and connected to the smart grid. The main goal is to reduce the average running costs (of procurement and operations) by optimizing the real-time scheduling of both electrical/thermal loads and electrical/thermal storage systems, managing their degradation and life cycle, and ensuring indoor user comfort. The optimization challenge is reduced to smaller sub-problems that can be solved one after the other in real time and are asymptotically optimal. They are particularly effective for real-time use in REC settings, especially when the input statistics are either unknown or highly variable. Simulation results under different scenarios and weather conditions, on both a daily and a monthly basis, indicate that the proposed method leads to average reductions in daily and monthly running costs of up to 13.53% and 19.37%, respectively, when benchmarked against other recent research, which reports similar decreases in the same scenarios.

Lithium-ion batteries are recognized as one of the most promising energy storage options for applications such as electrical vehicles and portable electronics. While graphite is the most commonly used anode material, with a theoretical capacity of 372 mAh/g, alternatives are being sought to improve performance. Silicon has emerged as a leading candidate due to its exceptionally high theoretical capacity of 4200 mAh/g ^[1].

However, silicon anodes face significant challenges, particularly severe volume expansion (up to 300%) during lithiation, which causes cracking, pulverization, and eventual degradation of the electrode ^[2]. To address this, the present work focuses on developing silicon–carbon yolk–shell anodes deigned to buffer the volume change and improve cycling stability. In this structure, a carbon shell encases the silicon core, providing both mechanical support and electronic conductivity. The yolk–shells are synthesized through dopamine polymerization at varying durations. To create the void between the carbon and silicon, the bare silicon is oxidized by different methods and selectively etched after the carbonization of the polymer.

Transmission electron microscopy (TEM) confirms successful synthesis of the yolk–shell morphology. X-ray diffraction (XRD) reveals characteristic silicon peaks. The percentage of silicon was evaluated by thermogravimetric analysis (TGA) and it is between 93 and 75%. Electrochemical testing of half-cell coin cells demonstrates that bare silicon shows a less stable profile than the yolk–shell samples, confirming the protective effect of the carbon shell.

References

[1] Yaodong Ma, Pengqian Guo, Mengting Liu, Pu Cheng, Tianyao Zhang, Jiande Liu, Dequan Liu, Deyan He, To achieve controlled specific capacities of silicon-based anodes for high- performance lithium-ion batteries, Journal of Alloys and Compounds, Volume 905, 2022, 164189,ISSN0925 8388

[2] Fan Zhang, Zirui Jia, Chao Wang, Ailing Feng, Kuikui Wang, Tianqi Hou, Jiajia Liu, Yi Zhang, Guanglei Wu, Sandwich-like silicon/Ti3C2Tx MXene composite by electrostatic self-assembly for high performance lithium ion battery, Energy, Volume 195, 2020, 117047, ISSN 0360-5442

Chitosan as a natural polymer has gained considerable attention in recent years due to its organic nature and unique chemical properties. In this study, chitosan extracted from prawn shells was studied as a green corrosion inhibitor in saline solution (3.5% NaCl) under different temperature conditions, 298 K, 303 K and 308K, and different saline solution concentration ranges of pH 6.5, 7.5, and 8.5. The prepared chitosan was characterized by Fourier Transform Infrared Spectroscopy (FTIR) and X-ray diffraction (XRD). Electrochemical studies (potentiodynamic polarization) were performed to study the corrosion rate (CR) and the inhibition efficiency (%IE). The metal surface was studied using Scanning Electron Microscopy (SEM) and X-ray Fluorescence (XRF) in order to determine the adsorption capacity of the inhibitor and confirm that a protective barrier was formed that protected the mild steel from rapid corrosion. Quantum simulations were carried out through density functional theory (DFT) using the Gaussian 09 program. Molecular dynamics was performed in Material studio software (2020) using the Forcite module and applying the COMPASS forcefield simulation, to study the interaction of the inhibitor with the metal surface at different temperatures, and therefore calculate the interaction energy. The study shows that an increase in temperature tends to speed up the corrosion rate, and an increase in the inhibitor amount shows a decline in the overall corrosion rate.

Advanced Reduction Processes (ARPs) have attracted increasing interest from the scientific community for the treatment of recalcitrant compounds from water. Such processes are based on the formation of highly reactive hydrated electrons (e⁻ₐ_q), which are capable to break strong chemical bonds. Per- and polyfluoroalkyl substances (PFASs) is a category of man-made chemicals. In particular, perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA), are highly persistent environmental contaminants, containing carbon–fluorine (C-F) bonds. Due to their widespread use in industrial and consumer products, including, amongst others, paints, textiles, non-stick coatings and firefighting foams, they have been frequently detected in water samples globally (PFOS: 0.002–187 ng L⁻¹; PFOA: 0.001–1371 ng L⁻¹). Their resistance to conventional treatment methods necessitates advanced remediation strategies. Based on the above, the present study evaluates the efficacy of ARPs, specifically the UV-C/sulfite system, for the degradation of PFOS and PFOA in aqueous solutions.

The UV-C/sulfite ARP system generates hydrated electrons (e⁻ₐ_q) and sulfur trioxide anion radicals (SO₃•⁻), which effectively cleave C–F bonds in PFASs. Experiments demonstrated that degradation efficiency was strongly dependent on sulfite concentration, with near-complete removal (>99%) achieved at optimal conditions within 210 minutes. These findings highlight UV/sulfite ARPs as a promising and efficient technology for PFAS remediation in contaminated water.

Origanum dictamnus L. is a medicinal plant known for its rich content in bioactive compounds. The plant cell wall consists of various structural polysaccharides such as cellulose, hemicellulose, and pectin, along with lignin, proteins, and bioactive compounds. These compounds are either trapped within the plant cell wall or free in the cytocol of the plant cell. Enzyme-assisted extraction (EAE) is a green technology that relies on the enzymes ability to selectively degrade the plant cell wall, thereby facilitating the release of bioactive compounds.

In the present study, EAE was applied to extract bioactive compounds from the leaves of Origanum dictamnus L. using the commercial enzyme preparation Cellic® CTec3 HS (Novozymes) and was compared to conventional extraction with ethanol–water mixtures of various concentrations. A Taguchi experimental design was applied to determine the optimal EAE conditions. The variables were enzyme loading (50, 100, and 200 U/mg), solid-to-liquid ratio (1, 4, and 7% w/v), and extraction time (1, 3, and 6 h). The responses were total phenolic content (TPC) and total flavonoid content (TFC). Conventional extraction was performed using ethanol–water mixtures of 0-100% v/v. TPC was determined using the Folin–Ciocalteu method and TFC with the aluminum chloride method. EAE achieved the highest TPC yield of 164.8 ± 5.2 mg GAE/g DW, at 1% w/v, 100 U/mg, and 6h. The maximum TFC yield reached 92.5 ± 5.7 mg CAE/g DW with 7% w/v, 100 U/mg, and 1h. In comparison, conventional extraction with 40% ethanol (v/v) gave the maximum TPC of 133.2 ± 0.5 mg GAE/g DW and TFC of 67.9 ± 0.9 mg CAE/g DW. These results support the potential of EAE as an efficient and sustainable method for the extraction of bioactive compounds from Origanum Dictamnus L.

Natural cooling and heating strategies have gained significant attention due to their ability to dissipate heat without electricity, particularly through radiative cooling (RC), especially in tropical climate areas, is essential in achieving sustainable and energy-efficient architecture. However, normal methods are often limited due to some environmental factors such as sunlight, rainfall and also snowfall constrained their effectiveness. To control these factors behaviour, we study about the thermal performance of two exterior wall materials—weather paint and red gutka bricks (Multani Tiles) during daytime or under high temperature through an integrated experimental and computational approach. The weather paint reflects almost 60-70% of sunlight depends upon their thickness and texture. Also, the red gutka bricks reduce temperature from 5°C - 10°C on similar properties. Experimental investigations include building-controlled sections of walls with both materials and capturing surface temperature variations using an infrared heat gun under different weather conditions. Simultaneously, four parametric building models have been developed in Revit, two for hot and two for cold climates, incorporating advanced passive design strategies such as optimized orientation, thermal mass integration, and adaptive insulation techniques. A comparative energy performance assessment is performed using Revit's simulation tools to assess the impact of wall material selection on thermal performance and indoor comfort levels. By combining experimental results with synthetic energy modelling data, this study offers a significant understanding of material efficacy in passive climate adaptation for sustainable environmentally friendly green practices aligned with climate-responsive design principles.

The intricate interaction between essential process parameters and their impact on gold–cyanide leaching recovery presents an environmental challenge for mining companies. Although polycarbonate materials demonstrate potential as adsorbents for gold-cyanide extraction, improving the adsorption process demands robust predictive models capable of accommodating variability among various operational parameters. This work proposed the development of a machine learning model that forecasts gold–cyanide leaching removal utilizing the architecture of Automated Machine Learning (AutoML) and feature engineering approaches in MATLAB. The experimental feed data comprised flow rate (900-1000 m³/hr), pH (10-11), and polycarbonate concentration (7-10 g/l) as the independent variables mapped against cyanide concentration (200-300 ppm) and gold concentration (0.5-2 ppm) as the dependent variables. The dimensionality reduction technique enabled the elimination of redundant input characteristics that exert negligible influence on gold cyanide leaching removal during the preprocessing phase. Evaluation of the regression model, following training and hyperparameter optimization, identified the decision tree (DT) as the most effective model, achieving a coefficient of determination (R²) of 0.998, a mean squared error (MSE) of 0.025, and a root mean square error (RMSE) of 0.1581. Feature selection employing the F-test identified pH as the predominant variable during model training. The proposed prediction model offers a deployable system designed to optimize mining operations and save operational expenses. This also underlines the reliability and rapid implementation of AutoML in these domains.

This study pioneers a novel Python-based computational framework to optimize the sustainable production of 1,3-butadiene (BD) from bioethanol. The core innovation lies in the synergistic integration of kinetic modeling, thermodynamics, and machine learning (ML) with an optimized K₂O:ZrO₂:ZnO/MgO-SiO₂ catalyst. This catalyst was selected for delivering the highest combined BD and acetaldehyde selectivity (72 mol%) while maintaining reasonable BD yield and productivity (0.12 g_BD·g_cat^-1·h^-1), outperforming Na and Li analogues primarily due to better surface area retention, thereby enhancing BD selectivity and minimizing byproducts. This holistic approach elucidates how temperature (300-400 °C), weight hourly space velocity (WHSV: 0.3-2.5 h^-1), and ethanol feed fraction (0.41-0.85) govern process efficiency. Key findings confirm the reaction's endothermic nature and a strong correlation between thermodynamic driving forces (Gibbs free energy ΔG = -25.3 to -10.5 kJ·mol^-1) and productivity. Optimal conditions (350-375 °C, WHSV 0.93-1.24 h^-1) maximized BD yield at 25.3%, significantly reducing byproducts compared to non-optimal settings where acetaldehyde selectivity reached 57.3%. Among ML models, Random Forest excelled (R² = 0.91 for ethanol conversion prediction), attributed to its superior handling of complex, nonlinear variable interactions, with temperature and feedstock composition identified as dominant factors. The methodology provides a practical computational toolkit for catalyst and reactor design, explicitly addressing the critical trade-off between productivity (reaching 0.49 g_BD·g_cat^-1·h^-1 at high WHSV) and yield. By enabling data-driven optimization of feed control and catalyst efficiency, this work offers a powerful strategy for advancing renewable chemical manufacturing and decarbonizing the production of critical precursors such as BD for synthetic rubber and plastics.

Machine learning models offer dynamic systems that can surpass the limits of quantitative models for estimating solar photovoltaic (PV) output, particularly under changing seasonal weather conditions. This study evaluates data-driven models, such as linear regression (LR), decision trees (DTs), and artificial neural networks (ANNs), to determine solar energy production. Historical metadata from May 2024 to December 2025 utilized as the input data pertained to climatic factors such as sun irradiance, temperature, humidity, wind speed, cloud cover, and precipitation. The ML systems were trained and validated in MATLAB, with integrated hyperparameter adjustment to enhance the model performance. Quantitative indicators comprising the mean squared error (MAE), root mean square error (RMSE), and coefficient of determination (R²) were applied to validate and compare the generalizability of the resultant models. The ANN model surpassed LR and the DT in identifying nonlinear relationships, with an average performance of an MSE = 0.7512, an RMSE = 0.8667, and R² = 0.9725. This was achieved on an optimized ANN architecture using the Bayesian Regularization training methodology, one hidden layer with eight neurons, a sigmoid activation function, and a learning rate of 0.001. These results demonstrate that the ANN provides a more efficient strategy for reliable PV output estimation, vital to improving energy planning and facilitating the larger integration of solar technology into weather-dependent power networks.

Introduction: The substantial global generation of eggshell waste, estimated at millions of metric tons annually, presents a significant environmental challenge, contributing to landfill burden and associated pollution. This necessitates innovative strategies for waste valorization, transforming this abundant byproduct into high-value biomaterials.

Methods: The valorization process involves initial cleaning and drying of eggshells to remove organic contaminants and moisture. Subsequently, eggshells undergo mechanical processing (crushing, grinding, sieving) to achieve desired particle sizes, and thermal treatments like calcination to convert calcium carbonate into calcium oxide or hydroxyapatite precursors. For real-time process control and quality assurance, advanced non-destructive analytical techniques such as Near-Infrared (NIR) spectroscopy and Hyperspectral Imaging (HSI) are integrated. These optical methods, coupled with machine learning algorithms, enable rapid and accurate assessment of critical parameters including chemical composition, moisture content, and purity, ensuring optimal material characteristics for specific applications.

Results: The application of these controlled processes, rigorously monitored by NIR/HSI and machine learning, yields high-purity eggshell-derived biomaterials. These materials find diverse applications, including advanced bone regeneration scaffolds and dental implants in biomedical engineering, natural exfoliants and skin mineralizers in cosmetics , calcium fortification in food and nutraceuticals, and adsorbents for environmental remediation.

Conclusions: Eggshell valorization represents a compelling example of circular economy principles in action, transforming a low-value waste stream into high-value products. The integration of advanced analytical techniques like NIR spectroscopy and HSI, powered by machine learning, is pivotal for ensuring the quality, consistency, and safety of these biomaterials, thereby facilitating their broader commercialization and contributing to a more sustainable future.