SCIENCE CHINA Materials•2026•DOI: 10.1007/s40843-026-4425-4
Flexible and weavable alternating-current electroluminescent (ACEL) fiber devices are pivotal for wearable displays and human-computer interfaces, yet their intrinsic lack of color tunability restricts high-density information interaction. This study presents a dynamically color-tunable electroluminescent fiber device with a coaxial winding structure that integrates multiple fiber electrodes emitting the three primary colors. Through simple voltage driving, the device achieves a color gamut covering 131.07% of the sRGB standard, enabling arbitrary full-color tunability, including standard white light with CIE coordinates of (0.31, 0.33). The emission peak is continuously tunable over a 161.7 nm range, a 4-fold enhancement compared to previously reported ACEL fibers. The coaxial winding architecture is compatible with large-scale fabrication, yielding hundred-meter-scale fiber devices with a luminance variation of only 2.76%. The electroluminescent performance remains stable under stringent industrial standards: 10,000 friction cycles, 20 accelerated washing cycles, and 10-day storage at 105 °C and −20 °C. Integration into a smart textile watchband demonstrates real-time heart rate visualization via progress color changes and gesture-controlled color switching, validating its potential as an effective human-computer interface.
SCIENCE CHINA Materials•2026•DOI: 10.1007/s40843-025-3687-6
P2-Na0.67Ni0.33Mn0.67O2 (NNMO) is a promising cathode for sodium-ion batteries (SIBs) due to its high energy density and operating voltage. However, severe P2-O2 phase transition at high cut-off voltage causes large volume variation, structural degradation, and rapid capacity decay. Ion doping has been explored to suppress this transition, but achieving both high capacity and stability remains challenging. Here, we demonstrate that precise composition regulation enables both. The designed P2-Na0.67Ni0.28Mg0.03Fe0.04Mn0.55Ti0.1O2 retains high electrochemical active element content while effectively suppressing phase transition, leading to outstanding structural stability and fast charge transfer kinetics. This cathode delivers a high specific capacity of 143.5 mAh g−1 at 0.1 C and maintains stable cycling over 1000 cycles. Our work provides a new strategy for rationally designing high-capacity, stable cathode materials for SIBs.
SCIENCE CHINA Materials•2026•DOI: 10.1007/s40843-025-3869-7
Iontronic capacitive pressure sensors (ICPSs) are pivotal for wearable technology, yet their performance is constrained by an inherent trade-off between sensitivity and detection range. Here, we introduce a micro-electric double layer (micro-EDL) engineering strategy to overcome this limitation. This is realized through a nanocomposite dielectric where multi-walled carbon nanotubes (MWCNTs) form a percolated network, generating a dense array of pressure-responsive nano-capacitors. Synergistically integrating a hierarchical MoS2/NiCo-LDH electrode provides abundant pseudocapacitive interfaces. The resulting sensor exhibits an ultrahigh sensitivity of 67,095 kPa−1 at 1 kHz, a broad detection range up to 1.3 MPa, rapid response and recovery times of 4 ms and 5 ms, respectively, and outstanding durability exceeding 18,000 cycles. Practical validation demonstrates 100% classification accuracy in recognizing complex gestures and gait patterns, underscoring its real-world applicability. These findings establish micro-EDL engineering as a promising route for advancing next-generation iontronic devices, offering insights into their electrochemical mechanisms.
SCIENCE CHINA Materials•2026•DOI: 10.1007/s40843-025-3925-5
Capacitive pressure sensors have garnered significant attention in electronic skin, human-machine interaction, health monitoring, and medical devices due to their remarkable properties like highly sensitive pressure perception, good repeatability, and rapid response capabilities. However, manufacturing capacitive pressure sensors that simultaneously achieve a broad linear detection range and high sensitivity remains a significant challenge. Herein, a novel hierarchically interlocked capacitive pressure sensor (HI-CPS) was designed by integrating a stretchable polyethylene glycol (PEG)-based nanofilm dielectric layer with hierarchically interlocked microstructures, demonstrating excellent linearity and high sensitivity over a wide sensing range. HI-CPS based on a one-layer nanofilm exhibits ultrahigh sensitivity (9.40 kPa−1) and an ultralow detection limit (0.1 Pa). When the dielectric layer comprises two layers of stacked nanofilms, the sensor not only maintains high sensitivity (3.17 kPa−1) but also achieves excellent linearity (R2 = 0.999) over a broad working range (<5 kPa), along with remarkable stability even after 10,000 cycles. Benefitting from the outstanding comprehensive performance, HI-CPS has been proven to be successfully implemented in monitoring various human biological signals, sign language recognition, and basketball shooting gesture correction. This strategy of assembling the tailored nanofilm with structural engineering has significant potential application in building high-performance pressure detection and recognition devices.
Journal of Environmental Engineering Technology•2026•DOI: 10.13205/j.hjgc.202607021
The continuous expansion of urban sewage treatment capacity has led to a sustained increase in sludge generation, making efficient treatment, disposal, and resource recovery critical in environmental engineering. Machine learning (ML) offers substantial potential for prediction and optimization in sludge treatment by extracting non-linear features from complex operational data. This review systematically examines the application of ML across typical sludge treatment processes, including dewatering, resource recovery (anaerobic digestion), and terminal disposal (incineration and landfill). The general modeling workflow is summarized across three dimensions: dataset preparation, algorithm selection, and model evaluation. A comparative analysis evaluates the applicability and limitations of support vector machines (SVM), random forests (RF), artificial neural networks (ANN), and other deep learning models. SVMs demonstrate greater stability with small-to-medium sample sizes and high-dimensional data, while RFs exhibit strong generalization and provide variable importance insights. ANNs and deep learning models excel in large-scale data and time-series or image tasks but require high data quality. Key findings from the literature include ANN achieving R²=0.99 and RMSE=0.02 in dewatering prediction, and R²=0.86 with NRMSE=0.31 in anaerobic digestion, while gradient boosting reached R²=0.90 and RMSE=0.33. Future directions emphasize multi-source data fusion, model interpretability (e.g., SHAP), and coupling ML with mechanistic models to enhance predictive accuracy and generalization, supporting intelligent and refined sludge treatment management.
Journal of Environmental Engineering Technology•2026•DOI: 10.13205/j.hjgc.202607022
Anaerobic sludge digestion is the core process for achieving energy recovery and sludge reduction in wastewater treatment plants. However, its complex biological reaction mechanisms and multivariable coupling characteristics pose persistent challenges to process optimization and stable control. Traditional mechanistic models, while theoretically clear, suffer from parameter calibration difficulties and insufficient adaptability under dynamic and nonlinear conditions. Machine learning (ML) has gained attention for its powerful data modeling capabilities. This review systematically examines ML applications in sludge anaerobic digestion, focusing on biogas production prediction, process monitoring and early warning, and process parameter optimization. For gas production, hybrid models and deep learning achieve high-precision methane yield predictions. Soft-sensing models using easy-to-measure parameters enable real-time estimation of volatile fatty acids and total ammonia nitrogen. At the optimization level, coupling surrogate models with optimization algorithms provides dynamic regulation strategies for co-digestion ratios and pretreatment conditions. Interpretable methods address the 'black-box' issue, enhancing engineering acceptability. Deep integration of these methods with dynamic optimization supports an intelligent decision-making framework. However, translation from laboratory to engineering faces constraints including data quality, model generalization, and implementation. This paper provides an analytical framework combining predictive capability with engineering reliability for sludge treatment optimization.
SCIENCE CHINA Materials•2026•DOI: 10.1007/s40843-025-4077-0
Flexible memristors are pivotal for advancing neuromorphic computing in wearable electronics, yet the intrinsic brittleness of inorganic oxides poses a critical challenge. Here, we employ an entropy-engineering strategy to control the amorphization of oxide compositions, yielding a precisely controlled crystalline/amorphous microstructure in a BaTi0.25Sn0.25Hf0.25Zr0.25O3 thin film. This film withstands bending angles up to 180°, enabling an Au/BaTi0.25Sn0.25Hf0.25Zr0.25O3/ITO/Mica device that functions as a memristor. Entropy engineering increases oxygen vacancy concentration, imparting stable resistive switching behavior under both flat and bent conditions. The device exhibits exceptional endurance and reproducibility over multiple bending cycles, demonstrating a significant strategy for advancing flexible memristor technologies and holding promise for next-generation high-performance flexible electronics.