Polyaniline (PANI) is regarded as a prospective sensing material for room-temperature NH3 detection owing to its high sensitivity and selectivity. In this study, WO3 nanoflowers were prepared via a simple chemical bath deposition method, and PANI/WO3 heterojunctions were subsequently fabricated through an in situ polymerization process. To evaluate the contribution of the heterojunction to gas-sensing performance, four sensors with different composite ratios were constructed. Among them, the optimized PW1-2 sensor exhibits a response of 3.8–100 ppm NH3 at room temperature, which is 1.6 times higher than that of pure PANI. Notably, it achieves an ultrafast response time of only 26 s, representing an 8.8-fold improvement over pure PANI (230 s), along with near-complete desorption capability. Furthermore, PW1-2 demonstrates excellent humidity resistance, with stable responses ranging from 3.7 to 3.97 under 40–80

Metal-organic-framework (MOF)-derived oxide is a promising gas-sensing material candidate because of its high surface area, porosity and structural diversity. Utilizing Aerosol-assisted chemical vapor deposition (AACVD), the fabrication of MOF-derived films performs characteristic morphology and controlled size, which greatly improve the gas-sensing performance. Herein, the heterojunction interface between SnO2 nanorods and MOF-derived Co3O4 nanoparticles was prepared by two step AACVD. Firstly, SnO2 nanorods were directly grown on flat electrode. Subsequently, size-controlled MOF-derived Co3O4 nanoparticles were coated on SnO2 nanorods. The hydrogen sensors were developed based on the above sensing materials with different deposition time (0, 30, 40 and 50 min) to demonstrate gas-sensing performance. The Co3O4/SnO2-40 sensor exhibits a response value of 305 % to 100 ppm of H2 at 310 degrees C, which is 4.4 times of the SnO2 sensor. Moreover, the Co3O4/SnO2-40 sensor retains a measurable response of 4 % even at an extremely low hydrogen concentration of 200 ppb. The superior sensing performance of this sensor can be attributed to the large surface area and the p-n heterojunction interface. Consequently, the construction of heterojunction via AACVD offers an approach for fabricating high performance hydrogen sensors.

Carbon monoxide (CO) is a colorless, odorless, and deadly gas that poses significant hazards even at low concentrations. Therefore, developing highly sensitive, low-power CO sensors are of critical importance. In this work, SnO2 nanosheets were directly fabricated on alumina planar by aerosol-assisted chemical vapor deposition (AACVD). Subsequently, MOF-derived cobalt oxide was grown on the SnO2 nanosheets through a chemical bath deposition method (CBD) to construct a Co3O4/SnO2 heterojunction. This heterojunction was utilized as a sensing layer for developing a CO gas sensor. The Co3O4/SnO2-2 sensor exhibited not only a higher response of 115% toward 100 ppm CO but also a lower detection limit of 100 ppb at 150 degrees C, comparing with the SnO2 sensor (38%, 200 ppb, 310 degrees C). The improved sensing capabilities of the Co3O4/SnO2-2 sensor arise from the combined influence of a pronounced heterojunction effect, abundant oxygen vacancies and a unique mesoporous structure. Combining AACVD and CBD methods, the fabrication of Co3O4/SnO2 heterojunction not only enables CO sensor with Low-temperature operating and low-concentration detection, but also offers a feasible approach for developing high-performance gas sensors.

Hydrogen is a clean energy carrier essential for carbon neutrality, but its invisible and odorless nature poses significant safety risks, particularly during low-concentration leaks. Although metal oxide semiconductor (MOS) sensors offer fast response and high sensitivity, their ability to detect ppb-level hydrogen remains limited. In this work, we present a high-performance hydrogen gas sensor based on nickel-doped tin dioxide (Ni-SnO2) nanorods, directly grown on planar electrodes via aerosol-assisted chemical vapor deposition (AACVD). By optimizing the Ni doping ratio and nanorod morphology, the 3 wt% Ni-SnO2 sensor achieves a low detection limit of 100 ppb for H2, demonstrating promising potential for low-concentration hydrogen detection. Moreover, the sensor exhibits outstanding selectivity, with a response to 100 ppm H2 nearly six times higher than that to the next most responsive interfering gas (NH3). Comprehensive XPS and Raman analyses reveal that Ni doping introduces abundant oxygen vacancies and lattice defects, which are the key origins of the enhanced sensing performance. Notably, the 3 wt% Ni-SnO2 sensor strikes an optimal balance between lattice defects and structural stability, delivering both high sensitivity and good moisture resistance with minimal baseline drift over weeks of operation. This work establishes a facile and scalable AACVD strategy for engineering defect-rich SnO2 nanostructures, enabling sub-ppm hydrogen detection with high selectivity and long-term stability—addressing a critical gap in practical hydrogen safety monitoring.

Covalent organic frameworks (COFs) are promising crystalline porous polymers for photocatalysis, yet their strong excitonic effects and rapid carrier recombination limit efficiency. However, strong excitonic effects and rapid electron-hole recombination remain key challenges. Herein, we employ density functional theory (DFT) and time-dependent density functional theory (TD-DFT) to systematically investigate the structure-activity relationships of three D-pi-A-type COFs (COF-alkene, TapbBtt-COF, and TtaTpa-COF) for photocatalytic overall water splitting. Benchmarking identifies the M06L functional, SMD solvent model, and 6-311+G(2d,p) basis set as optimal. Our results reveal that molecular planarity, D-pi-A configuration, and charge separation collectively govern performance. TtaTpa-COF exhibits the narrowest Eex (2.47 eV), longest absorption wavelength (502.15 nm), and lowest hole-electron overlap (0.51), enabling efficient carrier separation. For the hydrogen evolution reaction (HER), TtaTpa-COF shows the most favorable *H adsorption free energy (0.04 eV) and lowest LUMO level (-2.8 eV), yielding the highest activity. Notably, the D-pi-A system governs active-site selectivity: COF-alkene favors the alkene-linked carbon, whereas the other two favor imine nitrogen. For the oxygen evolution reaction (OER), all follow the adsorbate evolution mechanism with *OOH formation as the rate-determining step. TtaTpa-COF exhibits the lowest limiting potential (4.33 eV), indicating superior water oxidation kinetics. This work establishes a clear structure-activity relationship linking D-pi-A architecture to photocatalytic performance, providing a rational design framework for high-activity COF-based photocatalysts.

Polyimides (PIs) are indispensable high-performance polymers for advanced flexible electronics, owing to their remarkable thermal stability, mechanical strength, and optical properties. To accelerate the design of colorless, transparent, and thermally stable PIs, this work introduces a machine learning (ML)-driven inverse design framework. By integrating molecular fingerprinting with ML models, we first developed accurate quantitative structure-property relationship (QSAR) models. Gradient Boosting (GB) model with RDKit fingerprints was identified as the optimal predictor (R-2 > 0.94) for optical absorption (lambda(max)), charge transfer (CTe), and thermal decomposition temperature (T-d5). SHAP interpretability analysis further elucidated the roles of key substructures such as flexible ether linkers, fluorinated groups and conjugated structure. Guided by polycondensation chemistry, PIs disassembled into dianhydride (A) and diamine (B) building units and combinatorially expanded through flexible ether linkages (O) to create a virtual library consisting of 1530 AB-type and 175,664 ABOC-ype hypothetical monomers. High-throughput screening under stringent criteria (lambda(max) < 400 nm, CTe < 0.2 e, T-d5 > 530 degrees C) using a multi-task learning sparse encoder (MTL-SE) model successfully identified top candidates, including the outstanding h-BPDA-162 and 287 high-performance h-ABOC-type PI monomers. This work provides a scalable computational framework and specific structural guidelines as leads for the rational design and synthesis of advanced PI materials.

Recent advances have highlighted olefin-linked triazine-based covalent organic frameworks (NKCOFs) as promising photocatalysts for hydrogen evolution, due to their intrinsic donor-acceptor architectures providing an ideal platform for manipulating charge separation and transfer behavior. However, a systematic molecular-level understanding of how donor-acceptor (D-A) architectures and it-conjugation length cooperatively modulate the photoelectric properties in COFs remains elusive, particularly regarding structure-property relationships. In this work, we investigated the photoelectric properties of three NKCOFs with varying it-conjugated lengths using density functional theory (DFT) and time-dependent density functional theory (TD-DFT) calculations. Our findings revealed that the D-it-A configuration in olefin-linked NKCOFs facilitates charge transfer within it-*it* transitions, driven by their symmetric electronic structures. The introduction of cyano groups and it-conjugated units (such as olefins and benzene rings) enhances the planarity of NKCOFs, promoting charge separation and transfer during photocatalysis. The extension of the it-conjugated building blocks caused a redshift in absorption spectra and narrows the HOMO-LUMO gap, optimizing light-harvesting in the visible range. Crucially, we identified an optimal it-conjugated length in NKCOFs that influences the charge separation efficiency. Through hole-electron analysis, we confirmed that charge delocalization decreases with extended it-conjugated chains, while local excitation characteristics (14 %) become more pronounced. Several quantitative parameters further demonstrate NKCOF-113's superior charge transfer capability (Sr = 0.77 a.u., D = 0.28 & Aring;, CT = 0.14 e, t = -0.60 & Aring;) among the three NKCOFs. This work established a quantitative relationship between it-conjugated length, polarity, and planarity, offering a strategy approach for design COFs toward solar energy conversion (e.g., H2 evolution) and environmental remediation.

Carbon monoxide is a colorless and odorless, highly toxic gas, posing a significant risk to both human health and industrial safety. Zinc oxide has garnered significant interest for its outstanding chemical stability and costeffectiveness in the development of CO sensors. However, sensors based on a single ZnO has the shortcomings of low sensitivity and high detection limit, rendering them challenging to meet the precise detection requirements for CO. To address this issue, ZnO nanoflower arrays were prepared by chemical bath deposition (CBD), followed by modification of varying ratios of Co3O4 and NiO onto the ZnO nanoflower arrays. The response value of the sensor based on Co3O4-NiO/ZnO-2 to 100 ppm CO was 104% at 310 degrees C. Remarkably, the sensor achieved an exceptionally low detection limit of 1 ppb. Further, the crystal structure and morphology characteristics of Co3O4-NiO/ZnO were systematically analyzed by XRD, SEM, TEM, XPS, Raman, EDS, EPR, and BET. The results show that the improvement in gas-sensing performance is mainly ascribed to the ZnO nano-flower array and the heterojunctions formed among Co3O4, NiO and ZnO. In this study, Co3O4 and NiO decorated ZnO nanoflower arrays were prepared by a two-step CBD, offering a novel approach for advancing highperformance CO sensors.

Metal oxide semiconductor (MOS) sensor, one of the most widely used gas sensors, faces significant challenges in achieving lower detection limit. Addressing this issue, the utilization and strategic design of bimetallic catalysts present a promising avenue. In this paper, Pd-Pt was in-situ grown on ZnO nanoflowers directly prepared on planar substrate and a sensor for detecting acetone was developed. The test results reveal that the Pd-Pt/ZnO nanoflowers sensor exhibits a remarkable 107.6 response to 100 ppm acetone at 350 degrees C, while presenting a rapid response (15s). Moreover, this sensor demonstrates excellent capability in detecting acetone at concentrations as low as 1 ppb, showcasing impressive gas-sensing performance. The response of the sensor to 100 ppm acetone is twice that of ethanol at the same concentration. The exceptional performance of the sensor results from its distinctive nanoflower structure and the catalytic effect of the Pd-Pt, which offers an extensive surface area and enhances the rate of redox reaction for gas molecules. This in-situ grown method has great potential for improving the gas-sensing performance in gas sensor.

Covalent organic frameworks (COFs) have emerged as unique catalysts for photocatalysis; however, the relationship between their building block units and optoelectronic properties remains elusive. Herein, we explored the influence of building blocks on the optoelectronic properties of benzotrithiophene-based COFs (BTT-COFs) using density functional theory (DFT) and time-dependent DFT (TD-DFT) calculations. The calculation results suggested that three critical factors—the conjugated structure, planarity, and the introduction of nitrogen heteroatoms—significantly influenced charge separation and transfer within BTT-COFs. Structure–property relationships were established through several critical quantitative parameters, such as Sr, t, and CT. Among seven BTT-COFs, BTT-Tpa (Tpa: 4,4′,4″-triaminotriphenylamine) exhibited the most efficient charge separation and the highest charge transfer capability due to the electronegativity of triphenylamine, the delocalization of its lone pair electrons, and its unique star-shaped configuration. These theoretical results will provide an essential foundation for selecting donor–acceptor units in the design of novel COF materials for photocatalytic reaction applications.

In metal oxide semiconductors, oxygen vacancies play a crucial role in modulating the performances of gas sensor. Methods for regulating oxygen vacancies are diverse and widely applicable, which can enhance the charge transfer capability of semiconductors and are frequently employed as a strategy to improve gas-sensing performance. Tuning the carrier gas during tube furnace sintering to regulate oxygen vacancies represents a technical challenge, which holds significant research significance, particularly for the development of sensors, chips, and other related fields. This paper employs CBD (chemical bath deposition) technology to prepare sensitive materials. By adjusting the nitrogen-to-air ratio in the carrier gas during tube furnace sintering, oxygen vacancy regulation is achieved, thereby enhancing the gas-sensing characteristics of the sensor. The as-prepared Co3O4/SnO2 with different oxygen vacancies were characterized using XRD, SEM, TEM, XPS, BET and FT-IR. The gas-sensing results demonstrate that the sensor of Co3O4/SnO2-50% exhibits a response value of 281% to 100 ppm NH3 at 370 degrees C, along with a response value of 14% to 1 ppm NH3. Furthermore, the response time of the sensor to 1 ppm NH3 is 4 s, indicating a relatively fast response. This study provides an effective and convenient strategy for regulating oxygen vacancies, which greatly improves the gas-sensing performance of the Co3O4/ SnO2 sensor.

Defect engineering is regarded as an effective strategy for enhancing gas-sensing performance in metal-organic frameworks (MOFs). However, precise control over defect types and their specific impact on gas-sensing properties remains a significant challenge. Herein, we propose a representative water-treatment approach to induce and regulate different defect types in various MOFs. Comparative structural analysis of ZIF-8 and ZIF-67, differing in metal centers, before and after water treatment, reveals that water molecules disrupt metal-ligand bonds, leading to metal defects in ZIF-8 via metal detachment and ligand defects in ZIF-67 through partial ligand loss. Gas-sensing results demonstrate that defect concentrations and gas-sensing capabilities in MOFs can be effectively modulated by controlling water treatment time. Notably, the presence of metal defects enhances the NO2 response of ZIF-8 (20 ppm) by 2.63 times, while ligand defects improve the C2H4 response of ZIF-67 (25 ppm) by 3.96 times. Additionally, metal defect formation in MOF-74 is evidenced by a 2.97-fold enhancement in its response to 100 ppm acetone. Density functional theory calculations confirm that the defect sites enhance gas adsorption and sensing performance. This study offers new insights into defect engineering in MOFs, expanding the potential of defect-engineered MOFs for diverse applications.

Aiming at the problems of traditional electronic nose in the recognition of complex gas mixtures, such as lack of long-range dependence modeling, difficulty in capturing dynamic time dependence and limitation in multi-scale feature extraction, an end-to-end deep learning model, Gaussian Dependent Attention Dual-scale Temporal Convolutional Long Short-Term Memory Network (GDAM-DSTCN-LSTM), is proposed in this paper. The model uses Gaussian Dependent Attention Module (GDAM) to explicitly model the feature-space dependence between time steps and enhance the feature representation, uses Dual-Scale Temporal Convolutional Network (DSTCN) to extract multi-scale local temporal features in parallel, and combines Long Short-Term Memory Network (LSTM) to capture the long-term evolution of sensor data. Experimental results on UCI electronic nose dataset show that GDAM-DSTCNLSTM achieves $98.48 \%$ classification accuracy in mixed gas classification task. It is significantly better than traditional machine learning methods (such as Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Extreme Learning Machine (ELM)) and a variety of single or combined deep learning models (such as Convolutional Neural Network (CNN), Long Short-Term Memory network (LSTM), Temporal Convolutional Network (TCN), TCN-LSTM). Ablation experiments verify the effectiveness of GDAM and DSTCN modules, and the cosine annealing scheduling strategy effectively improves the stability and convergence rate of model training. In summary, GDAM-DSTCN-LSTM provides an efficient, stable and scalable end-to-end solution for high-precision identification of complex gas mixtures, which has a good prospect for industrial and environmental monitoring applications.

In this work, SnO2 nanostructures were fabricated directly on flat electrodes using chemical bath deposition (CBD). The SnO2-based sensor was then designed to enhance hydrogen gas-sensing performance by regulating the content of oxygen in the carrier gas through chemical vapor deposition (CVD). The as-prepared SnO2 nanostructure was characterized using SEM, TEM, XRD, XPS and BET. The gas-sensing performance of the SnO2based sensor was evaluated for different concentrations of hydrogen. The results demonstrated that the sensor of SnO2 under 50 % oxygen (50 % O2 CVD decorated SnO2) exhibited a response value of 271 % to 100 ppm H2 at 350 degrees C. More important, the sensor based on 50 % O2 CVD decorated SnO2 exhibited a response value of 6 % to 1 ppb H2, demonstrating an exceptionally low detection limit. The sensor exhibited a response value of 271 % to 100 ppm H2, which was 6.2 times higher than its response to 100 ppm NH3 (44 %). Furthermore, the sensor based on 50 % O2 CVD decorated expressed a response time of 2 seconds to H2. In summary, this sensor possessed a good selectivity and rapid response. These results clearly indicate that manipulating the content of oxygen in the carrier gas by CVD technology is an effective strategy for enhancing the gas-sensing performance.

The in-situ growth of sensitive materials has progressively emerged as a prevalent trend. In this paper, SnW3O9 nanosheet arrays are synthesized by aerosol-assisted chemical vapor deposition (AACVD), which is a method for in-situ preparation of sensitive films for gas sensors. Subsequently, Pt-Au is loaded onto the SnW3O9 nanosheet arrays by the wet impregnation process. Ultimately, a H2 sensor with high selectivity was successfully fabricated. At the optimum operating temperature of 310 °C, the response value of PtAu/SnW3O9-30 sensor to 100 ppm H2 is 6.5 times of 100 ppm NH3, 7.6 times of 100 ppm CO and 10.1 times of 100 ppm CH4. Meanwhile, the sensor performs excellent anti-interference capability under the mixed gases between 100 ppm H2 and 100 ppm CO, CO2 and CH4, respectively. In addition, the sensor has good stability with different relative humidity. The excellent gas sensing performance of the sensor can primarily be attributed to the unique morphology of SnW3O9 nanosheet arrays and the introduction of Pt-Au. The in-situ growth of PtAu/SnW3O9 nanosheet arrays on planar substrate can be great potential for application in fabricating highly selective H2 gas sensor.

The application of polyimides (PIs) is extensive in flexible electronics, artificial intelligence, and chip technology. However, the relationship between the structure of PIs and their colorless transparency remains unclear. In this study, we established a data set to investigate the colorless transparency of PIs by combining machine learning (ML) with time-dependent density functional theory (TD-DFT) calculations. The quantitative structure-activity relationship (QSAR) models were constructed to represent the colorless transparency of PIs, using 46 molecular descriptors derived from TD-DFT calculations. Comparing various ML algorithms, gradient boosting (GB), and support vector regression (SVR) models exhibited optimal performance for the colorlessness and transparency of PIs, respectively. Pearson correlation coefficient and SHAP model were employed to illustrate the role of significant descriptors in determining the colorlessness and transparency of PIs. Furthermore, the key substructures serving as ten diamine units were identified based on the data-driven analysis of the target properties of PIs. The research approach will provide theoretical guidance for the targeted synthesis of colorless and transparent PIs and introduce an innovative concept for the design of novel materials.

Cardiovascular disease, as the world's highest mortality disease category, is prone to acute myocardial infarction and stroke, posing a serious threat to human health. In recent years, breath analysis technology, with its significant advantages of being non-invasive, easy to operate, cost-effective, and portable, has emerged as a research hotspot and frontier in the field of early warning and diagnosis of cardiovascular diseases by detecting the concentration of CO, a hallmark gas of cardiovascular diseases. Among various detection methods, metal oxide semiconductor (MOS) sensors have demonstrated significant development potential in breath analysis applications due to their comprehensive advantages, including high sensitivity, rapid response characteristics, excellent environmental adaptability, and cost-effectiveness, gradually establishing themselves as the most promising technical solution in this field. In this study, we successfully developed a high-performance CO sensor based on nickel-doped zinc oxide (Ni-ZnO) nanosheets using the aerosol-assisted chemical vapor deposition (AACVD) technique. By optimizing the nickel doping ratio and the nanosheet structure, the sensor achieved a detection limit as low as 5 ppb for CO, while its response to 100 ppm of CO was more than 20 times higher than 100 ppm of other gases, demonstrating excellent selectivity. Therefore, CO sensors based on Ni-doped ZnO provide important technical support for the prevention and early diagnosis of cardiovascular diseases.

For the on-line stable detection of TEA, mixed potential gas sensor based on GDC (gadolinium doped cerium oxide, Ce0.8Gd0.2O1.95) solid electrolyte and MFe2O4 (M = Ni, Cu, Zn) sensing electrode has been fabricated in this work. The influence of element type and sintering temperature of the sensing materials on the sensing performance were mainly discussed, the sensor manufactured with CuFe2O4 sintered at 800 degrees C generates the highest response (-50.7 mV) to 20 ppm TEA compared with other sensors. The response value changed proportionally with the logarithm of TEA concentration, with the sensitivity of -21.9 and -43.6 mV/decade in the range of 0.5-5 and 5-100 ppm, respectively. The sensor exhibited excellent repeatability towards 2 and 10 ppm TEA for 7 continuous tests, and great selectivity for 10 various gases. Moreover, the sensor also demonstrated good moisture resistance and terrific long-term stability at 500 degrees C for 30 days continuous work. Based on the excellent sensing properties, this work supplied a stable TEA sensor fabricated with CuFe2O4 sintered at 800 degrees C, which has a great potential application prospect for the accurate detection towards TEA.

Intelligent wearable sensors play a crucial role in the detection of toxic gases and monitoring physiological signals, thereby effectively ensuring environmental and personal health safety. Nonetheless, achieving the requirements for antibacterial properties, comfortable wear, and multifunctional detection remains a major challenge. In this study, a novel Def-ZIF-8/PPY/BC-based flexible sensor is developed by in situ growth of zeolitic imidazolate frameworks-8 (ZIF-8) and polypyrrole (PPY) on bacterial cellulose (BC), followed by water immersion. The Def-ZIF-8/PPY/BC-based flexible sensor demonstrates effective dual-sensitivity responses to nitrogen dioxide (NO2) toxic gas and stress-strain behaviors at room temperature. The structural characterization and theoretical calculations affirm that the innovative water treatment method successfully introduces defects into ZIF-8, resulting in a significant 2.57-fold improvement in response intensity to 80 ppm NO2. Stress-strain sensing performance analysis reveals that the Def-ZIF-8/PPY/BC-based flexible sensor exhibited high sensitivity (6.44 kPa(-1)), rapid response and recovery times (0.7/0.4 s), and exceptional cyclic stability (8000 cycles). Further, the inhibitory effect of ZIF-8 on common bacterial strains contributes to the exceptional antibacterial properties (antibacterial rate exceeding 99%) of the Def-ZIF-8/PPY/BC based on the flexible sensor. This study offers a significant advancement in metal-organic framework defect engineering and provides an effective strategy for developing multifunctional wearable sensors.