RESUMEN

This study introduces a novel deep learning framework for lung health evaluation using exhaled gas. The framework synergistically integrates pyramid pooling and a dual-encoder network, leveraging SHapley Additive exPlanations (SHAP) derived feature importance to enhance its predictive capability. The framework is specifically designed to effectively distinguish between smokers, individuals with chronic obstructive pulmonary disease (COPD), and control subjects. The pyramid pooling structure aggregates multilevel global information by pooling features at four scales. SHAP assesses feature importance from the eight sensors. Two encoder architectures handle different feature sets based on their importance, optimizing performance. Besides, the model's robustness is enhanced using the sliding window technique and white noise augmentation on the original data. In 5-fold cross-validation, the model achieved an average accuracy of 96.40%, surpassing that of a single encoder pyramid pooling model by 10.77%. Further optimization of filters in the transformer convolutional layer and pooling size in the pyramid module increased the accuracy to 98.46%. This study offers an efficient tool for identifying the effects of smoking and COPD, as well as a novel approach to utilizing deep learning technology to address complex biomedical issues.

RESUMEN

Manipulation of confined water dynamics by voltage keeps great importance for diverse applications. However, limitations on the membrane functions, voltage-control range, and unclear dynamics need to be addressed. Herein, we report an anomalous electrically controlled gating phenomenon on cation-intercalated multi-layer Ti3C2 membranes and reveal the confined water dynamics. The water permeation rate was improved rapidly following the application and rise of voltage and finally reached a maximum rate at 0.9 V. The permeation rate starts to decrease from 0.9 V. Below 0.9 V, the electric field affects the charge and polarity of water molecules and then leads to ordered and denser rearrangement in the two-dimensional (2D) channel to accelerate the permeation rate. Above 0.9 V, with the assistance of metal cations, the surge in current induced aggregation of water molecules into clusters, thereby limiting the water mobility. Based on these findings, a high-performance humidity sensor was developed by simultaneously optimizing the response and recovery speeds through electric manipulation. This work provides flexible strategies in intelligent membrane design and nanofluidic sensing.

RESUMEN

Organic memristors based on covalent organic frameworks (COFs) exhibit significant potential for future neuromorphic computing applications. The preparation of high-quality COF nanosheets through appropriate structural design and building block selection is critical for the enhancement of memristor performance. In this study, a novel room-temperature single-phase method was used to synthesize Ta-Cu3 COF, which contains two redox-active units: trinuclear copper and triphenylamine. The resultant COF nanosheets were dispersed through acid-assisted exfoliation and subsequently spin-coated to fabricate a high-quality COF film on an indium tin oxide (ITO) substrate. The synergistic effect of the dual redox-active centers in the COF film, combined with its distinct crystallinity, significantly reduces the redox energy barrier, enabling the efficient modulation of 128 non-volatile conductive states in the Al/Ta-Cu3 COF/ITO memristor. Utilizing a convolutional neural network (CNN) based on these 128 conductance states, image recognition for ten representative campus landmarks was successfully executed, achieving a high recognition accuracy of 95.13% after 25 training epochs. Compared to devices based on binary conductance states, the memristor with 128 conductance states exhibits a 45.56% improvement in recognition accuracy and significantly enhances the efficiency of neuromorphic computing.

RESUMEN

Herein, we report a high-density dual-structure single-atom catalyst (SAC) by creating a large number of vacancies of O and Ti in two-dimensional (2D) Ti3C2 to immobilize Pt atoms (SA Pt-Ti3C2). The SA Pt-Ti3C2 showed excellent performance toward the pH-universal electrochemical hydrogen evolution reaction (HER) and multimodal sensing. For HER catalysis, compared to the commercial 20 wt % Pt/C, the Pt mass activities of SA Pt-Ti3C2 at the overpotentials of ∼30 and 110 mV in acid and alkaline media are 45 and 34 times higher, respectively. More importantly, during the alkaline HER process, an interesting synergetic effect between Pt-C and Pt-Ti sites that dominated the Volmer and Heyrovsky steps, respectively, was revealed. Moreover, the SA Pt-Ti3C2 catalyst exhibited high sensitivity (0.62-2.65 µA µM-1) and fast response properties for the multimodal identifications of ascorbic acid, dopamine, uric acid, and nitric oxide under the assistance of machine learning.

RESUMEN

Online monitoring fatigue damage and remaining fatigue life (RFL) prediction of engineering structures are essential to ensure safety and reliability. A data-driven online prediction method based on nonlinear ultrasonic monitoring was developed to predict the RFL of the structures in real-time. Nonlinear ultrasonic parameters were obtained to monitoring the fatigue degradation. A Bayesian framework was employed to continuously compute and update the RFL distributions of the structures. Nonlinear ultrasonic experiments were performed on the fatigue damaged Q460 steel to validate the developed prediction methodology. The result indicates that the developed method has high prediction accuracy and can provide effective information for subsequent decision-making.

RESUMEN

Noncontact sensors have demonstrated significant potential in human-machine interactions (HMIs) in terms of hygiene and less wear and tear. The development of soft, stable, and simply structured noncontact sensors is highly desired for their practical applications in HMIs. This work reports on electret-based self-powered noncontact sensors that are soft, transparent, stable, and easy to manufacture. The sensors contain a three-layer structure with a thickness of 0.34 mm that is fabricated by simply stacking a polymeric electret layer, an electrode layer, and a substrate layer together. The fabricated sensors show high charge-retention capability, keeping over 98% of the initial surface potential even after 90 h, and can accurately and repeatedly sense external approaching objects with impressive durability. The intensity of the detected signal shows a strong dependence on the distance between the object and the sensor, capable of sensing a distance as small as 2 mm. Furthermore, the sensors can report stable signals in response to external objects over 3000 cycles. By virtue of the signal dependence on distance, an intelligent noncontact positioning system is developed that can precisely detect the location of an approaching object. Finally, by integrating with eyeglasses, the transparent sensor successfully captures the movements of blinks for information translation. This work may contribute to the development of stable and easily manufactured noncontact soft sensors for HMI applications, for instance, assisting with communication for locked-in syndrome patients.

RESUMEN

Two-dimensional (2D) transition metal nitrides and carbides (MXenes), represented by Ti3C2Tx, have broad applications in flexible electronics, electromechanical devices, and structural membranes due to their unique physical and chemical properties. Despite the Young's modulus of 2D Ti3C2Tx has been theoretically predicted to be 0.502 TPa, which has not been experimentally confirmed so far due to the measurement is extremely restricted. Here, by optimizing the sample preparation, cutting, and transfer protocols, we perform the direct in-situ tensile tests on monolayer Ti3C2Tx nanosheets using nanomechanical push-to-pull equipment under a scanning electron microscope. The effective Young's modulus is 0.484 ± 0.013 TPa, which is much closer to the theoretical value of 0.502 TPa than the previously reported 0.33 TPa by the disputed nanoindentation method, and the measured elastic stiffness is ~948 N/m. Moreover, during the process of tensile loading, the monolayer Ti3C2Tx shows an average elastic strain of ~3.2% and a tensile strength as large as ~15.4 GPa. This work corrects the previous reports by nanoindentation method and demonstrates that the Ti3C2Tx indeed keeps immense potential for broad range of applications.

RESUMEN

Developing in situ/operando spectroscopic techniques with high sensitivity and reproducibility is of great importance for mechanistic investigations of surface-mediated electrochemical reactions. Herein, we report the fabrication of highly ordered rhombic gold nanocube superlattices (GNSs) as substrates for surface-enhanced infrared absorption spectroscopy (SEIRAS) with significantly enhanced SEIRA effect, which can be controlled by manipulating the randomness of GNSs. Finite difference time domain simulations reveal that the electromagnetic effect accounts for the significantly improved spectroscopic vibrations on the GNSs. In situ SEIRAS results show that the vibrations of CO on the Cu2O surfaces have been enhanced by 2.4 ± 0.5 and 18.0 ± 1.3 times using GNSs as substrates compared to those on traditional chemically deposited gold films in acidic and neutral electrolytes, respectively. Combined with isotopic labeling experiments, the reaction mechanisms for C-C coupling of CO electroreduction on Cu-based catalysts are revealed using the GNSs substrates.

RESUMEN

The traditional nonlinear ultrasonic technique, as typified by the second-harmonic generation and the frequency mixing response, can be employed to identify and characterize the micro-damage. However, the research on micro-damage characterization using nonlinear Lamb wave imaging technique remains an ongoing challenge and is rarely reported. A method called standardized amplitude difference is proposed for nonlinear feature enhancement, and further for fatigue crack imaging based on the wavefield data. Wavefield data contain abundant information on the spatial and temporal variation of propagating waves in the damaged structure. The nonlinearity index ß' of the signal difference under the high and low incident wave amplitudes is calculated for fatigue crack imaging. Two scanning methods, including local scanning and global scanning, are introduced to image the fatigue crack tip and visualize the wave field of the harmonics respectively. The experimental validation, based on the imaging results of an aluminum alloy plate specimen with a barely visible fatigue crack and a steel plate with a blind hole, manifests that the proposed method can be used to enhance and extract the nonlinear features and suppress the fundamental frequency, so as to improve the signal-to-noise ratio (SNR) of the micro-damage imaging results.

RESUMEN

Understanding the structure-stability relationship of catalysts is imperative for the development of high-performance electrocatalytic devices. Herein, we utilize operando attenuated total reflectance surface-enhanced infrared absorption spectroscopy (ATR-SEIRAS) to quantitatively monitor the evolution of Cu single-atom catalysts (SACs) during the electrochemical reduction of CO2 (CO2RR). Cu SACs are converted into 2-nm Cu nanoparticles through a reconstruction process during CO2RR. The evolution rate of Cu SACs is highly dependent on the substrates of the catalysts due to the coordination difference. Density functional theory calculations demonstrate that the stability of Cu SACs is highly dependent on their formation energy, which can be manipulated by controlling the affinity between Cu sites and substrates. This work highlights the use of operando ATR-SEIRAS to achieve mechanistic understanding of structure-stability relationship for long-term applications.

RESUMEN

Ultrasonic phased array imaging using full-matrix capture (FMC) has raised great interest among various communities, including the nondestructive testing community, as it makes full use of the echo space to provide preferable visualization performance of inhomogeneities. The conventional way of FMC data postprocessing for imaging is through beamforming approaches, such as delay-and-sum, which suffers from limited imaging resolution and contrast-to-noise ratio. To tackle these difficulties, we propose a deep learning (DL)-based image forming approach, termed FMC-Net, to reconstruct high-quality ultrasonic images directly from FMC data. Benefitting from the remarkable capability of DL to approximate nonlinear mapping, the developed FMC-Net automatically models the underlying nonlinear wave-matter interactions; thus, it is trained end-to-end to link the FMC data to the spatial distribution of the acoustic scattering coefficient of the inspected object. Specifically, the FMC-Net is an encoder-decoder architecture composed of multiscale residual modules that make local perception at different scales for the transmitter-receiver pair combinations in the FMC data. We numerically and experimentally compared the DL imaging results to the total focusing method and wavenumber algorithm and demonstrated that the proposed FMC-Net remarkably outperforms conventional methods in terms of exceeding resolution limit and visualizing subwavelength defects. It is expected that the proposed DL approach can benefit a variety of ultrasonic array imaging applications.

RESUMEN

Metal-organic frameworks (MOFs), synthesized by the self-assembly of organic ligands and metal centers, are structurally designable materials. In the current study, first-principles calculation based on density functional theory (DFT) was performed to investigate the intrinsic mechanical and electrical properties and mechanical-electrical coupling behavior of MOF-5. To improve the conductivity of MOF-5, homologous elements of Cu, Ag, and Au were adopted to replace the Zn atom in MOF-5, reducing the band gap and improving its electrical performance. Cu-MOF-5 and Au-MOF-5, with stable structures, exhibit better conductivity. The intrinsic mechanical properties such as independent elastic constants of MOF-5 and M-MOF-5 (M = Cu, Ag, Au) were obtained. MOF-5 and Cu-MOF-5 were experimentally synthesized to demonstrate the reduction in the band gap after metal substitution. The study of the strain effect of MOF-5 and Cu-MOF-5 proves that strain engineering is an effective method to regulate the band gap and this modulation is repeatable. This study clarifies the tunability of the band gap of MOF-5 with metal substituents and provides an efficient strategy for the development of new types of MOFs with desired physical properties using the combination of theoretical prediction and experimental synthesis and validation.

RESUMEN

Total focusing method (TFM) attracts much interest because of high image resolution and large inspection coverage. However, the synthetic focusing approach based on delay-and-sum beamforming employs only the defect information contained in the dataset while ignoring the spatial information of the array signals, leading to limited imaging performance mixed with artifacts and noise. In addition, the signal-to-noise ratio (SNR) suffers due to single-element emission of full matrix capture. This work combines a modified delay-multiply-and-sum (DMAS) beamforming approach with conventional synthetic focusing in the TFM algorithm, to achieve optimization of TFM imaging performance. DMAS-based TFM is able to take full advantage of the defect and spatial information in the array dataset, and to generate new frequency components for better image reconstruction. As demonstrated on a series of comparative simulation and experimental results, the imaging results of the optimized TFM provide a considerable improvement in SNR. Better lateral spatial resolution is also achieved due to the increased number of equivalent transducer elements and second harmonic component. Therefore, this work provides a quite promising alternative solution for the post-processing of ultrasonic phased array with improved imaging performance.

Asunto(s)

RESUMEN

Single-atom catalysts (SACs) attract extensive attention in the field of heterogeneous catalysis in recent years due to the maximum atom utilization and unique physical and chemical properties. The gas sensing is actually a heterogeneous catalysis process but the SACs are new to this area. Although SACs show huge potential in gas sensing, the SACs gas sensing area currently is still at the infancy stage. This work critically reviews the recent advances and current status of single-atom gas sensing materials. General synthesis routes, characterization methods, and sensing performance indexes are introduced. At the end, the challenges and future prospects on SACs gas sensing are presented from the authors' perspectives. This work is anticipated to provide insights and guideline for the chemical sensing community.

RESUMEN

As the key half reaction of water-splitting electrolysis, the hydrogen evolution reaction (HER) that occurs at the cathode directly determines the overall efficiency of hydrogen production. To improve the efficiency of electrochemical water splitting for hydrogen generation, efficient and robust catalysts need to be developed. Strain engineering, which represents an effective and promising category of strategies, can regulate the electronic structures of catalysts by modulating the lattice strain and ultimately optimizing the HER dynamics. This work critically reviews the recent progress of strain engineering in HER and provides future perspectives for this area. The methods and characterization techniques are also introduced in detail. Hopefully this review can provide guidelines for the design and manufacturing of advanced catalysts for HER and other heterogeneous catalysis reactions such as chemical sensing, CO2 reduction and NH3 synthesis.

RESUMEN

Characterization of grain microstructures of metallic materials is crucial to materials science and engineering applications. Unfortunately, the universal electron microscopic methodologies can only capture two-dimensional local observations of the microstructures in a time-consuming destructive way. In this regard, the nonlinear ultrasonic technique shows the potential for efficient and nondestructive microstructure characterization due to its high sensitivity to microstructural features of materials, but is hindered by the ill-posed inverse problem for multiparameter estimation induced by the incomplete understanding of the complicated nonlinear mechanical interaction mechanism. We propose an explainable nonlinearity-aware multilevel wavelet decomposition-multichannel one-dimensional convolutional neural network to hierarchically extracts multilevel time-frequency features of the acoustic nonlinearity and automatically model latent nonlinear dynamics directly from the nonlinear ultrasonic responses. The results demonstrate that the proposed approach establishes the complex mapping between acoustic nonlinearity and microstructural features, thereby determining the lognormal distribution of grain size in metallic materials rather than only average grain size. In the meantime, the integration of the designed nonlinearity-aware network and the quantitative analysis of component importance provides an acceptable physical explainability of the deep learning approach for the nonlinear ultrasonic technique. Our study shows the promise of this technique for real-time in situ evaluation of microstructural evolution in various applications.

RESUMEN

The movements of soft living tissues, such as muscle, have sparked a strong interest in the design of hydrogel actuators; however, so far, typical manmade examples still lag behind their biological counterparts, which usually function under nonequilibrium conditions through the consumption of high-energy biomolecules and show highly autonomous behaviors. Here, we report on self-resettable hydrogel actuators that are powered by a chemical fuel and can spontaneously return to their original states over time once the fuels are depleted. Self-resettable actuation originates from a chemical fuel-mediated transient change in the hydrophilicity of the hydrogel networks. The actuation extent and duration can be programmed by the fuel levels, and the self-resettable actuation process is highly recyclable through refueling. Furthermore, various proof-of-concept autonomous soft robots are created, resembling the movements of soft-bodied creatures in nature. This work may serve as a starting point for the development of lifelike soft robots with autonomous behaviors.

Asunto(s)

RESUMEN

In recent years, flexible pressure sensors have caused widespread concern for their extensive applications in human activity and health monitoring, robotics and prosthesis, as well as human-machine interface. Flexible pressure sensors in these applications are required to have a high sensitivity, large detective limit, linear response, fast response time, and mechanical stability. The mechanisms of capacitive, piezoresistive, and piezoelectric pressure sensors and the strategies to improve their performance are introduced. Sensing layers with microstructures have shown capability to significantly improve the performances of pressure sensors. Various fabrication methods for these structures are reviewed in terms of their pros and cons. Besides, the interference caused by environmental stimuli and internal stress from different directions leads to the infidelity of the signal transmission. Therefore, the anti-interference ability of flexible pressure sensors is highly desired. Several potential applications for flexible pressure sensors are also briefly discussed. Last, we conclude the future challenges for facilely fabricating flexible pressure sensors with high performance and anti-interference ability.

RESUMEN

The combination of air-coupled ultrasonic testing (ACUT) and ultrasonic Lamb wave is featured with long-distance propagation and high sensitivity to discontinuities, which is a promising method for rapid and accurate inspection of plate-like materials and lightweighted structures. However, dispersive nature of Lamb wave, signal attenuation plus inevitable noises would lead to low signal-to-noise ratio (SNR). To address this problem, phase coded excitation and pulse compression technique are proposed in this paper to achieve higher SNR by over 10 dB in received signals. 13-bit and 1-carrier-period Barker code is employed as both main lobe peak and Peak Side-lobe Level (PSL) are relatively high. It is demonstrated that A0 mode Lamb wave has good localization ability for defects based on these SNR-enhanced signals. Furthermore, Damage Index (DI) and modified Reconstruction Algorithm for the Probabilistic Inspection of Damage (RAPID) are applied to realize ultrasonic imaging based defect evaluation. Results show that the imaging results agree well with the actual artificial defects in terms of size and shape. Lamb-wave-based air-coupled ultrasonic testing, combined with DI and ultrasonic imaging algorithm, could be a potential way in the NDT of lightweighted structures.

RESUMEN

Dielectric elastomers (DEs) are polymer materials consisting of a network of polymer chains connected by covalent cross-links. This type of structural feature allows DEs to generate large displacement outputs owing to the nonlinear electromechanical coupling and time-dependent viscoelastic behavior. The major challenge is to properly actuate the nonlinear soft materials in applications of robotic manipulations. To characterize the complex time-dependent viscoelasticity of the DEs, a nonlinear rheological model is proposed to describe the time-dependent viscoelastic behaviors of DEs by combining the advantages of the Kelvin-Voigt model and the generalized Maxwell model. We adopt a Monte Carlo statistical simulation method as an auxiliary method, to the best knowledge of the author which has never reportedly been used in this field, to improve the quantitative prediction ability of the generalized model. The proposed model can simultaneously describe the DE deformation processes under step voltage and alternating voltage excitation. Comparisons between the numerical simulation results and experimental data demonstrate the effectiveness of the proposed generalized rheological model with a maximum prediction error of 3.762% and root-mean-square prediction error of 9.03%. The results presented herein can provide theoretical guidance for the design of viscoelastic DE actuators and serve as a basis for manipulation control to suppress the viscoelastic creep and increase the speed response of the dielectric elastomer actuators (DEA).